Background

In the Phase III SOLO2 trial (NCT01874353), maintenance treatment with the poly(ADP-ribose) polymerase inhibitor olaparib (Lynparza) was shown to significantly improve progression-free survival (PFS) vs placebo in pts with PSROC and a BRCA1/2 mutation (BRCAm; hazard ratio [HR] 0.30, 95% CI 0.22–0.41; P<0.0001; median 19.1 vs 5.5 months; Pujade-Lauraine et al. SGO 2017). A previous retrospective analysis of data pooled from 6 studies of olaparib in pts with a germline BRCAm suggested that olaparib activity declined as the number of prior lines of chemotherapy received increased (Matulonis et al. Ann Onc 2016). We report an analysis of PFS in SOLO2, grouped by number of prior lines of platinum-based chemotherapy (PBC) received by pts, performed to identify the most appropriate use of olaparib in the maintenance setting.

Methods

SOLO2 enrolled pts who had received ≥2 prior lines of PBC before being in response to their most recent regimen. Pts were randomized 2:1 to receive olaparib tablets (300 mg bid) or placebo. PFS was investigator assessed with modified RECIST v1.1. For the PFS subgroup analyses, subgroups were predefined; HRs were calculated using a Cox proportional hazards model.

Results

Of 295 randomized pts, 195 received olaparib and 99 received placebo. 85 pts in the olaparib arm (43.4%) had received ≥3 prior lines vs 37 pts (37.4%) in the placebo arm.

Table. PFS subgroup analysis by number of prior lines of platinum-based chemotherapy received

*Number of prior lines of platinum-based chemotherapy was unknown for one olaparib-arm patient; ^†Progression or death using modified RECIST version 1.1

Pts who had received 2 prior lines of PBC were more likely to have had a platinum-free interval of >12 months (70.9% vs 48.3% and 40.0% for 3 and ≥4 prior lines, respectively in the olaparib arm; 69.4% vs 60.0% and 23.5% placebo) and a complete response at baseline (50.9% vs 36.7% and 48.0% olaparib; 54.8% vs 35.0% and 35.3% placebo) vs pts who had received ≥3 prior lines.

Conclusions

In SOLO2, olaparib maintenance monotherapy improved PFS in pts with PSROC irrespective of the number of prior lines of PBC received.

Clinical trial identification

NCT01874353, 1 June 2017

Legal entity responsible for the study

AstraZeneca

Funding

AstraZeneca

Disclosure

R. Penson: Honoraria: Amgen, Genentech, AstraZeneca, Endocyte, Eisai, Vascular Biogenics, Baxalta, AbbVie, Clovis, Tesaro; research: Genentech, ImClone, Endocyte, AstraZeneca, Eisai, Amgen, Vascular Biogenics; commercial: Advance Medical. J. Ledermann: Honoraria from AstraZeneca and Pfizer, and consulting fees from AstraZeneca, Clovis Oncology, Pfizer and Roche. M. Friedlander: Research grants and advisory board honoraria from AstraZeneca. N. Colombo: Reports grants and personal fees from AstraZeneca, and personal fees from Roche, Pharmamar, Clovis, Pfizer and Tesaro. M. Gropp-Meier: Honoraria from AstraZeneca. G.S. Sonke: Institutional research funding by AstraZeneca, Merck, Novartis and Roche. A. Allen: Employee of AstraZeneca. E. Pujade-Lauraine: Received advisory board membership and honoraria from AstraZeneca and Pfizer, and advisory board membership, honoraria and speakers\' bureau membership from Roche. All other authors have declared no conflicts of interest.

Background

Niraparib (ZEJULA™) is a selective PARP1/2 inhibitor approved for maintenance treatment of adults with recurrent ovarian cancer who are in a complete or partial response to platinum-based chemotherapy. In preclinical studies, niraparib concentrates in the tumor versus plasma, delivering ≥90% durable PARP1/2 inhibition and a persistent antitumor effect. We report the relationship between exposure and response of niraparib in patients (pts) enrolled in the ENGOT-OV16/NOVA trial.

Methods

Preliminary modeling for niraparib was performed using Phase 1 study data (N = 104) to identify the initial parameters for the pharmacokinetic model, which was developed using combined Phase 1 and Phase 3 data (N = 512 pts) and the first-order conditional estimation with interaction method within NONMEM. Exposure-efficacy relationships were evaluated in the gBRCAmut and non-gBRCAmut cohorts separately; the efficacy was compared in pts with high exposure (> median exposure) vs low exposure (≤ median exposure). Maximum plasma concentration (C_max) and area under the curve over the dosing interval (AUC_tau) at steady state (SS) were the exposure metrics. Progression-free survival (PFS) was used as efficacy endpoint. Hazard ratios (HRs) and 95% confidence intervals (CI) for the low and high niraparib exposure groups in each cohort were provided, with exposure group as the independent variable and PFS as the dependent variable. Exposure and safety data of gBRCAmut and non-gBRCAmut cohorts were combined to evaluate exposure-safety relationships using logistic regression.

Results

In the gBRCAmut cohort, the HR was 0.91 (95% CI, 0.54-1.52) for niraparib exposure as measured by the SS AUC_tau for pts in the high exposure vs. the low exposure group. In the gBRCAmut cohort, the HR was 0.70 (95% CI, 0.49-0.99). Logistic regression analysis did not show any significant relationship between the incidence of thrombocytopenia or anemia Grade ≥ 3 and the SS Cmax or AUC increase.

Conclusions

Observed exposure-response relationships support the selection of 300 mg as the starting dose in both gBRCAmut and non-gBRCAmut pt populations. A trend towards increased efficacy associated with increased exposure was observed in the non-gBRCAmut cohort.

Clinical trial identification

NCT01847274

Legal entity responsible for the study

TESARO, Inc.

Funding

TESARO, Inc.

Disclosure

J. Wang, Z-Y. Zhang, X. Wang, S. Agarwal: Employment: Tesaro Stock: Tesaro. M.R. Mirza: Consulting or Advisory Role: Roche, Clovis Oncology, AstraZeneca, Tesaro. L. Gilbert: Honoraria: AstraZeneca, Advaxis Consulting/Advisory Role: AstraZeneca, Advaxis. A.V. Tinker: Consulting/Advisory: AstraZeneca. J.S. Berek: Consulting/Advisory: Atara Biotherapeutics. H.S. Pentikis, V. Kansra: Advisory board or board of directors: Tesaro Consulting: Tesaro. B. Benigno: Honoraria: AstraZeneca, Insys Therapeutics Research funding: Tesaro. S.J. Hazard: Employment: Tesaro, Genentech/Roche Stock: Tesaro, Genentech/Roche Travel, Accommodations, Expenses: Tesaro, Genentech/Roche. B.J. Rimel: Consulting/Advisory Role: AstraZeneca, Tesaro, Genentech/Roche Honoraria: Genentech. U.A. Matulonis: Consulting/Advisory: Merck KGaA, AstraZeneca, Immunogen, Tesaro, Genentech. All other authors have declared no conflicts of interest.

Background

Niraparib (Zejula™) is a selective poly(ADP-ribose) polymerase 1/2 inhibitor (PARPi) approved for maintenance therapy in adults with recurrent OC who are in response to platinum-based therapy. Here, we report safety and efficacy of niraparib in the subgroup of pts from the ENGOT-OV16/NOVA trial who were aged ≥65 years (y).

Methods

Pts were assigned to one of two independent cohorts based on germline BRCA mutation (gBRCAmut) status and randomized 2:1 within each cohort to receive either niraparib (300 mg) or placebo once daily. Pts were stratified by age (<65 vs ≥ 65 y) to analyze efficacy (measured by progression-free survival [PFS]) and safety. Efficacy and safety were also tested in patients <70 vs ≥ 70 y.

Results

Efficacy of niraparib was comparable in pts <65 vs ≥ 65 y in both gBRCAmut and non-gBRCAmut cohorts (Table). Efficacy was also similar in pts <70 vs ≥ 70 y in both cohorts (gBRCAmut: <70 y, HR = 0.30; ≥70 y, HR = 0.09. Non-gBRCAmut: <70 y, HR = 0.47; ≥70 y, HR = 0.35), although the sample size of pts who were ≥70 y in the gBRCAmut cohort was small (14 niraparib vs 7 placebo). The most common adverse events (AEs; nausea, constipation, fatigue, hypertension, anemia, thrombocytopenia, neutropenia) in the niraparib arm occurred with similar incidence in pts <65 vs ≥ 65 y as well as in pts <70 vs ≥ 70 y. Grade 3/4 AEs occurring in > 10% of niraparib-treated pts were consistent in pts <65 vs ≥ 65 y (thrombocytopenia, 27% vs 31%; anemia, 27% vs 20%; neutropenia, 12% vs 10%) and in pts <70 vs ≥ 70 y (thrombocytopenia, 28% vs 31%; anemia, 27% vs 13%; neutropenia, 11% vs 10%, respectively). There were no Grade 5 events.Table:

934PD

Conclusions

Niraparib was safe and highly effective in elderly patients.

Clinical trial identification

NCT01847274

Legal entity responsible for the study

TESARO, Inc.

Funding

None

Disclosure

A.V. Tinker: Consulting or Advisory Role: AstraZeneca. S. Mahner: Consulting: Roche, Clovis, Sensor Kinesis, MEDAC, AstraZeneca; Grants from: Roche, PharmaMar, Tesaro, MEDAC, AstraZeneca; Honoraria/reimbursement from: Roche, PharmaMar, Clovis, Tesaro, Sensor Kinesis, MEDAC, AstraZeneca. S. Banerjee: Travel, Accommodations, Expenses: AstraZeneca, Clovis Oncology. R.M. Wenham: Honoraria: Genentech, Janssen, Tesaro; Speakers\' Bureau: Genentech, Janssen; Research Funding: Merck. D.M. Provencher: Consulting: AstraZeneca; Speakers\' Bureau: AstraZeneca. I. Palacio Vázquez: Research Funding: Novartis, Tesaro; Expert Testimony: AstraZeneca; Travel, Accommodations, Expenses: PharmaMar, Roche. M.R. Mirza: Consulting or Advisory Role: Clovis Oncology, AstraZeneca, Tesaro. S.J. Hazard: Employment: Tesaro; Stock: Tesaro. U.A. Matulonis: Consulting/Advisory: Merck KGaA, AstraZeneca, Immunogen, Tesaro, Genentech. All other authors have declared no conflicts of interest.

Background

Olaparib is a PARP inhibitor approved in Europe as maintenance treatment after response to platinum-based chemotherapy in patients (pts) with relapsed, epithelial ovarian cancer (EOC) who harbour a mutation in BRCA. Approval was based on the results of study 19 (NCT00753545). We tested whether TP53 mutations categorized according to their functional effects influence overall survival (OS) in a cohort of 195 EOC pts randomized to olaparib or placebo after platinum-based chemotherapy in study 19.

Methods

TP53 status was assessed in anonymized archival tumor samples at Foundation Medicine.TP53 mutations were classified, blinded to treatment and clinical outcome, as “disruptive” (D) and “non disruptive” (ND) according to the degree of disturbance of p53 protein function and structure. Particularly, TP53 (D) mutations are associated with a severe disturbance in the DNA repair activities of the p53 protein. Overall survival (OS) and progression-free survival (PFS) were explored by TP53 and BRCA mutation status. Data cut-off was May 2016.

Results

95 pts had TP53(D) and 100 pts TP53(ND) tumours. Hazard ratios (HR) for PFS in the overall, BRCAm and BRCAwt populations were comparable between TP53(D) and TP53(ND). Pts with TP53(D) derived a statistically significant benefit from olaparib treatment, with median OS of 39.5 months (m) for olaparib and 24.0 m for placebo and a HR = 0.57 (95%CI = [0.35, 0.92]) versus HR = 0.73 (95%CI = [0.46, 1.15]) in the case of pts with TP53(ND). When including BRCA status in the analysis; all pts with BRCA mutations (n = 108) derived a benefit from olaparib in terms of OS, irrespective of TP53 status. In the BRCA wt group, pts with TP53(D) treated with olaparib had an OS of 35.0 m vs. 25.5 m for those receiving placebo (n = 47, HR = 0.80 [0.40, 1.52]). Finally, only pts with TP53(ND) and BRCAwt did not derive any benefit in median OS from olaparib treatment (n = 40, HR = 1.58 [0.77, 3.35]).

Conclusions

EOC patients BRCA wt with TP53 disruptive mutations may derive a survival benefit of olaparib treatment. Further studies are required to confirm this finding.

Legal entity responsible for the study

Instituto Oncologico Dr Rosell. Astra Zeneca

Funding

None

Disclosure

A. Fielding, B.A. Dougherty, Z. Lai, D. Hodgson, P. Rowe, S. Spencer, A. Gasco Hernandez: Astra Zeneca employee J. Lobera: Astra Zeneca All other authors have declared no conflicts of interest.

Background

PAOLA-01 is a randomized placebo-controlled, international phase III study, assessing olaparib in maintenance therapy in advanced high grade ovarian carcinoma patients responding to 1st line platinum-taxane-based chemotherapy plus bevacizumab. Stratification is performed on treatment outcome and on tumoral BRCA1/2 status (tBRCA) at screening. As secondary objective, the consistency between germline (gBRCA) and tBRCA testing is being explored.

Methods

This study is planned to recruit 762 pts in Europe and 24 in Japan. tBRCA is tested on FFPE tumor block within 5 French national institutional platforms using 2 different next-generation sequencing methods based either on capture or on re-sequencing technology. For French pts, gBRCA testing is performed in parallel in the same platforms.

Results

Since May 2015, 1181 pts have been screened and 662 pts randomized. tBRCA status was assessed in 962 samples with a median turn around time of 40 days (range 8-260). Only 44(4.6%) tumor samples were non-informative (too low tumor cellularity), 8 using capture method and 36 by re sequencing respectively. A deleterious variant (DV) was reported in 279 (29%) samples (191 (68%) in BRCA1, 87 (31%) in BRCA2 and one in both genes). Twelve variants of unknown significance were identified for BRCA2 and 13 for BRCA1. For the 384 French pts, where both gBRCA & tBCRA testing was performed in parallel we report the mutation rate detection in the Table below. Of note, only one large genomic rearrangement of BRCA1 was detected in blood sample exclusively.

Conclusions

tBRCA testing is a reliable tool for clinical trials with acceptable delay for clinical practice. Proportion of tBRCA testing failure is low and consistency with germ line testing adequate for routine practice.Table:

935PD

Clinical trial identification

EudractCT: 2014-004027-52 NCT02477644 First received: June 18, 2015

Legal entity responsible for the study

ARCAGY Research

Funding

AstraZeneca and Roche

Disclosure

I. Soubeyran: Fees from Astra-Zeneca, Roche and ThermoFisher; advisory boards fees from Astra-Zeneca, Pfizer and MSD for intervening expert. P. Harter: Consulting or advisory role: Roche, AstraZeneca; research funding: AstraZeneca. A. Gonzalez Martin: Consultant and speaker for Roche and AstraZeneca. K. Fujiwara: AstraZeneca as an advisory board, travel expense and research grant. Also has COI with Chugai-Roche for research grant. S. Pignata: Honoraria: AstraZeneca, Roche; consulting or advisory role: AstraZeneca, Roche; research funding: Roche. N. Colombo: Advisory board: Roche and AstraZeneca; Out of this trial, advisory board: Tesaro, Clovis, Pharmamar, Advaxis, Pfizer. C. Marth: Honoraria: Roche, AstraZeneca, Pfizer; consulting or advisory role: Roche, AstraZeneca, Pfizer Travel, accomodations, expenses: Roche, AstraZeneca and Pfizer. I. Vergote: Consulting or advisory board: AstraZeneca Z. J. Mäenpää: Consulting or advisory role: Roche, AMGEN, SOBI, AstraZeneca Travl, accomdations, expenses: GSK, SOBI, Roche, AstraZeneca. E. Pujade-Lauraine: Consulting or advisory role: Roche, AstraZeneca, Pfizer; speaker\'s bureau: Roche, AstraZeneca, Pfizer; travel, accomodations, expenses: Roche, AstraZeneca. I. Ray-Coquard: Honoraria: Roche, Pharmamar, AstraZeneca; consulting or advisory board: Roche, Pharmamar, AstraZeneca, Advixis; travel, accomodations, expenses: Roche, Pharmamar, AstraZeneca. All other authors have declared no conflicts of interest.

Background

Despite progresses in precision medicine, very few molecularly-targeted agents (MTA) are available for gynecologic cancer patients (pts). Objectives were to characterize tumor genomic alterations in the gynecological subpopulation of the ProfiLER program, to identify actionable alterations and to report clinical efficacy of MTA if used outside approved indications.

Methods

The ProfiLER program is a multicentric prospective trial to implement molecular profiling in pts with advanced disease. All potential pts with advanced gynecologic cancers were eligible. DNA extracted from either archival or fresh collected tumor samples was analyzed by targeted exon sequencing (NGS) of 59 cancer related genes and whole genome array comparative genomic hybridization (CGH). Genomic profiles were presented in a dedicated molecular tumor board (MTB) for recommendation of MTA when applicable.

Results

Out of the 2184 included pts in the ProfiLER program, 242 pts with advanced gynecologic cancer were recruited from March 2013 to April 2016. For 211 (87%) pts (ovary, n = 136; cervix, n = 22; uterus, n = 45; others, n = 8), molecular analyses have been performed (median delay 2.8 months). 101 pts (48%) had at least one actionable alteration: PIK3CA (n = 23), KRAS (n = 10), ERRB2 (n = 9) and 50 other alterations. 109 MTA were recommended in 83 pts: PI3K/AKT/mTOR inhibitors (n = 46), sorafenib (n = 19), lapatinib (n = 7) and 17 other MTA (n = 37). 54 pts have not received MTA at the time of the analysis (poor PS or death, n = 25; stable disease, n = 15; no dedicated clinical trial, n = 7; other, n = 7). Currently, 29 pts initiated MTA. Partial response was reported for 8 pts (28%) and a stable disease at 3 months for 6 pts (21%). 12 pts (41%) progressed and 3 were non-evaluable. Median progression-free survival was 2.7 months (95% CI 1.9-5.4). Median overall survival was 64.8 months (38-101) for pts receiving MTA and 49.8 months (39-63) for pts not receiving MTA.

Conclusions

CGH and NGS identified actionable alterations on 48% of pts with gynecologic cancer and are feasible in routine practice. Nearly half of patients treated derived benefit from the recommended MTA, but these represent a minority of the whole screened population.

Clinical trial identification

NCT01774409

Legal entity responsible for the study

Prof. Jean-Yves Blay, Centre Léon Bérard, Lyon (France)

Funding

Work funded by LYric (DGOS-INCa-4664). Aid granted by Bpifrance Financement abounded by European Community (E8983 – PREDICTIV).

Disclosure

O. Tredan: Honoraria, Consulting/Advisory, Travel funds: Lilly, Roche, Novartis, AstraZeneca, Pfizer, Celldex. P. Cassier: Research funding: Astrazeneca, Roche/Genentech, Novartis, Celgene, Blueprint, Lilly, Toray, MSD, Merck Serono, Transgene; Honoraria: Amgen, Astrazeneca, Elsalys. D. Perol: Corporate-sponsored research for Roche. All other authors have declared no conflicts of interest.

Background

Recent data suggest that programmed death ligand 1 (PD-L1) expression may predict response to anti–programmed death 1 (PD-1) therapy. This retrospective observational study evaluated the prognostic effect of PD-L1+ expression in patients with histologically confirmed epithelial ovarian, primary peritoneal, or fallopian tube cancer (OCa).

Methods

Patients diagnosed with FIGO stages II-IV OCa from 2004-2012, at Aarhus University Hospital and Rigshospitalet, Copenhagen, Denmark, were included. PD-L1 expression was measured in tissue collected at OCa surgery, using immunohistochemistry with anti–PD-L1 22C3 antibody. PD-L1⁺ expression was defined as staining in ≥ 1% of tumor or inflammatory cells. Patients partially sensitive (PPS; treatment-free interval [TFI] of 6-12 mo) or fully sensitive (FPS; TFI >12 mo) to platinum therapy were considered platinum sensitive, whereas those refractory (TFI <3 mo) or resistant (TFI of 3-6 mo) to platinum were platinum insensitive. Data were analyzed using Cox proportional hazard model, adjusting for age, stage, histology, residual tumor, surgery type, performance status, and/or TFI.

Results

Median age of the 376 patients at diagnosis was 63 years (range, 26-86). 77% had histologic grade 2/3 serous adenocarcinoma, 46% had residual tumor after surgery, and 9%, 70%, and 21% had FIGO stages II, III, and IV disease, respectively. FPS, PPS, platinum-resistant, and platinum-refractory disease comprised 49%, 27%, 15%, and 9% of patients, respectively. 50.5% of patients were PD-L1⁺, with prevalence increasing with increased platinum sensitivity (P for linear trend <0.05). Median overall survival (OS) was 43 mo (50.4 mo in PD-L1⁺ vs 38.3 mo in PD-L1^– patients). A statistically significant association was seen between PD-L1⁺ tumors and longer OS (adjusted hazard ratio [aHR], 0.71 [95% CI, 0.55-0.91]). The association was not significant in platinum-insensitive patients (aHR, 0.82 [0.50-1.36]), but there was a tendency towards significance in platinum-sensitive patients (0.77 [0.57-1.06]), driven by those with a TFI of 6-12 mo.

Conclusions

PD-L1 was frequently expressed in advanced OCa patients, and expression may be prognostic, particularly in those with partially platinum-sensitive OCa.

Legal entity responsible for the study

Merck & Co., Inc.

Funding

Merck & Co., Inc.

Disclosure

T.T. Vo: Employed by, own stock in, and have received research grants from Merck & Co., Inc. W. Zhou, M. Busch-Sørensen, D. Chappell: Employed by and own stock in Merck & Co., Inc. T. Steiniche: Received research funding from and have been reimbursed for travel and accommodation expenses by Merck & Co., Inc. All other authors have declared no conflicts of interest.

Background

Little is known about the immune microenvironment of OCCCs and its impact on outcomes. We studied the expression of a panel of immune response genes in OCCC to identify the presence and prognostic relevance of irGES in these tumours.

Methods

Immune response gene profiling was performed on 84 FFPE OCCC samples with matched clinical outcomes, collected between 2003 – 2016, using the nanoString nCounter PanCancer Immune Profiling Panel. Unsupervised hierarchical clustering analysis was performed and each sample underwent analysis for protein levels of PD-1, PD-L1, MMR and ARID1A via immunohistochemistry (IHC).

Results

Total of 74/84 samples were successfully profiled. Median age at diagnosis was 53 yrs. 41 (55.4%) were stage 1, 7 (9.5%) stage 2, 24 (32.4%) Stage3, 2 (2.7%) stage 4. 64/74 (86.5%) of pts received adjuvant chemotherapy post debulking surgery with 38% recurrence rate (median PFS 27 months (m)). Median follow up was 36m. Based on irGES, 4 distinct molecular subgroups of OCCCs were identified. G1 was hallmarked by high NK cell markers/PD-1 expression, G2 by increased CTLA-4/PD-L1 expression, G3 by adhesion cell markers, and G4 by increased levels of pro - angiogenic genes. G1 was observed to have significantly poorer PFS (median PFS 20m vs 68m, p= 0.011) and a trend towards poorer OS when compared with G2/3/4 (median OS 38.8m vs undefined, p = 0.0501). G4 carried the best prognosis (median PFS 108m vs 26m, p= 0.0515; median OS undefined, p= 0.0726). This difference in OS and PFS was reflected in stage 1 pts (p < 0.02 and <0.05; respectively) with a similar trend observed but not significant in non-stage 1 pts. Significant correlation was observed between gene and protein expression of tumour PD-1, tumour and stromal PD-L1, and tumour IL-6 using IHC (p < 0.0001 for all comparisons). 6 (8.1%) pts were MMR deficient with no significant association between ARID1A and MMR expression across each irGES.

Conclusions

OCCCs are heterogeneous and can be classified into 4 molecular subgroups based on their irGES profiles with distinct clinicopathological characteristics and prognostic outcomes. If validated in larger datasets, these signatures may serve to inform a clinical trial.

Clinical trial identification

not applicable

Legal entity responsible for the study

Institute Review Board SIngapore

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

At diagnosis, tumor-infiltrating lymphocytes (TILs) are prognostic in epithelial ovarian cancer (EOC). We recently demonstrated that neoadjuvant chemotherapy (NACT) significantly increased stromal TILs and stromal TILs remained prognostically significant after NACT. Here, we investigated the impact of NACT on different immune subpopulations and their relationship with clinical outcome.

Methods

Tissue microarrays of EOC (145 pre-NACT, 139 post-NACT, including 83 matched samples) were analyzed for CD3+,CD8+ and FOXP3+ by immunohistochemistry. Stromal TILs scored as percentage of stromal area, intraepithelial TILs as number of TILs in contact with tumor cells/HPF. Differences were evaluated by Mann-Whitney or Wilcoxon-signed-rank for unpaired or paired analyses, respectively and by Log-rank for PFS and OS.

Results

NACT significantly increased stromal CD3 + (p = 0.005) and CD8 + (p = 0.009) and intraepithelial CD8 + (p = 0.02) infiltration in unmatched samples and remained significant among paired samples for stromal CD3+ and CD8 + (p = 0.03 and p = 0.009). Neither CD3+ nor CD8+ expression correlated with outcome at diagnosis or post-NACT, however reduced accumulation of FOXP3+ post-NACT (<5%) was significantly associated with improved PFS (HR = 0.59; p = 0.016). A high stromal CD8+/FOXP3+ ratio post-NACT strongly correlated with improved PFS (median 30.9 vs.18.85 mos, p = 0.005) and OS (median 50.70 vs 37.10mos, p = 0.029). In contrast, at diagnosis, CD8+/FOXP3+ was not associated with prognosis.

Conclusions

NACT has a significant impact on the balance of cytotoxic versus suppressive T cells and a high ratio of CD8+/FOXP3+ post-NACT was most significantly associated with improved PFS and OS. Whether this could select patients for immune therapies in the post-operative setting should be investigated.

Clinical trial identification

none

Legal entity responsible for the study

Institut Gustave Roussy

Funding

INCA and MERUS

Disclosure

All authors have declared no conflicts of interest.

Background

If epithelial ovarian cancer (EOC) had a poor prognosis, more than 20% of patients (pts) can expected long remission. Few data are available on long term quality of life (QoL) in these pts and results reported were usually not compared to those of healthy controls. VIVROVAIRE was a large national case-control study comparing pts reported outcomes (PROs) among EOC pts without relapse within 3 years after first line treatment and a group of healthy women.

Methods

Pts were recruited in 27 French centers and clinical characteristics were issued from medical charts. Controls were randomized from electoral lists. Pts and controls matched on age. They filled in a form including PROs questionnaires: QoL, neurotoxicity and fatigue (FACT/G, FACT/O, FACT/GOG-Ntx, FACT/F), anxiety and depression (HADS), sleep disturbance (ISI) and Physical activity (IPAQ).

Results

318 pairs were analysed (from 349 pts and 327 controls included). Median age: 65 (20-86), high level of education: 52% and 58%, respectively. Pts characteristics: FIGO stage I/II (49%), III/IV (47%) unknown (4%); major histology, serous (50%), endometrioid (16%), clear cells (8%), mucinous (4%). BRCA1/2 mutations (n = 21; 15%), unknown (n = 168). 99% of the pts had a surgery and 96% received platinum based chemotherapy, associated with antiangiogenic agent (14%) Interval from first line therapy: median 5 years [2 to 24]. Pts reported lower physical and functioning QoL scores (p = 0.03 and p = 0.0002), higher score of fatigue (p < 0.0001), and poorer quality of sleep (p = 0.0003) than controls. No difference of scores of anxiety and depression was observed between the 2 groups. TOI score (related to ovarian cancer and treatment) and score of neurotoxicity were higher among patients (p < 0.0001); 26% of pts reported severe fatigue, more than 70% of the pts were concerned about digestive symptoms and severe neurotoxicity. Only 18% of the pts and controls had an active physical activity.

Conclusions

Compared to healthy women, EOC pts presented poorer long term QoL, fatigue with important neurotoxicity and digestive symptoms. Physicians have to take in count of the late effects of treatments to help the pts to cope with the sequelae.

Clinical trial identification

Not applicable (Observational study)

Study approved by CCTIRS and CNIL

Legal entity responsible for the study

Centre François Baclesse

Funding

Fondation de France; Ligue Nationale Contre le Cancer

Disclosure

All authors have declared no conflicts of interest.

Background

In the OVHIPEC study (NCT00426257), the addition of HIPEC to interval cytoreductive surgery (CRS) significantly improved recurrence-free (HR 0.66, 95% CI 0.50-0.87) and overall survival (HR 0.67, 95% CI 0.48-0.94) in patients with stage III ovarian cancer. We report the results of the health-related quality of life (HRQOL) analysis of patients in the OVHIPEC study.

Methods

Patients who had at least stable disease after 3 cycles of carboplatin and paclitaxel were randomly assigned to receive interval CRS with (n = 122) or without (n = 123) HIPEC using cisplatin. Patients in both arms received 3 additional cycles of carboplatin/paclitaxel post-operatively. HRQOL was assessed using the EORTC core questionnaire (the QLQ-C30) and the ovarian and colorectal cancer modules (QLQ-OV28 and QLQ-CR38). HRQOL assessments were planned at least 2 weeks prior to surgery, after surgery, following the 6th chemotherapy cycle, and every three months thereafter until a maximum of two years of follow-up. HRQOL data were analysed using a mixed effect model. Current preliminary results pertain to the QLQ-C30 summary score, and the fatigue, neuropathy, micturition and gastrointestinal symptom scale scores for up to 1 year of follow-up or time of recurrence.

Results

Of 245 randomized patients, 201 (82%) had at least one evaluable HRQOL assessment (93 for CRS arm, 108 for CRS+HIPEC arm). A linear mixed effects analysis indicated a significant group difference in QLQ-C30 summary scores over time (p = 0.047). The CRS group had modest improvement in overall HRQOL over time, while the CRS+HIPEC group remained largely unchanged. In both arms, the overall HRQOL scores were relatively high, both at baseline and during follow-up. No significant group differences over time were observed for fatigue, neuropathy, micturition or gastrointestinal symptoms.

Conclusions

Preliminary results indicate that patients in this trial had relatively high overall HRQOL scores, that tended to improve more over time in the CRS arm compared to the CRS+HIPEC arm. Future analyses will yield additional information about the specific patterns of HRQOL and symptom effects over the course of the trial.

Clinical trial identification

NCT00426257

Legal entity responsible for the study

The Netherlands Cancer Institute - Antoni van Leeuwenhoek Hospital

Funding

None

Disclosure

All authors have declared no conflicts of interest.

Background

EGFR inhibitors have proven to improve the efficacy of anticancer treatments in lung, colon, pancreas, or HNC. Results from studies show EGFR expression in cervical Ca to improve survival outcomes in the same. This study was designed to assess safety and efficacy of Nimotuzumab with concurrent cisplatin and RT in patients with Ca Cervix.

Methods

In this open-label randomized controlled multicentric study 100 patients with histologically confirmed Ca Cervix were recruited over a 4 year period with an equal allocation of 1:1 for intervention (Standard arm- concurrent CTRT (Cisplatin 40mg/m2 weekly IV +RT) plus 200mg weekly BIOMAB IV (n = 50) on same day of cisplatin infusion) vs. standard arm only (n = 50). At 2 years data were available for 39 patients in the intervention arm and 35 patients in standard arm. The size of the lesion was documented using CT/PETCT at baseline. The response was analyzed using RECIST criteria at the following treatment every 3 months for subsequent 2 years. Toxicity was assessed using CTCAE v4 toxicity criteria.

Results

There were 74 evaluable patients at end of 2 years. The mean age was 49.6± 10.2 years. The complete response following treatment was seen in 37.8% (BIOMAB arm) of patients at two years following treatment compared to 38.2% in standard arm. However, progressive disease was seen more in standard arm (52.9%) compared to BIOMAB arm (35.1%). Best overall response was seen in 64.9% patients in the intervention arm compared to 47.1% patients in the standard arm at two years following treatment which is significant. At 2 years 60% progressed on standard arm compared to 37.5% in the intervention arm. The mean estimate of progression-free survival being 36 months vs. 56.5 months (BioMab arm) (Log rank Mantel-Cox χ2= 3.9, p = 0.05). Except for one patient with Biomab sensitivity, there was no additional toxicity compared to the standard arm.

Conclusions

Nimotuzumab appears to be safe and effective targeted therapy in cervical cancer patients with long-term benefits.

Clinical trial identification

TS-01-2009

Legal entity responsible for the study

Healthcare Global Enterprises Ltd.

Funding

BIOCON

Disclosure

All authors have declared no conflicts of interest.

Background

VAN is a bi-specific human IgG1 antibody, simultaneously blocking two key angiogenic factors, Ang-2 and VEGF-A. VAN as a single agent showed an objective response rate (ORR) of 29% in bevacizumab- naïve PROC. The anti-PD-L1 agent, ATEZO demonstrated a 22% ORR in advanced OC. Preclinical data suggested additive antitumor activity of VAN when combined with anti−PD-L1. Hence, treatment with VAN plus ATEZO has the potential to reverse pro-angiogenic and immune-suppressive signals, thereby resulting in improved clinical benefit.

Methods

Eligible patients (pts) had PROC measurable by RECIST 1.1. Pts with history of bowel obstruction, > 2 prior lines of systemic chemotherapy, or previous treatments with VEGF-A inhibitors or agents targeting Ang/Tie2 receptor axis were ineligible. Pts received VAN 2000 mg and ATEZO 840 mg, each IV Q2W, until disease progression or unacceptable toxicity. Primary efficacy endpoint was ORR as per RECIST 1.1, with tumor assessments every 8 weeks.

Results

17 pts with median age of 63 years (range 45-74) were treated. Serous histology was present in all pts, except one clear cell subtype. 4 pts (24%) achieved confirmed PR, 8 pts (47%) experienced SD and 4 (24%) had PD, while one patient was not evaluable. The achieved RECIST ORR of 24% remained unchanged when evaluated as per immune-related response criteria. 9/17 pts were evaluable for CA-125 per GCIG criteria; three and two achieved a response with and without normalization respectively. The current estimate of PFS rate @ 6 months is 65% (median follow-up: 162 days). The most common adverse events (AE) of any grade (G) were decreased appetite, diarrhea (41% each), asthenia and constipation (35% each). AEs ≥ G3 included abdominal pain, LFT increase, asthenia, dyspnea, health deterioration, hypertension, GI obstruction, GI perforation (GIP), subileus, lymphedema, pleuritis and tonsillitis (6% each). One AE of GIP and asthenia each were fatal.

Conclusions

Our data suggest that VAN plus ATEZO does not improve upon monotherapy with VAN or ATEZO in PROC. The safety profile of this combination is consistent with reports for the single agents in this setting.

Clinical trial identification

NCT01688206

Legal entity responsible for the study

F. Hoffmann-La Roche Ltd

Funding

Roche

Disclosure

A. Oaknin: Advisory Board and/or Board of Directors for Roche, AstraZeneca, Clavis, PharmaMar. I. Vergote: Research (via KULeuven) for Amgen, Exelixis, Lilly, Morphotek, Pronota, Roche; Advisory Boards and/or Board of Directors for Roche, Genentech and multiple other pharmaceutical companies exceeding the character limit and thus not listed here. I. Ray-Coquard: Advisory Board for Roche. A. Leary: Advisory board and/or board of directors for AstraZeneca, Clovis, GamaMabs; Research funding: GamaMabs, Merus. A. Lahr, I. Franjkovic, S. Rossomanno, A. Sahbi, K. Longauer: Employment Roche. P. Gerber, F. Heil, C. Boetsch, O. Krieter: Stock options and employment Roche. T. Nayak: Stock options. All other authors have declared no conflicts of interest.