A multicentre randomised phase III trial comparing pembrolizumab versus single-agent chemotherapy for advanced pre-treated malignant pleural mesothelioma: the European Thoracic Oncology Platform (ETOP 9-15) PROMISE-meso trial

BACKGROUND

Malignant pleural mesothelioma (MPM) is an aggressive malignancy characterised by limited treatment options and a poor prognosis. At relapse after platinum-based chemotherapy, single-agent chemotherapy is commonly used and single-arm trials of immune-checkpoint inhibitors have demonstrated encouraging activity.

PATIENTS AND METHODS

PROMISE-meso is an open-label 1:1 randomised phase III trial investigating the efficacy of pembrolizumab (200 mg/Q3W) versus institutional choice single-agent chemotherapy (gemcitabine or vinorelbine) in relapsed MPM patients with progression after/on previous platinum-based chemotherapy. Patients were performance status 0-1 and unselected for programmed cell death ligand 1 (PD-L1) status. At progression, patients randomly assigned to receive chemotherapy were allowed to crossover to pembrolizumab. The primary end point was progression-free survival (PFS), assessed by blinded independent central review (BICR). Secondary end points were overall survival (OS), investigator-assessed PFS, objective response rate (ORR), and safety. Efficacy by PD-L1 status was investigated in exploratory analyses.

RESULTS

Between September 2017 and August 2018, 144 patients were randomly allocated (pembrolizumab: 73; chemotherapy: 71). At data cut-off [20 February 2019, median follow-up of 11.8 months (interquartile range: 9.9-14.5)], 118 BICR-PFS events were observed. No difference in BICR-PFS was detected [hazard ratio = 1.06, 95% confidence interval (CI): 0.73-1.53; P = 0.76], and median BICR-PFS (95% CI) for pembrolizumab was 2.5 (2.1-4.2), compared with 3.4 (2.2-4.3) months for chemotherapy. A difference in ORR for pembrolizumab was identified (22%, 95% CI: 13% to 33%), over chemotherapy (6%, 95% CI: 2% to 14%; P = 0.004). Forty-five patients (63%) assigned to chemotherapy received pembrolizumab at progression. With follow-up to 21 August 2019 [17.5 months: (14.8-19.7)], no difference in OS was detected between groups (HR = 1.12, 95% CI: 0.74-1.69; P = 0.59), even after adjusting for crossover. Pembrolizumab safety was consistent with previous observations. Exploratory efficacy analyses by PD-L1 status demonstrated no improvements in ORR/PFS/OS.

CONCLUSION

This is the first randomised trial evaluating the efficacy of pembrolizumab in MPM patients progressing after/on previous platinum-based chemotherapy. In biologically unselected patients, although associated with an improved ORR, pembrolizumab improves neither PFS nor OS over single-agent chemotherapy.

Efficacy of adoptive therapy with tumor-infiltrating lymphocytes and recombinant interleukin-2 in advanced cutaneous melanoma: a systematic review and meta-analysis

Adoptive cell therapy (ACT) using autologous tumor-infiltrating lymphocytes (TIL) has been tested in advanced melanoma patients at various centers. We conducted a systematic review and meta-analysis to assess its efficacy on previously treated advanced metastatic cutaneous melanoma. The PubMed electronic database was searched from inception to 17 December 2018 to identify studies administering TIL-ACT and recombinant interleukin-2 (IL-2) following non-myeloablative chemotherapy in previously treated metastatic melanoma patients. Objective response rate (ORR) was the primary end point. Secondary end points were complete response rate (CRR), overall survival (OS), duration of response (DOR) and toxicity. Pooled estimates were derived from fixed or random effect models, depending on the amount of heterogeneity detected. Analysis was carried out separately for high dose (HD) and low dose (LD) IL-2. Sensitivity analyses were carried out. Among 1211 records screened, 13 studies (published 1988 - 2016) were eligible for meta-analysis. Among 410 heavily pretreated patients (some with brain metastasis), 332 received HD-IL-2 and 78 LD-IL-2. The pooled overall ORR estimate was 41% [95% confidence interval (CI) 35% to 48%], and the overall CRR was 12% (95% CI 7% to 16%). For the HD-IL-2 group, the ORR was 43% (95% CI 36% to 50%), while for the LD-IL-2 it was 35% (95% CI 25% to 45%). Corresponding pooled estimates for CRR were 14% (95% CI 7% to 20%) and 7% (95% CI 1% to 12%). The majority of HD-IL-2 complete responders (27/28) remained in remission during the extent of follow-up after CR (median 40 months). Sensitivity analyses yielded similar results. Higher number of infused cells was associated with a favorable response. The ORR for HD-IL-2 compared favorably with the nivolumab/ipilimumab combination following anti-PD-1 failure. TIL-ACT therapy, especially when combined with HD-IL-2, achieves durable clinical benefit and warrants further investigation. We discuss the current position of TIL-ACT in the therapy of advanced melanoma, particularly in the era of immune checkpoint blockade therapy, and review future opportunities for improvement of this approach.

A retrospective cohort study of PD-L1 prevalence, molecular associations and clinical outcomes in patients with NSCLC: Results from the European Thoracic Oncology Platform (ETOP) Lungscape Project

INTRODUCTION

The PD-L1 biomarker is an important factor in selecting patients with non-small cell lung cancer for immunotherapy. While several reports suggest that PD-L1 positivity is linked to a poor prognosis, others suggest that PD-L1 positive status portends a good prognosis.

METHODS

PD-L1 positivity prevalence, assessed via immunohistochemistry (IHC) on tissue microarrays (TMAs), and its association with clinicopathological characteristics, molecular profiles and patient outcome- Relapse-free Survival (RFS), Time-to-Relapse (TTR) and Overall Survival (OS)- is explored in the ETOP Lungscape cohort of stage I-III non-small cell lung cancer (NSCLC). Tumors are considered positive if they have ≥1/5/25/50% neoplastic cell membrane staining.

RESULTS

PD-L1 expression was assessed in 2182 NSCLC cases (2008 evaluable, median follow-up 4.8 years, 54.6% still alive), from 15 ETOP centers. Adenocarcinomas represent 50.9% of the cohort (squamous cell: 42.4%). Former smokers are 53.7% (current: 31.6%, never: 10.5%). PD-L1 positivity prevalence is present in more than one third of the Lungscape cohort (1%/5% cut-offs). It doesn't differ between adenocarcinomas and squamous cell histologies, but is more frequently detected in higher stages, never smokers, larger tumors (1/5/25% cut-offs). With ≥1% cut-off it is significantly associated with IHC MET overexpression, expression of PTEN, EGFR and KRAS mutation (only for adenocarcinoma). Results for 5%, 25% and 50% cut-offs were similar, with MET being significantly associated with PD-L1 positivity both for AC (p < 0.001, 5%/25%/50% cut-offs) and SCC (p < 0.001, 5% & 50% cut-offs and p = 0.0017 for 25%). When adjusting for clinicopathological characteristics, a significant prognostic effect was identified in adenocarcinomas (adjusted p-values: 0.024/0.064/0.063 for RFS/TTR/OS 1% cut-off, analogous for 5%/25%, but not for 50%). Similar results obtained for the model including all histologies, but no effect was found for the squamous cell carcinomas.

CONCLUSION

PD-L1 positivity, when adjusted for clinicopathological characteristics, is associated with a better prognosis for non-metastatic adenocarcinoma patients.

Exploratory Analysis of Single-Gene Predictive Biomarkers in HERA DASL Cohort Reveals That C8A mRNA Expression Is Prognostic of Outcome and Predictive of Benefit of Trastuzumab

Purpose

The Herceptin Adjuvant study is an international multicenter randomized trial that compared 1 or 2 years of trastuzumab given every 3 weeks with observation in women with human epidermal growth factor 2-positive (HER2+) breast cancer after chemotherapy. Identification of biomarkers predictive of a benefit from trastuzumab will minimize overtreatment and lower health care costs.

Methods

To identify possible single-gene biomarkers, an exploratory analysis of 3,669 gene probes not expected to be expressed in normal breast tissue was conducted. Disease-free survival (DFS) was used as the end point in a Cox regression model, with the interaction term between C8A mRNA and treatment as a categorical variable split on the cohort mean.

Results

A significant interaction between C8A mRNA and treatment was detected (P < .001), indicating a predictive response to trastuzumab treatment. For the C8A-low subgroup (mRNA expression lower than the cohort mean), no significant treatment benefit was observed (P = .73). In the C8A-high subgroup, patients receiving trastuzumab experienced a lower hazard of a DFS event by approximately 75% compared with those in the observation arm (hazard ratio [HR], 0.25; P < .001). A significant prognostic effect of C8A mRNA also was seen (P < .001) in the observation arm, where the C8A-high group hazard of a DFS event was three times the respective hazard of the C8A-low group (HR, 3.27; P < .001). C8A mRNA is highly prognostic in the Hungarian Academy of Science HER2+ gastric cancer cohort (HR, 1.72; P < .001).

Conclusion

C8A as a single-gene biomarker prognostic of DFS and predictive of a benefit from trastuzumab has the potential to improve the standard of care in HER2+ breast cancer if validated by additional studies. Understanding the advantage of overexpression of C8A related to the innate immune response can give insight into the mechanisms that drive cancer.

Association of p27 and Cyclin D1 Expression and Benefit from Adjuvant Trastuzumab Treatment in HER2-Positive Early Breast Cancer: A TransHERA Study

Purpose: To assess the prognostic and predictive value of selected biomarkers involved in cell-cycle regulation or proliferation in patients with HER2-positive early breast cancer.Experimental Design: Protein expression of TOP2A, Ki67, cyclin D1, and p27 was immunohistochemically determined in tissue microarrays of surgical specimens from 862 patients randomized to trastuzumab (1 or 2 years; N = 561) and observation (N = 301) arms of the HERA trial. The primary analysis endpoint was disease-free survival (DFS). Biomarkers were examined as continuous or categorical variables (predefined cutoffs). Interaction terms between biomarkers and treatment were assessed in multivariate Cox models adjusted for variables of clinical interest.Results: A significant interaction was detected between p27 and treatment (adjusted P = 0.0049). Trastuzumab effect was significant in the p27-low subgroup (≤70% p27-positive tumor cells; N = 318). HR _{Comb Trast vs. Obs} 0.44, 95% CI, 0.29-0.65 (P < 0.001). No trastuzumab effect was observed in the p27-high subgroup N = 435; HR _{Comb Trast vs. Obs} 0.97, 95% CI, 0.66-1.44, P = 0.89), indicating that these patients derived little or no benefit from trastuzumab treatment. A prognostic effect of p27 on DFS was observed, with p27-high patients experiencing half the hazard of a DFS event compared with low ones (HR _{p27 High vs. Low} 0.49, 95% CI, 0.32-0.75). TOP2A, Ki67, and cyclin D1, as categorical variables were not predictive, whereas cyclin D1 as continuous variable was predictive of trastuzumab benefit.Conclusions: In TransHERA, patients with HER2-positive early breast cancer with low p27 expression in their tumors benefited from trastuzumab treatment, whereas patients with high p27 expression did not. Clin Cancer Res; 24(13); 3079-86. ©2018 AACR.

Prevalence and clinical correlation of programmed cell death 1 ligand (PD-L1) expression in patients with resected non-small cell lung cancer (NSCLC): Results from the European Thoracic Oncology Platform (ETOP) Lungscape cohort

8516

Background: Conflicting data exists on the potential prognostic impact of PD-L1 expression in NSCLC. The Lungscape project, a fully annotated large biobank of resected stage I-III NSCLC, allows detailed analysis of this issue. Methods: Prevalence of PD-L1 positivity and its association with clinicopathological characteristics and patient outcome - Relapse-free Survival (RFS), Time-to-Relapse (TTR) and Overall Survival (OS) - was explored in the ETOP Lungscape cohort. PD-L1 expression was assessed on tissue microarrays (TMAs) using the DAKO 28-8 immunohistochemistry assay. Positivity cut-off points of ≥1%, 5% and 50% for neoplastic cell membrane staining were considered. Results: PD-L1 data were available for 2182 patients, from 15 ETOP centers, with median follow-up 4.8 years; 1191 patients still alive; median age 66 years; 64% male, 32/54/11% for current/former/never smokers; 49/29/22% for stages I/II/III; 51/42/4/3% adenocarcinomas (AC)/squamous cell (SCC) /large cell and sarcomatoid (LCS)/other. Median RFS/TTR/OS were 53/99/69 months (AC: 52/84/72, SCC: 54/not reached/64; and LSC 52/103/74). PD-L1 prevalence with 1% cut-off was, overall: 43%, 95% confidence interval (95%CI): 41-46; (AC: 42%, 95%CI: 39-46; SCC: 44%, 95%CI: 40-47; and LCS: 53%, 95%CI: 42-65), while for 5% threshold, prevalence was 34%, 95%CI: 32-36. PD-L1 1% positivity was a significant predictor only for AC: HRRFS: + vs - = 0.82; 95%CI: 0.69-0.97, HRTTR: + vs - = 0.83; 95%CI: 0.68-1.01, HROS: + vs -= 0.83; 95%CI: 0.69-1.01 (adjusted p = 0.024, 0.064, 0.063 respectively). This effect is found also for the 5% cut-off, and preserved in the overall model including all histologies. Using the 50% cut-off, PD-L1 positivity was detected in 17% of patients; 95%CI: 15-18, but was no longer a significant predictor of outcome, overall and by histology type. Conclusions: PD-L1 positivity (1% and 5% cut-offs) was present in more than one third of resected NSCLC and was associated with a better prognosis for AC patients.

Abstract P5-18-01: TransHERA: The cell cycle regulator p27 predicts benefit from trastuzumab treatment in HER2-positive early breast cancer patients treated within the HERA trial

Purpose: Predictive biomarkers may help predicting adjuvant trastuzumab response and thus optimize the treatment of patients with HER2-positive breast cancer. The aim of the present study was to assess the prognostic/predictive value of various biomarkers involved in cell cycle regulation or proliferation.

Methods: Expression of p27, cyclin D1, TOP2a, and Ki67 was immunohistochemically determined in tissue micro arrays of specimens from 862 patients randomized to the trastuzumab (1 or 2 year; N=561) and observation (N=301) arms of the HERA trial. The primary endpoint of the analysis was disease-free survival (DFS). Biomarker expression status was determined as continuous variable or by pre-defined categories. The interaction terms between the four biomarkers and treatment were assessed in multivariate Cox proportional hazards regression models adjusted for variables of clinical interest. Associations were considered significant only if the false discovery rate (FDR) adjusted p-values remained significant.

Results: Baseline characteristics were well balanced between the two study arms. A total of 249 DFS events (28.9%) were observed in the TransHERA cohort, with an overall 8-year DFS of 70.5% (95% CI 67.2%-73.5%). None of the four biomarkers was significantly associated with DFS in the total study population. When biomarkers were categorized according to pre-defined cut-off levels, only p27 turned out to be highly predictive: Expression data for p27 were available in 753 TransHERA patients. A highly significant interaction was detected between p27 and treatment when adjusting for clinical parameters and the remaining three biomarkers (p=0.0039). For patients classified as p27 low (≤70% p27-positive tumor cells; N=318), a significant treatment effect was observed, with the hazard of a DFS event being greater for the observation group compared to patients treated with trastuzumab (HRTrast vs Obs=0.43, 95% CI 0.29-0.64, p&lt;0.001). In contrast, no statistically significant effect of trastuzumab treatment was detected in the p27 high group (N=435; HRTrast vs Obs=0.97, 95% CI 0.66-1.44, p=0.89), indicating that p27 high patients derived little or no benefit from trastuzumab treatment. Cyclin D1, TOP2a, and Ki67 used as categorical variables were not predictive, while cyclin D1 used as continuous variable was predictive of adjuvant trastuzumab benefit.

Conclusion: HER2-positive early breast cancer patients with low p27 expression in their tumors appear to benefit from trastuzumab treatment, whereas patients with high p27 expression do not.

This study was funded by Roche.

Citation Format: Martin Filipits, Michael Gnant, Urania Dafni, Varvara Polydoropoulou, Margaret J Hills, Brian Leyland-Jones, Martine Piccart-Gebhart, Mitch Dowsett. TransHERA: The cell cycle regulator p27 predicts benefit from trastuzumab treatment in HER2-positive early breast cancer patients treated within the HERA trial [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P5-18-01.

Abstract P3-06-02: Exploratory analysis of single gene predictive biomarkers in TransHERA DASL cohort reveals that C8A mRNA expression is prognostic of outcome and predictive of benefit of trastuzumab

HERA is an international multi-center randomized trial comparing 1 or 2 years trastuzumab, given every 3 weeks, with observation in women with HER2+ breast cancer after standard neoadjuvant or adjuvant chemotherapy. From December 2001 to June 2005, 5102 patients were randomized. Comparing 1 year trastuzumab with observation at 8 years of follow-up, statistically significant differences in disease-free survival (DFS) and overall survival, despite crossover to trastuzumab of observation arm patients, were found. No benefit in DFS of 2 years compared to 1 year trastuzumab was found[Goldhirsch 2013].

To determine possible predictive single-gene biomarkers, an exploratory ITT analysis, with DFS as the primary endpoint, was conducted using mRNA expression data from 610 TransHERA FFPE samplesprofiled on Illumina Whole-Genome DASL cubic spline normalization was applied to the data. Outcome was obtained from the HERA database with 8 years median follow-up and clinical cut-off date April 12, 2012. Characteristics were well balanced between treatment groups. The exploratory analysis of 20,464 genes using DFS as the endpoint identified C8A as a possible biomarker that is prognostic and predictive of response to treatment. Cox regression was used to model DFS, with the interaction term between treatment and C8A as a continuous and a categorical variable split on the cohort mean. The observation arm consists of 199 samples with 66 events and the trastuzumab arm(1&2-year combined) of 411 samples with 108 events.

A statistically significant interaction between C8A mRNA and treatment was detected (p&lt;0.001), indicating that C8A mRNA is predictive of response to trastuzumab treatment. For the C8A low subgroup (mRNA expression lower than the cohort mean) no significant treatment benefit is observed (p=0.73). On the other hand for the C8A high subgroup, patients in the trastuzumab arm experience a lower hazard of a DFS event by almost 75% compared to patients from the observation arm (HR=0.25; 95%CI:0.15-0.43, p&lt;0.001). A significant prognostic effect of C8A mRNA is also observed (p&lt;0.001) in the observation arm, where for the C8A high group the hazard of a DFS event is three times the respective hazard of the C8A low group (HR-=3.27; 95%CI:2.01-5.32, p&lt;0.001).

C8A is a member of the membrane attack complex and is part of the innate immune system. C8A inserts into the membrane of the target cell and binds with multiple copies of the pore-forming C9 leading to cell lysis. From the GeneAtlas, C8A is highly expressed in liver tissue suggesting an advantage for tumors with high expression of C8A and innate immune response. The Cancer Cell Line Encyclopedia indicates a wide range of C8A mRNA expression.

C8A as a single gene biomarker that is prognostic of DFS and predictive of benefit from trastuzumab has the potential to improve the standard of care in HER2+ breast cancer. Understanding the advantage of over expression of C8A related to the innate immune response can give insight into the mechanisms that drive cancer. We note with caution that this finding is the result of an exploratory analysis and is being pursued in additional trastuzumab cohorts for further validation.

Citation Format: Scooter Willis, Varvara Polydoropoulou, Zoi Tsourti, Urania Dafni, Brandon Young, Bradley Long, Pradip De, Nandini Dey, Casey Williams, Mitch Dowsett, Brian Leyland-Jones. Exploratory analysis of single gene predictive biomarkers in TransHERA DASL cohort reveals that C8A mRNA expression is prognostic of outcome and predictive of benefit of trastuzumab [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-06-02.