A randomized phase II trial of olaparib maintenance versus placebo monotherapy in patients with chemosensitive advanced non-small cell lung cancer.

e21649 Background: Impaired DNA damage response (DDR) is a common feature of cancer, however therapeutic exploitation has been limited to cancers harbouring somatic inactivation of BRCA1/2 which causes homologous recombination deficiency (HRD). We hypothesized that patients (pts) with metastatic non-small cell lung cancer responding to platinum doublet based chemotherapy, might enrich for impaired DDR encompassing HRD, rendering these tumours more sensitive to inhibition of poly-ADP ribose polymerase (PARP) inhibition by olaparib. Methods: PIN was a multicentre double-blind placebo controlled randomised phase II screening trial (alpha and beta = 0.2). Chemonaive pts had advanced (stage IIB/IV) squamous (Sq) or non-sq (NS) NSCLC, ECOG performance status 0-1, no EGFR nor ALK mutation. Prior immunotherapy with a PD1/ PDL1 inhibitor was allowed. If tumour shrinkage was observed after 3 cycles of platinum chemotherapy, pts were randomised 1:1 to olaparib (O, 300mg po bd q21) or placebo (P), which was continued until disease progression or withdrawal. Primary end point was progression free survival (PFS), with a one-sided test for significance. Secondary endpoints were overall survival (OS), response (RECIST v1.1) and safety (CTCAE4.0). Results: 70 pts were randomised to O (32) or P (38) between Aug 2014 and Nov 2017. There was no difference in median dose intensity (% (IQR), O vs P) was 86.4 (64.3-95.7) vs 93.3 (80.2-97.6). Pts receiving O had a longer, but not statistically significant median PFS (weeks (IQR) was O 16.6 (7.1-21.7) vs P 12 (5.6-18.7)); and hazard ratio (HR) was 0.83 (80% CI 0.6-1.15, p = 0.23), using intention to treat (ITT) unadjusted analysis. However, the ITT Cox model, adjusted for smoking history and histology, showed a significantly longer PFS for pts receiving O (HR 0.73 (80% CI 0.52-1.02, p = 0.11)). Pts receiving O also had greater, but not statistically significant OS: median (weeks (IQR) was O 59.4 (38.7-67.9) vs P 31.3 (22.4-58.6)). OS HR was 0.68 (95% CI 0.37-1.26; two-sided p = 0.22). 3 patients were completely withdrawn prior to progression. No complete responses were observed and treatment was well tolerated. Conclusions: Design parameters justifying a Phase III trial (p < 0.2) were met in the adjusted, but not unadjusted PFS analysis, with a trend towards longer PFS and OS in the O arm. We speculate that this signal may be driven by a DDR deficient molecular subgroup. Translational studies are warranted to investigate these possibilities. Clinical trial information: NCT01788332.