ABSTRACT
Purpose: Drug repurposing offers the opportunity for chemotherapy to be used to reestablish sensitivity to immune checkpoint blockade (ICB) therapy. Here we investigated the clinical and translational aspects of an early phase II study of azacitidine and carboplatin priming for anti-PDL1 immunotherapy (avelumab) in patients with advanced ICB-resistant melanoma. Experimental Design: A total of 20 participants with ICB-resistant metastatic melanoma received 2 × 4-week cycles of azacitidine and carboplatin followed by ICB rechallenge with anti-PD-L1 avelumab. The primary objective was overall response rate after priming and ICB rechallenge. Secondary objectives were clinical benefit rate (CBR), progression-free survival (PFS), and overall survival (OS). Translational correlation analysis of HLA-A and PD-L1 expression, RNA sequencing, and reduced representation bisulfite sequencing of biopsies at baseline, after priming and after six cycles of avelmuab was performed. Results: The overall response rate (ORR) determined after azacitidine and carboplatin priming was 10% (2/20) with two partial responses (PR). The ORR determined after priming followed by six cycles of avelumab (week 22) was 10%, with 2 of 20 participants achieving immune partial response (iPR). The CBR for azacitidine and carboplatin priming was 65% (13/20) and after priming followed by six cycles of avelumab CBR was 35% (n = 7/20). The median PFS was 18.0 weeks [95% confidence interval (CI): 14.87-21.13 weeks] and the median OS was 47.86 weeks (95% CI: 9.67-86.06 weeks). Translational correlation analysis confirmed HLA-A generally increased after priming with azacitidine and carboplatin, particularly if it was absent at the start of treatment. Average methylation of CpGs across the HLA-A locus was decreased after priming and T cells, in particular CD8+, showed the greatest increase in infiltration. Conclusions: Priming with azacitidine and carboplatin can induce disease stabilization and resensitization to ICB for metastatic melanoma. Significance: There are limited treatments for melanoma once resistance to ICB occurs. Chemotherapy induces immune-related responses and may be repurposed to reinstate the response to ICB. This study provides the first evidence that chemotherapy can provide clinical benefit and increase OS for ICB-resistant melanoma.
Subject(s)
Azacitidine , Carboplatin , Drug Repositioning , Drug Resistance, Neoplasm , Immune Checkpoint Inhibitors , Melanoma , Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Antibodies, Monoclonal, Humanized/administration & dosage , Antibodies, Monoclonal, Humanized/adverse effects , Antibodies, Monoclonal, Humanized/therapeutic use , Azacitidine/administration & dosage , Azacitidine/adverse effects , Azacitidine/therapeutic use , Biomarkers , Carboplatin/administration & dosage , Carboplatin/adverse effects , Carboplatin/therapeutic use , DNA Damage/drug effects , DNA Methylation/drug effects , Epigenesis, Genetic/drug effects , HLA-A Antigens , Immune Checkpoint Inhibitors/administration & dosage , Immune Checkpoint Inhibitors/adverse effects , Immune Checkpoint Inhibitors/therapeutic use , Immunotherapy/methods , Melanoma/drug therapy , T-Lymphocytes/metabolism , Translational Research, BiomedicalABSTRACT
PURPOSE: Survival with advanced pancreatic cancer is less than 12 months. Pancreatic exocrine insufficiency may contribute to pancreatic cancer-related cachexia, via nutrient malabsorption. We aimed to determine the feasibility of prescribing pancreatic extract (Creon®) for patients with advanced pancreatic cancer. METHODS: Patients with advanced pancreatic cancer, without frank malabsorption, were randomised in this feasibility study to pancreatic extract 50,000 units with meals and 25,000 units with snacks, or placebo. Standardised dietary advice was given. Anti-cancer and supportive care treatments were permitted. Outcomes included weight, body mass index (BMI), quality of life (QLQC30, PAN26), survival and nutritional assessment (PG-SGA). RESULTS: Eighteen patients were randomised before study closure due to slow recruitment. Baseline characteristics were well matched. Weight loss prior to randomisation was numerically greater in the pancreatic extract group (mean 0.7 vs 2.2 kg). Weight loss was numerically greater in the placebo group, however not significantly. No differences in BMI or nutrition score were seen. Quality of life did not differ between study groups. Median overall survival was 17 (95% CI 8.1-48.7) weeks in the control group, and 67.6 (95% CI 14.1-98.4) weeks in the pancreatic extract group (p = 0.1063). Only 17% (18/106) of potentially eligible patients were recruited, related to patient/family reluctance, rapid clinical deterioration and patients already prescribed pancreatic extract. A moderate pill burden was noted. CONCLUSION: Despite intriguing survival results, this study was not sufficiently feasible to proceed to a fully powered comparative study. A multi-centre study would be required to exclude a significant difference in outcomes.