EFFICACY AND SAFETY OF ENTINOSTAT AND PEMBROLIZUMAB IN PATIENTS WITH MELANOMA PREVIOUSLY TREATED WITH ANTI-PD1 THERAPY (CTPL03_CT072)
Ryan J. Sullivan, MD (Massachusetts General Hospital, Boston, MA, USA) presented data from the multicohort study ENCORE-601 (NCT02437136), evaluating entinostat + pembrolizumab in patients with advanced melanoma. He began his presentation outlining remarkable advances made in the last 10 years for patients with advanced melanoma, but there is still a clear unmet need for patients who fails on current standard of care. For patients who progress after Immune checkpoint inhibitors (ICIs) treatment, entinostat, a class I selective histone deacetylase (HDAC) inhibitor, in combination with pembrolizumab showed promising activity through alteration of the immunosuppressive tumor microenvironment. Total of 53 patients were enrolled (last enrolled in April 2018), among which 53% were PD-L1 positive. Majority (70%) of patients had received prior ipilimumab and/or anti-PD1 therapy and 13% of those patients had responded to the therapy. All patients with BRAF mutation had received BRAF/MEK inhibitor therapy prior to the trial entry. With entinostat + pembrolizumab, 10 out of 53 patients achieved a confirmed CR or PR (ORR = 19%, 95% CI: 9-32%) with median duration of response of 13 months. Additional 9 patients had SD more than 6 months with clinical benefit rate of 36%. Responses were observed regardless of prior PD-1 therapy. Any grade 3 or 4 immune related AEs were seen in 9.4% of patients and 6% of patients discontinued the therapy from AEs. Among 49 patients analyzed, increase in CD8+ T cells and decrease in M-MDSCs (characterized by CD14+, HLA-DR-/lo) were observed in the responders. Further analysis of gene signatures from pre- and on-treatment specimens in 4 responders and 4 non-responders indicated increased inflammation in the tumor microenvironment after treatment. The combination therapy with entinostat + pembrolizumab demonstrated significant clinical activity and acceptable safety in patients with melanoma who have progressed on prior PD-(L)1 blockade. Preliminary biomarker analysis supports the hypothesis that the addition of entinostat decreases MDSC and restores inflammation in the tumor microenvironment necessary for successful re-treatment with an anti-PD-(L)1.
TUMOR MUTATIONAL BURDEN AS A BIOMARKER OF SURVIVAL IN METASTATIC NON-SMALL CELL LUNG CANCER: MYSTIC, A STUDY OF FIRST-LINE DURVALUMAB ± TREMELIMUMAB VS. CHEMOTHERAPY (CTPL03_CT074)
Solange Peters, MD, PhD (Centre Hospitalier Universitaire Vaudois, Lausanne University, Switzerland) presented preliminary results from the phase III, randomized, open-label, multicenter trial MYSTIC (NCT02453282), evaluating first-line durvalumab (n = 374) ± tremelimumab (n = 372), vs. platinum-based chemotherapy (n = 372) in metastatic non-small cell lung cancer (NSCLC). Patients with EGFR and ALK wild-type, immunotherapy/ chemotherapy-naïve metastatic NSCLC were randomized (1:1:1) to durvalumab (20 mg/kg i.v. Q4W); durvalumab (20 mg/kg i.v. Q4W) + tremelimumab (1 mg/kg i.v. Q4W up to 4 doses); or chemotherapy. The MYSTIC trial provided the comprehensive data that supports TMB as a predictive biomarker of OS benefit with first-line durvalumab and durvalumab + tremelimumab treatment in metastatic NSCLC patients. In tissue TMB ≥10 mut/Mb population, OS was improved with durvalumab with or without tremelimumab vs. chemotherapy, though the dataset was small and requires a further investigation. Improved HR for OS was associated with increasing blood TMB levels (measured by GuardantOMNI) for durvalumab + tremelimumab vs. chemotherapy . A threshold of blood TMB ≥ 20 mut/Mb was associated with OS benefit with durvalumab vs. chemotherapy (HR 0.72 [95% CI: 0.50-1.05], 24-mo OS rate 33.8% vs. 19.4%). Markedly improved OS and clinical benefit with durvalumab + tremelimumab vs. chemotherapy (OS HR 0.49 [95% CI: 0.32-0.74], 24-mo OS rate 48.1% vs. 19.4%; PFS: HR 0.53 [95% CI: 0.34-0.81]; PFS HR 0.53 [95% CI: 0.34-0.81]; ORR 48.4% vs. 21.4%) suggest the potential contribution of tremelimumab to durvalumab treatment of this cohort. In this study, blood TMB values did not correlate with PD-L1 expression, indicating that blood TMB and PD-L1 are independent biomarkers. Also, tissue TMB values positively correlated with blood TMB values in 352 matched patient specimens (data for tissue TMB are shown in the abstract). These analyses support the use of appropriate biomarkers to select the patients most likely to benefit from immunotherapy in first-line.
PHASE I EVALUATION OF A BIFUNCTIONAL FUSION PROTEIN, M7824, TARGETING TGF-BETA AND PD-L1 IN PATIENTS WITH HUMAN PAPILLOMAVIRUS-ASSOCIATED MALIGNANCIES (CTPL03_CT075)
Julius Strauss, MD (National Cancer Institute, National Institutes of Health, Bethesda, MD, USA) presented data from the ongoing Phase I the dose-escalation trial of M7824 (NCT02517398) in patients with HPV-associated cancers (HACs). TGF-β pathway dysregulation may play a critical role in carcinogenesis and immune evasion in HACs, including anal, cervical, and p16+ squamous cell carcinoma of the head and neck (SCCHN). Targeting this pathway may also enhance clinical response to PD-(L)1 antibodies. Bintrafusp alfa (M7824) is an innovative first-in-class bifunctional fusion protein composed of two extracellular domains of TGF-βRII (a TGF-β “trap”) fused with a human IgG1 monoclonal antibody against PD-L1. Total of 43 patients treated on an ongoing phase I trial in the dose-escalation phase (n = 17) and in the dose expansion phase (n = 26). HPV status was confirmed in 36 patients (84%) by DNA PCR or RNA sequencing. Incidence of immune related AEs were similar to those of anti-PD-1 agents, although 16.3% patients had discontinued from treatment-related AEs. Notably, 9 patients (20%) developed squamous cell carcinoma (SCC) or keratoacanthoma (KA) which was treated with local measures and did not result in treatment discontinuation. Grade 1 or 2 mucosal bleeding was also reported as seen in trials with other TGF-β inhibitors. Overall response rate in all population was 27.9% with 2 CRs and 10 PRs and 30.6% in patients with confirmed HPV status. Three additional patients (2 SCCHN patients and 1 cervical cancer patient) achieved PR after initial PD, and total clinical response including late responders was 35% in HACs and 38.9% in confirmed HPV positive cancers. Median survival was 16.2 mo which was longer than reported median survival with a single agent anti-PD-1 therapy. Most responses were durable and median duration of response was not reached with range of 2-25 mo and 75% of responders were ongoing at the time of data cutoff. Notably, responses to bintrafusp alfa occurred irrespective of tumor type or PD-L1 expression status. Overall, bintrafusp alfa demonstrated a promising clinical activity in patients with HACs and safety profile was similar to anti-PD-1 therapy except for SCC/KA and low grade mucosal bleeding.
PEMBROLIZUMAB AFTER TWO OR MORE LINES OF PRIOR THERAPY IN PATIENTS WITH ADVANCED SMALL-CELL LUNG CANCER: RESULTS FROM THE KEYNOTE-028 AND KEYNOTE-158 (CTPL03_CT073)
Hyun Cheol Chung, MD, PhD (Yonsei Cancer Center, Yonsei University College of Medicine, Seoul, Republic of Korea) presented findings from the phase Ib basket study KEYNOTE-028 (NCT02054806) and the phase II basket study KEYNOTE-158 (NCT02628067), evaluating pembrolizumab in patients with advanced SCLC (n =131). Eligible patients had experienced progression/failure on standard therapy, were immunotherapy-naïve, and had received ≥ 2 lines of systemic therapy. Pembrolizumab (10 mg/kg Q2W [KEYNOTE-028] or 200 mg Q3W [KEYNOTE-158]) was administered for 2 years or until disease progression or intolerable toxicity. The ORR was 19.3% (95% CI: 11.4-29.4) with complete response in 2 patients and partial response in 14 patients. Median duration of response (DOR) was not reached (range, 4.1-35.8 mo). Nine of 16 responders had response lasting ≥ 18 mo. Median PFS was 2.0 mo (95% CI: 1.9-3.4) and median OS was 7.7 mo (95% CI: 5.2-10.1). 12- and 24-mo survival rates were 16.9% and 13.1% for PFS, and 34% and 21% for OS, respectively. Among all SCLC patients in KEYNOTE-028 and KEYNOTE-158 irrespective of prior therapies, treatment-related AEs occurred in ≥ 5% of patients. In this pooled analysis, pembrolizumab demonstrated promising anti-tumor activity in patients with extensive-stage SCLC who had received ≥ 2 lines of prior therapy. Responses were durable, with most patients estimated to have a response duration of at least 18 mo. Median OS and PFS in this pooled analysis were comparable to the SCLC cohort populations of KEYNOTE-028 and KEYNOTE-158. This study supports the use of pembrolizumab monotherapy for patients with advanced SCLC as third-line or later therapy.
EXPRESSION OF PD-L1 IN MACROPHAGES IS ASSOCIATED WITH BETTER OUTCOMES THAN PD-L1 POSITIVE TUMOR CELLS IN A PILOT COHORT OF NSCLC TREATED WITH PD-1 AXIS IMMUNOTHERAPY (MS.CL11.02_2674)
Yuting Liu from Dr. David L. Rimm lab (Yale University, New Haven, CT, USA) presented analysis of three retrospective studies of Yale NSCLC cohorts (457 non-immunotherapy-treated cases and 62 PD-1 axis immunotherapy-treated cases) to examine PD-L1 expression using molecular assessment of cellular localization. Currently, the immune checkpoint target and companion diagnostic, PD-L1 expression in tumor cells and immune cells are assessed by the pathologist to predict response to immunotherapy in different organ systems. In this study, PD-L1 expression and localization in multiple immune cell types were measured by the quantitative immunofluorescence methods and high resolution images. Interestingly, PD-L1 localization was significantly higher with macrophages compared to natural killer (NK) cells and CD8+ T cells in both tumor and stroma compartment. In stroma, 80% of PD-L1 positive cells were CD68+ macrophages, while 12%, 5% and 3% with CK+ tumor cells, CD56+ NK cells and CD8+ T cells, respectively. In tumor, PD-L1 positive cells were CD68+ macrophages (41%) and CK+ tumor cells (53%) as well as CD56+ NK cells (4%) and CD8 cytotoxic T cells (2%) (p < 0.05). High PD-L1 expression in macrophages was correlated longer OS (p = 0.036) while high PD-L1 expression in tumor cells was not (p = 0.72). In over 500 lung cancer cases, the predominant immune cell type expressed PD-L1 was CD68+ macrophages, as compared to NK cells and CD8 T cells. The level of PD-L1 in macrophages was significantly associated with the level of PD-L1 in tumor and with high levels of CD8+ T cells, suggesting a connection between high PD-L1 and immune-hot tumors. Together, high levels of PD-L1 expression in macrophages is suggested to be a positive predictive biomarker for PD-1 pathway blockade therapy. Further work is necessary to obtain the data from the large size of the treated cohort.
LYMPHODEPLETION PRIOR TO HER2-CAR T CELLS SAFELY AUGMENTS T CELL EXPANSION AND INDUCES CLINICAL RESPONSES IN PATIENTS WITH ADVANCED SARCOMAS (LBMS02_LB-147/4)
Shoba A. Navai, MD (Baylor College of Medicine, Houston, TX, USA) presented data from the phase I, dose-escalation clinical study HEROS 2.0 (NCT00902044), assessing the safety and anti-tumor efficacy of HER2-CAR T cells in combination with lymphodepletion chemotherapy in patients with advanced sarcoma. In this trial, autologous HER2-CAR T cells, 1x104-1x108/m2, were intravenously administered to 11 patients with recurrent/refractory HER2+ sarcoma. Up to three infusions of ≤ 1x108/m2 were safely administered in combination with lymphodepletion reagents either fludarabine (n = 3) or fludarabine + cyclophosphamide (n = 8). Fludarabine and fludarabine + cyclophosphamide induced lymphopenia with an absolute lymphocyte count of less than 100/ml on the day of the T-cell infusion. Fludarabine + cyclophosphamide induced neutropenia (absolute neutrophil count of less than 500/ml) for up to 14 days. Eight patients developed grade 1/2 cytokine release syndrome (CRS) within 24 hours of CAR T-cell infusion, followed by complete resolution after supportive care within 5 days of onset. No neurotoxicity was observed. T cells expanded in 9 patients with a median peak expansion on day 7 (range: 5 to 28). CAR T cells could be detected by qPCR in 10 patients at 6 weeks post infusion. One patient with Rhabdomyosarcoma (RMS) metastatic to the bone marrow had a complete response (CR) for 12 mo, and then relapsed after 6 mo off therapy. The patient was subsequently re-treated and achieved CR that has been maintained for 15 mo. One patient with osteosarcoma metastatic to the lungs has an ongoing CR for 32 mo. Three patients had stable disease (SD), and 5 had progressive disease (PD). In the RMS patient who attained CR, remodeling of the TCR-β repertoire was observed and distinct antibody responses to oncogenic pathway proteins were identified. Thus, administration of lymphodepletion chemotherapy followed by autologous HER2-CAR T cells infusion is tolerated and associated with objective clinical benefit in some patients with advanced HER2+ sarcoma. Immune profiling data suggest that HER2-CAR T cells in combination with fludarabine/cyclophosphamide lymphodepletion induce immune reactivity. These findings warrant further evaluation in a phase 2 study as a single agent or in combination with other approaches.
CHK1 INHIBITOR, SRA737, WITH GEMCITABINE ENHANCES ANTI-PD-L1-MEDIATED MODULATION OF THE IMMUNE MICROENVIRONMENT AND TUMOR REGRESSIONS IN SMALL CELL LUNG CANCER MODEL (LBMS02_LB-148/5)
Triparna Sen, MD (University of Texas, MD Anderson Cancer Center, Houston, TX, USA) presented the preclinical data of the efficacy and immune correlation profiles of a checkpoint kinase 1 (CHK1) inhibitor, SRA737, in combination with low dose gemcitabine (LDG) ± anti-PD-L1 in the small cell lung cancer (SCLC) model. SCLC is the most aggressive form of lung cancer with poor response to immunotherapy targeting the PD-(L)1 pathway. SCLC exhibits high expression of CHK1 and that the CHK1 inhibitor SRA737 activates the innate immune system via STING pathway, demonstrating robust anti-tumor activity and synergy in combination with anti-PD-L1 in an SCLC model. In immunocompetent B6129F1 mice, Trp53, Rb1 and p130 triple knockout SCLC cells were implanted to induce tumor growth to examine the treatment with single agents or various drug combinations. Without SRA737, anti-PD-L1 and LDG demonstrated minimal effect on tumor growth and only a modest effect as a combination. In contrast, SRA737 monotherapy demonstrated moderate anti-tumor activity in a dose-dependent manner. When anti-PD-L1 was combined with the SRA737 + LDG regimen, the profound and synergistic anti-tumor activity was observed with durable tumor regressions. Analysis of tumor infiltrating immune cells at the end of this treatment indicated a strong induction of cytotoxic T-cells and a reduction of exhausted and regulatory T cells. Similarly, pro-inflammatory M1 type macrophages and dendritic cells were increased while immunosuppressive M2 type macrophages and MDSC cells were decreased. These data implicate that the combination of anti-PD-L1 with the SRA737+LDG regimen leads to an effective anti-tumor activity accompanied by modulation of immune microenvironment. Clinical trials of SRA737 monotherapy (NCT02797964) and in combination with LDG (NCT02797977) in multiple solid tumors are ongoing.
Glossary
AE = adverse event
CAR = chimeric antigen receptor
CHK1 = checkpoint kinase 1
CI = confidence interval
CR = complete responses
DOR = duration of response
HACs = HPV-associated cancers
HDAC = histone deacetylase
HPV = human papillomavirus
HR = hazard ratio
ICIs = Immune checkpoint inhibitors
IV = Intravenous
KA = keratoacanthoma
LDG = low dose gemcitabine
Mb = megabase
mo = month(s)
NK = natural killer
NSCLC = non-small cell lung cancer
ORR = overall response rates
OS = overall survival
PD = progressive disease
PD-1 = programmed cell death protein-1
PD-L1 = programmed death-ligand 1
PFS = progression-free survival
RMS = Rhabdomyosarcoma
PR = partial response
Q3W = every 3 weeks
Q4W = every 4 weeks
QIF = quantitative immunofluorescence
SCCHN = squamous cell carcinoma of the head and neck
SCLC = small cell lung cancer
SD = stable disease
TC = tumor cell
TMB = tumor mutational burden