Background

HR+/HER2+ BC is heterogeneous and there is a need to identify predictive biomarkers. We previously reported the ability of the PAM50-based CES to predict chemoendocrine sensitivity in HR+/HER2-negative BC beyond intrinsic subtype and risk of relapse (ROR) (Prat.CCR.2017). Here we evaluate the association of CES with pCR and DFS following different anti-HER2 combinations in HR+/HER2+ BC.

Methods

Intrinsic subtype and clinicopathologic data were obtained from 8 independent studies for a total of 485 HR+/HER2+ early BC. Patients (pts) were treated with anti-HER2 therapy either with endocrine therapy (PAMELA and PER-ELISA) or with chemotherapy (CHERLOB, OptiHER, LPT109096, ICO, HCB, PER-ELISA and CALGB 40601 [Alliance]). CES was evaluated as a continuous variable and categorically (CES-E [endocrine-sensitive], CES-U [uncertain] and CES-C [chemosensitive]) using previously reported cutoffs. We first performed statistical analyses in each dataset individually, and then all 8 combined. Multivariable analyses were used to test the association of the CES score with pCR and DFS.

Results

In the combined cohort, CES-E, CES-U and CES-C were identified in 16%, 22% and 62% of the pts, respectively. CES (continuous variable) was associated with higher pCR rates independent of clinical characteristics, treatment type, intrinsic subtype and study (adjusted Odd Ratio [OR]=0.42; p = 0.02). In the PER-ELISA trial, CES was also found associated with response (decrease in ki67) following 2 weeks of letrozole alone (OR=29.1, p 0.01). 295 pts (CHERLOB, ICO, HCB and CALGB40601) were analyzed for DFS with a median follow-up of 66 months (IQR 37-82). The adjusted DFS hazard ratio of the CES (continuous variable) was 0.13 (p < 0.01) independent of pCR, clinical characteristics, ROR and intrinsic subtype. In pts without pCR, disease recurrence occurred in 4% of CES-E, 19% of CES-U and 34% of CES-C pts (p < 0.01).

Conclusions

In HR+/HER+ BC, CES shows clinical validity for predicting chemoendocrine sensitivity in combination with anti-HER2 targeted therapies and is a good prognostic factor beyond intrinsic subtype and clinicopathologic characteristics.

Clinical trial identification

CALGB-40601: NCT00770809; Per-ELISA: NCT02411344; SOLTI-PAMELA: NCT01973660; SOLTI-Opti-HER:NCT01669239; LPT109096: NCT00524303; CHERLOB: NCT00429299.

Legal entity responsible for the study

Instituto de Investigaciones Biomédicas August Pi i Sunyer (IDIBAPS).

Funding

Becas FSEOM para Formación en Investigación en Centros de Referencia en el Extranjero 2018 (TP). Fundación SEOM, Becas FSEOM para Formación en Investigación en Centros de Referencia en el Extranjero 2016 (AF-M). Conquer Cancer Foundation -YIA in Breast Cancer 2019 (GG). BCRF, Susan G Komen, NCI SPORE (P50-CA58823), R01-CA229409 and Alliance U10CA180821(LAC, CMP). Instituto de Salud Carlos III - PI16/00904 (AP), Pas a Pas (AP), Save the Mama (AP), Breast Cancer Now - 2018NOVPCC1294 (AP).

Disclosure

M.V. Dieci: Honoraria (self): Genomic Health; Honoraria (self): Eli Lilly; Honoraria (self): Celgene. S. Pernas Simon: Travel/Accommodation/Expenses: Roche; Travel/Accommodation/Expenses: Novartis; Advisory/Consultancy: Polyphor; Advisory/Consultancy: TFS. J. Gavila Gregori: Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Honoraria (self), Advisory/Consultancy: Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy: MSD Oncology. V. Guarneri: Honoraria (self): Eli Lilly; Honoraria (self): Roche/Genentech; Honoraria (self): Novartis. P. Villagrasa: Speaker Bureau/Expert testimony: NanoString Technologies. M.J. Vidal: Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Honoraria (self): Daiichi Sankyo; Honoraria (self), Advisory/Consultancy: Novartis; Honoraria (self), Travel/Accommodation/Expenses: Pfizer. M. Oliveira: Advisory/Consultancy, Research grant/Funding (institution): AstraZeneca; Research grant/Funding (institution): Philips Healthcare; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Genentech/Roche; Honoraria (self), Research grant/Funding (institution), Travel/Accommodation/Expenses: Novartis; Research grant/Funding (institution): Immunomedics; Honoraria (self), Research grant/Funding (institution): Seattle Genetics; Advisory/Consultancy, Research grant/Funding (institution): GSK; Research grant/Funding (institution): Boehringer-Ingelheim; Advisory/Consultancy, Research grant/Funding (institution): PUMA Biotechnology; Research grant/Funding (institution): Zenith Epigenetics; Travel/Accommodation/Expenses: Pierre-Fabre; Travel/Accommodation/Expenses: GP Pharma; Travel/Accommodation/Expenses: Grünenthal; Travel/Accommodation/Expenses: Eisai. C.M. Perou: Advisory/Consultancy, Shareholder/Stockholder/Stock options: BioClassifier LLC; Licensing/Royalties: Breast PAM50 assay. A. Prat: Honoraria (self), Travel/Accommodation/Expenses: Daiichi Sankyo; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Novartis; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Roche; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self): MSD Oncology; Honoraria (self): Lilly; Advisory/Consultancy: NanoString Technologies; Advisory/Consultancy: Amgen; Advisory/Consultancy: Bristol-Myers Squibb. All other authors have declared no conflicts of interest.

Background

Changes in ctDNA levels may predict response to a variety of drugs, including CDK4/6i; however, the best assay and method are still to be defined.

Methods

This is a prospective single-center study in hormone receptor-positive/HER2-negative advanced breast cancer pts treated with CDK4/6i and endocrine therapy (ET). Paired plasma samples were collected at cycle 1 day 1 (C1) and cycle 2 day 1 (C2). Somatic alterations and variant allele fraction (VAF) were assessed using the 74-gene Guardant360 assay (Guardant Health). A VAF ratio (VAFR) was calculated for each alteration with a VAF of ≥ 0.4% at C1 or C2. Molecular response was defined for each patient as the mean of all VAFRs (mVAFR). Pts with VAFs < 0.4% at C1 and C2 were considered to have low-shedding tumors. Progression free survival (PFS) hazard ratios (HR) were calculated using a univariate Cox model. PAM50 subtypes and tumor infiltrating lymphocytes (TILs) were determined in a subset.

Results

48 pts treated with ET and palbociclib (89%) or ribociclib (11%) were analyzed with a median follow-up of 12.0 months (IQR 6.7-14.6). Clinical characteristics: 65% had visceral disease, 48% were treated as 1st-line, 35% as 2nd-line, 57% used fulvestrant and 33% an aromatase inhibitor. PAM50 subtype distribution (n=27): Luminal A (n=9), Luminal B (n=10), HER2-enriched (n=4), Normal (n=3) and Basal-like (n=1). ctDNA was detected in 96% of pts. mVAFR < 0.3 (high-ctDNA responders) (n=12) and low-shedding tumors (n=13) correlated with significantly improved PFS (HR=0.39, p=0.025), especially when compared to pts with ctDNA mVAFR > 1 (HR=0.27, p=0.010, n=12). Within PAM50 tested tumors, non-Luminal (n=5) were low-ctDNA responders (mVAFR > 0.3) (n=3) or low-shedding (n=2); Luminal A or B were high-ctDNA responders (n=8), low-ctDNA responders (n=7) and low-shedding (n=4). TILs were increased in low relative to high-ctDNA responders (mean 3.3% vs 1.8%).

Conclusions

ctDNA dynamics are an early surrogate of CDK4/6i + ET efficacy. The clinical utility of this biomarker should be tested in prospective clinical trials in which pts with unfavorable ctDNA responses are randomized to alternative treatment strategies.

Legal entity responsible for the study

The authors.

Funding

European Union’s Horizon 2020 Research and Innovation Programme under Grant agreement No. 847912.

Disclosure

M.J. Vidal: Honoraria (self), Travel/Accommodation/Expenses: Pfizer; Honoraria (self), Advisory/Consultancy: Novartis; Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Honoraria (self): Daiichi Sankyo. E.M. Ciruelos: Advisory/Consultancy: Roche; Advisory/Consultancy: Lilly; Advisory/Consultancy: Novartis; Advisory/Consultancy: Pfizer. A. Prat: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Full/Part-time employment, An Immediate Family Member: Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Roche; Honoraria (self): MSD Oncology; Honoraria (self), Advisory/Consultancy: Amgen; Honoraria (self), Advisory/Consultancy: Daiichi Sankyo; Advisory/Consultancy: Bristol-Myers Squibb; Advisory/Consultancy: Boehringer; Advisory/Consultancy, Research grant/Funding (institution): PUMA; Advisory/Consultancy: Oncolytics Biotech; Advisory/Consultancy: AbbVie; Advisory/Consultancy: NanoString Technologies; Research grant/Funding (institution): Incyte. All other authors have declared no conflicts of interest.

Background

Resistance to CDK4/6i is inevitable. CTC count is prognostic in ABC, but its role in pts treated with CDK4/6i is not well defined. Genetic loss of RB1 is a known yet infrequent marker of CDK4/6i resistance. We assessed the prognostic role of CTC count and gene-expression (GE) levels of RB1 in CTCs in pts receiving P.

Methods

The TREnd trial (NCT02549430) randomized pts with endocrine resistant ABC to either P alone or P plus the endocrine therapy received in the prior line of treatment. In TREnd, blood samples were prospectively collected in CellSave® tubes before starting P (T0), after the first cycle (T1) and at disease progression (T2). CTCs were isolated and counted by CellSearch® System (CS) using CellSearch^TM Epithelial Cell kit. Samples with ≥5CTCs were sorted by DEPArray system® (DA). RNA extraction and retro-transcription for GE experiments were performed by Cell Lysis Two-Step RT-qPCR. RB1 and GAPDH GE levels were measured by ddPCR, with a multiplex assay with a sensitivity of 30-10 pg of cDNA, set up on three different cell lines sensitive and resistant to P.

Results

46 pts were suitable for CTC analysis. CTC count at T0 did not show significant prognostic value in terms of progression free survival (PFS). However, pts with at least 1 detectable CTC at T1 (n=26) had a worse PFS than those with 0 CTCs (n=16) (p=0.02). Similar results were observed with a cut-off of 5 CTCs (p=0.04). At T1, 7 out of 39 pts had an increase of at least 3 CTCs which proved prognostic (p=0.01). Pts with ≥5CTCs at T2 (n=6/23) who received chemotherapy as post-study treatment had a shorter time to treatment failure (p=0.02). DA sorting was conducted on 20/46 pts and GE data for RB1 were obtained from 19 pts. CTCs showed heterogeneous RB1 expression. Pts with detectable expression of RB1 in at least one time-point had better, but not significant, outcomes than those with undetectable levels.

Conclusions

Persistence or an increase in CTCs after one cycle of P may identify pts with worse outcome. High CTC counts at disease progression on P may indicate poor post-treatment prognosis. Measuring RB1 GE levels on CTCs by ddPCR is feasible, but its clinical significance is yet unclear.

Clinical trial identification

NCT02549430.

Legal entity responsible for the study

Fondazione Sandro Pitigliani per la Lotta Contro i Tumori.

Funding

Pfizer.

Disclosure

G. Curigliano: Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Seattle Genetics; Speaker Bureau/Expert testimony, writing engagement: Novartis; Advisory/Consultancy, Speaker Bureau/Expert testimony: Lilly; Advisory/Consultancy, Speaker Bureau/Expert testimony: Pfizer; Advisory/Consultancy, Speaker Bureau/Expert testimony: Foundation Medicine; Speaker Bureau/Expert testimony, Cosultancy: NanoString; Advisory/Consultancy, Speaker Bureau/Expert testimony: Samsung; Advisory/Consultancy, Speaker Bureau/Expert testimony: Celltrion; Advisory/Consultancy, scientific affairs group: Ellipsis; Speaker Bureau/Expert testimony, writing engagement: BMS; Speaker Bureau/Expert testimony: MSD; Advisory/Consultancy: Mylan. M. Benelli: Advisory/Consultancy: Novartis. L. Biganzoli: Honoraria (self), Advisory/Consultancy: AstraZeneca; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Celgene; Honoraria (self), Advisory/Consultancy: Eisai; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Genomic Health; Honoraria (self), Advisory/Consultancy: Ipsen; Honoraria (self), Advisory/Consultancy: Lilly; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy: Pierre Fabre; Honoraria (self), Advisory/Consultancy: Roche. A. Di Leo: Honoraria (self), Advisory/Consultancy: Amgen; Honoraria (self), Advisory/Consultancy: AstraZeneca; Honoraria (self), Advisory/Consultancy: Bayer; Honoraria (self), Advisory/Consultancy: Daiichi-Sankyo; Honoraria (self), Advisory/Consultancy: Eisai; Honoraria (self), Advisory/Consultancy: Genentech; Honoraria (self), Advisory/Consultancy: Genomic Health; Honoraria (self), Advisory/Consultancy: Lilly; Honoraria (self), Advisory/Consultancy: Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy: Roche; Honoraria (self), Advisory/Consultancy: Seattle Genetics. L. Malorni: Advisory/Consultancy, Research grant/Funding (self): Novartis; Advisory/Consultancy, Research grant/Funding (self): Pfizer; Advisory/Consultancy: Lilly. All other authors have declared no conflicts of interest.

Background

Although the survival benefit of dual epidermal growth factor receptor 2 (HER2) blockade with trastuzumab and pertuzumab was definitely demonstrated in HER2-amplified early breast cancer, sufficient biomarkers are urgently required to explain the heterogeneous response to dual HER-2 blockade therapy. The prognostic significance of immune infiltration in TRYPHAENA trial was investigated to tailor treatment in current analysis.

Methods

Among the 225 HER2-amplified early breast cancer patients randomly assigned to trastuzumab/pertuzumab concurrently or sequentially with standard chemotherapy as neoadjuvant therapy in TRYPHAENA trial, 162 patients with available gene expression profile and complete follow-up data were enrolled. The normalized gene expression matrix (GSE109710) based on the NanoString nCounter array was downloaded from Gene Expression Omnibus database and further used to estimate the immune infiltration score (IIS) for each patient by the Immune Cell Abundance Identifier tool. A cut-off of IIS to stratify patients was determined by the R-based survminer package. Multivariable Cox proportional event-free survival (EFS) hazard ratios were preformed.

Results

Among the 162 women included in the analysis (median [range] age, 49.0 [27-81] years), the pathologic complete response (pCR) rate was 50.0% (21/42) in patients with a high IIS (>0.628) and 66.7% (80/120) in patients with a low SII (≤0.628). At a median follow-up of 4.7 years, the multivariable-adjusted hazard ratio for EFS was 2.933 (95%CI, 1.223-7.033) for the high IIS and 0.356 (95%CI, 0.127- 0.999) in patients who achieved pCR, respectively. Table: 7P

Cox regression for EFS

Conclusions

Our analysis demonstrates an independent prognostic value of IIS in patients receiving trastuzumab/pertuzumab-based neoadjuvant chemotherapy.

Clinical trial identification

NCT00976989.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Despite its recognized role in breast cancer (BC), the complexity of immune microenvironment remains largely unexplored. Multiplex immunohistochemistry (mIHC) holds opportunity to more comprehensively assess BC immunity, potentially providing information to improve immunotherapy.

Methods

In the neoadjuvant phase II SOLTI-1114 PAMELA trial (NCT01973660), 151 HER2+ BC patients received lapatinib and trastuzumab, plus hormonal therapy if HR+. Baseline (BSL, n=66) and day-15 (D15, n=54) biopsies from 76 patients were analyzed using a custom mIHC 6-plex panel, including immune subtyping (CD3, CD4, CD8, FOXP3), localization (keratin for tumor mask), and activity (Ki67 for proliferation). Immune cell density (cells/mm²) and localization were determined by digital image analysis and classified in: intratumor (A), proximal peritumor (B - < 10um; C - 10 to 30um from tumor) and distal peritumor stroma (D). Intrinsic subtyping was determined at the same timepoints using the PAM50 predictor (nCounter).

Results

Both at BSL and D15, no significant difference in immune subpopulation densities was observed according to PAM50 subtype. At both timepoints, fraction of proliferating (Ki67+) immune cells (all subpopulations) differed significantly according to subtype (basal-like tumors showing the highest and luminal tumors showing the lowest fraction of proliferating cells, p varying from <0.001 to 0.031). Tumors achieving a pCR showed numerically higher densities of CD3+, CD8+ and FOXP3+ cells. Association with pCR was stronger at D15 and for immune cells intratumor/more proximal to tumor (Table). Overall, at D15, tumors achieving pCR showed higher CD3+ density (p=0.03) and higher ratio in Ki67+CD8+/Ki67+FOXP3+ (p=0.03). Table: 8P

Odds ratios (95% CI) for pCR for 1000 cells/mm2 increases in immune cell density according to subpopulation, timepoint, and localization

Conclusions

In early HER2+ BC, immune cells show differential proliferation patterns according to tumor biology. Number of immune cells spatially interacting with tumor after priming by anti-HER2 therapy was associated with pCR.

Clinical trial identification

NCT01973660.

Legal entity responsible for the study

The authors.

Funding

Vall d'Hebron Institute of Oncology (VHIO).

Disclosure

G. Griguolo: Travel/Accommodation/Expenses: Pfizer. A. Prat: Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), immediate family member being employed by Novartis: Novartis; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Roche; Honoraria (self): MSD Oncology; Honoraria (self): Lilly; Honoraria (self), Travel/Accommodation/Expenses: Daiichi Sankyo; Advisory/Consultancy: NanoString Technologies; Advisory/Consultancy: Amgen; Advisory/Consultancy: Bristol-Myers Squibb. P. Nuciforo: Advisory/Consultancy: Bayer; Advisory/Consultancy, Speaker Bureau/Expert testimony: MSD Oncology; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis. All other authors have declared no conflicts of interest.

Background

Breast cancer is the most common type of malignancy in pregnant women and it occurs in 1-4 per 10,000 pregnancies. The incidence of pregnancy associated breast cancer (PABC), accounting for up to 3.8% of all breast cancers, is expected to rise, especially in developed countries. Whether these cancers arise before or during pregnancy, and whether they are stimulated by the high hormonal environment of pregnancy, is currently unknown. This study assesses the histopathological profile of PABC in a large Dutch population-based cohort.

Methods

We identified 744 patients with PABC (defined as breast cancer diagnosed during or within six months after pregnancy), between 1988 and 2019, in the nationwide Dutch Pathology Registry (PALGA). An age-matched PALGA cohort of unselected breast cancer patients (≤ 45 years), diagnosed between 2013 and 2016, was used as a control (n=4,314). Histopathologic features of both cohorts were compared.

Results

Mean age at diagnosis of PABC patients was 34.5 years and most breast cancers were diagnosed during pregnancy (70.7%). As compared to age-matched controls, PABC patients had tumors of higher Bloom-Richardson grade (grade I: 1.2% vs. 17.9%, grade II: 14.7% vs. 35.3%, grade III: 71.9% vs. 31.1%). Furthermore, estrogen (ER) and progesterone-receptor (PR) expression was more often reported as negative (ER: 45.0% vs. 20.2%, PR: 48.8% vs. 29.4%), while a higher percentage of PABC patients was reported as overexpressing HER2 (22.0% vs. 15.1%). The most observed subtype in PABC was triple-negative breast cancer; 34.1% compared to 15.1%, and luminal subtypes represented only 17.2% v.s. 60.6% in the non-PABC cases. There were no differences in grade or hormone receptor-status between pregnant -and postpartum-PABC patients.

Conclusions

This study, based on a large population-based cohort of 744 PABC patients, underlines a different, more aggressive histopathologic profile compared to age-matched breast cancer patients ≤ 45 years. RNA-and DNA sequencing of breast tumors will be conducted to unravel the genetic background and find opportunities for prevention and optimal treatment (more targeted and less toxic).

Legal entity responsible for the study

University Medical Center Utrecht.

Funding

A Sister’s Hope for Breast Cancer Research.

Disclosure

All authors have declared no conflicts of interest.

Background

The recurrence score (RS), which is derived from the results of an assay of 21 genes, predicts the likelihood of recurrence in patients with breast cancer, thus potentially helping clinicians decide when to recommend chemotherapy. However, non-genomic clinicopathologic prognostic markers may also be able to distinguish patients with low, intermediate, and high risk of recurrence without the added cost of genetic testing.

Methods

We developed a novel non-genomic tool called predicted RS (pRS) and investigated the relationship between RS and pRS among patients with stage I estrogen receptor-positive, human epidermal growth factor receptor 2-negative (HER2) breast cancer. We reviewed the records of 1055 patients at The University of Texas MD Anderson Cancer Center with estrogen receptor-positive, HER2-negative stage I breast cancer who had RS results available. We used multivariable linear regression to develop pRS in this population. We then validated our models in a cohort of 242 patients from Anne Arundel Medical Center with the same characteristics.

Results

The pRS model is pRS = 26.089 – 0.071ER – 0.092PR + 0.132Ki67 + 1.08I[HG=II] + 7.129I[HG=III] where I is an indicator function. The pRS had a Pearson correlation coefficient of 0.7352 with RS in our validation cohort (p < 0.0001). Two (1.2%) of the 170 patients with low/intermediate RS (RS ≤25) were classified into the high pRS group (P-value <0.0001). None of the patients with a pRS >30 were considered low/intermediate risk (RS ≤ 25) as defined in the TAILORx trial. Table: 10P

Correlation between recurrence score (RS) and predicted RS (pRS) in the Anne Arundel Medical Center validation cohort using two risk categories∗

Conclusions

pRS accurately identified patients who had an RS >25 and thus were at high risk for recurrence. These patients could therefore be prescribed chemotherapy without first undergoing genetic testing.

Editorial acknowledgement

Department of Scientific Publications at The University of Texas MD Anderson Cancer Center for manuscript editing.

Legal entity responsible for the study

The authors.

Funding

90 National Cancer Institute (P30CA016672) for all MD Anderson investigators CRLCC Eugène Marquis for F.LE DU.

Disclosure

F. Le Du: Advisory/Consultancy: Myriad. N.T. Ueno: Research grant/Funding (institution): Genomic Health; Research grant/Funding (institution): Sysmex. All other authors have declared no conflicts of interest.

Background

BIOITALEE aims to study circulating tumor DNA (ctDNA) alterations, their evolution and association with clinical outcome in patients (pts) receiving ribociclib+letrozole as 1^st-line treatment for aBC. Here we report the analysis of baseline liquid biopsy (LB) and tumor tissue samples (TS), and their comparison.

Methods

287 postmenopausal pts with HR+, HER2– aBC were enrolled in 47 Italian centers. 271 pts were suitable for Next Generation Sequencing (NGS) on LB and 144 had a paired TS valid for NGS. LBs and TSs were analyzed by a Custom panel (average coverage 23000x). LBs were also analyzed by Oncomine Pan-Cancer Assay (Thermo Fisher Scientific). The agreement was determined by Cohen’s kappa statistic (K) and McNemar test (p).

Results

Of the 144 paired TSs, 116 (80.5%) were collected in the 6 months preceding study entry. In custom panel analysis, 72.9% TS and 44.4% LB, exhibited at least one alteration. The concordance was moderate (K=0.51, CI: 0.44-0.58, p<1e-4) mostly due to negative findings in LB. For PIK3CA, 21.5% of pts had TS+/LB- and discordant cases showed significantly lower allele frequencies (AFs) (Wilcoxon p<1e-4). The concordance for the 3 most frequently altered genes is detailed below: Table: 11P

25 distinct PIK3CA variants with different AFs were observed, suggesting both clonal and subclonal alterations. For LB, the concordance between Custom and Oncomine panel was good (K=0.70, CI: 0.64-0.76) (for PIK3CA, K=0.79, CI: 0.70-0.88).

Conclusions

In our study, mutations were more frequently found in TS rather than LB, supporting the strategy of querying the tissue to complement ctDNA results. The ultra-deep NGS of TS in this study, enabled improved comparison between TS and LB. LB+ findings with TS- results were infrequent. Discordancy in PIK3CA status is associated with lower AFs in TS, likely due to subclonal events. Further analyses are ongoing and will be presented.

Clinical trial identification

NCT03439046.

Legal entity responsible for the study

Novartis Farma SpA, Italy.

Funding

Novartis Farma SpA, Italy.

Disclosure

G. Bianchini: Advisory/Consultancy: Roche; Advisory/Consultancy: Novartis; Advisory/Consultancy: Eisai; Advisory/Consultancy: AstraZeneca; Advisory/Consultancy: Pfizer; Advisory/Consultancy: Lilly; Advisory/Consultancy: Genomic Health; Advisory/Consultancy: Amgen; Advisory/Consultancy: MSD; Advisory/Consultancy: Sanofi; Advisory/Consultancy: Daiichi Sankyo; Advisory/Consultancy: Chugai. M. De Laurentiis: Advisory/Consultancy, Speaker Bureau/Expert testimony: Pfizer; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis; Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Celgene; Advisory/Consultancy, Speaker Bureau/Expert testimony: AstraZeneca; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eisai; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eli Lilly; Advisory/Consultancy, Speaker Bureau/Expert testimony: Amgen; Advisory/Consultancy: MSD; Advisory/Consultancy, Speaker Bureau/Expert testimony: Pierre Fabre. G. Arpino: Honoraria (self), Research grant/Funding (institution): Novartis; Honoraria (self), Research grant/Funding (institution): Roche; Honoraria (self): Lilly; Honoraria (self): AstraZeneca; Honoraria (self): MSD; Honoraria (self): Amgen; Honoraria (self), Research grant/Funding (institution): Eisai; Honoraria (self), Research grant/Funding (institution): Pfizer. A. Zambelli: Advisory/Consultancy: Novartis; Advisory/Consultancy: Roche; Advisory/Consultancy: AstraZeneca; Advisory/Consultancy: Pfizer; Advisory/Consultancy: Lilly; Advisory/Consultancy: Biogen; Advisory/Consultancy: Genomic Health. F. Puglisi: Research grant/Funding (institution): AstraZeneca; Research grant/Funding (institution): Eisai; Research grant/Funding (institution): Roche; Advisory/Consultancy: Novartis; Advisory/Consultancy: MSD; Advisory/Consultancy: Eli Lilly; Advisory/Consultancy: Roche; Advisory/Consultancy: Pfizer; Advisory/Consultancy: Pierre Fabre; Advisory/Consultancy: Eisai. L. Del Mastro: Honoraria (self): Roche; Honoraria (self): Pfizer; Honoraria (self): Ipsen; Honoraria (self): Eli Lilly; Honoraria (self): Novartis; Honoraria (self): Takeda; Honoraria (self): MSD; Honoraria (self): Genomic Health; Non-remunerated activity/ies: Celgene; Honoraria (self): Seattle Genetics. M.A. Colleoni: Advisory/Consultancy: AstraZeneca; Advisory/Consultancy: Pierre Fabre; Advisory/Consultancy: Pfizer; Honoraria (self): Novartis; Advisory/Consultancy: OBI Pharma; Advisory/Consultancy: Puma Biotechnology; Advisory/Consultancy: Celldex. F. Montemurro: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis; Speaker Bureau/Expert testimony: Pfizer; Advisory/Consultancy: Pierre Fabre; Speaker Bureau/Expert testimony: Daiichi Sankyo. G.V. Bianchi: Advisory/Consultancy: Novartis; Speaker Bureau/Expert testimony: Eli Lilly. C. Zamagni: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses, Non-remunerated activity/ies: Roche; Advisory/Consultancy, Research grant/Funding (institution): Eisai; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses, Non-remunerated activity/ies: Novartis; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Non-remunerated activity/ies: AstraZeneca; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses, Non-remunerated activity/ies: Pfizer; Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: PharmaMar; Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Celgene; Advisory/Consultancy, Research grant/Funding (institution): Lilly; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Amgen; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Tesaro; Honoraria (self), Advisory/Consultancy: QuintikesIMS; Research grant/Funding (institution), Travel/Accommodation/Expenses: Pierre Fabre; Research grant/Funding (institution), Travel/Accommodation/Expenses: Istituto Gentili; Research grant/Funding (institution): Takeda; Research grant/Funding (institution): Teva; Research grant/Funding (institution): Medivation; Research grant/Funding (institution): AbbVie; Research grant/Funding (institution): Array BioPharma; Research grant/Funding (institution): Morphotek; Research grant/Funding (institution): Synthon, Seattle Genetics. S. Bianchetti: Full/Part-time employment: Novartis. D. Castelletti: Full/Part-time employment: Novartis. M. Benelli: Honoraria (self): Novartis. L. Malorni: Advisory/Consultancy, Research grant/Funding (institution): Novartis; Advisory/Consultancy, Research grant/Funding (institution): Pfizer. All other authors have declared no conflicts of interest.

Background

In the CORALLEEN trial, R+L achieved similar response rates to multi-agent CT. We present a comprehensive gene expression analysis done before, during, and after therapy to fully characterize the biology behind the primary results of the trial.

Methods

CORALLEEN was a randomized study in postmenopausal women with stage I-IIIA hormone receptor positive (HR+)/HER2-negative Luminal B breast cancer by PAM50. Patients (pts) received either 6 cycles of R+L or 4 cycles of AC followed by 12 doses of paclitaxel. Primary endpoint was rate of PAM50 Risk of Relapse (ROR) low disease at surgery. Baseline, week 2, and surgical specimens were collected. Expression of 770 genes and 31 signatures were determined using the Breast360^TM nCounter-based codeset. Response was defined as ROR-low disease at surgery, relative/absolute changes in ROR between baseline/week 2 and surgery, RCB-0/I or levels of Ki67 at surgery. To identify genes associated with response, a significance of microarrays (SAM) analysis with a false discovery rate (FDR) <5% was performed.

Results

A total 297/318 (93.4%) samples were available. No genes or signatures at baseline, or week 2, were found to be associated with response at surgery in each arm. At week 2, 146 (14.6%) genes or signatures were found significantly up-regulated (n=47) and down-regulated (n=99) in the R+L arm compared to CT. R+L induced higher expression of genes related to DNA damage repair and immune activation (e.g. TP53, RAD52, GZMM and CD19) and lower expression of cell-cycle and hormone-related genes (e.g. PGR, CDK1 and MKI67). At surgery, 102 (10.2%) genes or signatures were found significantly up-regulated (n=4) and down-regulated (n=98) in the R+L arm compared to CT. R+L induced higher downregulation of estrogen- and proliferation-related genes and signatures (e.g. PGR, ER signaling, and MKI67).

Conclusions

No genes were able to predict response. Compared to CT, R+L induced higher downregulation of proliferation-related genes and signatures at week 2 and surgery. These results support the strategy to use the neoadjuvant setting to select patients who achieve a large molecular downstaging following CDK4/6 inhibition.

Legal entity responsible for the study

SOLTI Breast Cancer Research Group.

Funding

Novartis, Nanostring, Breast Cancer Research Foundation-AACR Career Development Award (to AP), PhD4MD grant (to NC), Fundación Científica Asociación Española Contra el Cáncer - Ayuda Postdoctoral AECC 2017 (to FB-M), This project has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant agreement No. 847912.

Disclosure

N. Chic: Travel/Accommodation/Expenses: Novartis. C. Saura: Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Puma biotechnology; Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Pfizer; Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy, Travel/Accommodation/Expenses: AstraZeneca; Advisory/Consultancy, Travel/Accommodation/Expenses: Celgene; Advisory/Consultancy, Travel/Accommodation/Expenses: Daiichi-Sankyo; Advisory/Consultancy, Travel/Accommodation/Expenses: Eisai; Advisory/Consultancy, Travel/Accommodation/Expenses: Genomyc Health; Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Novartis; Advisory/Consultancy, Travel/Accommodation/Expenses: Pierre Fabre; Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Synthon biopharmaceuticals; Research grant/Funding (institution): Roche-Genentech; Research grant/Funding (institution): Macrogenics; Research grant/Funding (institution): Piqur; Research grant/Funding (institution): BMS. M. Muñoz: Speaker Bureau/Expert testimony, Research grant/Funding (institution): Novartis; Research grant/Funding (institution): Breast Cancer Research Foundation-AACR; Research grant/Funding (institution): Breast Cancer Now Career Catalyst; Travel/Accommodation/Expenses: Roche; Speaker Bureau/Expert testimony: Pfizer; Speaker Bureau/Expert testimony: Lilly. X. González-Farré: Travel/Accommodation/Expenses: Roche; Travel/Accommodation/Expenses: Eisai. M. Oliveira: Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy: GlaxoSmithKline; Advisory/Consultancy: Puma Biotechnology; Research grant/Funding (institution): Philips Healthcare; Travel/Accommodation/Expenses: Grünenthal Group; Travel/Accommodation/Expenses: Novartis; Travel/Accommodation/Expenses: Pierre Fabre; Travel/Accommodation/Expenses: GP Pharm. M. Gil-Gil: Honoraria (self): Genentech; Honoraria (self): Novartis; Honoraria (self), Travel/Accommodation/Expenses: Pfizer; Honoraria (self), Travel/Accommodation/Expenses: Daiichi; Honoraria (self): Eisai; Travel/Accommodation/Expenses: Roche. E.M. Ciruelos: Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy: Lilly; Advisory/Consultancy: Novartis; Advisory/Consultancy, Travel/Accommodation/Expenses: Pfizer. P. Villagrasa: Speaker Bureau/Expert testimony: NanoString. J. Gavila Gregori: Advisory/Consultancy, Research grant/Funding (institution): Novartis; Advisory/Consultancy, Research grant/Funding (institution): Pfizer; Advisory/Consultancy, Research grant/Funding (institution): Roche. A. Prat: Advisory/Consultancy, Research grant/Funding (institution): Novartis; Advisory/Consultancy, Research grant/Funding (institution): Pfizer; Advisory/Consultancy: Lilly; Advisory/Consultancy: NanoString; Advisory/Consultancy, Research grant/Funding (institution): Amgen; Advisory/Consultancy, Research grant/Funding (institution): Roche; Advisory/Consultancy: Oncolytics; Advisory/Consultancy: Daiichi; Advisory/Consultancy: Puma; Advisory/Consultancy: BMS. All other authors have declared no conflicts of interest.

Background

15-20% of breast cancers are HER2 over-expressing (IHC 3+), or over-amplified (IHC2+/ISH+) while 45% have lower levels of HER2 expression (IHC 2+/ISH- or IHC 1+); there are currently no approved HER2-targeted therapies for patients with low levels of HER2 expression. Measuring HER2 expression levels is critical in the management of patients with breast cancer yet IHC results are often confounded by multiple variables, including fixative conditions (pre-analytical) and staining procedures (analytical). In addition, results are often difficult to interpret since a number of cases show only moderate overexpression of the protein and the analysis of the IHC staining are subject to interobserver variability (post-analytical). Therefore, more sensitive and objective assays are needed to better identify patients who may benefit from anti-HER2therapies including those patients with lower levels of HER2. To address this gap, we evaluated non-antibody-based methods to quantify targets from FFPE tissue and liquid biopsy.

Methods

Here, we have analyzed 107 breast carcinomas for ERBB2 RNA and protein expression using, QRT-PCR (Fluidigm) and SRM-MS (mProbe) and compared between patients with HER2 expression levels of IHC 0(34.6%), 1+(15.9%), 2+(27.1 %) or 3+(22.4%) (ANOVA).

Results

ERBB2 RNA and protein expression were progressively increased according to HER2 IHC grouping (i.e. lowest concentration in HER2 0 samples, highest in HER2 3+ samples). ERBB2 RNA and protein levels were significantly elevated in 2+ vs. 0 samples (2-fold increase, p < 0.05). No significant trend was observed in 2+ vs. 1+ and in 1+ vs. 0. Moreover, no correlation was seen between plasma and tissue ERBB2 RNA expression or IHC status. In this work, SRM-MS revealed a 100-fold difference in HER2 expression between the HER2 IHC 0 versus IHC 3+ tumor tissue samples. PCR-based methods were not reliable enough to detect a similar range of expression.

Conclusions

These results are encouraging and favor SRM-MS profiling as a more sensitive method to detect HER2 protein levels from tumor tissue samples. Future studies will include comparative analysis of SRM-MS versus other methods to detect HER2 expression and require confirmation on a larger series of tumors, notably for the tissue samples expressing lower levels of HER2 (0, 1+ and 2+).

Legal entity responsible for the study

Translational Medicine.

Funding

AstraZeneca.

Disclosure

F. Cecchi, D. Carrolle, M. Gustavson, S. Sridhar, M. de los Reyes, S. Thyparambil, A. Bhalkikar, W-L. Liao, S. Coats, T. Hembrough: Full/Part-time employment, Employee of AstraZeneca.

Background

There are more pathological complete responses (pCR) after neoadjuvant treatment in breast cancer with predominance of tumor infiltrating lymphocytes (TILs). The objective is to analyze immunosuppressive (regulatory T) and cytotoxic (CD8+ T) TILs before and after neoadjuvant treatment and the pathological response achieved in breast carcinoma.

Methods

Translational study of 50 breast carcinoma patients with neoadjuvant treatment. Measurement of cytotoxic CD8 + and regulatory T lymphocytes (CD25H or FOXP3 +) was performed in peripheral blood (before, during and after treatment), and before (biopsy) and after (surgical specimens) neoadjuvant in tumor tissue. The pathological response was assessed according to Miller & Payne (M&P: G1: minimal changes, G2: <30%, G3: 30-90%, G4:> 90%, G5: pCR). Peripheral blood lymphocytes were measured by flow cytometry (cells/microliter) and lymphocytes from tissue were measured by immunohistochemistry using the Ladoire classification (G0: 0 cells in 5f/20x, G1: 1-5, G2: 5-15, G3: > 15).

Results

Peripheral blood CD8+ T lymphocytes decreased significantly after treatment in patients with a <30% tumor response (M&P grade 1-2), median of 239 cells/ul in cycle 1 (C1) vs 133 cells/ul in C6, p 0.041. However, they remained constant (200-300 cells/ul) in 30-90% tumor response (M&P grade 3-4) and in pCR (M&P grade 5). Median CD8+ T lymphocytes in M&P grades 1-2 vs 5 were 184 vs 258 cells/ul (p 0.044) in C4, 180 vs 276 cells/ul (p 0.023) in C5 and 133 vs 285 cells/ul (p 0.012) in C6. The percentage of CD8+ T from tissue in M&P grade 5 is focused on Ladoire grade 3, while M&P grade 1-2 highlights a lower gradation of CD8+ T (Ladoire grade 0-2). There are high levels of FOXP3+ from tissue both before and after treatment in M&P grade 1-2. However, a low FOXP3+ percentage is expressed in M&P grade 5, and even that percentage decreases drastically in Ladoire grade 2-3 after treatment. The peripheral blood regulatory T (CD25H) cells descrease in M&P grade 3-4 and do not vary in M&P grade 1.

Conclusions

There is a significant descent of CD8+ T cells in non-pCR patients, while remaining elevated in pCR. There are more FOXP3+ T cells in non-pCR. CD8+ T and regulatory T cells are potential predictive biomarkers in breast carcinoma.

Legal entity responsible for the study

Hospital Universitario Virgen Macarena, Seville, Spain.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Tumor cellularity and tumor-infiltrating lymphocyte score (CelTIL) measured at day (D) 15 following anti-HER2 therapy was found associated with pathologic complete response (pCR) (Nuciforo. Ann Oncol 2018). Here, we explored the dynamics of CelTIL across 3 trials to determine its value as an early read-out of NAT efficacy.

Methods

Samples and clinical information from 369 patients (pts) across 3 trials (CORALLEEN, PAMELA and NEOERIBULIN) were explored. In CORALLEEN, 106 pts with Luminal B/HER2- breast cancer (BC) were randomized to 6 months of letrozole + ribociclib (LTZ+RIB) or AC x 4 + paclitaxel x 12. In PAMELA, 151 pts with HER2+ BC were treated with lapatinib + trastuzumab (hormonal therapy if hormone receptor [HR]-positive) for 18 weeks. In NEOERIBULIN, 101 HR+/HER2- and 73 triple-negative pts were treated with eribulin x 4. In each trial, TILs and tumor cellularity were determined at baseline and at D15 (PAMELA and CORALLEEN) or D21 (NEOERIBULIN) of treatment. CelTIL is calculated following this formula: −0.8 × tumor cellularity (%) + 1.3 × TILs (%). Changes in CelTIL between baseline and D15/21 samples and associations with response (RCB) were explored.

Results

In CORALLEEN, LTZ+RIB (n=49) or one dose of AC (n=47) did not show a significant change in CelTIL (mean -7.9; p=0.14 and +2.2; p=0.62). In NEOERIBULIN (n=132), eribulin significantly increased CelTIL in all pts (MD +10.0; p=0.03), both in HR+ (mean difference +4.8) and HR- (MD +11.1) BC. CelTIL changes were found significantly associated with both pCR and RCB0/1, independently of HR status. In PAMELA (n=141), NAT significantly increased CelTIL in all pts (MD +31.4; p<0.01), both in HR+ (MD +32.2) and HR- (MD +40.4) BC. CelTIL changes were found significantly associated with pCR and RCB0/1, independently of HR status. Finally, difference in CelTIL (D15-baseline) was found significantly associated with RCB0/1 (odds ratio=1.02, p<0.001) independently of trial and HR status.

Conclusions

Early and absolute changes in CelTIL following NAT are associated with tumor shrinkage at surgery. This biomarker could be used as an early read-out of drug activity and these data should help estimate power and sample size for future trials.

Legal entity responsible for the study

SOLTI Breast Cancer Research Group.

Funding

Pas a Pas, Save the Mama.

Disclosure

P. Nuciforo: Advisory/Consultancy: Bayer; Advisory/Consultancy, Speaker Bureau/Expert testimony: MSD; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis. J. Cortés: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Roche; Honoraria (self), Advisory/Consultancy: Celgene; Advisory/Consultancy: Cellestia; Advisory/Consultancy, Research grant/Funding (institution): AstraZeneca; Advisory/Consultancy: Biothera Pharmaceutical; Advisory/Consultancy: Merus; Advisory/Consultancy: Seattle Genetics; Advisory/Consultancy: Daiichi Sankyo; Advisory/Consultancy: Erytech; Advisory/Consultancy: Athenex Polyphor; Honoraria (self), Advisory/Consultancy: Lilly; Advisory/Consultancy: Servier; Advisory/Consultancy, Research grant/Funding (institution): Merck Sharp & Dohme; Honoraria (self): Novartis; Honoraria (self), Research grant/Funding (institution): Eisai; Honoraria (self), Research grant/Funding (institution): Pfizer; Honoraria (self): Samsung; Research grant/Funding (institution): Ariad pharmaceuticals; Research grant/Funding (institution): Baxalta/Servier Affaires; Research grant/Funding (institution): Bayer Healthcare. A. Llombart Cussac: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Novartis; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Roche; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: AstraZeneca; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Eli Lilly; Research grant/Funding (institution), Travel/Accommodation/Expenses: EISAI; Advisory/Consultancy, Research grant/Funding (institution): Genomic Health; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Pfizer; Advisory/Consultancy: MSD. J. Gavila Gregori: Advisory/Consultancy, Research grant/Funding (institution): Novartis; Advisory/Consultancy, Research grant/Funding (institution): Pfizer; Advisory/Consultancy, Research grant/Funding (institution): Roche. N. Chic: Travel/Accommodation/Expenses: Novartis. M. Vidal: Honoraria (self), Travel/Accommodation/Expenses: Pfizer; Honoraria (self), Advisory/Consultancy: Novartis; Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Honoraria (self): Daiichi Sankyo. M. Muñoz: Honoraria (self): Lilly; Honoraria (self), Travel/Accommodation/Expenses: Roche Genentech; Honoraria (self): Eisai; Honoraria (self): Merck; Honoraria (self), Travel/Accommodation/Expenses: Novartis. P. Villagrasa: Speaker Bureau/Expert testimony: NanoString. A. Prat: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Research grant/Funding (institution): Roche; Honoraria (self): MSD Oncology; Honoraria (self): Lilly; Honoraria (self), Travel/Accommodation/Expenses: Daiichi Sankyo; Advisory/Consultancy: NanoString; Advisory/Consultancy: Amgen; Advisory/Consultancy: Bristol-Myers Squibb. All other authors have declared no conflicts of interest.

Background

Checkpoint proteins regulate the immune system. Breast cancer (BC) cells exploit the up-regulation or down-regulation of these proteins to evade anti-tumor immune responses. Soluble forms of immune checkpoint molecules (ICM) can be measured in human plasma, however their biological and clinical significance remains essentially unknown. The aim of the present analysis was to measure the levels of pre-treatment ICM in newly- diagnosed BC patients (pts) and compare them to healthy controls.

Methods

Soluble forms of ICM, as well as cytokines and chemokines, were measured using Multiplex® bead array and ELISA technologies. Plasma samples from 98 BC pts and 45 healthy controls were analyzed for each protein. Data was prospectively obtained. Measured levels were compared between BC pts and healthy controls using a non-parametric test (Mann-Whitney).

Results

Soluble stimulatory molecules GITR (p<0.000002), GITRL (p< 0.007), CD27 (p< 0.002), CD28 (p<0.003), CD40 (p< 0.003), CD80 (p< 0.009), ICOS (p< 0.0006), as well as inhibitory molecules PD-L1 (p< 0.0000001), CTLA-4 (p< 0.005), TIM-3 (p< 0.00006), HVEM (p< 0.00002) TLR-2 (p< 0.05), levels were significantly lower in early BC pts compared to healthy controls. When analyzed according to BC characteristics (TNBC vs. non-TNBC, tumor size, stage, nodal status and age) no significant difference was detected between the soluble levels of these ICM between the different subsets. Additionally, serum levels of CXCL5 (p< 0.000001), CCL23 (p< 0.04), IL-16 (p< 0.00005), interferon-α (p< 0.03) and IL1-RA (p< 0.03) were significantly lower compared to healthy controls. Serum CX3CL1 or fractalkine (p< 0.024465) was significantly higher compared with healthy controls.

Conclusions

We identified low levels of both the stimulatory and inhibitory immune checkpoint molecules in newly diagnosed, non-metastatic BC pts compared to healthy controls. These results indicate that early BC is associated with a down-regulation of both soluble stimulatory and inhibitory immune-checkpoint pathways. Newly diagnosed early BC pts have a generalized immune-suppression independent of subtype and stage, which, to our knowledge, is the first study to describe soluble immune checkpoints in early BC pts.

Legal entity responsible for the study

Department of Immunology, University of Pretoria, South-Africa.

Funding

CANSA (Cancer Association of South Africa), Roche South-Africa.

Disclosure

All authors have declared no conflicts of interest.

Background

How the breast cancer genome evolves during malignant progression has been highly debated. To gain insight into the landscape of genomic aberrations occurring in breast cancers from the onset to metastasis, we compared the molecular profile of matched primary and relapsed tumors, focusing on the additional alterations developed in metastases.

Methods

The study population included a mono-institutional cohort of 128 patients with breast cancers (n=72 Luminal B - LUM, n=56 Triple-negative - TN) and with at least one recurrence in a timeframe of 17 years. 289 samples, comprising primary tumors (P) and matched metastases (M), were subjected to a comprehensive genomic profile using Next-Generation Sequencing (NGS) panels (FoundationeOne Dx or Oncomine Comprehensive Cancer Assay v.3). Genomic data were available for 106 P and 82 M that reached the NGS quality parameters.

Results

The most recurrent genomic alterations were TP53 mutations (P=49%, M=49%), PIK3CA mutations (P=33%, M=30%) and MYC copy number gain (P=25%, M=23%). TP53, PIK3R1, and NF1 aberrations were more frequently identified in TN breast cancer whereas ESR1 mutations and CCND1, FGF3, FGF19, and FGFR1 copy number gains in LUM cases (p-value < 0.05). The total number of P alterations, TP53 mutations, and MYC amplification were significantly and independently associated with a shorter time to relapse (p-value < 0.05). Considering matched P and M samples, we found a molecular subtype change in 10 of 128 (7.8%) cases. 55.8% of driver alterations were shared between P and matched M, including the most frequently mutated genes. In 27/61 (44.3%) cases 65 driver alterations were detected in M only, including 20 alterations classified as level1-level4 by OncoKB ranking and affecting mostly ESR1 (n=8) and PIK3CA (n=8) genes.

Conclusions

Prognostic biomarkers associated with time to relapse could be identified in primary tumors. A core of driver alterations was shared between P and M samples but novel and clinically relevant alterations could be identified in M only.

Legal entity responsible for the study

Giuseppe Viale.

Funding

ERA-NET on Translational Cancer Research (TRANSCAN) European Commission/DG Research and Innovation on: “Translational research on human tumour heterogeneity to overcome recurrence and resistance to therapy”.

Disclosure

C. Fumagalli: Advisory/Consultancy: Biocartis. M. Barberis: Advisory/Consultancy: Roche; Advisory/Consultancy: MSD; Advisory/Consultancy: AstraZeneca. E. Guerini Rocco: Advisory/Consultancy: ThermoFisher; Advisory/Consultancy: Novartis; Honoraria (self): AstraZeneca; Honoraria (self): Roche. All other authors have declared no conflicts of interest.

Background

PIK3CA gene mutations are a source of heterogeneity in HER2-positive breast cancer, with potential impact on prognosis and treatment sensitivity. We explored the frequency, association with clinicopathological features and prognostic impact of PIK3CA mutations in the randomized adjuvant ShortHER trial.

Methods

The ShortHER trial randomized 1254 patients with HER2-positive early breast cancer to 9 weeks or 1 year of adjuvant trastuzumab combined with chemotherapy. Non-inferiority of the short arm was not demonstrated with the frequentist approach. PIK3CA hot-spot mutations in exon 9 (E542K, E545K-A-G, Q546E-K) and exon 20 (M1043I, H1047R-L-Y, G1049R-S) were analysed by using Pyrosequencing method on DNA extracted from centralized FFPE tumor samples.

Results

A mutation of the PIK3CA gene was detected in 21.7% of the 803 genotyped patients (n=174 mutated; n=629 wild type). Mutations in exon 9 and 20 occurred in 78 (9.7%) and 95 (11.8%) cases, respectively. PIK3CA mutation occurred more frequently in hormone receptor-positive vs hormone-receptor negative cases (23.5% vs 17.4%, p=0.057) and in post-menopausal vs premenopausal patients (23.8% vs 17.7%, p=0.050). No association with stage, age, grade, TILs and treatment arm was observed. At a median follow up of 7.8 years, 5-yr DFS rates were 90.6% for PIK3CA mutated and 86.2% for PIK3CA wild-type patients (HR 0.84, 95% CI 0.56-1.27, p=0.417). PIK3CA mutation showed a favorable prognostic impact in the subgroup of patients owing to the PAM50 HER2-enriched subtype (n=232): 5-yr DFS 91.8% vs 76.1% for PIK3CA mutated and wild-type patients (log-rank p=0.049; HR 0.46 95% CI 0.21-1.02). There was no significant difference in DFS according to PIK3CA mutation in subgroups defined by hormone receptor status: HR 0.87 (95% CI 0.39-1.95, p=0.734) for hormone receptor-negative and HR 0.84 (95% CI 0.52-1.36, p=0.471) for hormone receptor-positive.

Conclusions

Within the HER2-enriched molecular subtype, PIK3CA mutated patients showed better DFS as compared to PIK3CA wild-type patients. These results highlight the need to integrate multiple biomarkers in order to dissect the heterogeneity of HER2-positive breast cancer.

Clinical trial identification

NCT00629278.

Legal entity responsible for the study

University of Padua.

Funding

Italian Medicines Agency (AIFA, grant FARMS5KR); Italian Association for Cancer Research (AIRC, grant MFAG-15938).

Disclosure

V. Guarneri: Advisory/Consultancy, Research grant/Funding (institution): Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eli Lilly. M.V. Dieci: Advisory/Consultancy: Genomic Health; Advisory/Consultancy: Eli Lilly. A. Frassoldati: Advisory/Consultancy: Roche; Advisory/Consultancy: Novartis; Honoraria (self): Pfizer. A. Musolino: Advisory/Consultancy: Roche; Advisory/Consultancy: Eli Lilly; Advisory/Consultancy: Pfizer; Advisory/Consultancy: Eisai; Advisory/Consultancy: Macrogenics. O. Garrone: Advisory/Consultancy: Eisai; Advisory/Consultancy: Amgen; Advisory/Consultancy: Novartis. F. Piacentini: Advisory/Consultancy: Eisai. A. Prat: Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Roche; Honoraria (self): MSD Oncology; Honoraria (self): Lilly; Honoraria (self), Travel/Accommodation/Expenses: Daiichi Sankyo; Advisory/Consultancy: NanoString Technologies; Advisory/Consultancy: Amgen; Advisory/Consultancy: BMS. P.F. Conte: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Novartis; Honoraria (self), Advisory/Consultancy: Eli Lilly; Honoraria (self), Advisory/Consultancy: AstraZeneca; Advisory/Consultancy: Tesaro; Honoraria (self), Research grant/Funding (institution): BMS; Honoraria (self), Research grant/Funding (institution): Roche; Research grant/Funding (institution): Merck KGaA. All other authors have declared no conflicts of interest.

Background

Mutations in DDR genes, most notably BRCA1/2 and have been predictive to respond to PARP inhibitors and other DNA-damaging systemic therapies. However, the reports of detected germline and somatic variants in other DDR-related genes in early-stage, untreated breast cancers is limited. Here we report variants identified in multiple DDR-related genes in patients with untreated TNBC and association with pathologic complete response (pCR).

Methods

Pretreatment core biopsies were obtained from 193 patients with early-stage (I-III) TNBC enrolled on the ARTEMIS trial (NCT02276443). DNA was extracted from blood for germline as well as from tumor samples for somatic testing and underwent whole exome sequencing and RNAseq. Enrichment of gene alterations were compared using a Fisher exact test in patients (pts) with and without pCR. Whole exome sequencing was performed and pair-end sequencing reads in FASTQ format were generated and aligned to the hg19 human reference genome. Platypus was used to call germline mutations on DDR genes. MuTect was used to identify somatic point mutations, and Pindel was used to identify somatic insertions and deletions. A series of post-calling filtering were applied for somatic mutations.

Results

Overall DDR variants were identified in 42% (82/193) of the patients and 110 total variants identified (Table). Almost all of were somatic (101/110). Out of 21 genes considered, 16 had an identifiedvariant in at least 1 patient. Somatic + germline mutations in BRCA was associated with an increase in pCR (p=0.03). Although filtering against known databases were completed to exclude polymorphisms as much as possible, some of these findings may still be consistent with variants of uncertain significance. Table: 21P

Conclusions

Both germline and somatic variants in DDR-related genes were identified in patients with early-stage, untreated TNBC. Further somatic variant analysis and response to neoadjuvant chemotherapy will be reported.

Clinical trial identification

NCT02276443.

Legal entity responsible for the study

The authors.

Funding

The Moonshot Program at The University of Texas MD Anderson Cancer Center.

Disclosure

J.K. Litton: Research grant/Funding (institution),: Genentech; Research grant/Funding (institution): AstraZeneca; Research grant/Funding (institution): Pfizer/Medivation; Advisory/Consultancy, uncompensated: AstraZeneca; Advisory/Consultancy, uncompensated: Pfizer; Research grant/Funding (institution): zenith; Research grant/Funding (institution): EMD-Serono. All other authors have declared no conflicts of interest.

Background

Novel antibody-drug conjugates against HER2 are showing high activity in clinically HER2-negative (cHER2-) breast cancer (BC) with low HER2 expression. However, the clinical and molecular features of cHER2-/HER2-low BC are yet to be elucidated.

Methods

We collected retrospective data from 8 multicenter cHER2- BC datasets, including 4 clinical trials. HER2 status in each study was determined using standard FDA-approved antibodies and ISH-techniques and classified according to the ASCO/CAP guidelines. For this study, tumors were regrouped in HER2 0 (0 score) and HER2-low (1+ or 2+ with ISH-negativity). The following variables were compared between the 2 groups in all patients and according to hormone receptor (HR) status: age, grade, ki67, histotype, tumor size, HR, HER2 and nodal status. nCounter-based PAM50 subtypes distribution and the expression of the 50 PAM50 genes, including ERBB2, were also compared.

Results

A total of 3,136 patients with cHER2- disease (57.4% HER2-low and 42.6% HER2 0) were evaluated. Overall, 888 (28.3%) tumor samples came from metastatic sites. No statistically significant differences were found regarding clinicopathological variables between HER2-low and HER2 0. Within HR-positive (+) disease (n=2,497), 63.2% and 36.8% of tumors were HER2-low and HER2 0, respectively. Subtype distribution was similar across HR+/HER2-low and HR+/HER2 0. A total of 45/50 PAM50 genes were found differentially expressed between HR+/HER2-low and HR+/HER2 0 (False Discovery Rate [FDR]<5%). High expression of luminal (e.g. ESR1 and FOXA1) and ERBB2, and low expression of proliferation-related genes (e.g. MKI67) was found in HER2-low compared to HER2 0. Within triple negative BC (TNBC) (n=622), 34.2% were HER2-low and 65.8% were HER2 0. Subtype distribution was similar across TNBC/HER2-low and TNBC/HER2 0. No PAM50 gene was found differentially expressed between TNBC/HER2-low and TNBC/HER2 0 (FDR≥5%). Finally, ERBB2 mRNA levels were higher in HER2-low/HR+ tumors than HER2-low/TNBC (p<0.001).

Conclusions

HER2-low disease within clinically HER2- BC is frequent. However, significant differences exist according to HR status. Compared to HER2-low/TNBC, HER2-low/HR+ disease is a more distinct biological entity and has higher ERBB2 expression.

Legal entity responsible for the study

The authors.

Funding

Instituto de Salud Carlos III - PI16/00904, Pas a Pas, Save the Mama, Breast Cancer Now - 2018NOVPCC1294. Fundación Científica Asociación Española Contra el Cáncer - Ayuda Postdoctoral AECC 2017. Fundación SEOM, Becas FSEOM para Formación en Investigación en Centros de Referencia en el Extranjero 2016 and 2018.

Disclosure

C.H. Barrios: Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): AstraZeneca, Novartis, Roche, GSK, Pfizer, Libbs, Daiichi Sankyo and MSD. A. Lluch: Research grant/Funding (self): Amgen, AstraZeneca, Boehringer-Ingelheim, GSK, Novartis, Pfizer, Roche/Genentech, Eisai, Celgene, Pierre Fabre; Advisory/Consultancy: Novartis, Pfizer, Roche/Genentech, Eisai, Celgene. M. Martin Jimenez: Research grant/Funding (self): Roche, PUMA and Novartis; Advisory/Consultancy: AstraZeneca, Amgen, Taiho Oncology, Roche/Genentech, Novartis, PharmaMar, Eli Lilly, PUMA, Taiho Oncology, Daiichi Sankyo and Pfizer; Speaker Bureau/Expert testimony: AstraZeneca, Amgen, Roche/Genentech, Novartis and Pfizer. A. Prat: Honoraria (self): Pfizer, Novartis, Roche, MSD Oncology, Lilly and Daiichi Sankyo; Research grant/Funding (self): Novartis and Roche; Travel/Accommodation/Expenses: Daiichi Sankyo; Advisory/Consultancy: NanoString Technologies, Amgen, Roche, Novartis, Pfizer and Bristol-Myers Squibb; Licensing/Royalties: patent PCT/EP2016/080056. All other authors have declared no conflicts of interest.

Background

To investigate the influence of metabolic syndrome and its components on the risk of breast cancer.

Methods

Retrospective nationwide cohort study analyzing data of 13,377,349 women older than 19 years who were enrolled between 2009 and 2014 from the Korean National Health Insurance Service was performed. Cox proportional hazards model was used to calculate hazard ratio (HR) and 95% confidence interval (CI) of breast cancer risk according to metabolic syndrome and its components.

Results

The presence of metabolic syndrome decreased the risk of all breast cancer types in all subjects (HR: 0.954; 95% CI: 0.939-0.970). In women with age ≤ 50 years, metabolic syndrome decreased the risk of all breast cancer types, with similar findings for all subject groups (HR: 0.915; 95% CI: 0.892-0.939). In women with age >50 years, metabolic syndrome increased the risk of all breast cancer types (HR: 1.146; 95% CI: 1.123-1.170), especially in age groups of more than 55 years. The effect of metabolic syndrome was more prominent as age subgroups became older, especially in postmenopausal women. In women with age > 50 years, HRs increased as the number of metabolic syndrome components increased, while HRs decreased as the number of metabolic syndrome components increased in women with age ≤ 50 years.

Conclusions

The presence of metabolic syndrome increased the risk of breast cancers in postmenopausal women, but decreased the risk in premenopausal women. The effect of metabolic syndrome was more prominent as age subgroups became older, especially in postmenopausal women. Every metabolic syndrome component played similar roles on the risk of breast cancer. Their effects became stronger when the number of components increased.

Editorial acknowledgement

This study was supported by the Korean Breast Cancer Society and the National Health Insurance Corporation.

Legal entity responsible for the study

Ki-Tae Hwang.

Funding

Has not received any funding.

Disclosure

The author has declared no conflicts of interest.

Background

Invasive breast cancers with HER2 gene amplification are associated with poor outcome. Heterogeneous HER2 amplification has been observed in up to 41% of breast cancers, depending upon its definition. Carcinogenesis is driven by intra-tumour heterogeneity. Molecular diversity enables cancer cells to circumvent specific targeted treatment. In this study, we compared the genetic differences between admixed HER2-positive and HER2-negative breast cancer components. This in-depth analysis investigated the heterogeneity in their somatic mutational landscape.

Methods

Formalin-fixed, paraffin-embedded tissue of 10 breast carcinomas with at least one HER2-negative and at least one HER2-positive component was micro-dissected. Each component was subjected to targeted next-generation sequencing using a 53-gene panel. Somatic mutations and copy number variations were investigated.

Results

We identified 3 splice site alterations, 32 missense variants, 12 deletions, 9 insertions, and 7 nonsense variants in 26 different genes, which are (likely) pathogenic. Overall, these molecular anomalies were heterogeneously distributed among the different tumour components. The HER2-negative tumour components did not yield common alternative drivers. One patient had a CCND1 copy number gain limited to a HER2-negative tumour component. Two patients had an 8q24 gain in at least one cancer component, resulting in increased copy numbers of the MYC and PVT1 genes. Two patients had an FGFR1 copy number gain in at least one tumour component. One patient had an EGFR copy number gain in a HER2-negative DCIS component, resulting in EGFR protein overexpression.

Conclusions

This series of 10 heterogeneously HER2-amplified breast tumours demonstrates that not all breast cancer cells require HER2 as a driver of tumour growth. Several other molecular anomalies are able to act as alternative or collaborative drivers. This study illustrates that breast carcinogenesis is characterized by a diverse and heterogeneous molecular landscape, of which some genetic anomalies drive cancer progression, and others are mere ‘passenger’ molecular aberrations.

Legal entity responsible for the study

The authors.

Funding

Mathilde Horlait Dapsens Foundation (Brussels, Belgium).

Disclosure

All authors have declared no conflicts of interest.

Background

Invasive breast cancer (IBC) subtypes, which are subject to different treatments, are identified in clinical routine by expression of estrogen receptor (ER), progesterone receptor (PR), Ki-67 and HER2 status by immunohistochemistry (IHC) and/or in situ hybridization (ISH). Yet, IHC evaluation might be hampered by (pre-)analytical errors and optimal cut-offs are still under discussion. Gene expression assays may offer a reliable way to measure mRNA expression of these four markers (ESR1, PGR, ERBB2 and MKI67). Here, we investigated the correlation of the commercially available “four-marker” Xpert® Breast Cancer STRAT4 (CE-IVD) mRNA assay with the gold standard (IHC/ISH) in different pathologic laboratories across Europe.

Methods

Ten pre-therapy breast core biopsies with IBC [six ER+/PR+ with varying Ki-67, two HER2+, two triple negative IBC diagnosed in the coordinating center (CC)] with sufficient formalin fixed paraffin embedded tissue were evaluated. IHC/ISH data for ER, PR, HER2 and Ki-67 were extracted from the original pathology report. For each case, STRAT4 (ESR1, PGR, ERBB2 and MKI67 mRNA assay) was performed in the CC and STRAT4 results matched IHC subtyping. Five European pathology laboratories participated in the harmonization study. Each site received one H&E stained slide and one unstained slide for STRAT4 testing. Binary mRNA results of each marker (positive vs. negative) were compared with the gold standard IHC/CISH of the CC. 80% of all results tested at each site had to be in agreement with the gold standard to pass the EQA.

Results

All centers passed the EQA study. Sensitivity, specificity and accuracy of ESR1 and ERBB2 mRNA were 100% for all ten samples. Instead, PGR mRNA was falsely reported as negative for one case (case #2) by two sites and MKI67 was falsely negative for two cases (case #2 by four sites, #10 by one site). Case #2 was a pleomorphic invasive lobular BC with heterogeneous PR (IHC staining 10%) and Ki-67 IHC (up to 30%), whereas case #10 displayed homogenous Ki-67 IHC.

Conclusions

The results of our study showed that STRAT4 might offer a reliable alternative for the evaluation of ER, PR, HER2 and Ki-67 in IBC. However, prognostic and predictive value of STRAT4 should be further validated in clinical cohorts.

Legal entity responsible for the study

Institute of Pathology, University Hospital Erlangen, Erlangen, Germany.

Funding

Cepheid, Sunnyvale, CA 94089, United States.

Disclosure

R. Erber: Honoraria (self): Roche Pharma AG; Honoraria (self): Eisai; Honoraria (self): Novartis; Research grant/Funding (institution), Travel/Accommodation/Expenses: BioNTech; Research grant/Funding (institution): Cepheid; Research grant/Funding (institution): NanoString Technologies. A. Hartmann: Research grant/Funding (self): Cepheid. P. Fasching: Research grant/Funding (self): Cepheid; Honoraria (self), Advisory/Consultancy: Roche; Honoraria (self), Advisory/Consultancy: Amgen; Honoraria (self), Research grant/Funding (institution): Novartis; Honoraria (self), Advisory/Consultancy: Pfizer; Honoraria (self), Advisory/Consultancy: Celgene; Honoraria (self), Advisory/Consultancy: Puma Biotechnology; Honoraria (self), Advisory/Consultancy: Daiichi Sankyo; Honoraria (self), Advisory/Consultancy: Eisai; Honoraria (self), Advisory/Consultancy: Merck Sharp & Dohme; Honoraria (self), Advisory/Consultancy: AstraZeneca; Honoraria (self), Advisory/Consultancy: Hexal; Honoraria (self), Advisory/Consultancy: Myelo Therapeutics GmbH; Honoraria (self): Lilly; Honoraria (institution), Research grant/Funding (institution): Cepheid; Honoraria (institution), Research grant/Funding (institution): BioNTech AG; Advisory/Consultancy: Teva; Advisory/Consultancy: Macrogenics. R. Stöhr: Research grant/Funding (institution): Cepheid. M.W. Beckmann: Research grant/Funding (institution): Cepheid. M. Zentgraf: Research grant/Funding (institution): Cepheid. M. Ruebner: Research grant/Funding (institution): Cepheid. H. Huebner: Research grant/Funding (institution): Cepheid. E. Guerini Rocco: Research grant/Funding (institution): Cepheid. G. Viale: Honoraria (self), Advisory/Consultancy: MSD Oncology; Honoraria (self): Pfizer; Advisory/Consultancy: Dako; Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (self), Travel/Accommodation/Expenses: Roche/Genentech; Advisory/Consultancy: Astellas Pharma; Advisory/Consultancy: Novartis; Advisory/Consultancy: Bayer; Advisory/Consultancy: Daiichi Sankyo; Advisory/Consultancy: Menarini; Research grant/Funding (institution): Ventana Medical Systems; Research grant/Funding (institution): Dako/Agilent Technologies; Research grant/Funding (institution): Cepheid; Travel/Accommodation/Expenses: Celgene. A. Cayre: Research grant/Funding (institution): Cepheid. F. Penault-Llorca: Research grant/Funding (institution): Cepheid. T. Caniego Casa: Research grant/Funding (institution): Cepheid. J. Palacios Calvo: Research grant/Funding (institution): Cepheid. P. Jank: Research grant/Funding (institution): Cepheid. C. Denkert: Research grant/Funding (institution): Cepheid. L. Khoury: Research grant/Funding (institution): Cepheid. T. Mairinger: Research grant/Funding (institution): Cepheid. F. Ferrazzi: Research grant/Funding (institution): Cepheid. All other authors have declared no conflicts of interest.

Background

Breast cancer (BC) shows a high incidence both in Kazakhstan and around the world, among the diseased the proportion of young women in the last 10 years is increasing, some women have a burdened family history. Approximately 20-30% of cases of hereditary breast cancer are caused by presence of BRCA1 and BRCA2 genes defects. Also, there are additional genes which can increase the risk of BC and they are still under study.

Methods

The study included 235 unrelated patients from Kazakh population(the average age 34.25 ± 4.56) with BC. Genomic DNA was obtained from peripheral blood and sequencing was performed using TruSight Cancer Kit on the MiSeq platform. Studio Variant was used to annotate and interpret genetic variants. 26 (11,1%) patients had a maternal family history of BC, 21 (80.8%) of them had first- degree relatives diagnosed with BC at different age.

Results

Bioinformatics analysis of NGS data identified 23,915 variants in 83 genes, 8030 evaluated as missense variants, 13212 as synonymous nucleotide substitutions, 753 variants in the 3′-untranslated region (3′-NTO) , 1221 variants in the 5′-untranslated region (5′-NTO), 35 variants leading to a shift of the reading frame (deletion / insertion), nonsense (stop codon variants) - 17, variants in the acceptor splicing site - 7, variant splicing donor site - 2, inframe deletion /insertion-10, variants that violate start codons- 3, as well as 622 intron variants / variants in non-coding regions. Pathogenic mutations in the genes BRCA1 (24 variants (37.5%) and BRCA2 (18 (28.1%) were most often found. Also, 22 additional pathogenic variants were identified in the non-for BRCA genes (APC, ATM, BLM, CHEK2, PALB2, TP53, ERCC2, FANCA, FANCM, NBN, PMS1, PMS2, SDHB and XPA). A hereditary history was recorded in 29.1% and 27.8% of representatives of the group with pathogenic mutations in the BRCA1 and BRCA2, respectively, higher compared to the group of patients without pathogenic mutations and to the group of patients with mutations in the BRCA-negative genes.

Conclusions

NGS showed frequent and novel germline mutations in BRCA1/2 and non-BRCA genes. After final statistic data processing, diagnostic and prevention tools for key genes will be developed and included in the National guidelines of Republic of Kazakhstan.

Legal entity responsible for the study

Ministry of Healthcare, Republic of Kazakhstan.

Funding

Ministry of Healthcare, Republic of Kazakhstan.

Disclosure

All authors have declared no conflicts of interest.

Background

Common sites of distant metastases from BC include the skeleton and a number of studies have long attempted to identify a specific osteotropism signature to select patients for preventive therapeutic options. Here, we evaluated the expression of a gene panel potentially involved in BM onset in CTCs from metastatic BC patients and found a putative correlation of such a gene profile with sites of distant relapse.

Methods

Following approval by local Ethics Committee and written informed consent, CTCs were isolated from 39 stage IV BC patients, either treatment naïve or in progressive disease, through immunomagnetic pre-enrichment with autoMACS Separator® and sorting by DEPArray®. A panel of 136 genes involved in BC progression and BM development was derived from literature data to assess their expression levels in CTCs by RNAseq. The panel was first verified on subclones of the MDA-MB231 BC cell line with different organotropism (P0:parental population; P7:osteotropic subclone; LM:subclone with lung tropism) and then validated in CTCs from patients grouped in relation to their metastatic sites, namely (a) BM-only, (b) Other, (c) BM and Other, at the time of collection.

Results

The median number of recovered CTCs was 56 (range 9–106). No correlation was found between CTC number and BC histopathological features. The transcriptome heatmap of unsupervised hierarchical clustering of BC cell lines, based on normalized read counts, clustered distinct profiles in relation to their tissue tropism. The feasibility of this approach was then validated on CTCs and 31 differentially expressed genes (DEGs) were revealed between CTCs from (a) and (b) groups. By Gene Ontology evaluation of these DEGs, we found that most of them were enriched with greater statistical significance in different biological processes enrolled in bone tissue development and morphogenesis.

Conclusions

CTCs isolated from BC patients with different sites of metastases harbor distinct GEP that can be successfully evaluated by our assay. Prospective investigation is desirable to assess the prognostic role of identified DEGs in earlier BC stages.

Legal entity responsible for the study

The authors.

Funding

AIRC Associazione Italiana per la Ricerca sul Cancro [grant number 17536].

Disclosure

J. Brown: Honoraria (self), Research grant/Funding (institution): Amgen; Honoraria (self), Research grant/Funding (institution): Novartis; Honoraria (self), Research grant/Funding (institution): Bayer; Honoraria (self): BMS. R.E. Coleman: Advisory/Consultancy, Speaker Bureau/Expert testimony: Amgen; Advisory/Consultancy, Speaker Bureau/Expert testimony: Astellas; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eisai; Advisory/Consultancy, Speaker Bureau/Expert testimony: Genomic Health; Advisory/Consultancy, Speaker Bureau/Expert testimony, patent holder for a biomarker developed by Inbiomotion: Inbiomotion; Advisory/Consultancy, Speaker Bureau/Expert testimony: Scancell. All other authors have declared no conflicts of interest.

Background

BRCA1 and BRCA2 are tumor suppressor genes that play an important role in prevention, monitoring, and use of targeted therapy in cancer. The mutation types include single-nucleotide variation (SNV), small insertions and deletions (InDel) and copy number variations (CNV). Traditional methods for SNV and InDel detection are multiplex polymerase chain reaction (PCR) plus Sanger sequencing or Next Generation Sequencing (NGS). For CNV detection, the gold standard method is Multiplex Ligation-dependent Probe Amplification (MLPA). A new NGS method based on Halo-shape ANnealing and Defer-Ligation Enrichment (HANDLE) technology developed by AmoyDx can detect all mutation types within one reaction tube, and a turn-around time of 5 h for library preparation with hands-on time of 1 hour.

Methods

There were 2686 Chinese samples (2563 whole blood samples and 123 FFPE tissue samples) detected in AmoyDx Medical Institute, including 1357 breast, 754 ovarian, and 575 other samples (prostate, pancreatic, patients of unknown cancer types and family members of cancer patients). All samples were tested following the instructions of the AmoyDx BRCA1 and BRCA2 gene mutation detection kit combined with the automatic AmoyDx NGS Data Analysis System (ANDAS). The MLPA method was used for confirmation of CNV results of whole blood samples.

Results

In total, there were 476 samples detected with a pathogenic or likely pathogenic BRCA mutation. The mutation rate was 17.7% for all samples, 12.2% in breast and 26.4% in ovarian cancer samples. There were 11 samples with multiple breast cancer types or combined breast cancer and other cancer types, 6 of which were detected with pathogenic or likely pathogenic mutations. There were 20 out of 476 (4.2%) samples that demonstrated pathogenic or likely pathogenic BRCA CNVs by MLPA. 18 (90%) of these mutations were from BRCA1 gene, and 19 (95%) of them were concordant with the BRCA mutation detection according to the AmoyDx NGS test kit based on HANDLE technology.

Conclusions

BRCA1/2 gene mutations were detected in multiple cancer types. CNVs of BRCA1 were much more frequent than CNVs of BRCA2 in Chinese cancer patients. Testing of BRCA 1/2 by NGS based on HANDLE technology is a valid and fast solution for detection of BRCA1/2 mutations.

Legal entity responsible for the study

AmoyDx Medical Institute.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Preclinical evidence indicates that cyclin-dependent kinase (CDK) 4/6 inhibitors stimulate antitumor immunity, which may contribute to their anticancer activity. The neutrophil-to-lymphocyte ratio (NLR) and the platelet-to-lymphocyte ratio (PLR) reflect systemic inflammation and immune system functional status, and could be associated with CDK 4/6 inhibitor efficacy in patients (pts) with hormone receptor-positive advanced breast cancer (HR+ aBC).

Methods

We performed a retrospective, monocentric study to investigate the association between NLR or PLR, as measured at baseline and after the first three treatment cycles, and progression free survival (PFS) in HR+ aBC pts treated with CDK 4/6 inhibitors in combination with endocrine therapies (ETs). The thresholds for NLR and PLR were defined using the maximally selected rank statistics. The impact of these parameters on PFS was evaluated at univariate and multivariable analysis by using Cox proportional hazard model.

Results

A total of 162 pts treated with palbociclib (n=142), ribociclib (n=16) or abemaciclib (n=3) plus ETs between January 2017 and December 2019 at our Institution were included. NLR and PLR at baseline were not associated with PFS. Conversely, high NLR (>3) and high PLR (>323.6) after three treatment cycles were associated with significantly lower PFS (p=0.011 and p=0.013, respectively). Multivariable analysis confirmed an independent association between high NLR or PLR and lower PFS (aHR 3.66, 95% CI 1.44-9.33, p=0.007 and aHR 2.79, 95% CI 1.36-5.70, p=0.005, respectively). Another factor associated with worse PFS was the presence of liver metastases (aHR 3.02, 95% CI 1.53-6.00, p=0.003).

Conclusions

This is the first study to show a significant association between high NLR or PLR values, as measured during CDK 4/6 inhibitor treatment, and lower PFS in HR+ aBC pts. Our results suggest that the NLR and PLR could be used as precocious biomarkers of treatment efficacy. A multicenter observational study to confirm these data in a larger cohort of pts is ongoing.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

G.V. Bianchi: Advisory/Consultancy: Novartis; Advisory/Consultancy: Eli Lilly. F.G.M. De Braud: Advisory/Consultancy: Roche; Advisory/Consultancy: EMD Serono; Advisory/Consultancy, principal investigator: Novartis; Advisory/Consultancy, principal investigator: Tesaro; Advisory/Consultancy, principal investigator: Loxo Oncology; Advisory/Consultancy, principal investigator: Celgene; Advisory/Consultancy, principal investigator: Nektar; Advisory/Consultancy, principal investigator: BMS. All other authors have declared no conflicts of interest.

Background

Triple negative breast cancer (TNBC) is a subtype of breast cancer with poor survival. Tumor-Infiltrating Lymphocytes (TILs) have been identified as a potential biomarker in TNBC. However, some tumors lack of TILs completely. This could be explained by the confinement of lymphocytes in the nearby lymph nodes (LN). This project aims to identify immune checkpoints that could be retaining those lymphocytes. We also analyzed other immune biomarkers in LN and tumor; and performed mutation, neoantigen analysis and gene expression profiling in tumor.

Methods

We identified 102 TNBC patients with localized tumors and no metastasis (T1c-T2N0M0), that did not receive neoadjuvant treatment, and had tumor and LN available in paraffin blocks. We scored TILs as indicated in the latest published guidelines (2017), and included only patients with >=50% (high TILs) or =<5% (low TILs). Total n was 35 patients, 15 high and 20 low. We measured CTLA-4, PD-1, PD-L1, PD-L2 and OX-40 by immunohistochemistry, and the expression of 50 immune genes by the nCounter platform in tumor and LN. We performed Whole Exome Sequencing (WES) and the Breast Cancer 360TM (BC360) panel (770 genes) in tumor.

Results

CTLA-4 is significantly more expressed in LN of low TIL (p=0.01, median 4% vs. 7%), PD-L1 in tumor cells of high TILs, PD-L2 in germinal centers of high TILs, and OX-40 in tumor, LN and GC of low TILs. Gene expression panel showed that CD274, CD273 and VEGF are significantly more expressed in low TILs tumors. In LN, only CD68 and CD8 were significantly more expressed in high TILs. Despite non-significant median mutation load (p=0.76) by WES analysis, we found significant neoantigen load (p=0.027) in low TILs. BC360 panel showed significant expression of stromal signatures in low TILs.

Conclusions

Low TIL tumors demonstrate higher neoantigen load, but incapable of recruiting T cells at the tumor. On the contrary, high TILs, with less neoantigens, have more antigen presentation (CD68) and T cell proliferation in LN. These suggest another mechanism of avoiding immune infiltration. We observed higher CTLA-4 expression in LN of low TILs, indicating a potential first brake for lymphocyte migration to the tumor. PD-L1 could be a secondary brake at the tumor site.

Legal entity responsible for the study

The authors.

Funding

Pompadour through Intheos Foundation.

Disclosure

A. Quintana: Travel/Accommodation/Expenses: Merck Sharp & Dohme. V. Peg: Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche; Advisory/Consultancy: AstraZeneca; Advisory/Consultancy: MSD; Advisory/Consultancy: Genomic Health; Honoraria (institution): Sysmex. A. Prat: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): Novartis; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (self): Pfizer; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): Eli Lilly; Advisory/Consultancy, Speaker Bureau/Expert testimony: NanoString Technologies; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): Amgen; Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (self): Roche; Honoraria (self), Advisory/Consultancy: Oncolytics Biotech; Honoraria (self), Leadership role: Daiichi Sankyo; Advisory/Consultancy, Speaker Bureau/Expert testimony: PUMA; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: BMS; Leadership role, Research grant/Funding (institution): Boehringer Ingelheim. G. Villacampa: Speaker Bureau/Expert testimony: MSD. R. Dientsmann: Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche; Speaker Bureau/Expert testimony: Ipsen; Speaker Bureau/Expert testimony: Amgen; Speaker Bureau/Expert testimony: Sanofi; Speaker Bureau/Expert testimony: Servier Laboratories; Speaker Bureau/Expert testimony: MSD; Advisory/Consultancy: Roche; Advisory/Consultancy: Boehringer Ingelheim; Research grant/Funding (institution): Merck; Research grant/Funding (institution): Pierre Fabre. J. Perez: Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy: Eli Lilly. E. Muñoz: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Novartis; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: BMS; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: MSD; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Pier Fabre; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Sanofi. L. de Mattos: Speaker Bureau/Expert testimony: Roche; Advisory/Consultancy: NanoString. J. Cortés: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): Roche; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Celgene; Honoraria (self), Speaker Bureau/Expert testimony: Novartis; Advisory/Consultancy: Cellestia; Advisory/Consultancy, Research grant/Funding (institution): AstraZeneca; Advisory/Consultancy: Biothera Pharmaceutical; Advisory/Consultancy: Merus; Advisory/Consultancy: Seattle Genetics; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Daiichi Sankyo; Speaker Bureau/Expert testimony: Erytech; Advisory/Consultancy: Athenex; Advisory/Consultancy: Polyphor; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Lilly; Advisory/Consultancy: Servier; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): MSD; Advisory/Consultancy: GSK; Advisory/Consultancy: Leuko; Advisory/Consultancy: Bioasis; Advisory/Consultancy: Clovis Oncology; Research grant/Funding (institution): Ariad Pharmaceuticals; Research grant/Funding (self): Baxalta; Research grant/Funding (institution): GMBH/Servier Affaires; Research grant/Funding (self): Bayer Healthcare; Research grant/Funding (institution): Eisai; Shareholder/Stockholder/Stock options: MedSIR; Research grant/Funding (institution): F Hoffman-La Roche; Research grant/Funding (institution): Guardant Health; Research grant/Funding (institution): Pfizer; Research grant/Funding (institution): Piqur Therapeutics; Research grant/Funding (institution): Puma; Research grant/Funding (institution): Queen Mary University of London. All other authors have declared no conflicts of interest.

Background

Estrogen receptor-positive, and human epidermal growth factor receptor 2-negative (ER+HER2-) breast cancer account for 60∼70% of all breast cancer patients, whose grade III cases are rare but respond poorly to endocrine therapy. We systematically analyzed clinical and multi-omics data to find a potential avenue for personalizing therapy for them.

Methods

SEER database (n=25,629) and another two Chinese cohorts (WCCCG (n=546); FUSCC (n=348)) were used to assess the association between different subtypes (grade III vs. I/II ER+HER2-) and clinicopathological and survival significance. The remaining three multi-omics cohorts came from TCGA (n=88), METABRIC (n=404) and MSKCC (n=272), and we analyzed multi-omics data to describe the molecular features of grade III ER+HER2- cases.

Results

Grade III ER+HER2- cases harbored higher proportions of large tumor size (>5cm), lymph nodes metastasis, chemotherapy rate and luminal B subtype than I/II cases, where inferior survival outcomes were also observed. We detected increased mutation prevalence of TP53, CSMD3 and TTN in grade III cases with enrichments of mutation signatures linked to DNA repair deficiency. DNA methylation (HM450) data and methylation specific PCR indicated that cg18629132 located in promoter of MKI67 was hypermethylated in grade I/II cases and normal tissue, but hypomethylated in grade III cases, who harbored higher expression of mRNA MKI67. GISTIC2.0 identified 42 and 20 focal copy number variation events in non-metastatic and metastatic grade III cases, respectively, either CDKN1B on 12p12.3 or MDM2 on 12q15 amplification event has an independent prognostic effect on grade III cases. For transcriptional profiling between PAM 50 defined luminal and non-luminal grade III cases, the differential expressions of mRNAs were enriched in IL-17 and estrogen signaling pathways. Recursive partitioning analysis (RPA) used to construct a decision tree with two genes (non-luminal: GATA3+/GATA3- and AGR+), where this classifier was validated in our IHC-based WCCCG cohort.

Conclusions

Grade III ER+HER2- tumors have distinct clinical and molecular characteristics compared to grade I/II tumors, particularly with respect to non-luminal subgroup, and we should tailor and escalate therapies for them.

Legal entity responsible for the study

Kang Wang.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

In Taiwan, breast cancer has been rapidly jumped to the first place in the incidence of women-specific malignancies. It will be urgent to discover some critical biomarkers predicting patient prognosis and outcome in Taiwan. Because Taiwanese breast cancer has a relatively lower incidence and earlier onset than Western population, the subset of clinical biomarkers and treatments applied in Western population may not be suitable for all Taiwanese population.

Methods

Recently microRNAs (miRNAs) were found to be frequently deregulated in breast cancer, and some specific miRNAs were found to be associated with the DNA damage response (DDR). We measured miR-133a, both circulating and intra-tumor levels, to identify the relationship among miR-133a, HER2, and sruvival. We also detected the protein level of Mre11, one of double-strand break repair protein complex, to fingure out the correlation with miR-133a.

Results

Our preliminary results showed that circulating miR-133a levels of breast cancer patients are significantly higher than those of health subjects by using the receiver operating characteristic (ROC) curve (p=0.001). Both circulating and intra-tumor miR-133a levels were inversely correlated with the HER2 status (p=0.015 and 0.011), suggesting that Her2 modulate the expression of miR-133a. The Kaplen-Meier survival curve further showed that high circulating miR-133a can increase the overall survival rate of patients who didn’t receive chemotherapy, implicating that miR-133a might also modulate the response of chemotherapy, which has been linking to DNA repair ability. We found that ectopic miR-133a overexpression suppressed the protein levels of Mre11 but did not impact its mRNA levels.

Conclusions

In this project, we aim to disclose the interplay of miR-133a with Her2 and Mre11 in breast cancer development. Furthermore, we will establish the minimally invasive screen platform to improve early detection, prognosis, and follow up of breast cancer. We believe that the accomplishment of this project will not only provide better understanding of the carcinogenesis of breast cancer but also a prescreening tool to facilitate decisions about which individuals to be recommended for further diagnostic tests or treatments.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The aim of this study is to report the particularities of breast cancer molecular subtypes and their correlations with the clinicopathological characteristics in a large cohort of South Tunisia.

Methods

We analyzed 617 breast cancer cases collected at Habib Bourgiba Hospital of Sfax between 2016 and 2019. Clinicopathological features were studied. Molecular subtypes were determined based on four parameters: human epidermal growth factor receptor 2 (HER2), proliferation index (Ki 67), estrogen and progesterone receptors. Five subgroups were reported: luminal A, luminal B HER2 positive, luminal B HER2 negative, triple negative and HER positive. Chi-squared test was performed to evaluate the correlation between pathology and molecular subtype classifications.

Results

Hormone receptor (ER and or RP) were positive in 80.8% and 35.3% of cases were HER positive. Luminal B HER2 negative was the most frequent molecular subtype (30%) followed by luminal B HER2 positive (27.4%), luminal A (12.9%), triple negative (11.6%) and HER positive (7.3%). Seventy eight percent of T4 stage was in luminal B (p=0.03). Histological grade was lower with Luminal A (20.8 %, p < 0.001) with more negative lymph nodes status (49.4%, p=0.01).Triple negative were significantly associated with high histological grade (80%, p <0.001) and tended to significance for HER2 positive (66%, p=0.07).

Conclusions

Our data demonstrated that the luminal B subtype was the most frequent subtype in South Tunisian population; however, the proportion of luminal A subtype was less than reported in other studies. Moreover, HER2 positive breast cancer subtype occurred at a high incidence. According to this molecular profile, we suggest that breast cancer in our region seems to be aggressive tumor that needs more systematic treatment intensification.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Risk stratification by genomic signatures has been shown to improve prognostication and guide treatment decisions among patients with hormone-receptor positive tumors. However, their role in young women breast cancer (YWBC) has witnessed some controversy.

Methods

A systematic search was performed in the MEDLINE, EMBASE and CENTRAL databases for studies that evaluated the use of commercially available genomic signatures OncotypeDx, Mammaprint, Endopredict, Breast Cancer Index, Genomic Grade Index and Prosigna in YWBC (i.e. patients aged ≤40 years at diagnosis). Eligible studies were those that included YWBC and disclosed the number of patients per risk category. The Fisher’s test for independence was used to assess differences between age groups.

Results

Out of 752,935 women that underwent genomic testing, the minority (3.7%) were YWBC. 742,671 were tested with OncotypeDx, 10,053 with MammaPrint and 211 with Endopredict. Analysis of this age-group was not available for the other tests. Compared to older patients, YWBC were more likely to be subjected to genomic testing (33% vs 29%, p=0.02) and had a higher proportion of intermediate- to high-risk tumors when classified by OncotypeDx (61% vs 50%, p<0.01), MammaPrint (65% vs 38%, p<0.01), and Endopredict (68% vs 49%, p=0.06). Only three studies specifically exploring the prognostic value of genomic tests in YWBC were found, all using OncotypeDx. In patients with genomic low-risk, 6-year distant recurrence-free survival was 92%, while 5-year overall survival and breast cancer specific survival were nearly 100%. Nonetheless, YWBC were more likely to receive chemotherapy than older patients when classified as low- (24 vs 3%, p<0.01) or intermediate- (57 vs 32%, p<0.01) risk.

Conclusions

Only a small proportion of YWBC were included in genomic signature studies, with approximately one third classified as low-risk. Although the prognostic value of genomic tests for young women is currently available only for OncotypeDx, data support that patients with low genomic-risk have an excellent prognosis. Hence, genomic tests could be a useful tool for identifying young patients in which chemotherapy omission is appropriate.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

M. Lambertini: Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche; Honoraria (self), Speaker Bureau/Expert testimony: Takeda; Honoraria (self), Speaker Bureau/Expert testimony: Theramex. H.A. Azim Jr: Advisory/Consultancy: NanoString; Full/Part-time employment: Innate Pharma. C. Villarreal-Garza: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy: Novartis; Advisory/Consultancy, Travel/Accommodation/Expenses: Pfizer; Advisory/Consultancy: AstraZeneca; Speaker Bureau/Expert testimony: Myriad Genetics; Travel/Accommodation/Expenses: MSD. All other authors have declared no conflicts of interest.

Background

Ductal carcinoma in situ (DCIS) of the breast is a heterogeneous disease in terms of morphology and genetics. Tumor architecture and cellular morphology classify DCIS into subgroups to guide clinical management. This leaves a large portion of histopathological features (HPF) uncharacterized. This study explored the extent of heterogeneity of nuclear atypia, DCIS architecture, necrosis, calcifications, stromal architecture and stromal inflammation between biopsies and their subsequent resection specimen in DCIS.

Methods

The percentage of ducts containing a particular HPF was assessed on the biopsy slide and 3 slides of the resection specimen. A histoscore was determined to allow assessment of histopathological heterogeneity. For instance, the histoscore for nuclear grade was calculated as follows: (% grade 1 nuclei) + (% grade 2 nuclei)*2 + (% grade 3 nuclei)*3, with a score ranging from 100-300. Heterogeneity was arbitrarily defined as a mean histoscore with a standard deviation > 10% of the maximum histoscore. Statistical analysis comprised the Friedman test, post hoc Wilcoxon tests with Bonferroni corrections and Spearman’s correlation tests.

Results

Fifty-one DCIS biopsies were correlated with their subsequent resection specimen: 24 cases of 51 (47%) showed heterogeneity for nuclear atypia, 25/51 (49%) for tumoral architecture, 23/51 (45%) for calcifications, 29/51 (57%) for necrosis, 22 /51 (43%) for myxoid stromal architecture, and 21/51 (41%) for stromal inflammation. Heterogeneity was not associated with patient age, DCIS size or type of surgery except for one weak association (p=0,048) between heterogeneity in stromal inflammation and DCIS size. No relationship between heterogeneity in nuclear atypia and heterogeneity in other HPF was found. All HPF in the biopsy correlated significantly with the HPF in the resection specimen, except for necrosis (p=0,004).

Conclusions

DCIS lesions present a heterogeneous histoscore for each HPF (41%-57%). All HPF determined in the biopsy correlated well with the surgical specimen, except for necrosis. We therefore conclude that overall morphological heterogeneity in DCIS has only a limited impact, and that biopsies of pure DCIS are generally representative for the entire DCIS lesion.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Circulating tumor cells (CTC) are associated with clinical outcome in metastatic breast cancer (MBC). We explored the ability of early CTC monitoring (after 2 cycles) for predicting clinical benefit (CB), overall response rate (ORR) and outcome in terms of progression-free survival (PFS) and overall survival (OS) in patients (pts) with HER-2 negative MBC receiving first-line treatment with bevacizumab and paclitaxel.

Methods

Multicenter prospective observational study. Centralized CTC determination was performed at baseline (C0) and after cycle 2 (C2) using CellSearch. A cutoff of 5 CTCs/7.5ml was used to stratify pts into treatment-sensitive (<5 CTCs) and resistant (≥5 CTCs). The optimal CTC level cutoff for predicting CB and PFS was assessed using ROC curves.

Results

111 pts were enrolled: median age: 54 years; ECOG 0/1: 50.5/39.6%; triple-negative: 29%; metastatic sites (median): 3; metastases location: liver (62%), bone (58%), lung (40%). The clinical benefit rate (CBR), ORR, and median PFS for the whole cohort were 65%, 41%, and 16.6 months, respectively. With a median follow-up of 17.5 months, the median OS was not achieved. At C0, 43/87 (49%) and 44/87 (50.6%) pts presented with <5 and ≥5 CTCs, respectively. The CTC level after C2 was in 73/85 (86%) and 12/85 (14%) pts <5 and ≥5, respectively. Among pts with CTCs ≥5 at C0, 78% had <5 CTCs after C2. The CTC level after C2 was predictive for CBR (73% vs 59%, p=0,046), ORR (48% vs 17%, p=0.043), and PFS (17 vs. 5 months, p=0.026). Median OS was 13 months in pts with ≥5 CTCs after C2 and it was not achieved in pts with <5 CTCs (p <0.001). A cutoff point of 1 CTC after C2 yielded 65% sensitivity and 61% specificity for prediction of PFS (p=0.021). Pts with 0 CTC achieved a significantly longer PFS vs. those with at least 1 CTC after 2 cycles (22.6 vs. 8.1 months; p=0.005). Grade 3-4 toxicity rate (39%) was not significantly different according to CTCs after 2 cycles (p =0.531).

Conclusions

CTC monitoring after 2 cycles of first-line bevacizumab-paclitaxel treatment may predict tumor response and clinical outcome in terms of PFS and OS in HER-2 negative MBC.

Editorial acknowledgement

Cristina Vidal, from the CRO (Contract Research Organization) Dynamic Science (Spain) has provided editorial and medical writing support funded by Roche Farma S.A.

Legal entity responsible for the study

Roche Farma S.A.

Funding

Roche Farma S.A.

Disclosure

I. Álvarez-López: Advisory/Consultancy: Roche, Pfizer, Novartis, and Genomic Health; Speaker Bureau/Expert testimony: Pfizer, Novartis, and Roche; Research grant/Funding (institution): Pfizer, Roche, and AstraZeneca; Travel/Accommodation/Expenses: Pfizer, Roche, and Eisai. F. Ayala de la Pena: Speaker Bureau/Expert testimony: AstraZeneca, Celgene, Eisai, Novartis, Roche, Pfizer, Pierre Fabre; Research grant/Funding (institution): Celgene, Roche; Travel/Accommodation/Expenses: Roche, Pfizer, Celgene, Eisai, Pierre Fabre, and MSD. A. Perelló: Advisory/Consultancy: Roche, Pfizer, Novartis, MSD, and Eisai; Speaker Bureau/Expert testimony: Pfizer, Novartis, and Roche; Travel/Accommodation/Expenses: Roche, Novartis. A. Llombart Cussac: Advisory/Consultancy: Lilly, Roche. Pfizer, Novartis, Pierre-Fabre, Genomic Health, and GSK; Speaker Bureau/Expert testimony: Lilly, AstraZeneca, MSD; Leadership role: Eisai, Celgene, Lilly, Pfizer, Novartis, Roche, and MSD; Research grant/Funding (institution): Roche, Foundation Medicine, Pierre-Fabre, Agendia; Travel/Accommodation/Expenses: Roche, Lilly, Novartis, Pfizer, AstraZeneca; Shareholder/Stockholder/Stock options: MedSIR; Initia-Research. All other authors have declared no conflicts of interest.

Background

The next generation sequencing technology has the advantages of high speed, high throughput and high accuracy. Because of these advantages, it is used in various cancer fields. Several gene pannels have been applied to breast cancer to assess risk and determine treatment direction accordingly. The purpose of this study was to improve the prognosis and future treatment of patients with breast cancer by applying NGS.

Methods

From January 2018 to December 2018, we studied patients who underwent surgery at Kosin University Gospel Hospital. The study patients were from stage 1 to stage 3 of breast cancer. Patients who were not able to undergo surgery or who had more than stage 4 patients were excluded. This study included patients who underwent Neo-systemic therapy(NST). NGS was performed postoperatively. And in patients who underwent NST, NGS proceeded to pre-chemotherapy specimens.

Results

The expression of somatic mutations varies with the type of breast cancer. In all type of breast cancers, TP53 was 26%, PIK3CA was 24%, and ERBB2 was 8%. In luminal type, PIK3CA was 32%, GATA 13% and TP53 12%. In the HER2 type, TP53 was 35%, ERBB2 was 23%, and PIK3CA was 17%. TNBC had 44% TP53, 13% PIK3CA, and 4% GAS6. In most cases, two or more mutations were combined. Table: 40P

Conclusions

Different types of somatic mutations have been observed, depending on the subtype of breast cancer. Each type exhibited a different mutation distribution. Some treatments for mutations are being developed, and we expect to be effectively applied through NGS. The limitations of this study did not determine how these various mutations correlate with the prognosis of breast cancer. Long term follow up is needed in the future.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

A previous phase 2 study in metastatic breast cancer compared treatment with intravenously delivered oncolytic reovirus, pelareorep (pela), in combination with paclitaxel (PTX) versus PTX alone. This study demonstrated a statistically significant improvement in overall survival (OS), without differences in objective response or progression-free survival. We hypothesized that the OS benefit from pela + PTX may be attributed to an adaptive immune response triggered by pela. To test this hypothesis, and examine if pela can mediate the priming of an anti-tumor immune response, we designed a study called AWARE-1 (A window-of-opportunity study of pela in Early Breast Cancer), which is currently enrolling and for which initial translational research results are presented.

Methods

AWARE-1 is evaluating the safety and effect of pela ± atezolizumab on the tumor microenvironment (TME) in 38 women with early breast cancer. Patients are treated with pela on days 1, 2, 8, and 9, while atezolizumab is administered on day 3. Tumor biopsies are collected at diagnosis, day 3, and day ∼21. The primary endpoint of the study is CelTIL score, a metric for quantifying the changes in tumor cellularity and infiltration of TILs, where an increase in CelTIL is associated with a favorable response to treatment. Tumor tissue was examined for pela replication, and changes to the TME were assessed by immunohistochemistry and TCR-seq. Peripheral blood was also examined by TCR-seq.

Results

Analysis of CelTIL show an increase in four of the six patients to date. Productive viral replication in day 3 and day ∼21 biopsies was very high, as measured by in situ detection of viral capsid protein in tumor cells. Immunohistochemistry analysis revealed an increase in CD8+ T cells and upregulation of PDL1 on day 3 and day 21 biopsies for all patients. TCR-seq from blood showed that levels of T cell clonality correlate with changes in the TME and CelTIL.

Conclusions

Overall, the degree of viral replication was consistent with changes in CelTIL and changes within the TME. Preliminary data from the first six patients in AWARE-1 demonstrate pela-mediated priming of an adaptive immune response.

Clinical trial identification

NCT04102618.

Legal entity responsible for the study

Oncolytics Biotech Inc.

Funding

Oncolytics Biotech Inc.

Disclosure

L. Manso: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Astra; Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Novartis; Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Tesaro; Advisory/Consultancy, Speaker Bureau/Expert testimony: Pfizer. P. Villagrasa: Speaker Bureau/Expert testimony: NanoString. Y. Izarzugaza: Honoraria (self), Research grant/Funding (institution), Non-remunerated activity/ies: Novartis; Research grant/Funding (institution): Breast Cancer Research Foundation-AACR Career Development Award; Research grant/Funding (institution): Breast Cancer Now Catalyst Grant; Honoraria (self), Non-remunerated activity/ies: Roche; Non-remunerated activity/ies: Pfizer; Honoraria (self): AstraZeneca; Honoraria (self): Eisai. M. Juan: Advisory/Consultancy: Grifols. G. Wilkinson: Shareholder/Stockholder/Stock options, Licensing/Royalties, Full/Part-time employment: Oncolytics Biotech Inc. M. Coffey: Honoraria (self), Licensing/Royalties, Full/Part-time employment, Non-remunerated activity/ies: Oncolytics Biotech Inc. X. González-Farré: Honoraria (self): SOLTI; Honoraria (self), Non-remunerated activity/ies: Roche; Honoraria (self): Eisai. A. Prat: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): Pfizer; Honoraria (self), Advisory/Consultancy: Lilly; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: NanoString; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): Amgen; Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (institution): Roche; Honoraria (self), Advisory/Consultancy: Oncolytics Biotech; Honoraria (self), Speaker Bureau/Expert testimony: Daiichi Sankyo; Honoraria (self), Advisory/Consultancy: Puma; Honoraria (self), Advisory/Consultancy: BMS. J. Gavila Gregori: Honoraria (self), Research grant/Funding (institution): Novartis; Honoraria (self), Research grant/Funding (institution): Pfizer; Honoraria (self), Research grant/Funding (institution): Roche. All other authors have declared no conflicts of interest.

Background

There are three main cascades - WNT, Hedgehog and NOTCH, occur in cancer stem cells during the cancerogenesis. However, the conducted studies indicate the existence of other signaling mechanisms of regulation - NF-kB and PI3K signaling pathways. In our work, we investigated the expression of NF-kB, PI3K, PTEN molecules, as well as WNT, Hedgehog, NOTCH in cells of triple negative and HER-2 overexpressed breast cancer with high and low content of cancer stem cells (positive and negative ALDH1A1 expression).

Methods

We studied a material of 110 cases of invasive breast cancer. To determine the stem cells in the tumor population, the presence of ALDH1A1 protein in cancer cells was investigated. In all cases, expression of estrogen, progesterone receptors, as well as expression of HER-2 and Ki-67 protein was studied by immunohistochemistry to determine a subtype of breast cancer. The expression of signaling pathways molecules PI3K, NF-kB, PTEN, WNT, Notch, Hedgehog was also determined by immunohistochemical method.

Results

The results are shown in the table: Table: 42P

Groups with high/low expression of ALDH1A1

Conclusions

It was found, that cancer cells of all cases of triple negative and HER-2 overexpressed subtypes expressed NF-kB, PI3K signaling pathways molecules. Activation of Notch and WNT signaling pathways was more common for cells of HER-2 overexpressed subtype than Triple-negative subtype (p<0.05) in the group with high level of ALDH1A1, while in the group with low level of ALDH1A1, expression of NOTCH and WNT in cancer cells of both subtypes was not significantly different (p>0.05). We did not found any activity of Hedgehog (HH) signaling pathway in the group with high level of ALDH1A1, while in the group of low level of ALDH1A1 expression of HH signaling was positive in some cases, and besides, it was higher in cancer cells of Triple negative subtype than HER-2 overexpressed subtype (p<0.05).

Legal entity responsible for the study

Ministry of Healthcare of Sverdlovsk Region of Russian Federation.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Programmed death-liangd-1(PDL1) is a molecule involved in immune evasion in breast cancer. In the phase III clinical trial of impassion 130, the PDL1 expression is a predictive biomarker for the metastatic triple-negative breast cancer therapy. To determine PD1/PDL1 expression in early stage of triple-negative breast cancer, and to analyze the relationship between their expression and prognosis.

Methods

Immunohistochemistry (IHC) was performed on paraffin-embedded tumor samples. Logistic regression was used to analyze the associations between PDL1 protein expression and long-term prognosis. Kaplan-Meier plot and log-rank test were used to compare disease-free survival (DFS) between groups. A cox proportional hazards model was used to calculate the adjusted hazard ratio (HR) with 95% confidential interval (95%CI).

Results

205 triple-negative patients were enrolled in this study from 1 June 2009 to 31 Oct 2015. Patients had a representative tumor specimen (formalin-fixed, paraffin-embedded archival) for testing of PDL1 expression. We collected the clinicopathological characteristics of the patients. The median follow-up time was 66.9 months. The 5-year DFS rate was 86.1% (95% CI 81.4%-90.8%) and the 5-years OS rate was 93.6% (95% CI 91.0%-97.6%). In the univariate analysis, we found that lymph nodes, Ki67 index and PDL1 expression were associated with DFS and OS; however, in the multivariate analysis, patients with PDL1 expression showed significantly more favorable prognosis in DFS (HR 2.875, 95%CI 1.216-6.796, p=0.016) and improve the OS compared with the PDL1 negative group (HR 3.157, 95%CI 0.844-11.809, p=0.088).

Conclusions

PDL1 protein expression is a predictive biomarker of good prognostic factor for survival in triple-negative breast cancer patients.

Legal entity responsible for the study

The authors.

Funding

NASF.

Disclosure

All authors have declared no conflicts of interest.

Background

Approximately one third of breast cancer (BC) patients are obese, a fact that has been associated with an increased frequency of aggressive clinicopathological features and worse disease outcomes. The aim of this study was to evaluate if obesity plays a selective role in disease stage according to molecular subtype.

Methods

Medical records of women diagnosed with BC between January 2013 and December 2015 in a center located in Monterrey, Mexico were retrospectively reviewed. The patients were grouped by body mass index (BMI; obese: ≥30 kg/m² and non-obese: <30 kg/m²) and clinical stage (early-stage: 0-II and late-stage: III-IV). Associations between the variables were examined using Fisher's exact test of independence. The significance level was set at p < 0.05.

Results

A total of 821 patients were diagnosed with BC, of which 363 were excluded since information about disease stage was missing. Median age was 51 years (range: 29-88), with 14% classified as young. The predominant molecular subtype was luminal-A (55%), followed by triple-negative (23%), and luminal-B and HER2+ (11% each). Most patients presented with early-stage disease (56%). Overall, obesity was not associated with disease stage. However, when stratified by molecular subtype, obesity was associated with diagnosis at a late-stage among women with triple-negative disease (55% vs 35%, p=0.0495). None of the other molecular subtypes demonstrated an association between stage and BMI.

Conclusions

In this cohort, obesity was only associated with a late-stage diagnosis in women with triple-negative disease. Although obesity and related comorbidities have been associated with unfavorable characteristics in triple-negative tumors, the molecular factors involved have not been elucidated. Current evidence indicates that increased availability of steroid hormones, insulin-like growth factors, adipokines and inflammatory cytokines could all play an important role in the aggressiveness of these tumors. Major limitations are present in this study, mainly its retrospective nature, limited number of patients and lack of information regarding clinical outcomes. Future research is needed to confirm these findings and analyze their clinical relevance.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

H. Diaz-Perez: Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Pfizer; Travel/Accommodation/Expenses: Roche; Travel/Accommodation/Expenses: Novartis; Travel/Accommodation/Expenses: Lilly; Travel/Accommodation/Expenses: Amgen; Travel/Accommodation/Expenses: Asofarma. C. Villarreal-Garza: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche Mexico; Advisory/Consultancy: Novartis; Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Pfizer; Advisory/Consultancy: AstraZeneca; Speaker Bureau/Expert testimony: Myriad Genetics; Advisory/Consultancy: Lilly. All other authors have declared no conflicts of interest.

Background

It is well established that the regulation of pre-mRNA splicing is highly tissue specific. However, whether breast cancer-predisposing splicing defects observed in blood cells also occur in other cell types, namely in stem cells from which tumors are likely to originate, remains to be established. The aim of this study was to characterize BRCA1/2 mRNAs expressed in different primary cell types derived from normal individuals and heterozygous carriers of the pathogenic Portuguese BRCA2 c.156_157insAlu founder mutation that leads to an in-frame skipping of exon 3.

Methods

Upon informed consent, we collected and cultured primary human fibroblasts and isolated peripheral blood mononuclear cells (PBMCs) from whole blood of 6 donors with and without the Portuguese founder mutation. For understanding BRCA1/2 transcriptomic differences between proliferating and terminally differentiated cells, we used a sendai virus-based approach to reprogram human fibroblasts into induced pluripotent stem cells (iPSCs). RNA was extracted and analyzed using a nanoliter-sized droplet technology paired with digital PCR (ddPCR) that allows for high-throughput, absolute nucleic acid quantitation and detection of alternative mRNA processing events.

Results

We found mRNA isoforms generated by alternative splicing of normal BRCA2 transcripts, including skipping of exon 3, in all cell types analyzed. However, the proportion of exon 3 skipping was higher in cells carrying the c.156_157insAlu mutation. Remarkably, we detected significantly higher levels of BRCA2 transcripts in iPSCs compared to PBMCs and fibroblasts, in both normal and mutant cells.

Conclusions

Our results indicate that the Portuguese BRCA2 founder mutation induces quantitative rather than qualitative differences in mRNA splicing. Moreover, we observed upregulation of BRCA1/2 gene expression upon reprogramming of fibroblasts to iPSCS. This result is consistent with previous findings suggesting that BRCA genes are more expressed in stem cells compared to differentiated cells.

Legal entity responsible for the study

Carmo-Fonseca Lab at Instituto de Medicina Molecular João Lobo Antunes.

Funding

Fundação para a Ciência e a Tecnologia and FEDER through POR Lisboa 2020 - Programa Operacional Regional de Lisboa PORTUGAL 2020 (LISBOA-01-0145-FEDER-029469).

Disclosure

The author has declared no conflicts of interest.

Background

The presence of high levels of tumor infiltrating lymphocytes (TILs) has been associated with better prognosis in early triple-negative breast cancer (TNBC). The Immunoscore® (IS) is a prognostic tool, which categorizes the densities of spatially positioned CD3 and CD8 cells in both invasive margins (IM) and the center of the tumor (CT), yielding a five-tiered classification (0–4). High IS values have been reported to predict improved outcomes in colorectal cancer patients (pts).

Methods

We performed the IS in 103 breast cancer (BC) pts who previously received neo-adjuvant anthracycline and taxane +/- trastuzumab based chemotherapy TNBC=53, Luminal=32, HER2+=18. Pre-treatment tumor samples were immune-stained for CD3 and CD8 T-cell markers. Quantitative analysis of the immune cells was carried out using a computer-assisted image analysis in different tumor locations.

Results

The pathological complete response (pCR) rate of the entire cohort was 44%. On univariate analysis, factors associated with higher pCR included primary tumor size (p<0.005), nodal status (p<0.069), ER (p<0.000), PR (p<0.000), molecular subtype (TNBC=62%, HER2+=50% and Luminal A+B=9%, p<0.000), Ki67 (>40=56% vs 15-39=40% vs <15=0%, p<0.000) and Stage (p<0.028). A high density of CD3 (> than 800mm²) and CD8 (> than 400mm²) positive T-cells in the CT was associated with higher pCR (CD3 CT: 60% vs 25%, p=0.000 and CD8 CT: 64% vs 27%, p=0.000). Analysis of CD3 (> than 1400mm²⁾ (CD3 IM: 63% vs 19%, p=0.00) and CD8 in the IM (> than 500mm²) was also significant for an association with pCR (CD8 IM:63% vs 15%, p=0.000). High IS (3+4= 63%) vs intermediate (2=35%) vs low (0+1=24%) was significantly associated with pCR (p=0.006). In a logistic regression model Ki-67 (p<0.005) and IS (p<0.021) and molecular subtype (p<0.010) retained significance. DFS: At 3 years 94% of IS high pts did not relapse compared to 80% IS intermediate or low pts (p<0.07).

Conclusions

This study shows a significant prognostic and potentially predictive role for the IS in BC pts, particularly in TNBC. This study shows a significant prognostic and potentially predictive role for the IS in BC pts, particularly in TNBC.

Legal entity responsible for the study

The Rosebank Clinical and Translational Research Unit.

Funding

The Rosebank Clinical and Translational Research Unit.

Disclosure

J. Galon: Shareholder/Stockholder/Stock options: HalioDx. A. Fugon: Full/Part-time employment: HalioDx. M. Martel: Full/Part-time employment: HalioDx. All other authors have declared no conflicts of interest.

Background

Triple negative breast cancer (TNBC) is characterized by aggressive behaviour, with high morbimortality. Androgen Receptors (AR) have a role in the activation of cellular proliferation, migration and invasiveness. The AR are expressed in 7-75% of TNBC. This study evaluates the relationship between positive AR with clinical data, relapse and overall survival.

Methods

All TNBC cases from the Pathology Service of Instituto Nacional del Cáncer were studied between 2015 to 2019 (n=69). Eligible patients (n=32) had at least 18 months of follow-up, ending at the 31/01/2020 or death. From clinical files and biopsy reports, age, TNM, treatment, relapse and Ki67 were extracted. Date and cause of death were obtained from death certificates. AR status was determined by immunohistochemistry and defined as AR-positive (>1%) or AR-negative (0%). The Kaplan-Meier method was used to estimate survival rates. Crude and adjusted Hazards Ratios (HR) were estimated by proportional Cox regression model.

Results

At diagnosis the average age was 54.6 years (DE:14.7), tumor size 3.9 cm (DE:2.2), 59.4% had positive lymph nodes, positive skin 18.8% and 87.5% without metastasis. Stage I - II 53.1%, Stage III - IV 46.9%. 37.5% received lumpectomy, 50% mastectomy. Neoadjuvant chemotherapy 44%, 59.4% adjuvant or palliative, and 50% radiotherapy. Average follow-up was 38.9 months (DE:15.7). 50% relapsed in an average of 14.9 months (DE:10.9) and 28% died (months to death 32.1 (DE:7.9)). 56.3% had AR+. There was no difference between AR+ and AR- regarding size nor Ki67 expression. There was an increased rate of death with AR+ (33.3% vs 21.4%) and metastasis (16.7% vs 7.1%), but that was not significant. In adjusted Cox regression models, AR+ showed higher risk of relapse (HR:3.54 IC95%:0.65,19.3, p=0.144) and mortality (HR:3.85 IC95%:0.82,18.1, p=0.088).

Conclusions

This study confirmed a high mortality in TNBC. Expression of AR-positive appears to be correlated with increased recurrence and mortality. AR-positive could be a prognostic and predictive marker. TNBC AR-positive patients could potentially apply to anti-androgen targeted therapy. More prospective studies with larger samples size and longer follow-up are required.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The sensitivity to endocrine therapy (SET2,3) index measures transcriptional activity of genes related to the estrogen and progesterone receptors (ER, PR) adjusted for baseline prognostic index (cT, cN, RNA4). Higher SET2,3 predicts greater intrinsic tumoral sensitivity to endocrine therapy (ET). SET2,3 was prognostic in patients receiving chemotherapy (CT) and ET. In this study, we measured SET2,3 performed using a new clinical assay platform, QGP, in a prospective registry study to compare SET2,3 with disease-free survival (DFS).

Methods

This was a single institution, prospective registry study of women ≥ 18 years newly diagnosed from 2011-2016 with Stage I-III hormone-receptor positive (HR+), Her2-negative invasive breast cancer treated with adjuvant ET with or without neoadjuvant or adjuvant CT. SET2,3 is continuous and has a pre-defined cutpoint that defined high vs low score in the setting of chemo-endocrine treatment. Kaplan-Meier method was used to estimate the DFS distributions. Log-rank test and multivariable Cox proportional hazards models were performed to associate DFS with SET2,3 and other clinical factors.

Results

The population included 278 subjects with median age 56 (25-84) and follow-up 70 months (65-72). Most had clinical stage T2 (57.6%), N0 (65.8%). SET2,3 was performed on the clinical QGP platform in 256 samples with 59.4% high and 40.6% low score. 5-year DFS in the overall cohort was 90.8% (95% CI 87.3-94.5). In the high SET2,3 group, 5-year DFS was 95.3% (95% CI 92.0-99.1) and in the low SET2,3 group, 85.7% (95% CI 79.0-92.9). SET2,3 was significantly asscociated with DFS in the overall cohort (p=0.0005) and in patients receiving both CT and ET (p=0.013). There was a trend toward significant association of SET2,3 and DFS (p=0.079) in a subset of 96 patients who received ET only. A multivariable model including receipt of CT, neoadjuvant treatment, and SET2,3 demonstrated SET2,3 to be the only independent predictor of DFS (HR 0.51, 95% CI 0.27-0.96, p=0.037).

Conclusions

SET2,3 was independently predictive of DFS in this registry study of HR+/Her2-negative breast cancers using a clinical-grade QGP assay. Future studies incorporating SET2,3 as a prognostic factor in clinical decision-making are indicated.

Legal entity responsible for the study

Fraser Symmans, MD.

Funding

Has not received any funding.

Disclosure

F. Symmans: Licensing/Royalties, Co-inventor on a patent application regarding the SET index: United States Patent and Trademark Office; Advisory/Consultancy, Unpaid scientific advisor: Delphi Diagnostics; Shareholder/Stockholder/Stock options, Co-founder shares of Delphi Diagnostics (that licensed the IP from MDACC): Delphi Diagnostics. All other authors have declared no conflicts of interest.

Background

Genetic counselling and testing are recommended for selected patients with breast cancer (BC). We aimed to asses genetic counselling referral rates in an academic center of patients with BC that fulfilled clinical criteria and/or family history for BRCA status testing.

Methods

In this retrospective study, all patients diagnosed with breast cancer at the Integrated Cancer Care Centre of Charles-Le Moyne Hospital between January 1^st and December 31^st 2017 were included. The clinical characteristics and family history from the patient’s charts were reviewed in order to determine if genetic counselling was indicated according to the NCCN Version 2.2015 of “Clinical practice guidelines in oncology. Genetic/Familial High-Risk Assessment: breast and ovarian” and to assess if the patients were subsequently referred. The main objective was to determine the proportion of patients with BRCA status testing criteria referred for genetic counselling in an academic center.

Results

266 patients diagnosed with BC were included. 81 (30.4%) patients met one or more criteria for BRCA status testing. Among those, 48 (59.2%) were referred for genetic counselling and 33 (40.8%) were not. Among the patients not referred, 11 (33.3%) met clinical criteria and 22 (66.6%) met family history criteria. The family history criteria most often overlooked was of two or more BC diagnoses in close family (15 patients), followed by BC diagnosis of a relative under the age of 50 (4 patients). The clinical criteria of BC diagnosis equal or under the age of 45 was most often missed (6 patients), followed by triple negative BC at age equal or under 60 years in 2 patients. There were also 25 patients which had genetic testing but who did not appear to meet referral criteria (34% of patient of all patients referred).

Conclusions

A lower than expected genetic counselling referral rate of patients meeting criteria for BRCA status testing was observed. A significant number of patients were also referred without meeting testing criteria. These findings support the need to raise awareness and better inform clinicians treating breast cancer patients in order to improve the rates of genetic counselling and testing.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The mainstay of neoadjuvant chemotherapy for HER2-positive breast cancer (BC) is the combination of highly cytotoxic drugs such as doxorubicin (DOX) and the anti-HER2 therapy with Trastuzumab (TZ), which allows a pathological complete response in up to 50% of patients. Unfortunally, their co-administration is strongly limited by intrinsic cardiotoxicity, therefore only a sequential administration of DOX and TZ is allowed in clinical practice. However, the concurrent use of DOX and TZ has been demonstrated to be more useful both for responder and non-responder patients representing an unmet clinical need in BC oncology. Nanomedicine could fix this issue developing smart drug delivery systems specifically targeted toward BC cells, which display limited off-target toxicity. Here, we propose nanoformulation of DOX in H-Ferritin-nanocages (HFn-DOX), exploiting its capability to increase doxorubicin (DOX) anticancer efficacy while reducing its cardiotoxicity.

Methods

A murine model of HER2+ BC has been treated twice a week for 2 weeks and half with placebo, TZ alone (5 mg/Kg), DOX or HFn-DOX (1 mg/Kg), DOX+TZ and HFn-DOX+TZ. Tumors have been evaluated for size, apoptosis grade, Granzyme release, angiogenesis, Tumor Infiltrating Leucocytes enumeration, amount of Cancer-Associated Fibroblasts, DOX and DOXol content, TZ accumulation and penetration using different approaches such as immunohistochemistry, western blot, flow cytometry, immunofluorescence and mass spectrometry. Cardiotoxicity has been evaluated by morphological analysis of cardiac tissue by transmission electron microscopy. TZ accumulation has been assessed also in heart lysates by western blot.

Results

The coadministration of HFn-DOX with TZ displays increased antitumor potential combined with a cardio protective effect against TZ-induced mitochondrial cardiotoxicity, which is attributable to its capability to reduce TZ accumulation in heart improving in the meantime its penetration in tumour.

Conclusions

HFn-DOX could be a valuable strategy to allow safe co-administration of DOX and TZ exploitable for clinical application.

Legal entity responsible for the study

The authors.

Funding

Regione Lombardia and Fondazione Cariplo (grant number 2016-0919).

Disclosure

All authors have declared no conflicts of interest.

Background

About half of luminal and triple negative breast cancers (TNBC) show low HER2 expressions. These HER2-low breast cancers (BC), with a HER2 IHC score of 1+ or 2+ with negative ISH, are emerging as a new druggable entity, related to the development of novel anti-HER2 antibody-drug conjugates. Little data is available on the evolution of low HER2 expressions between early and metastatic BC (mBC).

Methods

We retrospectively reviewed clinical-pathological data of mBC patients consecutively referred to our New Drugs Division (from Jan2014 to Dec2019), for whom both primary tumor and a metastatic biopsy (performed at any time) were characterized. We divided HER2-negative cases by ASCO/CAP 2018 guidelines into an IHC 0 subgroup and a HER2-low subgroup (1+ and 2+/ISH-negative). χ2-test was used for comparisons between categorical parameters.

Results

267 patients were included in the analysis (64% HR+/HER2-neg, 19% HER2+, 17% TNBC at diagnosis). Among primary tumors, 42% showed low HER2 expression, with a higher rate in the luminal-like BC population compared to TNBC (60% vs 39%, p=0.04). There was a significant enrichment for HER2-low cases in mBC compared with primary lesions (50% vs 42%, p=0.02). Late-relapsers (DFS ≥ median) showed a higher relative increase compared to early relapsers (DFS < median) (+35% vs +0%), with a similar trend in both luminal-like and TNBC. Overall, the increase in HER2 expression between primary and mBC was mainly driven by IHC 0 cases shifting to HER2 low, with a decrease from 39% IHC 0 cases on primary to 32% on metastatic (p=0.12). More details on HER2 expression evolution are provided in the table. Table: 51P

Evolution of HER2 IHC expression on primary vs mBC

Conclusions

Low HER2 expressions are more common among luminal-like cancers than TNBC, and appear enriched in mBC compared with primary tumours, enlarging the cohort of advanced patients eligible for trials of novel anti-HER2 compounds. Validation on larger populations is warranted.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Invasive lobular carcinoma (ILC) is the second most commonly diagnosed histological subtype of breast cancer, comprising up to 15% of all cases. Considerable data suggest there are fundamental clinical and biological differences between ILC and the more commonly diagnosed invasive carcinoma no special type. We sought to characterise clinical, pathological and biological parameters that could predict prognosis in ILC that would aid in the understanding of tumour subtype and in the management of patients diagnosed with this subtype of disease.

Methods

Patient cohorts were assembled from the Royal Brisbane and Women’s Hospital and Wesley Hospital, Brisbane, Australia. Clinical and pathology data were obtained from medical records and re-review of radiology (mammography and ultrasound) and pathology features. Clinical follow up data were collected from the Queensland Cancer Registry. Tissue microarrays were constructed from resected specimens in the form of formalin-fixed, paraffin-embedded tumour tissue. Immunohistochemistry (IHC) was performed for a series of putative novel biomarkers predictive of prognosis that were identified during the development of LobSig, a novel gene signature with prognostic potential in ILC.

Results

Features of known prognostic nature in ILC were evident in this cohort (i.e. tumour size, grade and lymph node status). By mammography, most ILC were classified as localised, spiculated lesions (124/260, 48%), while 30% showed no detectable abnormality. ILC were classified by ultrasound as localised (80%) or diffuse (20%) lesions. ILC not detected by mammography were more likely to be patients <50 years and/or with dense parenchyma in the surrounding breast tissue. In both mammography and ultrasound, localised lesions had a significantly better breast cancer specific survival (BCSS) relative to diffuse lesions (p<0.05). Preliminary IHC data indicate the expression of ARGLU, FBXL3 and RNF135 might be associated with BCSS in ILC (p<0.05, Log rank (Mantel-Cox) test).

Conclusions

We illustrate some novel clinicopathological and biological features that contribute to the understanding of ILC as a tumour entity and may aid in the management of patients in the future.

Legal entity responsible for the study

The University of Queensland.

Funding

Wesley Research Institute; National Health and Medical Research Council, Australia; National Breast Cancer Foundation, Australia.

Disclosure

All authors have declared no conflicts of interest.

Background

Few studies have evaluated the prognostic role of immune biomarkers in HER2 positive Breast cancer (HPBC) patients irrespective of clinical stage. There is paucity of studies evaluating their predictive and prognostic value in advanced stage HPBC subjected to upfront systemic therapy.

Methods

A total 57 advanced stage HPBC patients, who received upfront NACT and antiHER2 therapy were evaluated for immune cell (CD8+ T-cells and FOXP3+ Treg) infiltration (per mm2) in tumor microenvironment (TME) and peritumoral stroma (PTS). The PDL1 expression score (SP263) was evaluated in tumor cells (TPS), immune cells (ICS) and as combined positive score (CPS). PD-L1: TPS, ICS and CPS of > 1% were considered positive. These parameters were correlated with response to systemic therapy and prognosis: responders(R)/non-responders(NR).

Results

Out of 57 patients, 25 were concomitantly hormone positive. Patients were classified as NR (n=24) and R (n=33). There was no significant association of PDL-1 positivity with treatment response (TPS: p=0.489; ICS: p=0.910 and CPS: p=0.977). Increased CD8+ T cell infiltration significantly correlated with treatment response (p=0.013) in TME as well as in PTS (p=0.013). FOXP3+ Treg infiltration did not significantly correlate with treatment response in both TME (p=0.569) and PTS (p=0.412). Ratio of CD8+Tcells/FOXP3+ Treg infiltration significantly correlated with treatment response in both TME (p=0.05) and PTS (p=0.045). In the R category, within the TME, increased CD8+Tcells were statistically associated with positive PDL-1 (TPS:p=0.017; CPS:p=0.015) and similar findings in PTS (TPS:p=0.017; CPS:p=0.015).

Conclusions

We found statistical significant association of treatment response with CD8+Tcells infiltration and ratio of CD8+ Tcells/FOXP3+ Treg infiltration in both TME and PTS in advanced stage Her-2 Positive Breast cancers.

Legal entity responsible for the study

Rajiv Gandhi Cancer Institute and Research Center.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

CDK 4/6 inhibitors have been recently approved in combination with letrozole in hormone receptor-positive breast cancer. Although this combination therapy has been found effective in some patients, resistance often develops. To aid in developing new therapies for palbociclib-resistant breast cancer and better understand resistance mechanisms, we established two PDX models from patients with luminal A breast cancer at time of progression, following acquired resistance to palbociclib therapy. Two models, designated ST3932 and ST4378, were developed in athymic nude mice and characterized for receptor expression, genomic alterations, and in vivo drug sensitivity.

Methods

ST3932 was established from a biopsy taken from a 62-year-old woman pretreated with various therapies including tamoxifen, fulvestrant/palbociclib, and paclitaxel. ST4378 was established from fluid taken from a 66-year-old woman pretreated with various therapies including letrozole/radiation, docetaxel/cyclophosphamide, and letrozole/palbociclib. The resulting models were passaged and challenged with palbociclib and other CDK4/6i to confirm resistance. Receptor expression was determined immunohistochemically. Genomic analysis, including WES and RNAseq, were performed to characterize models and identify mechanisms of resistance. For in vivo studies, endpoints included tumor volume and time from treatment initiation with %T/C values and tumor regression reported at study completion.

Results

The ST3932 and ST4378 models retained ER expression over tested passages with similar histology compared with an archival clinical sample. Sequencing identified several conserved variants; however, none have been currently identified as known mechanisms for palbociclib resistance. Both models demonstrated resistance to palbociclib with %T/C values >90%; however, ST3932 reported moderate sensitivity to both ribociclib and abemaciclib (%T/C∼50%).

Conclusions

We have established and characterized two palbociclib-resistant breast PDX models, designated ST3932 and ST4378, which can be utilized as a valuable tool in better understanding CDK4/6i resistance and in developing novel therapies for CDK4/6i-resistant patients.

Legal entity responsible for the study

Michael J Wick.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Male breast cancers (MBCs) are rare and account for 1% of all breast cancers. Radiomics is evolving as tool for precision medicine in various cancers, in particular breast cancer. This paper aims to establish the benefit of using radiomics data derived from PET images in the categorization of breast cancer immunophenotype in male population.

Methods

One radiologist having 6 years experience manually drew the region of interest on the tumor for purposes of segmentation and textural analysis. The segmented tumor was validated independently by two radiologists having 30 years of experience each. 851 radiomics features were extracted from the available imaging (PET-CT; Total n = 18) using slicer 3D software and Pyradiomics. Feature selection was done by taking union of the features which had Pearson correlation > 0.5 with p53 and Ki-67. Models were trained to predict the biomarkers using leave one out cross validation and different algorithms for different biomarkers.

Results

All cases were hormonal receptor positive, Estrogen receptor (ER) expression range was 80-100%, and progesterone receptor (PR) expression range was 0-100%. p53 was overexpresssed in 16% cases (n=3), while low heterogenous expression was seen in 84% (n=15). Ki-67 was high (>14%) in 33.3% (n=6), and low in 66.6% (n=12). The various other markers namely cyclin D1 , GCDFP15 , Bcl2 , and AR were not utilized for radiomic model building in view of skewness of expression in above cases. Using nearest shrunken centroids model for evaluation of p53 in the subset population we were able to achieve an accuracy of 75%. The Kappa score was 0.30 and area under the curve was 0.75. Using same model in the subset population of Ki-67 we were able to achieve an accuracy of 81%. The Kappa score was 0.58 and area under the curve was 0.80.

Conclusions

Radiomic features could be useful in deciphering male breast cancer immunophenotypes and serve as potential imaging biomarkers. This modality can potentially address tumoral heterogenity that may be missed on a single biopsy from the most feasible site. Also, imaging biomarkers can be used as a real time estrimate of the dynamics of biomarker expression in an individual patient.

Legal entity responsible for the study

Rajiv Gandhi Cancer Institute and Research Center.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

CDK4/6 inhibitors (CDK4/6i) in combination with hormone therapy is the standard treatment of hormone-receptor-positive, HER2-negative metastatic breast cancer (mBC). Resistance mechanisms are unknown and constitute an unmet medical need. In PALOMA-3 trial the PIK3CA mutations (PIK3CA mut) were not associated with resistance to palbociclib (p=0.34). The aim is to assess the role of PIK3CA mut in routine clinical practice as a mechanism of resistance to CDK4/6i in patients with luminal mBC.

Methods

We conducted a retrospective and bi-centric study, between Hospital Clínico Universitario de Valencia and Hospital Universitario ‘12 de Octubre, to evaluate the impact of PIK3CA mut on CDK4/6i treatment in patients with HR+/HER2- mBC. The relationship between PIK3CA mutational status and progression-free survival (PFS) was analyzed by Cox's proportional hazards model and the log rank test. With the aim of homogenizing the sample, patients treated in the first line were also analyzed separately.

Results

All patients (n=92) were diagnosed with a luminal mBC. Forty patients (43.5%) presented PIK3CA mut and 52 (56.5%) were wild type (WT). The median PFS was 12.0 months (95% CI 9.3-14.6). No significant difference in PFS was found based on PIK3CA mutational status (10.9 months in PIK3CA mut, 95% CI 7.9-14.0; vs 12.7 months in PIK3CA WT, 95% CI 8.6-16.8) HR 1.05 p=0.84 (logrank test). The incidence of PIK3CA mutations were higher among patients treated at first line for ≤ 6 months (46.67%), however only 26.92% of long-term responders presented PIK3CA mutations. This effect was not identified at second line. Table: 57P

Patient baseline characteristics

Conclusions

The presence of PIK3CA mutations was not associated with resistance to CDK4/6 inhibitors in terms of PFS. Nevertheless, the frequency of PIK3CA mutations was lower in patients with extended benefit (more than 6 months) at first line of treatment. Future studies to explore the impact of triplet combination therapy (PIK3CA and CDK4/6i plus endocrine treatment) are needed.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

A. Lluch: Research grant/Funding (institution): Amgen, AstraZeneca, Boehringer-Ingelheim, GSK, Novartis, Pfizer, Roche/Genentech, Eisai, Celgene, Pierre Fabre; Advisory/Consultancy: Novartis, Pfizer, Roche/Genentech, Eisai, Celgene. E.M. Ciruelos: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche Pfizer. A. Cervantes: Advisory/Consultancy: Merck Serono, Roche, Beigene, Bayer, Servier; Speaker Bureau/Expert testimony: Merck Serono, Roche, Amgen, Bayer, Servier, Foundation Medicine; Research grant/Funding (institution): Genentech, Merck Serono, Roche, Beigene, Bayer, Servier, BMS, MSD. L. Manso: Advisory/Consultancy, Speaker Bureau/Expert testimony: Roche, Astra, Tesaro, Pfizer, Novartis; Research grant/Funding (self): Tesaro; Travel/Accommodation/Expenses: Roche, Tesaro Novartis. All other authors have declared no conflicts of interest.

Background

Alpelisib (a PI3Kα-specific inhibitor) has demonstrated clinical benefit in combination with fulvestrant and is approved for PIK3CA-mutated, ER +,HER- metastatic breast cancer (MBC) patients who relapsed after aromatase inhibitor (Andre F et al, NEJM, 2019). The PI3CA status may be determined either in tumour biopsy and/or cell-free DNA (cfDNA) (Andre F et al, NEJM, 2019). In our center we use a multiplex digital droplet PCR assay (ddPCR) for the detection of circulating PIK3CA mutations.

Methods

MBC ER+/HER2− patients were tested at the time of disease progression. cfDNA was extracted from 5 mL of plasma with a QIAamp circulating nucleic acid kit and quantified using a fluorometer-based quantification. Our ddPCR assay allows the detection and quantification of 26 mutations located in exons 4, 7, 9 and 20, with a high sensitivity and specificity, using a three-colour detection system (Stilla Technologies).

Results

Thus far, 116 patients have been tested. They had previously received a median of two lines of treatment (min-max: 0-14) in the metastatic setting. Plasma cfDNA concentrations ranged from 4 to 897 ng/ml (median: 16 ng/ml). Forty-seven patients (40%) harboured at least one mutation; among them, 8 (6.9%) patients had 2 PIK3CA mutations and 2 (1.7%) patients had 3 PIK3CA mutations. H1047R, E545K, H1047L, E542K, Q546K, C420R and N345K mutations were identified in, respectively, 19 (16.4%), 16 (13.8%), 9 (7.8%), 6 (5.2%), 4 (3.4%), 3 (2.6%) and 2 (1.7%) patients. The median number of mutant copies per ml of plasma was 128 (min–max: 1–41953) and the median allele frequency was 2% (min–max: 0.01%–59.12%). PIK3CA-mutated patients had significantly higher levels of cfDNA (mean = 80.4ng/ml for mutated patients, mean = 28.8ng/ml for non-mutated patients, p<0.0001). The mutation detection rate was higher for patients with visceral metastases (47.3%) compared from those with non-visceral metastases (27.5%).

Conclusions

The ddPCR assay is a very sensitive, rapid, and cost-effective technique for the selection of alpelisib treatment in routine clinical practice. We plan to compare these results to those obtained from matched tumour biopsies.

Legal entity responsible for the study

The authors.

Funding

La Vannetaise.

Disclosure

All authors have declared no conflicts of interest.

Background

Approximately 60% of pts with early-stage TNBC who undergo NAC without attaining pathologic complete response (pCR) will eventually relapse. Here we assessed the relationship between primary tumor (t-CNAs) and CTC-CNAs and clinical outcome in TNBC pts at the time of initial diagnosis and following NAC.

Methods

CNA analysis was performed by the targeted NGS IonAmpliSeq Comprehensive Cancer Panel on pre-NAC tumor specimens, and low-pass whole genome sequencing on CTCs enriched by the marker-independent Parsortix approach and selected through the DEPArray system. t- and CTC-CNAs were tested for association, and compared to the publically available TCGA-BRCA dataset.

Results

Starting from a case study of 19 pts, we successfully analyzed up to 58 CNA events per pt (median 47, range 31-58) in 16 pre-NAC tumor specimens from 4 and 12 women with stage I-III TNBC achieving or not pCR. Global genomic gain of chromosomes 8, 6 and 17 occurred in 7%, 6%, and 5% of cases; loss of chromosomes 1, 2 and 9 in 12%, 8% and 2% of cases. Overall DAXX (69%), MYC (62%) and SOX11 (56%) were the most commonly altered genes. MSH2 and PRDM1 amplifications were found exclusively in pts attaining pCR (4/4 vs 0/12). Loss of PAX3 occurred in tumor samples of all patients with pCR and in 1 case with residual disease at surgery (4/4 vs 1/12). MSH2 and PRDM1 gains and PAX3 loss were found in 28%, 28% and 23% respectively of the TCGA-BRCA dataset, largely comprising of naïve pts. None of these CNAs significantly correlated with overall survival (Hazard Ratio [HR] 1.2, 95%CI 0.47-3.2; HR 1.7, 95%CI 0.65-4.2, and HR 2.1, 95%CI 0.6-6.4, for MSH2, PRDM1 and PAX3, respectively), suggesting their potential to reflect NAC response. A total of 18 CTCs were retrieved from 4 patients at the time of first relapse (min-max CTC number per patient 2–8). A common feature of all these CTCs was the absence of PAX3 deletion.

Conclusions

In TNBC t- and CTC-CNAs may represent genomic markers of poor response to NAC that are possibly maintained during metastatic dissemination. Further investigations are required to assess their potential to identify pts at high risk of recurrence for whom alternative therapies and more frequent monitoring are advised.

Legal entity responsible for the study

The authors.

Funding

5x1000 Ministry of Health.

Disclosure

All authors have declared no conflicts of interest.

Background

Positive estrogen and progesterone receptors (ER, PR) breast cancers (BC) require endocrine therapy and were associated to favorable outcomes. Hormone receptors (HR) status is considered a crucial predictive factor. The ER−PR+ group is reported to be only 1–5% of all BCs and remains poorly understood and was associated to poor outcomes The aim of this study was to compare the clinical outcomes of patients with single HR positive BC to those with ER+PR+ tumors and TNBC.

Methods

It is a retrospective study including 700 women with invasive Her2 negative breast carcinoma were included. Patients were stratified according to ER and PR expression as double HR+ (ER + PR+), single HR+ (ER + PR- and ER-PR+) and triple negative HR-negative (HR-, ER-PR-) We reviewed the clinicopathologic characteristics of patients, including biologic factors, such as ER, PR, and Ki-67 Survival was analysed using the Kaplan-Meier method. Hazard ratios were estimated using a Cox regression for OS in a multivariate analysis.

Results

700 patients with negative Her2 disease were included, 421 (60.1%) were ER + and PR +, 111 (15.8%) cases were single HR+ tumors, of which 48 (43.2%) ER-PR+ and 63 (56.7%) were ER + PR- Single HR+ tumors were grade III in 45%, KI67 were > 20% in 70%, T3 in 23% and T4 in 35% and metastatic in 25% of cases. The multivariate analysis revealed that patients with ER + PR- tumors were associated with shorter survival compared with ER + PR+ tumors, with a hazard ratio of 3.69 for overall survival (OS). Patients with ER-PR+ tumors had higher risk of death compared with ER + PR+ tumor, with a hazard ratio of 6.23 for OS. While OS of ER- PR+ patient was statistically non different from the one of triple negative breast cancer.

Conclusions

Single HR+ tumors without HER2 overexpression (ER + PR-or ER-PR + ) were associated with more aggressive clinic-pathological features and poorer survival than ER + PR + tumors, and had comparable poor survival to triple-negative breast cancer.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Breast cancer is identified as a leading cancer in Kyrgyz females. Poor diagnostic approaches lead to high rate of advanced breast cancer cases and consequently to high mortality rate. Genetic testing is a promising method of prevention and early diagnosis of breast cancer.

Methods

This was a case-control study of 201 women of the Kyrgyz ethnic group with a morphologically verified breast cancer (N=99) and 102 controls age-matched with BC cases. The mean age of the patients was 48 years (minimum 24, maximum 74, STD=9.83). The genotyping was performed by using restriction fragment length polymorphism assay. The extraction of DNA was carried out from venous blood.

Results

Genotype CT of the HMMR V353A polymorphism is associated with low risk of breast cancer in the Kyrgyz females (OR=0,481, 95%CI 0,272 – 0,850, р=0,011). Combination of the allele 194Trp (XRCC1 Arg194Trp) and genotype СТ (HMMR V353A) (OR=0,302, [95% CI 0,128-0,713], р=0,005), combination of genotypes CT (HMMR V353A) //TT (Palb2 T1100T (3300T>G) (OR=0,459, 95% CI [0,259-0,814], р=0,007), combination of genotypes CT (HMMR V353A) //GG (TNF aG3080A) (OR=0,546, 95% CI [0,298-0,999], р=0,048) are also associated with low risk of breast cancer in the Kyrgyz ethnic group. Furthermore, the allele 194Trp is associated with late age of onset of breast cancer when comparing to 194Arg allele of XRCC1 gene (p=0,017). Allele 194Trp significantly more often occurs in postmenopausal women (p=0,005) and in women with high BMI (>25) (p=0,003).

Conclusions

These results may contribute to further genetic research aimed at identifying genes associated with breast cancer in the Kyrgyz population.

Legal entity responsible for the study

Research Institute of Molecular Biology and Medicine.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Breast cancer is one of the most common malignancies in women worldwide. The development and progression of breast cancer is very complicated process and involves both genetic and epigenetic factors. Genetic variation of both thymocyte selection associated high mobility group box family member 3 (TOX3) and fibroblast growth factor receptor 2 (FGFR2) genes is a newly-described risk factor for breast cancer.

Methods

Eighty participants were enrolled in this study; 40 women with breast cancer and 40 age- matched healthy controls. Detection of (TOX3) rs12443621 and (FGFR2) rs2981582 SNPs was done for all the included subjects using Taq Man Allelic Discrimination assay technique by real time PCR. For patients, all clinical and pathological data, chemotherapy toxicity and survival were studied and analyzed in relation with (TOX3) and (FGFR2) SNPs.

Results

A statistically significant difference was observed between cases and controls regarding the frequency of rs122443621 in TOX3 gene with GG genotype (45% in patients compared to 25% in controls) (P=0.04) and rs2981582 of FGFR2 gene polymorphisms AA genotype (42.5% in patients compared to 17.5% in controls) (P=0.02). There was a significant statistical difference among different genotypes of TOX3 rs122443621 regarding molecular subtypes& TNM stage. There was a significant statistical difference among different genotypes of TOX3 (rs12443621) as 58.3 % of patients with GG genotype developed toxicity grade (G) II & III compared to 15.4% of AG while AA genotype developed only GI (P=0.02) , for FGFR2 (rs2981582) G II&III toxicity was 53.8 % in patients with AA genotype while GA and GG developed lower incidence of toxicity (11.1% and 9.1% respectively) (P=0.03). Regarding survival, the mean progression free survival (PFS) 20.59 months with 95% CI (19.42 – 21.76) months, during this follow up duration there was a non significant statistical difference either among different genotypes of TOX3 (rs12443621) or FGFR2 (rs2981582) (P>0.05).

Conclusions

TOX3 rs122443621 and FGFR2 rs2981582 SNPs are associated with an increased risk for breast cancer, advanced stages, molecular subtype of the disease and with chemotherapy toxicity.

Legal entity responsible for the study

Menoufia University.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Triple negative breast cancer immunohistochemically defined by negative estrogen receptor (ER), progesterone receptor (PR) and HER2 expression, is highly heterogenous. As there is a lack of routine targeted therapies, chemotherapy in neoadjuvant setting remains a standard of treatment. Genomic alterations are extensively investigated to determine the drivers of tumor evolution, new treatment targets and to identify the ways of resistance to therapies.

Methods

28 patients with stage II – III triple negative breast cancer were assigned to receive neoadjuvant chemotherapy with paclitaxel (80mg/m²) and carboplatin (AUC 1.5-2) 12 cycles weekly followed by 4 cycles of AC (doxorubicin 60 mg/m² plus cyclophosphamide 600 mg/m² 3-weekly). Hybrid capture-based next-generation sequencing was performed by using FoundationOne®CDx test on DNA from formalin-fixed paraffin-embedded (FFPE) primary tumor tissue obtained prior to the treatment.

Results

The median age of 28 patients was 52 years (ranged 32 to 72 years). One or more mutations were found in all cases. The number of detected gene mutations was an average 5.4 (range 1 to 14 per case). The most frequent somatic mutations were TP53 92.9% (26 of 28), PIK3CA 28.6% (8 of 28), RAD21 21.4% (6 of 28), PTEN and MYC 17.9% (5 of 28), NSD3 14.3% (4 of 28), NF1, FGFR2, CCNE1, AKT2 and ZNF703 10.7% (3 of 28). Germline BRCA1 mutation was found in 2 cases, while germline BRCA2 mutation - in 1 case.

Conclusions

Comprehensive genomic profiling of primary triple negative breast cancer identifies potentially actionable mutations in a large set of tumors and might be clinically important for treatment individualization.

Legal entity responsible for the study

National Cancer Institute, Vilnius, Lithuania.

Funding

National Cancer Institute.

Disclosure

M. Drobniene: Research grant/Funding (self), Travel/Accommodation/Expenses: Roche. All other authors have declared no conflicts of interest.

Background

Management of breast cancer banks upon hormone receptor (HR) status, whose absence is associated with features like poor prognosis and aneuploidy. A newly identified protein, CUEDC2 was found maintaining HR level by promoting its degradation, while its moonlighting job is detected in ploidy maintenance during mitosis. The HR members (ER/PR) function as transcription factors. However, their involvement in transregulation of mitotic checkpoint protein(s) is largely unexplored. We hypothesize that HR controls mitosis and maintains ploidy by employing its transregulatory activity, while upregulated CUEDC2, as seen in several cancer types, rendering HR reduction, promotes mitotic deregulation and aneuploidy in HR-negative breast cancer.

Methods

Expression analyses were performed by real-time PCR, immunohistochemistry, and/or Western blot. Plasmid or siRNA constructs were transfected with lipofectamine reagent. The phosphorylation of CUEDC2 was blocked by Cdk1 blocker, RO3306 or site directed mutagenesis. Mitotic synchronization was made with nocodazole treatment. Mitotic progression was examined by immunofluorescence of Ser10-phosphorylated-Histone3. Cellular ploidy was examined by fluorescence in situ hybridization.

Results

The data revealed high expression of mitotic-checkpoint proteins in HR+ve tumors, compared to HR-ve cases, in both primary tumors as well as in cell lines, where ectopic ER-α induced their endogenous levels. Concurrently, HR-ve lines showed abnormal mitotic progression post release from mitotic arrest. Further, CUEDC2 showed higher expression in HR-ve breast malignancies, compared to HR+ve cases, in both primary tumors and cell lines. Additionally, mitotic ubiquitin ligase and Ser110-phosphorylated-CUEDC2 were found crucial in ER-α regulation. Finally, promotion of mitotic deregulation and aneuploidy, upon CUEDC2 upregulation, were rescued by ectopic ER-α.

Conclusions

We found a novel molecular connection between hormone receptor and ploidy maintenance in breast cancer. Furthermore, upregulated CUEDC2, at this crossroad, deregulates this balance, promoting aneuploidy in HR-ve breast malignancies.

Legal entity responsible for the study

The author.

Funding

Early Career Award (ECR/2015/000206), Scientific and Educational Research Board (SERB), Govt. of India.

Disclosure

The author has declared no conflicts of interest.

Background

Despite molecular classification, patients with luminal early breast cancers have different clinical outcomes. In order to select whose patients would be more benefit from adjuvant chemotherapy, despite clinico-pathological features, genomic signatures help clinician to decide which adjuvant treatment is the most appropriate. EndoPredict is one of them.

Methods

Since November 2016, for patients treated in Brittany’s institutions, we proposed EndoPredict assay for unclear cases of adjuvant treatment. For patients treated in our Comprehensive Cancer Center, we retrospectively reported decision of adjuvant treatment before and after EndoPredict assay and compare to PREDICT’s tool scores.

Results

From November 2016 to July 2018, 274 breast cancer tumors were analyzed with EndoPredict assays. 131 patients were treated in Rennes, and presented in multidisciplinary breast tumor board before and after EndoPredict assay. Before EndoPredict results, clinicians, recommend chemotherapy for 37 patients (28%). 91 patients (70%) were classified as EndoPredict high risk. Finally, 76 (58%) received chemotherapy. PREDICT tool recommend chemotherapy for 7 patients (5%).

Conclusions

Although genomic tests were developed in order to de-escalate adjuvant treatment, in our Comprehensive Cancer Center, the use of EndoPredict assay lead to an increase of 30% of prescription of chemotherapy.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Adipocytes and cancer-associated adipocytes (CAAs) are under-investigated cells from the tumour microenvironment. Different image analysis software exist for counting and measuring these cells, however, it is unclear which is the best for breast cancer (BC) samples. The objectives are: (1) to identify the best software for counting and measuring distant adipocytes and CAAs, and, (2) to apply the identified software for comparing distant adipocytes and CAAs in a series of BC.

Methods

Firstly, we compared adipocyte counts, diameters and areas on three BC test slides using HALO®, Adiposoft, and AdipoCount. Secondly, we analysed a series of 10 BC samples with HALO®. For each patient, we analysed ∼ 500 distant adipocytes and ∼ 500 CAAs. Distant adipocytes were defined as at least two mm away from cancer cells, whereas CAAs were defined as the three first lines of adipocytes close to the invasive front.

Results

All three methods performed equally good with regard to area and diameter measurement (all estimated concordance correlation coefficient values > 0.97 and > 0.96, respectively). HALO® clearly outperformed the other two methods with regard to adipocyte counting, reaching higher sensitivity and specificity (Table). When applying this software to the BC samples, CAAs presented smaller areas (median fold-change: 2.4, IQR: 2.13 – 2.63) and diameters (median fold-change: 1.6, IQR: 1.51 – 1.66) compared to distant adipocytes (p<.001, p<.001, respectively). Both CAAs size and distant adipocytes size were associated with BMI as continuous (rho = 0.8 p = .008; rho = 0.73 p= .029, respectively) and as categorical variable (Kendall’s tau = 0.60 p = .038, Kendall’s tau = 0.55, p = .062, respectively). Table: 67P

Conclusions

Quantifying adipocytes in BC sections is feasible by digital software analysis. This study is the first digital analysis that demonstrates a clear reduction in size between CAAs and distant adipocytes supporting the concept of an interaction between CAAs and cancer cells.

Legal entity responsible for the study

Institut Jules Bordet.

Funding

Fondation Belge contre le Cancer, Fondation Cancer Luxemburg.

Disclosure

All authors have declared no conflicts of interest.

Background

As cancer cells strongly rely on glycolysis, targeting cancer cell glucose metabolism could be a suitable approach to find novel therapeutic targets. The role of RNA export factors and long non-coding RNAs (lncRNAs) in cancer cell metabolism have not been fully uncovered yet. Therefore, we comprehensively examined the interplay between the RNA export factor paraspeckle-associated protein 1 (PAP1) and the lncRNA NEAT1 in triple negative breast cancer cell metabolism.

Methods

In order to uncover PAP1-associated genes, RNA sequencing was performed. After identifying a correlation with NEAT1, cellular growth and apoptosis assays were conducted. Mitochondrial morphology was visualized after cellular staining with MitoTracker® on a confocal microscope. Key metabolic pathways such as mitochondrial respiration and glycolysis were measured with the Seahorse XF Analyzer.

Results

In a preceding study, we established the RNA export factor PAP1 as mediator of breast carcinogenesis - regulating cell growth by apoptosis induction. These findings motivated us to further investigate this topic. RNA sequencing data identified a significant correlation between PAP1 and the BC-related lncRNA NEAT1. Strikingly, we could elaborate PAP1 as transcriptional regulator of NEAT1 and mediator of NEAT1-dependent paraspeckle formation. These data were corroborated by the finding, that silencing NEAT1 phenocopies PAP1 silencing in terms of cellular growth and apoptosis induction. As mitochondria are crucially involved in apoptotic processes, glucose metabolism and ATP production, we examined the effect of PAP1 and NEAT1 silencing on mitochondrial morphology and function. Reducing the expression of either one of these genes changes mitochondrial shape to a more spherical, apoptotic phenotype and significantly decreases mitochondrial respiration and ATP production.

Conclusions

In this study, we could elaborate PAP1 as novel regulator of NEAT1 in breast carcinogenesis. Furthermore, our data point towards a crucial involvement of PAP1 and NEAT1 in cancer cell metabolism – an interesting finding justifying a more detailed investigation on the role of RNA export factors and lncRNAs in cancer metabolism.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The identification of potential biomarkers capable of detecting, predicting or monitoring treatment response is currently one of the main objectives in oncology. Our research group had previously identified the epigenetic inactivation of Transforming Growth Factor β-Induced (TGFBI) in different trastuzumab resistance HER2-positive breast cancer (HER2+ BC) cell models. In the present work, our objective was to validate the TGFBI promoter methylation in a cohort of HER2+ BC patients before and after neoadjuvant treatment.

Methods

The study cohort included 24 patients with HER2+ early BC treated with neoadjuvant anthracycline-taxane-based chemotherapy plus trastuzumab, of which 20 patients presented partial or no response and 4 patients complete treatment response. TGFBI methylation was analyzed in formalin-fixed paraffin-embedded (FFPE) biopsy (pre-treatment) and tumor samples (post-treatment) from each patient. All the DNA samples were analyzed using bisulfite pyrosequencing. The correlation between TGFBI methylation and clinical-histopathological characteristics has been analyzed and characteristic curves (ROC) were used to assess the predictive capacity of TGFBI as a marker.

Results

Similar TGFBI promoter hypermethylation levels were observed in pre-treatment samples from patients with complete response to trastuzumab and from the non-responders. In contrast, non-responsive patients showed significantly higher methylation levels in post-treatment (30.26%±3.52) than pre-treatment samples (6.08%±1.51). In particular, significant TGFBI hypermethylation after trastuzumab (80%) compared to the pre-treatment samples was observed in non-responsive patients with pre- and post- treatment paired samples. Importantly, the ROC curve analysis showed an area under the curve (AUC) of 0.9502 (95% CI: 0.8716 to 1.029). No significant association between TGFBI methylation levels before and after treatment and their clinical-histopathological characteristics was identified.

Conclusions

These preliminary results provide a basis for further studies to validate TGFBI hypermethylation as a potential epigenetic monitoring biomarker for trastuzumab resistance in HER2+ BC patients.

Legal entity responsible for the study

University of Girona and Catalan Institute of Oncology (ICO), Dr. Josep Trueta Hospital.

Funding

University of Girona.

Disclosure

All authors have declared no conflicts of interest.

Background

The endocrine therapy (ET) has established itself as the basis for drug treatment in MLBC due to its high efficiency and low toxicity. Aromatase inhibitors (AI) are initial therapy in the most cases. However, a third of them have a progression of the disease due to primary or acquired resistance.The role of ESR1 mutations is actively discussed as an early and practically achievable (especially when repeat biopsies are impossible) marker for predicting the development of insensitivity to standard ET.

Methods

Treatment results of 53 patients with MLBC, who received first-line ET with nonsteroidal AI - letrozole (2.5 mg daily) or anastrozole (1 mg daily) were analyzed. We provided the baseline and post-treatment molecular genetic testing of ESR1 in peripheral blood. ESR1 (A-351G, T-397C) polymorphism was detected by analysis of DNA restriction fragment length polymorphism. The treatment response assessed according to RECIST 1.1. In a year of treatment all patients were divided into 2 groups depending on the progression of the disease. The first group included 19 patients with progression earlier than one year after AI and in the second was 34 patients without progression in one year of AI.

Results

We identified that ESR1 gene heterozygous variant A/G351 occured in 51,7 % and T/C397 in 68,8 %. Analysis of associations between genetic variants ESR1 A-351G and ESR1 T-397C showed significant relationship between themselves (p < 0.05). It has been established that the presence of polymorphic genotypes ESR1 A-351G (odds ratio (OR) 2.81 [95% CI = 1.16 –6.82], p=0.05) and ESR1 T-397C (OR 3.33 [95% CI =1.00 –11.90], p=0.05) was associated with early disease progression (up to 1 year). There was no statistically significant association of the polymorphisms ESR1 (A-351G, T-397C) gene with receptor status (ER%, progesterone receptor %) and Ki-67% (p < 0.05).

Conclusions

Further research is required to provide evidence of the ESR1 gene polymorphisms (A-351G, T-397C) as an additional risk factor of early resistance to AI in patients with MLBC.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Radiomics is a promising tool for imaging biomarker discovery in the era of precision medicine. This paper aims to establish the benefit of using multi-modality radiomics data from PET and MR images in the categorization of breast cancer molecular phenotype.

Methods

One radiologist having 6 years of experience manually drew the region of interest on the tumor for purposes of segmentation and textural analysis. The segmented tumor was validated independently by two radiologists having 30 years of experience each. A total of 851 radiomics features were extracted from the available imaging (Total n = 63; PET-CT 25/MRI 38) using slicer 3D software and Pyradiomics. Harmonization was achieved in the different imaging modalities and hence the removal of batch effect was done with the help of the combat-a R package. Feature selection was done by taking the union of the features which had Pearson correlation >3 with any of the biomarker ER, PR or Her2neu, or > 2.5 with BRCA. Models were trained to predict the biomarkers using leave one out cross-validation and different algorithms for different biomarkers.

Results

Using KNN (K- Nearest neighbor) model in the subset population of ER (22 positives and 41 negatives) we were able to achieve an accuracy of 80%. The Kappa score was 0.45 and the area under the curve (AUC) was 0.72. Using the KNN model in the subset population of PR (18 positives and 45 negatives) we were able to achieve an accuracy of 82%. The Kappa score was 0.43 and AUC was 0.70. Using KNN (K- Nearest neighbor) model in the subset population of ER (22 positive and 41 negatives) we were able to achieve an accuracy of 80%. The Kappa score was 0.45 and the area under the curve (AUC) was 0.72. Using the KNN model in the subset population of PR (18 positive and 45 negatives) we were able to achieve an accuracy of 82%. The Kappa score was 0.43 and AUC was 0.70.

Conclusions

Radiomic features on imaging could be useful in deciphering breast cancer phenotypes and serve as potential imaging biomarkers. This modality can potentially address tumoral heterogeneity that may be missed on a single biopsy from the most feasible site. Also, imaging biomarkers can be used as a real-time estimate of the dynamics of biomarker expression in an individual patient.

Legal entity responsible for the study

Rajiv Gandhi Cancer Institute and Research Center.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Non-mass enhancing breast lesions pose a diagnostic dilemma and a clinical challenge that are encountered commonly in clinical practice. Suspicion of breast cancer makes it an important entity for a radiologist and medical oncologist.

Methods

All patients with clinical suspicion of Breast disease were included in the study. MRI reveals MR-BIRADS 4 and 5 non mass enhancing lesions and further underwent biopsy/post MRM/BCS/lumpectomy in our hospital. After inclusion/exclusion criteria clinically meaningful 128 non-mass lesions were assessed further for distribution pattern, internal enhancement pattern and kinetics.

Results

128 non-mass lesions were analyzed in n= 127 patients. 83 lesions were malignant and MR-BIRADS 5 category dominated (66/128, 51.5 %). Most common pattern of distribution/internal enhancement/curve type were segmental (47, 36.72%), clumped (64, 50%) and washout curve type (99, 77.34%) respectively. Regarding association with malignancy, odds ratio of lesions with segmental/regional/multiple regional distribution pattern was 13.5 (95% CI= 5.6-32.5), clumped/clustered ring internal enhancement pattern was 43.07 (95% CI= 14.3-129.6) and washout curve type was 17.8 (95% CI= 6.0-52.3). The sensitivity of the washout curve type for diagnosis of malignancy was 93.9%. The specificity of clumped/clustered ring internal enhancement pattern was 88.9%.

Conclusions

Segmental/regional/multiple regional distribution patterns, clumped/clustered ring internal enhancement pattern and washout curve type was the most powerful indicator for malignant pathology in non-mass enhancing lesions. The study envisaged the unmet need for consensus on the characterization of non-mass enhancing lesions in most of the previously done studies.

Legal entity responsible for the study

Rajiv Gandhi Cancer Institute and Research Center.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The molecular landscape of single hormone receptor-positive breast cancers – ER(+)/PgR(-) and ER(-)/PgR(+), remains vague. Clinically, endocrine sensitivity and prognosis of these tumors are believed to be intermediate between ER(+)/PgR(+) and double-negative cancers. MicroRNAs (miRNAs) emerge as important players in breast cancer biology and may serve as potential therapeutic targets. In the current study, we aimed to compare the miRNA expression profiles of these cancers.

Methods

The group consisted of 32 breast cancer patients with thoroughly characterized status of ER and PgR expression [14 ER(+)/PgR(-) and 18 ER(-)/PR(+) cases]. The expression of 829 miRNAs was evaluated with nCounter Human v3 miRNA Expression Assay (NanoString). miRNAs differentiating between ER/PgR phenotypes were selected based on fold change, and the differences were estimated with Student's t-Test or Two Way ANOVA (considering also the HER2 status). The results were validated using the TCGA dataset.

Results

A trend of higher expression of miRNAs associated with ER-positivity (miR-149, miR-375, miR-26b, miR-425, miR-29c, miR-200a, miR-191, miR-144) was observed in the ER(+)/PgR(-) group. In contrast, ER(-)/PgR(+) cases tended to express higher levels of miR-92a-3p, miR-155-5p, miR-18a-5p, miR-222-3p, characteristic for triple-negative cancers. The last two miRNAs are known to directly down-regulate ER expression. Yet, due to the limited number of cases in our cohort, no differences between miRNAs expression in ER(+)/PgR(-) and ER(-)/PgR(+) breast tumors remained significant after correction for multiple comparisons. Four miRNAs validated in the TCGA dataset (miR-1180, miR-223, miR-30d, miR-99a) were down-regulated in HER2-overexpressed/amplified tumors of both ER/PgR phenotypes.

Conclusions

Our data support the role of miRNAs in the pathogenesis of ER(-)/PgR(+) breast tumors. Interestingly, the miRNA profiles of the single hormone receptor-positive breast cancers are mainly associated with the HER2 status. The high expression of miR-1180 in HER2-negative cases has not been reported before.

Legal entity responsible for the study

Medical University of Gdańsk.

Funding

National Science Centre, Poland (grant 2017/25/B/NZ5/00656).

Disclosure

A. Lacko: Honoraria (self), Travel/Accommodation/Expenses: Roche; Honoraria (institution), Travel/Accommodation/Expenses: Pfizer; Honoraria (self), Travel/Accommodation/Expenses: AstraZeneca; Honoraria (self), Travel/Accommodation/Expenses: Egis. B.S. Radecka: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution): BMS; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Lilly; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Merck; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): MSD; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Novartis; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Pfizer; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Pierre-Fabre Medicament; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Roche; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Servier. J. Pikiel: Honoraria (self): Tesaro; Travel/Accommodation/Expenses: Roche; Honoraria (self): Regeneron Pharmaceuticals; Honoraria (self): Pfizer; Honoraria (self): Novartis; Honoraria (self): Amgen; Honoraria (self): Clovis; Honoraria (self): Incyte; Honoraria (self): Odonate; Honoraria (self): Roche. M. Litwiniuk: Honoraria (self): Amgen; Advisory/Consultancy: AstraZeneca Poland; Travel/Accommodation/Expenses: Teva Poand; Honoraria (self): Roche Poland; Honoraria (self), Travel/Accommodation/Expenses: Pfizer. K. Pogoda: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony, Research grant/Funding (self), Travel/Accommodation/Expenses: Novartis; Advisory/Consultancy, Research grant/Funding (self): Lilly; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self), Travel/Accommodation/Expenses: Pfizer; Honoraria (self), Travel/Accommodation/Expenses: Amgen; Travel/Accommodation/Expenses: Pierre Fabre; Honoraria (self), Research grant/Funding (self): AstraZeneca; Honoraria (self): Alvogen; Honoraria (self): Egis Pharmaceuticals; Research grant/Funding (self): Merck; Honoraria (self): Teva. A. Szwajkosz: Honoraria (self), Research grant/Funding (self): Brass Research and Development Sp z o.o.; Honoraria (self), Advisory/Consultancy: ETET Sp z o.o.; Research grant/Funding (self): Merck & Co; Travel/Accommodation/Expenses: Accord Heathcare Polska sp. z o.o.; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): MSD Polska; Honoraria (self), Advisory/Consultancy: Novartis Investments S.A.R.L.; Honoraria (self), Advisory/Consultancy: EGIS Polska Spółka Z O.O.; Honoraria (self), Research grant/Funding (self): Biostat sp z o.o.; Honoraria (self): Teva pharmaceuticals Polska sp. z o.o.; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Accord Heathcare Polska sp. z o.o.; Research grant/Funding (self): Amgen Sp. z o.o. E. Senkus-Konefka: Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Amgen; Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: AstraZeneca; Advisory/Consultancy: Clinigen; Honoraria (self), Advisory/Consultancy: Egis; Honoraria (self), Advisory/Consultancy: Eli Lilly; Honoraria (self), Advisory/Consultancy: Genomic Health; Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Novartis; Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Pfizer; Honoraria (self): Pierre Fabre; Honoraria (self), Advisory/Consultancy, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy: Sandoz; Advisory/Consultancy: TLC Biopharmaceuticals. All other authors have declared no conflicts of interest.

Background

Human epidermal growth factor receptor 3 (HER3) is a member of family of receptor tyrosine kinases, including EGFR, HER2 and HER4. HER3 is thought to function as a signaling protein promoting tumorigenesis, proliferation, migration and metastasis. Overexpression in tumors has been associated with worse clinical outcomes. HER3 serum level was shown to corelate with disease progression. The aim of this study is to assess clinical value of HER3 serum level determination in patients with breast cancer qualified for neoadjuvant chemotherapy.

Methods

57 patients with confirmed breast cancer before treatment were qualified for the study, aged 31-77 (median 53 years), including 28 premenopausal and 29 postmenopausal. The control group consisted of 26 healthy women aged 16-80. Clinical and pathological features were determined in a selected group of patients with breast cancer who subsequently underwent preoperative chemotherapy, i.e. tumor size (T), lymph node status (N), presence of distant metastases (M), estrogen receptor status (ER) and progesterone (PgR), HER2 receptors and Ki 67 proliferative index. The blood serum of the examined patients and healthy women was determined by the enzyme-linked ELISA method in doublets of the HER3 biomarker concentration. Mann-Whitney test and ROC curve analysis were used for statistical calculations.

Results

Significant differences between HER3 concentrations in breast cancer patients and in the control group were shown as preliminary results of the study (p = 0.035). In the examined group of patients, no differences in HER3 levels were found depending on the menopausal status. In the ROC curve analysis in patients vs healthy women, the diagnostic sensitivity for HER3 was AUC (area under curve) 0.652; p = 0.023. Considering the clinical-pathological features, no significant correlation was found between biomarker concentration and tumor size (T), lymph node status (N), receptor status and Ki67 index.

Conclusions

Based on the preliminary results of the study, a relatively high diagnostic sensitivity of ErB3 / HER3 concentrations was demonstrated, which shows its potential usefulness in breast cancer patients. The research is continuing.

Legal entity responsible for the study

The authors.

Funding

National Cancer Institute.

Disclosure

All authors have declared no conflicts of interest.

Background

With increased survival in breast cancer, resulting from advances in treatment, patients incur the possibility of subsequent primary malignancies, especially lung cancer. The aim of this study was to assess the frequency of lung cancer following breast cancer diagnosis, the associations between breast cancer and lung cancer, the pathological features of double primary cancer, and the status of epidermal growth factor receptor (EGFR) mutations in second primary lung cancer.

Methods

A review of medical charts at the Jiangsu Province Hospital (Jiangsu, China) revealed 8048 patients with pathologically confirmed breast cancer between January 2008 and December 2018. Clinical information, including pathology and immunohistochemistry of cancer tissues, EGFR status, date of GGO detection, and cancer stage, was collected. Statistical analysis was performed using SPSS version 21.0 (IBM Corporation, Armonk, NY, USA).

Results

Of the 8048 patients, 55 (0.7%) were diagnosed with a second primary lung cancer, which accounted for approximately 13.6% of the pulmonary ground-glass opacity (GGO) detected. The incidence was higher than in the general female population (standardized incidence ratio 1.4 [95% confidence interval (CI): 1.25-1.55]). Patients who experienced a second primary lung cancer exhibited a significantly higher rate of EGFR mutation (78.6%) than those with lung adenocarcinoma alone, with most exhibiting low-grade malignancy, older age, estrogen receptor negativity, low Ki67, and no lymph node metastasis.

Conclusions

Breast cancer patients, especially those with low-grade malignancy, were at high risk for developing primary lung cancer. For isolated GGO in patients with high-risk factors, clinicians should insist on close follow-up. Furthermore, EGFR may play an important role in primary lung adenocarcinomas and breast cancer.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

The National Key Research and Development Program of China (ZDZX2017ZL-01), High-level innovation team of Nanjing Medical University (JX102GSP201727), Wu Jieping Foundation (320.6750.17006), Key Medical Talents (ZDRCA2016023), 333 Project of Jiangsu Province (BRA2017534 and BRA2015470), The Collaborative Innovation Center for Tumor Individualization Focuses on Open Topics (JX21817902/008), and Project of China Key Research and Development Program Precision Medicine Research (2016YFC0905901).

Disclosure

All authors have declared no conflicts of interest.

Background

Among breast carcinomas, TNBC comprises a distinct disease entity with a unique microenvironment of TILs ,To clarify the biological and prognostic function of TILs, which is a host factor in tumor microenvironment, we focused on effector CD8⁺ T cells in our study as marker for TILs , CD8+ TILs represent a vital component of the local anti-cancer immune response.

Methods

We were evaluating the prevalence of CD8+ as a marker for TILs in the parrafin wax block of pre-treatment biopsies of 30 triple negative breast cancer patients ,and its prognostic value by correlating it with OS and DFS.

Results

Our study showed that All our patients (100%) were positive for CD8+, with a minimum range of 1% and a maximum range of 60 %, most of the patients (20 patients) had CD8 % between (10% to 20 %). high levels of CD8 + TILs are good prognostic indicators in TNBC. our study showed that there were associations of CD8+ TILs infiltrate status with longer progression free survival and better overall survival in triple-negative breast cancer, but were not statistically significant probably due to our small sample size. Our study showed no correlation between CD8+ level and some clinical-pathological variables (tumor size, nodal status, tumor stage, menopausal status, age, family history). Our findings as well as other studies demonstrate that quantification of CD8 + TILs is feasible using routine immunohistochemical techniques.

Conclusions

TNBC is the subtype that is most frequently associated with TILs, but only a minority of TNBCs demonstrate a high number of TILs, suggesting that IM therapy could be necessary to promote immunorecognition and increase the adaptive immune infiltrate to levels adequate for a survival benefit in the majority of patients with this BC subtype. Patients with high levels of TILs at the time of diagnosis might benefit from the use of drugs that can enhance antitumoral immune responses.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The treatment of hormone receptor positive metastatic breast cancer patients has changed dramatically over the last few years and combination strategies attempting to overcome resistance of the disease are gaining importance. After introducing CDK 4/6 inhibitors in the treatment one of the subsequent strategies is definitely targeting PI3 kinase pathway. Several drugs have been tested, but only recently, SOLAR1 phase III trial demonstrated the benefit of addition of alpelisib to fulvestrant, with acceptable tolerability. With this trial the importance of liquid biopsy testing was postulated.

Methods

Cell-free DNA from plasma from 23 patients with metastatic hormone receptor positive was screened for PIK3CAhotspot mutations. To this end a SimSen-seq assay (simple, multiplexed, PCR-based barcoding of DNA for sensitive mutation detection using sequencing) covering the 11 most frequent PIK3CA mutations (Table) was designed. Using Seraseq® ctDNA Reference Materials with various variant allele frequencies (VAFs) the assay enabled a detection of PIK3CA H1047R, down to a VAF of 0.125%, while the mutation could not be detected in the wild type sample. Nevertheless, since in a set of healthy controls, background noise was observed during assay validation, only samples with VAFs >1% were considered as positive.

Results

Out of 23 tested patients, a PIK3CA mutation with a VAF >1% could be detected in 14 patients (60.9%). Ten patients were identified with H1047R mutation and 2 patients were E542K positive. In one patient, we were able to identify co-ocurrence of both mutations. VAF ranged from to 1.1% to 49.9% with an average of 8.2%. It is of note though, that presences of PIC3Ca mutations below the detection limit cannot be excluded.

Conclusions

SimSen-seq based detection of PIK3CA mutations from plasma selected for alpelisib treatment has shown promising results and warrants further validation in a larger cohort of candidate patients. FDA recommendation is to initially carry out the mutation testing in ctDNA and if the test is negative for PIK3CA mutations in plasma, patients should undergo testing for PIK3CA mutations in tumour tissue. Tissue analysis is currently ongoing.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

This study aimed to describe surgical, neoadjuvant, and adjuvant treatment patterns among patients diagnosed with early-stage HR+/HER2- BC.

Methods

Data from a multinational (France [FR], Germany ([DE], Italy [IT], Japan [JP], Spain [ES], United Kingdom [UK], and United States [US]) sample of patients with stage I-III HR+/HER2- BC who received at least one treatment in the adjuvant setting were analysed. Surgery, chemotherapy, and endocrine (aromatase inhibitor or tamoxifen) treatment patterns in neoadjuvant and first adjuvant settings were summarized overall, by country, and age at initiation of first adjuvant treatment (≤50 and >50 years).

Results

A total of 2,447 patients were included in this analysis. Neoadjuvant therapy was given in 18% patients. Of these, 89% of regimens included chemotherapy (66% taxane, 66% anthracycline) and 52% included endocrine treatment (34% aromatase inhibitor, 12% tamoxifen). More than half of all patients (55%) received only one surgery (41% mastectomy, 38% lumpectomy). In the adjuvant setting, 1,261 (52%) patients received chemotherapy (36% anthracycline, 33% taxane). DE had lowest rate of adjuvant chemotherapy (38%) and IT the highest (68%). Overall, adjuvant endocrine therapy was prescribed in 1,509 (62%) patients, with the highest rate observed in DE (74%) and lowest rate in the UK (53%) and JP (54%). Among first adjuvant endocrine treatments, 19% and 38% of patients aged ≤50 at initiation received an aromatase inhibitor or tamoxifen, respectively. In patients aged >50 at initiation, aromatase inhibitor and tamoxifen use in endocrine therapy was 52% and 10%, respectively. Overall, rates of aromatase inhibitor use ranged from 33% (JP, DE, UK) to 48% (IT); tamoxifen use ranged from 8% (IT) to 32% (DE).

Conclusions

This study describes real-world treatment patterns among patients with HR+/HER2- early stage BC across seven countries. Chemotherapy use is generally high in the neoadjuvant setting but varies in the adjuvant setting across countries. Approximately 50-75% of patients received adjuvant endocrine therapy, indicating a potential treatment gap among this HR+ patient population.

Legal entity responsible for the study

Pfizer Inc.

Funding

Pfizer Inc.

Disclosure

C. Criscitiello: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Pfizer Inc; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eli-Lilly; Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis. D. Spurden: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. A. Rider: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. R. Williams: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. M. Corsaro: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. J. Pike: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. E.H. Law: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc.

Background

ExteNET, an international, randomized, placebo-controlled phase III trial, showed that extended adjuvant neratinib given for 12 months after trastuzumab-based adjuvant therapy significantly improved 2-year (HR=0.50, p=0.003) and 5-year (HR=0.58, p=0.002) invasive disease-free survival (iDFS) in early-stage HER2+ and hormone-receptor positive (HR+) breast cancer. We explored the impact of treatment duration on outcomes in the ITT population and in the subgroup of patients (pts) with HR+ disease who started neratinib <1 year after prior trastuzumab (EU label population).

Methods

Pts with early-stage HER2+ breast cancer received oral neratinib 240 mg/day or placebo for 12 months (or until disease recurrence or unacceptable toxicity) after trastuzumab-based adjuvant therapy. Pts who received neratinib for ≤3 or ≥11 months (including those who had recurrence prior to 11 months) were compared with the ITT placebo group. iDFS (primary endpoint) and secondary endpoints (DCIS, DDFS, time to distant recurrence, and 5-year CNS recurrence) were analyzed using Kaplan-Meier methods and Cox proportional-hazards models adjusted for prognostic factors. Data cut-off: March 1, 2017.

Results

There were 2840 pts in the ITT population and 1334 patients who were HR+, <1-year. The table shows iDFS findings for HR+, <1-year pts and stratified by treatment duration (≤3, ≥11 months). Greater benefit was seen in pts who stayed on treatment for ≥11 months when compared to pts who received ≤3 months of treatment. Similar findings were seen for secondary endpoints. Table: 83P

Conclusions

These exploratory data suggest that pts who received a longer duration of treatment with neratinib (≥11 months) derived greater benefit compared to those who stopped treatment early (≤3 months). Patients who receive recommended duration of treatment with neratinib of 12 months may have improved outcomes when compared to patients who discontinue early (within 3 months).

Clinical trial identification

NCT00878709.

Legal entity responsible for the study

PUMA Biotechnology.

Funding

PUMA Biotechnology.

Disclosure

M. Martin Jimenez: Advisory/Consultancy: Pierre Fabre. B. Zhang: Full/Part-time employment: PUMA Biotechnology. A. Wong: Full/Part-time employment: PUMA Biotechnology. B. Moy: Research grant/Funding (institution): PUMA Biotechnology. All other authors have declared no conflicts of interest.

Background

The addition of trastuzumab (T) in dose-dense (dd) sequential (s) chemotherapy (CT) has been found to improve the outcome of patients (pts) with HER2(+) BC. We aimed to evaluate the 5-year survival benefit of dds-CT in pts treated in the post-trastuzumab era.

Methods

We analyzed 3,026 pts diagnosed with operable BC between 2005 and 2013, treated with dds-CT within one randomized and two observational HeCOG trials. Pts with HER2(+) disease had received T for 1 year. Hormonal and radiation therapy were administered, as indicated.The primary endpoints were disease-free survival (DFS) and overall survival (OS).

Results

In total, 59.7% of pts had Hormone receptor (HR)(+)/HER2(-)(luminal) tumors, 25.5% had HER2(+) disease, and 14.5% had triple-negative breast cancer (TNBC). T was administered in 95.5% of pts with HER2(+) disease. At a median follow-up of 7.4 years, the 5-year DFS rate of pts with HER2(+) and luminal tumors was 88%, respectively, as compared to 83% for those with TNBC. The 5-year OS rate was 93% for pts with HER2(+) disease, 92% for luminal and 87% for TNBC. Pts with luminal disease had the greatest 5-year DFS rates both among women with positive and negative nodes (86% and 95%, respectively), followed by pts with HER2(+) BC (83% and 94%, respectively) and TNBC (76% and 89%, respectively).The 5-year OS rate of pts with HER2(+) and luminal disease was 91%, respectively among pts with positive nodes and 97%, respectively in pts with node-negative BC. The 5-year OS rate of TNBC pts with node-negative disease was 92%.

Conclusions

Trastuzumab is a part of one of success stories in Oncology, as its use dramatically improved the prognosis of HER2(+) pts. In this study, the outcomes of pts with HER2(+) BC were similar to those with luminal tumors, while TNBC remained the most unfavorable group.

Legal entity responsible for the study

Hellenic Cooperative Oncology Group (HeCOG).

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The 21-gene recurrence score (RS) assay is included in treatment guidelines for predicting adjuvant chemotherapy benefit for early-stage breast cancer patients with hormone-receptor (HR) positive disease. However, there has been no consideration regarding racial difference of that predictive value, although the assay had developed based on mainly Western women. This study aimed to demonstrate the racial differences in the predictive values of 21-gene RS assay.

Methods

T1-2N0 HR-positive breast cancer patients who had results of 21-gene RS were selected from the Surveillance, Epidemiology, and End Results (SEER) database. Breast cancer-specific mortality (BCSM) was compared between patients who had adjuvant chemotherapy (the “CTx group”) and who did not (the “no CTx group”) to estimate predictive value of the assay. That comparison was repeated within each racial group (Whites, Blacks, other races).

Results

The SEER database included 89,402 T1-2N0 HR-positive breast cancer patients who had results of 21-gene RS. Among them, 13,193 patients had RS 26-100, which includes 10,697 Whites, 1,282 Blacks, and 1,144 other races respectively. Adjuvant chemotherapy was given to 8,364 (63.4%) patients. Adjusted hazard ratio for BCSM in the CTx group (relative to the no CTx group) was 0.732 (95% confidence interval[CI]: 0.587–0.915) in Whites, 0.695 (95% CI: 0.396-1.217) in Blacks, and 1.422 (95% CI: 0.580-3.487) in other races respectively. No subgroup in non-White women showed predictive value of the 21-gene assay within patients with RS 26-100 except Black women with grade 3 disease.

Conclusions

The predictive value of the 21-gene RS assay assessing adjuvant chemotherapy benefit was validated in White women based on the SEER database, although that predictive value was not warranted in non-White women.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The Hannah trial proofed equivalent efficacy of T (600mg q3w) for subcutaneous (sc) injection into the thigh versus (vs) intravenous (iv) administration in early BC. This trial allowed sc injection only into the thigh. The aim of this analysis was to evaluate pts preference for different routes of administration of T.

Methods

Within the GAIN-2 trial a prospective, multicenter, randomized phase II T sc substudy was performed. HER2+ BC pts who have received T iv simultaneously to (neo)adjuvant chemotherapy were randomized (1:1) to receive postoperatively T 600mg sc either into the thigh or abdominal wall (AW). Co-primary endpoint was pts preference for previous iv administration vs sc injection (thigh/AW) and pharmacokinetic profiles of T sc (thigh/AW). Safety, compliance and factors influencing the preferences were also analyzed. Pts preference was assessed by patient interviews (PINT) before randomization (PINT1) and after 8 cycles of T sc (PINT2). A modified intention-to-treat (mITT) analysis was conducted for randomized pts with at least one dose of T sc.

Results

219 pts represented the mITT set (thigh 110; AW 109). Baseline characteristics were well balanced between the treatment groups. 182 pts (83.1%) replied to PINT2. Of them overall 83.5% (95% CI 78.1, 88.9) preferred administration of T sc (thigh 80.6% [95% CI 72.6, 88.7]; AW 86.5% [95% CI 79.4, 93.6], p=0.322). 23 pts (thigh 15; AW 8) had no preference. The expected preferences given in PINT1 (iv/no preference vs sc) showed a significant influence on the preference of the application site given in PINT2 (univariate analysis; OR 1.12 [95% CI 1.03-1.21]; p=0.007). No increased toxicity was observed and the study compliance was comparable (thigh vs AW). T sc injections were generally described as acceptable by the majority of patients.

Conclusions

We confirmed, as previously demonstrated in the PrefHer study (Pivot et al. 2014), that the sc regimen is preferred over iv regimen by pts. There were no safety signals or differences in compliance regarding the different areas of sc injection. Due to higher bioavailability (Möbus et al. 2017) thigh remains the preferred site of injection.

Clinical trial identification

NCT01690702.

Legal entity responsible for the study

GBG.

Funding

Roche.

Disclosure

M. Reinisch: Honoraria (self): Roche. M. Untch: Honoraria (institution): AbbVie; Honoraria (institution): Amgen GmbH; Honoraria (institution): AstraZeneca; Honoraria (institution): BMS; Honoraria (institution): Celgene GmbH; Honoraria (institution): Daiji Sankyo; Honoraria (institution): Eisai GmbH; Honoraria (institution): Lilly Deutschland; Honoraria (institution): Lilly Int; Honoraria (institution): MSD Merck; Honoraria (institution): Mundipharma; Honoraria (institution): Myriad genetics; Honoraria (institution): Odonate; Honoraria (institution): Pfizer GmbH; Honoraria (institution): PUMA Biotechnology; Honoraria (institution): Roche Pharma AG; Honoraria (institution): Sanofi Aventis Deutschland GmbH; Honoraria (institution): TEVA Pharmaceuticals; Honoraria (institution): Novartis; Honoraria (institution): Pierre Fabre. T. Reimer: Honoraria (self): Roche; Honoraria (self): Pfizer; Honoraria (self): AstraZeneca; Honoraria (self): Lilly; Honoraria (self): Novartis. C. Jackisch: Honoraria (self): Roche; Honoraria (self): Celgene. F. Marmé: Honoraria (self): AstraZeneca; Honoraria (self): Roche; Honoraria (self): Pfizer; Honoraria (self): Tesaro; Honoraria (self): Novartis; Honoraria (self): Amgen; Honoraria (self): PharmaMar; Honoraria (self): Genomic Health; Honoraria (self): CureVac; Honoraria (self): Eisai; Honoraria (self): Celgene; Honoraria (self): Clovis; Honoraria (self): Janssen-Cilag; Honoraria (self): Immunomedics; Honoraria (self): GSK; Honoraria (self): MSD. M. Schmidt: Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis; Honoraria (self), Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Pfizer; Research grant/Funding (self), Research grant/Funding (institution): Pierre Fabre. E. Stickeler: Honoraria (self): Roche. V. Möbus: Advisory/Consultancy: Roche; Advisory/Consultancy: Clovis; Advisory/Consultancy: Myelotherapeutics; Travel/Accommodation/Expenses: Celgene. S. Loibl: Honoraria (institution): AbbVie; Honoraria (institution): Amgen; Honoraria (institution): AstraZeneca; Honoraria (institution): Celgene; Honoraria (institution): Novartis; Honoraria (institution): Pfizer; Honoraria (institution): Roche; Honoraria (institution): Seattle Genetics; Honoraria (institution): PriME/Medscape; Honoraria (self): Chugai; Honoraria (institution): TEVA; Honoraria (institution): Vifor; Honoraria (institution): Daiichi -Sankyo; Honoraria (institution): Lilly; Honoraria (institution): Samsung; Honoraria (institution): Eirgenix; Licensing/Royalties: EP14153692.0 pending. All other authors have declared no conflicts of interest.

Background

To describe and characterize cancer-specific health-related quality of life (HRQOL) among patients diagnosed with early stage HR+/HER2- BC.

Methods

A multinational (France, Germany, Italy, Japan, Spain, UK and US) survey of patients diagnosed with stage I-III HR+/HER2- BC was conducted from June to October 2019. Patients were identified by their consulting physician and invited to complete a pen and paper questionnaire. Patients completed the General and Breast version of the cancer-specific Functional Assessment of Cancer Therapy (FACT-G and -B) questionnaires. Mean FACT-G and FACT-B scores (overall and specific subscales) were calculated and compared by disease status (active adjuvant treatment vs. surveillance vs. recurrence) at the time of questionnaire completion using Mann-Whitney and Chi-Squared tests.

Results

1,152 patients completed HRQOL questionnaires (mean age 59 years, treatment status at diagnosis: 76% active adjuvant treatment, 21% surveillance and 3% recurrence). Mean [standard deviation] FACT-G scores (62.2 [16.0]) for the recurrence group were significantly lower than mean scores for disease-free patients in active adjuvant treatment (72.8 [18.3]) or surveillance (71.3 [16.0]) groups. Mean FACT-B scores were also significantly lower for the recurrence group (86.4 [19.5]) compared to active adjuvant treatment (99.4 [22.5]) and surveillance groups (97.7 [19.7]). FACT subscale scores for physical, emotional and functional well-being were also the lowest among the recurrence group (all p-values <0.05).

Conclusions

Group-level differences in cancer-specific HRQOL were statistically significant, with disease-free patients in active adjuvant treatment or surveillance groups reporting higher HRQOL and well-being than patients in the recurrence group. These findings demonstrate the high burden of disease on HRQOL with recurrence among patients with early stage HR+/HER2- BC.

Legal entity responsible for the study

Pfizer Inc.

Funding

Pfizer Inc.

Disclosure

C. Criscitiello: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Pfizer Inc; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eli-Lilly; Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis. D. Spurden: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. A. Rider: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. R. Williams: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. M. Corsaro: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. J. Pike: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. E.H. Law: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc.

Background

To identify and describe clinically meaningful subgroups based on clinical and pathologic disease characteristics during primary diagnosis and treatment of HR+/HER2- eBC.

Methods

Data was analyzed from a multinational (France, Germany, Italy, Japan, Spain, United Kingdom, United States) sample of patients with an initial diagnosis of stage I-III HR+/HER2- BC who were disease-free or in recurrence. LCA was used to cluster patients based on tumour size, nodal status, progesterone receptor status, and histological grade. Kruskal-Wallis and Chi-Squared tested differences across clusters by patient characteristics, treatment, and current disease status (disease-free vs. recurrent disease).

Results

From 1,512 eligible patients, four distinct clusters were identified potentially corresponding to levels of risk for disease-recurrence: Lowest (38%), Moderate (52%), Highest (6%), and Uncertain risk (4%). Risk groups differed by mean age (55 [Highest] to 62 years [Lowest]); node positive status: Lowest (4%), Moderate (47%), Uncertain (33%), Highest (69%); tumour size >5cm: Lowest (0%), Moderate (2%), Uncertain (87%), Highest (67%); and histological grade 2/3 at diagnosis: Lowest (31% / 7%), Moderate (70% / 25%), Uncertain (83% / 2%), Highest (0% / 99%). Stage II and III disease at diagnosis differed by cluster: Highest (21% and 77%), Uncertain (35% and 61%), Moderate (63% and 23%) and Lowest (30% and 4%). Use of neoadjuvant and adjuvant chemotherapy also varied; most common for Highest Risk (93% and 71%, respectively), least for Lowest Risk (79% and 23%, respectively). Highest and Uncertain risk groups were most likely to experience disease recurrence compared to Moderate and Lowest groups. All p-values ≤0.0001.

Conclusions

Patient clusters based on key disease characteristics at initial diagnosis of eBC were identified across 7 countries and may represent clinically relevant subpopulations. Clusters exhibiting the most advanced disease characteristics (Highest and Uncertain) corresponded well to stage II and III disease classification. Future research is required to confirm the predictive validity of these risk groups.

Legal entity responsible for the study

Pfizer Inc.

Funding

Pfizer Inc.

Disclosure

A. Rider: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. D. Spurden: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. R. Williams: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstract: Adelphi Real World. M. Corsaro: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. J. Pike: Full/Part-time employment, Adelphi were paid consultants to Pfizer in connection with the development of this abstact: Adelphi Real World. E.H. Law: Shareholder/Stockholder/Stock options, Full/Part-time employment: Pfizer Inc. C. Criscitiello: Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Pfizer Inc; Advisory/Consultancy, Speaker Bureau/Expert testimony: Eli-Lilly; Advisory/Consultancy, Speaker Bureau/Expert testimony, Travel/Accommodation/Expenses: Roche; Advisory/Consultancy, Speaker Bureau/Expert testimony: Novartis.

Background

The objective of this cross-sectional study is to compare the prevalence of sexual dysfunction in premenopausal women with breast cancer in Greece, who receive adjuvant endocrine therapy with either tamoxifen or aromatase inhibitors (AI), with or without ovarian function suppression (OFS). A second endpoint is to investigate and compare the incidence of genitourinary symptoms and conditions.

Methods

A questionnaire was distributed on hardcopy and through an online platform from November 2018 to March 2019, to Greek-speaking women of 18 years of age or older, who had been receiving adjuvant endocrine therapy for breast cancer for at least three months and were deemed premenopausal/perimenopausal at diagnosis. The questionnaire included investigator-generated items regarding demographics, sexual and gynecologic history, as well as validated instruments for sexual functioning and urogenital tract disorders (FSFI, QLQ-BR23, UDI-6 and PFIQ-7).

Results

Of the 108 received responses, 70 were considered eligible for analysis. Most participants were currently on treatment with tamoxifen/OFS (N=35). Women on AI/OFS reported great deterioration of their sex life compared to women on tamoxifen with OFS (p=0.001). Sexual dysfunction was evident in 74% (N=49) of our participants, as defined by the cutoff value of 26 for the FSFI total score (median: 17.45, IQR=26.18). In particular, women on AI/OFS had significantly lower scores compared to those on tamoxifen with (20.8, 95%CI [14.58;22.67] vs 8.8, 95%CI [4.86;13.60], p=0.040) and without OFS (19.95, 95%CI [13;22.91] vs 8.8, 95%CI [4.86;13.60], p=0.039). Sexual enjoyment of women on AI/OFS was significantly affected compared to women on tamoxifen with or without OFS (p=0.019 and p=0.020, respectively). No differences in vaginal atrophy symptoms or gynecologic conditions were detected.

Conclusions

Sexual dysfunction is highly prevalent in premenopausal women on endocrine therapy, especially in those treated with aromatase inhibitors and ovarian function suppression. Health professionals should promote discussions with their patients in order to decide the optimal treatment choice.

Legal entity responsible for the study

Aristotle University of Thessaloniki, Thessaloniki, Greece Theagenio Cancer Hospital, Thessaloniki, Greece.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

This study reviewed published literature of patient-reported outcomes (PROs) associated with different treatment phases for patients with hormone receptor-positive (HR+), human epidermal growth factor 2-negative (HER2-) early breast cancer (BC).

Methods

A systematic literature review was conducted according to PRISMA guidelines to identify studies evaluating PROs among patients aged ≥18 years and diagnosed with HR+/HER2- eBC. Ovid MEDLINE, Embase, and Cochrane Evidence Based Medicine databases were searched between 2005 to 2019. Reported PRO measurements were summarized and categorized based on association with treatment (neoadjuvant, adjuvant, surgery), remission (or disease-free), and recurrence.

Results

Of 3,622 records evaluated, 10 studies reported PROs for HR+/HER2- eBC patients. Of these, 8 studies reported on cancer-specific health-related quality of life (HRQOL), 3 on general anxiety, and 5 on decisional conflict. Study treatments in the neoadjuvant setting included endocrine and targeted therapies. In the adjuvant/remission setting, studies investigated the use of either prognostic tests or hydrotherapy. Baseline and on-treatment HRQOL were available among patients receiving neoadjuvant therapy (4 studies) using the Functional Assessment of Cancer Therapy (General, Breast, and Endocrine Subscale modules) and European Organization for Research and Treatment of Cancer Quality of Life (EORTC QLQ) C30 and -BR23 questionnaires. In the adjuvant setting (5 studies), baseline and post-administration of prognostic tests using general anxiety (State-Trait Anxiety Inventory) and/or decision conflict (Decision Conflict Scale) were measured. One study of hydrotherapy used the EORTC QLQ-BR23 to measure HRQOL during remission. No studies evaluated the relationship between PROs and disease recurrence.

Conclusions

Current literature indicates that PROs are either measured prior to surgery, during adjuvant therapy, or during remission with a lack of studies evaluating PROs with disease recurrence. Future research is needed to fully articulate the disease- and treatment-related experiences across the patient journey in HR+/HER2- eBC.

Legal entity responsible for the study

Pfizer Inc.

Funding

Pfizer Inc.

Disclosure

E.M. Salvo: Full/Part-time employment, Eversana were paid consultants to Pfizer in connection with the development of the abstract’: Eversana. J. Cueto: Full/Part-time employment, Paid contractor to Pfizer in the development of this abstract: Pfizer Inc. E.H. Law: Shareholder/Stockholder/Stock options, Employee of Pfizer Inc. Shareholder/Stockholder/Stock options in Pfizer Inc.: Pfizer Inc. C. Aniceto Da Silva: Full/Part-time employment, Eversana were paid consultants to Pfizer in connection with the development of the abstract’: Eversana. C. Cameron: Full/Part-time employment, Eversana were paid consultants to Pfizer in connection with the development of the abstract’: Eversana. I.A. Samjoo: Full/Part-time employment, Eversana were paid consultants to Pfizer in connection with the development of the abstract: Eversana.

Background

Gene expression profiles (GEP) represents a relevant tool to evaluate the best adjuvant approach for localised luminal breast cancer (BC) patients. Several platforms have been already validated for menopausal N0 patients, but the use among pre-menopausal, N+ population still needs further validation. The results obtained with Mammaprint® has suggested the possibility to adopt GEP to drive treatment decisions also in N+ population. Nevertheless, Prosigna™ is not yet prospectively validated to assess risk and personalise treatment in these patients.

Methods

To evaluate the impact in adjuvant treatment decision of Prosigna™ among BC patients diagnosed with N1a disease, a retrospective analysis was performed in our institution. Between November 2015 and November 2019, 56 patients (P) were finally selected and analysed as they met all the inclusion criteria, clinical characteristics are shown in the table. The objectives of the study were to evaluate the role of Prosigna in the decision-making process, and to identify, among all the parameters used to obtain the risk of recurrence (ROR) score, such as group of risk, intrinsic subtype and probability of relapse at 10 years, the parameter which mostly affected treatment decision.

Results

All the 56 P included would have been candidate to receive chemotherapy (CT) according to clinical practice. Using Prosigna™, it was possible to stratify risk and personalize the treatment by avoiding chemotherapy in in 29 P (51,8%). After a median follow-up of 18 months [range 3-43] no relapses were detected. Table: 91P