Background

Analytical, functional, and pharmacokinetic similarity between ABP 980 and trastuzumab (TRAS) has been demonstrated. Here we report the results of primary efficacy analysis in the corresponding clinical study.

Methods

The objective of this randomized, multicenter, double-blind study was to compare ABP 980 with TRAS on pathologic complete response (pCR) in women with HER2 positive early breast cancer. After run-in anthracycline-based chemotherapy, patients were randomized 1:1 to intravenous ABP 980 or TRAS plus paclitaxel Q3W for 4 cycles. Patients had to complete a full cycle of run-in therapy to be eligible for randomization. Patients continued to the adjuvant phase on IP Q3W for up to 1 year. The co-primary endpoints were risk difference (RD) and risk ratio (RR) of pCR in breast tissue and axillary lymph nodes of tumor samples. Clinical similarity was confirmed if the 2-sided 90% CIs for RD and RR were within the bioequivalence margin of -13% to 13% for RD and 0.759 to 1.318 for RR. Secondary endpoints included safety.

Results

Of the 827 enrolled patients, 725 were randomized (ABP 980: n = 364; TRAS: n = 361); 696 (ABP 980: n = 358; TRAS: n = 338) were included in the pCR evaluable population. Based on local review, 48.0% and 40.5% of patients in the ABP 980 arm and TRAS arm, respectively, achieved pCR. RD and RR of pCR were 7.3% (90% CI: 1.2%, 13.4%) and 1.19 (90% CI: 1.033, 1.366), with the upper bound CI slightly exceeding the equivalence margin. Based on central independent review, 47.8% and 41.8% in the ABP 980 arm and TRAS arm achieved pCR. RD and RR of pCR were 5.8% (90% CI: -0.5, 12.0%) and 1.14 (90% CI: 0.993, 1.312), contained within the equivalence margin. 292 (80.2%) and 287 (79.5%) in the ABP 980 and TRAS arm, respectively, had ≥1 adverse event (AE); 54 (14.8%) patients in the ABP 980 arm and 51 (14.1%) in the TRAS arm had a Grade ≥3 AE. Most common AEs (ABP 980 vs TRAS) were arthralgia (17.3% vs 15.2%), asthenia (14.8% vs 16.3%), neutropenia (14.6% vs 12.5%), peripheral neuropathy (13.7% vs 11.9%), and anemia (11.0% vs 10.2%).

Conclusions

Results of this study show clinical equivalence of ABP 980 and TRAS in the neoadjuvant setting and add to the totality of evidence demonstrating similarity between ABP 980 and TRAS.

Clinical trial identification

20120283

Legal entity responsible for the study

Amgen Inc.

Funding

Amgen Inc.

Disclosure

G. von Minckwitz: Research grants to the institution from Pfizer, Roche, Celgene, Novartis, AstraZeneca, Teva, Amgen. N. Zhang, V. Hanes: Full-time employee and stockholder of Amgen Inc. All other authors have declared no conflicts of interest.

Background

CT-P6 is a proposed biosimilar to Reference Trastuzumab (RTZ). This trial (2013-004525-84) evaluated the similarity of CT-P6 and RTZ in efficacy and safety for HER2+ EBC.

Methods

549 patients with HER2+ EBC were randomized to receive CT-P6 (n = 271) or RTZ (n = 278) in combination with docetaxel (Cycles 1-4) and 5-fluorouracil, epirubicin and cyclophosphamide (Cycles 5-8). CT-P6 or RTZ was administered at 8 mg/kg (Cycle 1 only) followed by 6 mg/kg every 3 weeks. After surgery, patients received CT-P6 or RTZ monotherapy up to 10 cycles. The primary endpoint was pathological complete response (pCR) rate at surgery. Secondary endpoints were overall response rate (ORR), PK, PD and safety.

Results

The pCR rate was 46.8% in CT-P6 and 50.4% in RTZ. The 95% CIs for the estimate of treatment difference were within the equivalence margin (±0.15) in both PPS and ITT. The proportion of patients with at least 1 SAE was 7.4% in CT-P6 and 11.9% in RTZ over 1-year treatment. 6 patients (3 in CT-P6 and 3 in RTZ) withdrew treatment due to significant LVEF decrease. Infusion related reaction was reported for 11.4% of patients in CT-P6 and 10.4% of patients in RTZ.Table:

152PD Summary of efficacy endpoints

Conclusions

This study demonstrated the equivalence of efficacy between CT-P6 and Reference Trastuzumab in EBC patients. Secondary efficacy endpoints also supported the similarity for two study drugs. CT-P6 was well tolerated with a similar safety profile to that of Reference Trastuzumab through the neoadjuvant and adjuvant period.

Clinical trial identification

EudraCT Number: 2013-004525-84 NCT Number: NCT02162667

Legal entity responsible for the study

Celltrion, Inc.

Funding

Celltrion, Inc.

Disclosure

F.J. Esteva: Consulting Role: Celltrion, Inc. G. Dzagnidze: Research funding: Celltrion, Inc., Roche, AstraZeneca Pharmaceutical, Inc. A. Eniu: Research Funding: Roche, AstraZeneca, Celltrion, Inc., Pfizer. Travel, Accommodations, Expenses: Roche, AstraZeneca, Teva. G. Morar-Bolba: Consulting or Advisory Role: Pierre Fabre. Travel, Accodomations, Expenses: Roche. R.K. Li: Research Funding: Pfizer, GlaxoSmithKline, AstraZeneca, Novartis, Celltrion, Inc. S.J. Lee: Employment, Board of Directions, Stock ownership, Travel, Accommodations, Expenses: Celltrion, Inc. S. Yu: Employment, Stock ownership, Travel, Accommodations, Expenses: Celltrion, Inc. J. Stebbing: Consulting or Advisory Role: Celltrion, Inc. All other authors have declared no conflicts of interest.

Background

Equivalence for efficacy between SB3 (a proposed trastuzumab biosimilar) and originator trastuzumab (TRZ) in terms of breast pathologic complete response (bpCR) rates has been demonstrated and previously reported.¹ Here we present the one-year results on safety, immunogenicity, event-free-survival (EFS), and overall survival (OS).

Methods

Study compared neoadjuvant SB3 or TRZ for 8 cycles concurrently with chemotherapy (4 cycles of docetaxel followed by 4 cycles of 5-fluorouracil/epirubicin/cyclophosphamide). Patients then underwent surgery followed by 10 cycles of adjuvant SB3 or TRZ as randomised. The primary endpoint was bpCR rate and secondary endpoints included safety, immunogenicity, EFS, and OS up to the adjuvant period.

Results

A total of 875 patients were randomised with a median follow-up duration of 438 days, and 765 patients completed adjuvant therapy (SB3, N = 381; TRZ, N = 384). Incidences of treatment emergent adverse events (TEAEs) were comparable between arms (Table). Most frequently occurring TEAEs were alopecia, neutropenia, and nausea during the neoadjuvant period and radiation skin injury, procedural pain, and fatigue during the adjuvant period. EFS rates were 92.2% in SB3 and 91.6% in TRZ (hazard ratio 0.94; 95% CI, 0.59 to 1.51). There were a total of 6 deaths (SB3, N = 1; TRZ, N = 5). Immunogenicity was low and comparable, with anti-drug antibody positive for 3 patients (0.7%), in each arm.Table:

153PD Safety profile

Conclusions

One-year safety, immunogenicity, and survival results further support the biosimilarity established between SB3 and TRZ. Reference: 1. Pivot X et al. ASCO 2017, ID50

Clinical trial identification

EudraCT Number: 2013-004172-35 ClinicalTrials.gov NCT02149524

Legal entity responsible for the study

Samsung Bioepis Co., Ltd.

Funding

Samsung Bioepis Co., Ltd.

Disclosure

X. Pivot: Consultant with honorarium for Samsung Bioepis Co., Ltd. L. Younju, J. Lim: Employee of Samsung Bioepis Co., Ltd. Stock ownership of Samsung Biologics. All other authors have declared no conflicts of interest.

Background

This comparative clinical trial evaluated efficacy, safety, immunogenicity and pharmacokinetics (PK) of PF-05280014, a potential trastuzumab biosimilar, vs trastuzumab sourced from the EU (trastuzumab-EU), both given with D and C, as neoadjuvant treatment for operable HER2+ breast cancer.

Methods

Patients (pts; N = 226) were stratified by primary tumor size and hormone receptor status and randomized 1:1 to receive PF-05280014 or trastuzumab-EU (8 mg/kg at Cycle 1; 6 mg/kg thereafter), both with D (75 mg/m²) and C (target AUC 6), every 3 wks for 6 treatment cycles. The study was powered to test whether PF-05280014 was noninferior to trastuzumab-EU in the percentage of pts with Cycle 5 C_trough (pre-dose Cycle 6) >20 μg/mL. Efficacy was measured by the percentage of pts with pathological complete response (pCR), defined as the absence of invasive neoplastic cells in breast and lymph nodes after neoadjuvant therapy, and objective response rate (ORR). Safety and immunogenicity were also assessed.

Results

The percentage of pts with Cycle 5 C_trough >20 μg/mL was 92.1% for PF-05280014 and 93.3% for trastuzumab-EU; the lower limit of the 95% CI (-8.02% to 6.49%) for the stratified difference between groups was above the noninferiority margin (-12.5%). The pCR rate was 47.0% (95% CI: 36.9-57.2) for PF-05280014 and 50.0% (95% CI: 39.0-61.0) for trastuzumab-EU. Central radiology review-assessed ORR was 88.1% (95% CI: 80.2-93.7) for PF-05280014 and 82.0% (95% CI: 72.5-89.4) for trastuzumab-EU. All causality, grade 3-4 treatment-emergent adverse events were reported by 38.1% (PF-05280014) vs 45.5% (trastuzumab-EU) of pts. No pts in the PF-05280014 and 1 (0.89%) in the trastuzumab-EU group had positive anti-drug antibody titer.

Conclusions

PF-05280014 demonstrated similarity in efficacy, safety and immunogenicity, and noninferiority in PK to trastuzumab-EU. A separate comparative safety and efficacy study (NCT01989676) is evaluating PF-05280014 vs trastuzumab-EU, both given with paclitaxel, as first-line treatment for HER2+ metastatic breast cancer.

Clinical trial identification

NCT02187744; EudraCT No: 2013-004679-11

Legal entity responsible for the study

Pfizer Inc.

Funding

Pfizer Inc.

Disclosure

P.E. Lammers: Advisory boards with Pfizer Inc. M. Dank: Member of Biosimilars Oncology European Advisory Board with Pfizer Inc since 2013. R. Abbas, F. Hilton, J. Coiro, I. Jacobs: Full time employee and lares stock holdings and/or stock options from Pfizer Inc. All other authors have declared no conflicts of interest.

Background

Increased number of TILs at baseline is associated with pathological complete response (pCR) and improved outcomes in HER2+ early breast cancer (BC) treated with anti-HER2-based chemotherapy. The associations in the neoadjuvant setting in the absence of chemotherapy and the effect of on-treatment TILs changes on pCR in the breast (pCR_B) are unknown.

Methods

PAMELA is a prospective study in HER2+ BC designed to evaluate the ability of the PAM50 intrinsic subtypes (IS) to predict pCR_B following neoadjuvant lapatinib and trastuzumab (with hormonal therapy if hormone receptor-positive[HR+]). Levels of TILs as continuous and categorical (TILs-low = <50%, TILs-high = >50%) variables and their changes were correlated with pCR_B.

Results

TILs evaluation was available for 148 baseline (BS) and 134 Day-15 (D15) samples of 151 recruited patients. At BS, the median (interquartile range) levels of TILs were 10% (5-20). Median TILs distribution according to IS was: HER2-E (10%), Luminal (Lum) A (7.5%), LumB (5%), Basal-like (5%) (p = 0.02). Levels of TILs were higher in HR- (10%, 1-20) vs HR + (5%, 1-20) tumors, although not statistically significant (p = 0.07). pCR_B rates were 58.3% (7/12) for TILs-high and 27.2% (37/136) for TILs-low (p = 0.03). At baseline, TILs were significantly associated with pCR in univariate analysis. At D15, median levels of TILs were 15% (5-30) with an increase across all the different IS (p < 0.01). The distribution across IS at D15 was: HER2-E (20%), Lum A (10%), Lum B (7.5%), Basal-like (20%) and Normal-Like (10%) (p = 0.37). At D15, TILs-high tumors showed a pCR_B rate of 65% (13/20) vs 21.1% (24/114) of TILs-low (p < 0.01). As a continuous variable, higher TILs levels at D15 and changes of TILs levels from BS to D15 were associated with higher pCR_B rates independently of HR status and IS (p < 0.01). When analysis was performed for HR-negative and HR-positive patients, separately in both cohorts, TILs at D15 was significantly associated with pCR.

Conclusions

The presence of TILs at D15 is an independent predictive marker of pCR_B in HER2+ early BC treated with neoadjuvant anti-HER2 agents without chemotherapy.

Clinical trial identification

NCT01973660

Legal entity responsible for the study

SOLTI Breast Cancer Research Group

Funding

GlaxoSmithKline (now Novartis)

Disclosure

All authors have declared no conflicts of interest.

Background

Identifying early breast cancer (eBC) patients that require adjuvant chemotherapy is defined by clinical features, which do not accurately identify high-risk patients. The USO01062 Phase 3 study enrolled 2,611 high-risk patients based on clinical features, with a 13% event rate observed. Comparing the genomic landscape of tumors between patients who had a recurrence event and those who did not may uncover genomic aberrations associated with recurrence that may be used to identify high-risk patients. The genomic landscape of TNBC subtypes is also largely unknown and NGS profiling may shed light on novel therapeutic opportunities.

Methods

The USO01062 study failed to show a benefit for the addition of capecitabine to adjuvant chemotherapy (O'Shaughnessy J. et al. 2015). Arms were pooled and DNA and RNA were extracted from 1,181 tumor samples, of which 145 patients had a DFS event, and were matched demographically to a set of 146 patients without an event for targeted NGS profiling using FoundationOne®. Gene expression was previously run using a breast cancer specific 800-gene panel (Wilson T.R. et al. 2016).

Results

Analysis of somatic alterations within IHC subtypes identified unique prognostic factors, e.g. alterations in ATM, ERCC4 and IGF2R correlated with a worse HR in HR+ disease, whereas alterations in MAP3K1, RPTOR and LYN correlated with a worse HR in TNBC. Analysis of tumor mutational burden (TMB) revealed TNBC tumors had the highest burden, which did not correlate with clinical outcome or expression of PDL1 and CD8 genes. Molecular subtyping of TNBC (Lehmann B.D. et al. 2011) found distinct genetic drivers in each subtype, e.g. alterations in TP53 and MYC were the most frequent in BL1 and BL2 tumors. IM tumors expressed alterations in TP53, CREBBP and BRCA1. LAR tumors expressed alterations in PIK3CA and PTEN.

Conclusions

TMB was not prognostic and did not correlate with PDL1 or CD8 gene expression, suggesting that TMB in TNBC may not be a surrogate for the immune activated subtype. TNBC molecular subtyping identified different genomic drivers providing evidence for genomic heterogeneity within subtypes. Lastly, comparison of patients that experienced a DFS event identified genomic alterations that may be used to identify high-risk patients.

Clinical trial identification

Patients were enrolled onto the parent study USO01062, (NCT00089479).

Legal entity responsible for the study

Hoffmann-La Roche

Funding

Genentech, Inc.

Disclosure

T.R. Wilson, A.R. Udyavar, C-W. Chang, J. Giltnane, J.M. Spoerke, J. Aimi, H. Savage, A. Daemen, R. Bourgon, M.R. Lackner: Employed by Genentech, Inc. Stocks in Roche

Background

In 2004 the 70-gene signature, MammaPrint® (MP), developed to predict High or Low Risk of distant breast cancer (BC) recurrence, was introduced in the observational RASTER trial. Patients (cT1–3N0M0) and their doctors took the clinical Dutch guideline and MP in to account to decide on adjuvant systemic treatment (AST). Five years follow-up data confirmed the prognostic value of the MP (Drukker, Int J Cancer, 2013). In this analysis we report the outcome at 10 years.

Methods

Ten year survival data was available for all 427 Raster patients, age < 61. For the current analysis, clinical high (C-high) or low (C-low) risk was scored according to the modified version of Adjuvant! Online (Cardoso, N Engl J Med, 2016). 10-year distant-recurrence-free-interval (DRFI) probabilities were compared between risk groups based on the 70-gene signature and clinical assessment.

Results

The 70-gene signature identified 51.4% (219/427) patients with a genomic Low Risk of BC recurrence (G-low). 10-year DRFI in patients with G-low or genomic High Risk (G-high) was 93.7% and 86.8% respectively (HR 1.4; 95% confidence interval{CI] 1.0-1.9). Clinical assessment identified 57% as C-low. The 10-year DRFI was 91.7% in C-low and 88.2% in C-high (HR 1.4; 95%CI 0.8-2.6). The 10-year DRFI in the combined genomic and clinical riskgroups was 94.4% in patients with a C-low/G-Low profile, only 11.6% of them received AST. In the C-low/G-High group 10-years DRFI was 88.5%, over 90% of them received AST. For C-high risk patients 10-year DRFI was 90.9% if G-Low (n = 46) and 87.3% if G-High (n = 137). In ER-positive BC (ER+) (N = 342) 10-years DRFI was 93.6% (G-Low) versus 88.8% (G-High) (HR 1.6; 95%CI 0.8-3.3). With clinical risk assessment, 10-years DRFI in ER+ was 91.6% (C-low) versus 91.9% (C-high).

Conclusions

Patients who omitted chemotherapy based on MammaPrint Low Risk had an excellent 10 year DRFI, confirming the prognostic value of the MP. When C-high, the MP identified another 10.8% (46/426) of patients as G-Low who might forego adjuvant chemotherapy if ER+. In contrast to genomic risk stratification, the clinical risk assessment was unable to differentiate for survival between ER+ C-high and C-low risk patients.

Clinical trial identification

not applicable

Legal entity responsible for the study

S. Linn

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

Optimal prognostic markers and survival impact of A-containing CT in HER2-, particularly HR+ EBC, are still a matter of debate.

Methods

WSG PlanB is the first prospective trial compared A-free 6xTC vs. standard 4xECà4xDoc in high-risk pN0 or pN+ HER2- EBC and RS > 11 in pN0-1 HR+/HER2- EBC. Primary endpoint was DFS. Secondary endpoints included DDFS/LRFS and overall survival (OS). Multivariable analysis included continuous RS, IHC4, ER, PR, Ki67 (centrally measured), local/central grade, age, nodal status, tumor size, treatment, surgery type.

Results

From 2009 to 2011, PlanB enrolled 3198 pts. 2449 pts. were randomized. to 6xTC vs. 4xEC→4xDoc (n = 1222/1227). After 60 months median follow-up, similar 5y DFS of 90%, 5y DDFS of 94%/93% (TC/EC-Doc) and 5y OS of 95% were observed in both arms. Pts with RS 0-11 without CT (n = 348, 15.3%) had 5-year DFS of 94%, irrespective of nodal status. 5y DDFS was 97.8% in RS 0-11 pts (all and only ET-treated: pN0: 97.7%, pN1: 97.9%), 96.9% in RS12-25 and 89.7% in RS > 25 (p < 0.001 for RS > 25 vs. either RS 0-11 or RS 12-25). RS was the strongest independent predictor for DDFS (HR = 2.64; 95% CI: 1.67-4.17) in multivariable analysis (all and CT-treated); the model also included tumor size>2 cm, nodal status, local grade 3. 5y LRFS was 98.6% in RS 0-11, 99.3% in RS 12-25 and 98.0% in RS > 25 (p = 0.002 for RS > 25 vs. RS 12-25, p = 0.21 for RS > 25 vs. RS 0-11). In multivariable analysis Ki67 (HR = 3.62, 95% CI: 1.39-9.47), together with >pN2 and mastectomy (all and CT-treated) predicted LRFS.

Conclusions

We observed excellent 5y DDFS of 97.8% in clinically high/genomic low risk HER2- HR+ EBC pts treated by ET alone. TC and EC-Doc resulted in similar outcomes. High RS was predictive for poor DDFS and LRFS by univariable, and for DDFS by multivariable analysis; Ki67 was independently prognostic for LRFS. These first prospective data regarding the impact of RS on DDFS and LRFS in clinically intermediate/high-risk EBC underline the importance of using both clinicopathological and genomic markers for individualized therapy decisions.

Clinical trial identification

EUDRA-CT Number: 2008-004263-19

Legal entity responsible for the study

West German Study Group

Funding

Amgen; Sanofi-Aventis; Genomic Health

Disclosure

O. Gluz: Honoraria and/or travel support (GHI, Nanostring, Amgen, Roche). U. Nitz, S. Kümmel, C. Liedtke, S. Shak, H.H. Kreipe: Genomic Health. R. Wurstlein, N. Harbeck: Genomic Health, Agendia, Nanostring. All other authors have declared no conflicts of interest.

Background

In premenopausal pts with HER2+ EBC, the prognostic effect of TIA is unknown and the gonadotoxicity of trastuzumab (T) and lapatinib (L) remains largely uncertain. We aimed to assess the prognostic effect of TIA and the impact of T and/or L on the risk of developing TIA in premenopausal pts with HER2+ EBC.

Methods

ALTTO was an international, open-label, randomised phase 3 trial in pts with HER2+ EBC. Pts were randomised in 4 adjuvant anti-HER2 arms: T alone, L alone, a sequence of the 2 agents (T->L) and their combination (T+L). As per study protocol, menopausal status was collected in all pts at randomisation and at week 37. By selecting only premenopausal pts at randomisation, we investigated whether TIA in pts with hormone receptor-positive (HR+) and negative (HR-) EBC would impact on disease-free (DFS) and overall survival (OS), and the risk factors for developing TIA. Landmark and time-dependent modeling were used to account for guaranteed time bias.

Results

Out of 8381 pts randomised in ALTTO, 2862 were included in this analysis. Median age was 43 years (range 38-47); 1679 (59%) pts had HR+ EBC. Pts with HR+/HER2+ EBC who experienced TIA had significant better DFS (hazard ratio [HR] 0.64; 95% confidence intervals [CI] 0.52-0.79) and OS (HR 0.53; 95% CI 0.38-0.74) than those who did not have TIA. By contrast, pts with HR-/HER2+ EBC had similar DFS (HR 0.85; 95% CI 0.68-1.07) and OS (HR 0.89; 95% CI 0.64-1.25) regardless of whether they had TIA (interaction P for DFS=0.009 and for OS = 0.002). A similar TIA rate was observed in the T (72.6%), L (74.0%), T->L (72.1%) and 74.8% (T+L) arms (p = 0.644). Older age (p < 0.001), addition of taxanes to anthracycline-based chemotherapy (p < 0.001), and use of adjuvant endocrine therapy (p < 0.001) significantly increased the risk of TIA.

Conclusions

In premenopausal pts with HR+/HER2+ EBC, TIA was associated with significant survival benefits. Anti-HER2 agents did not impact the likelihood of developing TIA. These data are of great importance in oncofertility counseling and support the use of ovarian suppression as part of adjuvant endocrine therapy in premenopausal HR+/HER2+ EBC pts.

Clinical trial identification

The trial is registered with the clinicaltrial.gov identifier, number NCT00490139.

Legal entity responsible for the study

Novartis Pharma AG, Basel, Switzerland

Funding

Novartis Pharma AG and the National Cancer Institut of the National Institutes of Health.

Disclosure

E. De Azambuja: Honoraria from Roche. Travel grants from Roche and GlaxoSmithKline outside the submitted work.

All other authors have declared no conflicts of interest.

Disclosure

E. De Azambuja: Honoraria from Roche. Travel grants from Roche and GlaxoSmithKline outside the submitted work.

All other authors have declared no conflicts of interest.

Background

Exploration of neoadjuvant chemotherapy with anti-HER2 therapy to achieve higher pathological complete response (pCR) is important. We focused on trastuzumab emtansine (T-DM1), pertuzumab (P) and tailored response guided therapy in HER2+ primary breast cancer (cT1c–T3, cN0–N1, M0, tumor ≤ 7 cm).

Methods

This randomized phase II study evaluated efficacy and safety of 3 regimens, A) 6 cycles TCbHP, B) 4 cycles TCbHP followed by 4 cycles T-DM1+P, and C) 6 cycles T-DM1+P: responders to 4 cycles T-DM1+P were assigned to 2 more cycles [C1] or non-responders were assigned to 4 cycles FEC [C2] (Table). Responders: ≥30% decrease in tumor size (MRI) and Ki67 ≤10% or no cancer cells in core needle biopsy. ER(+) patients received anti-hormonal therapy concurrent to T-DM1+P. HER2, ER and Ki-67 were assessed in a central laboratory. Primary endpoint was pCR rate (yT0-is, yN0; centrally confirmed). Secondary endpoints were safety, ORR and breast conservation rate.

Results

A total of 206 patients were enrolled (Aug 2014-Feb 2016; full analysis set, n = 204). Patient characteristics were comparable among all groups (median age 53.0 y, post-menopause 53.9%, T2 70.6%, median tumor size 26 mm, N0 63.2%, ER(+) 57.8%). In group C, 80 (79.2%) patients continued T-DM1+P due to favorable response. pCR rate in group A, B, and C was 56.9%, 71.2%, and 57.4%. By exploratory analyses, pCR rate was higher for groups B and C1 than A in ER(+), but comparable in ER(-) patients. No significant differences in secondary endpoints. No treatment discontinuation due to AEs and similar drug-related SAE profile were seen among groups. Of specific mention: low drug-related alopecia in group C1 (5.0%) than A, B or C2 (81%–94%) and less febrile neutropenia in C1 (0%) than A, B or C2 (15%–33%).Table:

159PD Summary of response

Conclusions

Addition of T-DM1+P to standard TCbHP regimen may be possibly superior to TCbHP. Tailored T-DM1+P is a promising approach with mostly equal efficacy and less toxicity compared to TCbHP.

Clinical trial identification

UMIN-CTR: UMIN000014649

Legal entity responsible for the study

Japan Breast Cancer Research Group

Funding

Japan Breast Cancer Research Group and Chugai Pharmaceutical Co., Ltd.

Disclosure

N. Masuda: Honorarium from Chugai and AstraZeneca. T. Takano, M. Toi: Funding from Chugai. Y. Ito: Funding from MSD, AstraZeneca, Novartis, Parexel, Chugai, and Lilly. H. Kasai: Consulting fee/honorarium from Chugai. T. Takasuka: Employee and stock options with Chugai. S. Morita: Consultancy for Chugai. All other authors have declared no conflicts of interest.

Background

Breast cancer (BC) brain metastases (BCBM) are emerging as a major factor of morbidity and poor overall survival. We hypothesized that a clinical risk prediction model could predict BCBM and allow the selection of an enriched early-stage BC patient population, who are at higher risk for BCBM, for preventive trials.

Methods

Electronic medical records were retrospectively reviewed for patients diagnosed and treated for early-stage BC at MD Anderson Cancer Center between 1997 and 2014 under a study approved by the institutional review board. The clinicopathologic prognostic features selected for analysis are: age, HER-2 receptor status and hormone receptor (HR) status, tumor histology, grade, and stage, menopausal status, vascular and lymphatic invasion. A multivariate Cox proportional hazards regression analysis was conducted.

Results

A total of 15164 patients with complete data for key variables were studied. Patients were randomly split 2:1 into training and validation sets. Of the 10026 patients in the training set, 317 developed BCBM and of the 5138 in the validation set, 133 developed BCBM. The 10-year estimated risk of brain metastasis was 4.2% (95% CI, 3.7% to 4.7%) in the training set. Younger age, HER-2 negative and HR-negative receptor status, higher tumor stage and grade, were all significantly and independently associated with BCBM. The risk prediction model had an estimated Harrell’s concordance index of 81% (95%CI, 77% to 86%) in the validation set. In the 10% of validation set patients predicted to have the highest risk, the 10-year risk of brain metastasis was 15% (95% CI, 11% to 18%).

Conclusions

This risk prediction model for brain metastasis risk in early BC allows us to: (1) Identify the significant clinical risk factors for BCBM, (2) use these risk factors to develop an individualized risk score for BCBM, and (3) use this score to select patients at higher risk of BCBM to be prioritized for preventive trials.

Legal entity responsible for the study

Nuhad K. Ibrahim

Funding

Sheril Wynne Research Fund.

Disclosure

All authors have declared no conflicts of interest.

Background

In cancer randomized controlled trials (RCT), endpoints other than overall survival (OS) such as disease-free survival (DFS) are increasingly being used as primary endpoint. Their development is influenced by the need to reduce the number of patients, the duration and ultimately the cost of the trials. Their use as primary endpoint however require a rigorous validation as surrogate endpoints for OS. In adjuvant breast cancer, only a few studies evaluated surrogate endpoints for OS, limited by the use of aggregate data. We present the results of a meta-analysis of individual-patient data assessing surrogate endpoints for OS in the context of adjuvant breast cancer.

Methods

We evaluated four endpoints as potential surrogates for OS relying on a meta-analysis of 5 phase III trials: relapse-free survival (RFS), invasive DFS, locoregional RFS and distant DFS. At the patient level, we estimated the individual-level associations by jointly modeling each surrogate with OS using a copula function. At the trial level, we estimated the association between the treatment effects using (1) a linear regression model weighted by the trial size and (2) the two-step model proposed by Burzykowski, Molenbergh and Buyse.

Results

We gathered individual-patient data from 11676 patients from 5 RCT. The endpoints were highly associated with OS at the patient level (r ≥ 0.98). At the trial level, invasive DFS showed the higher association with OS (R²_WLR=0,78; R²_2SM=0,87).

Conclusions

This is the first meta-analysis on individual-patient data evaluating surrogate endpoints for OS in adjuvant breast cancer. These results suggest that the endpoints could be interesting candidate surrogates for OS, but further evaluation on a larger set of trials is required to improve the precision of our estimations of the trial-level associations.

Legal entity responsible for the study

ICM Regional Cancer Center of Montpellier

Funding

French National Insitute of Cancer (INCa)

Disclosure

All authors have declared no conflicts of interest.

Background

Triple-negative breast cancer (TNBC) is an aggressive subtype of BC in need for predictive biomarkers of response to neoadjuvant chemotherapy (NACT). We aimed to evaluate the predictive value of the TNBCtype-4 classifier in a population of TNBC treated with carboplatin and docetaxel (TCb).

Methods

Patients with stage I-III TNBC (ER and PR < 1%, HER2-negative) were accrued in a non-randomized trial of neoadjuvant carboplatin AUC 6 and docetaxel 75 mg/m2 q3 weeks (NCT01560663). Pathological complete response (pCR) was defined as the absence of invasive tumor in the breast and axillary lymph nodes (ypT0/is ypN0) and residual disease was evaluated using the residual cancer burden (RCB) method (Symmans et al 2007). RNAseq was performed from FFPE-extracted mRNA from the basal core biopsy, and RNAseq data was uploaded into the TNBCtype online tool (http://cbc.mc.vanderbilt.edu/tnbc). Correlation studies were analyzed with R studio v 3.2.1.

Results

RNAseq was available for 94 of the 121 patients enrolled. Patients included had a median age at diagnosis of 51 years (range 28-78), 69.1% had nodal involvement and 52.1% and 46.8% had stage II and III disease, respectively. pCR rate and pathological good response (pCR or RCBI) were 44.7% and 56.4%. TNBCtype-4 distribution was: 34.0% BL1, 20.2% BL2, 23.4% M and 14.9% LAR. An additional 7.4% were classified as ER-positive. BL1 was associated with a significant younger age at diagnosis and higher ki67 values. TNBCtype-4 showed a significant association with response to NACT (p = 0.027), even in multivariate analysis including tumor size and nodal status, with BL1 patients achieving the highest pCR rate (65.6%), followed by BL2 (47.4%). Conversely, LAR and ER-positive showed the lowest pCR rates, 21.4% and 14.3%. When compared to BL1, LAR and M subtypes had an OR of achieving a pCR of 0.14 and 0.30 respectively (p < 0.05).

Conclusions

TNBCtype-4 shows a significant predictive value of response in a TNBC cohort homogeneously treated with TCb, with BL1 and LAR displaying the best and worse responses to NACT respectively.

Clinical trial identification

NCT01560663

Legal entity responsible for the study

Hospital General Universitario Gregorio Marañon

Funding

Institute of Health Carlos III (PI12-02684)

Disclosure

All authors have declared no conflicts of interest.

Background

Patients (pts) with TNBC involved in the GeparSixto study showed an improved pCR rate (ypT0 ypN0) with the addition of carboplatin (Cb) to anthracycline/taxane-based neoadjuvant chemotherapy, which translated in an improved early disease-free survival (DFS). No difference was observed in the HER2+ subgroup for pCR and DFS by adding Cb. Here, we present the results on the long-term survival analysis.

Methods

In the GeparSixto trial, pts were treated for 18 weeks with paclitaxel 80mg/m² q1w and non-pegylated-liposomal doxorubicin (NPLD) 20mg/m² q1w (PM), concurrently with bevacizumab 15mg/kg q3w if TNBC or trastuzumab 6(8)mg/kg q3w and lapatinib 750mg daily if HER2+. 595 pts were randomised 1:1 to receive concurrently Cb AUC 1.5-2.0 q1w (reduced to 1.5 by an amendment after 330 pts) vs no Cb, stratified by subtype (HER+ vs TNBC), 588 pts started treatment. Primary objective was pCR (ypT0 ypN0). DFS, distant (DDFS), loco regional recurrence-free (LRRFS) and overall survival (OS) were secondary objectives.

Results

After a median follow-up of 47.3 months (range 1.7-62.8) overall no significant difference in DFS was seen with PMCb vs PM (HR = 0.83 [95%CI 0.58-1.20]; p = 0.327). However, Pts with TNBC had a significantly better DFS (HR = 0.56 [95%CI 0.34-0.93]; p = 0.024) and DDFS (HR = 0.50 [95%CI 0.29-0.86]; p = 0.013) when treated with PMCb. No difference was seen in pts with HER2+ disease (DFS HR = 1.34 [95%CI 0.77-2.34]; p = 0.295; interaction test p = 0.022 and DDFS HR = 1.56 [95% CI 0.86-2.83]; p = 0.145; interaction test p = 0.006). A trend towards a better OS was observed in pts with TNBC (HR = 0.60 [95%CI 0.32-1.12]; p = 0.110). OS was not different between the two arms, neither overall (HR = 0.72 [95%CI 0.43-1.21]; p = 0.223) nor in HER2+ disease (HR = 1.13 [95%CI 0.44-2.93]; p = 0.800). Multivariable analysis comfirms that pCR (pCR vs no pCR) independently predicted DFS (HR = 0.23, p < 0.001), DDFS (HR = 0.21, p < 0.001), and OS (HR = 0.29, p = 0.002).

Conclusions

Long-term survival analysis supports the neoadjuvant use of Cb in TNBC. The value of pCR as a strong predictor of DFS and OS was confirmed.

Clinical trial identification

NCT 01426880

Legal entity responsible for the study

German Breast Group

Funding

Teva, GSK, Roche, Hexal

Disclosure

G. von Minckwitz, S. Loibl: Research grant to the institution from Teva. All other authors have declared no conflicts of interest.