Durvalumab (D) in combination with olaparib (O) for advanced, previously treated non-small cell lung cancer (NSCLC).

Abstract

e21153 Background: Poly (ADP-ribose) polymerase inhibitors (PARPi) show variable clinical activity in individuals with advanced cancers. In the lung-MAP substudy S1400G, talazoparib alone was not sufficiently active in squamous lung cancers with homologous recombinant repair deficiency. Combinations of PARPi with immune checkpoint inhibitors (ICIs) are under investigation since preclinical studies show an immunostimulatory effect of PARPi that can potentially increase responsiveness to ICIs. To test this hypothesis, we evaluated the PARPi, O with the ICI, D in individuals with advanced NSCLC who had received at least one prior platinum-based regimen. Methods: Eligible participants were enrolled in an expansion cohort of a proof-of-concept phase I/II study (NCT02484404) and received O 300 mg PO BID with D 1500 mg IV every 4 weeks until disease progression or unacceptable toxicity. The primary endpoint was progression-free survival (PFS). Results: Between July 2016 and July 2021, 15 participants with recurrent NSCLC were enrolled (median age: 66 years; stage IV: 12; adenocarcinoma 11; prior ICI: 9). After median potential follow-up of 39.6 months (mo), median PFS (mPFS) was 3.2 mo (95% CI: 0.9-7.6). Among 14 subjects with known EGFR mutation status, mPFS was longer in 11 individuals with EGFR-WT tumors (5.7 vs. 0.9 mo; p = 0.001). DNA repair mutation ( APC, BAP1, BRCA1, BRCA2, CHEK2, FANCI, FANCL and PALB2) status was known for 9 subjects; mPFS of subjects with (n = 7) or without (n = 2) DNA repair mutations was 3.2 vs. 18.1 mo (p = 0.68), respectively. Tumor PD-L1 expression > 50% (n = 3/10) and prior ICI therapy (n = 9/15) was associated with longer, albeit not statistically significant, mPFS: 7.6 vs. 1.8 mo; p = 0.22, and 5.7 vs. 1.8 mo; p = 0.39, respectively. Two (13%) partial responses, 7 (47%) stable disease and 6 (40%) cases of progressive disease were observed. Four (27%) subjects, including 3 previously treated with ICIs, achieved durable response or stability for up to 3 years. Treatment-related adverse events (TRAEs) were generally mild (grade 1 or 2). The most common TRAEs were anemia (40%), thrombocytopenia (27%), anorexia, fatigue, pain, and hypophosphatemia (each observed in 20%). Grade > 3 TRAEs occurred in 4 (27%) subjects (all hematologic). Immune-related AEs occurred in 4 (27%) individuals. Two subjects developed myelodysplasia after 12 and 36 mo of treatment. Three (20%) subjects required dose reductions or discontinuation of O due to anemia. There were no treatment-related deaths. Accrual will continue until 20 eligible participants are enrolled. Conclusions: O+D has acceptable safety and modest efficacy in an unselected population with recurrent NSCLC. Sufficient clinical activity is not observed in individuals with EGFR-mutated tumors. Genomic and immunological analyses are being performed to identify predictive biomarkers in individuals with durable benefit and will be reported. Clinical trial information: NCT02484404.