Impaired double strand DNA repair by homologous repair deficiency (HRD) leads to sensitivity to poly ADP ribose polymerase (PARP) inhibition. Poly-ADP ribose polymerase (PARP) inhibitors target HRD to induce synthetic lethality and are used routinely in the treatment of BRCA1 mutated ovarian cancer in the platinum-sensitive maintenance setting. A subset of non-small cell lung cancers (NSCLCs) harbour impaired DNA double strand break repair. We therefore hypothesised that patients with metastatic non-small cell lung cancer exhibiting partial responses to platinum doublet-based chemotherapy, might enrich for impaired HRD, rendering these tumours more sensitive to inhibition of PARP inhibition by olaparib.

The Olaparib Maintenance versus Placebo Monotherapy in Patients with Advanced Non-Small Cell Lung Cancer trial (PIN) was a multicentre double-blind placebo controlled randomised phase II screening trial. This study was conducted at 23 investigative hospital sites in the UK. Patients had advanced (stage IIIB/IV) squamous (Sq) or non-squamous (NSq) NSCLC, and had to be chemo-naive, European Cooperative Oncology Group (ECOG) performance status 0-1. Prior immunotherapy with a PD1 or PDL1 inhibitor was allowed. Patients could be registered for PIN prior to (stage 1), or after (stage 2) initiation of induction chemotherapy. If any tumour shrinkage was observed (any shrinkage of RECIST target lesions), following a minimum of 3 cycles of platinum doublet chemotherapy, patients were randomised 1:1 using a centralised online system, to either olaparib (300 mg twice daily by mouth in 21-day cycles) or placebo, which was continued until disease progression, or unacceptable toxicity. Intention to treat (ITT) analyses of the primary endpoint included all randomised participants. Per protocol (PP) safety analysis included all participants who received at least one dose of study drug. Primary endpoint was progression-free survival (PFS), with a one-sided -value of 0.2 to demonstrate statistical significance. Hazard ratios (HR) for PFS were both unadjusted and adjusted for the randomisation balancing factors (smoking status and histology). The trial was registered with ClinicalTrials.gov (NCT01788332) and EudraCT (2012-003383-51).

A total of 940 patients were assessed for stage 1 eligibility of whom 263 were registered between Feb 24, 2014 and Nov 7, 2017. 194 patients were excluded prior to stage 2 (no tumour shrinkage or unevaluable) and 70 were randomised; 32 (46%) to Olaparib and 38 (54%) to placebo. 4% (3/70) of patients randomised had a CR and 96% (67/70) had a PR (or other evidence of tumour response/mixed stable) during induction therapy. A total of 36 patients were registered in stage 2 only, i.e., post induction therapy. Intention to treat (ITT) unadjusted analysis showed a PFS hazard ratio (HR) of 0.83 (one-sided 80% CI upper limit 1.03, one-sided unadjusted log rank test -value=0.23). ITT Cox-adjusted model showed a HR 0.73 (one-sided 80% CI upper limit 0.91, one sided -value 0.11). Adverse events were reported in 31/32 subjects (97%) in the olaparib arm and 38/38 (100%) in the placebo group. The most commonly reported adverse events in the olaparib group were fatigue (20/31; 65%), nausea (17/31; 55%), anaemia (15/31; 48%) and dyspnea (13/31; 42%). In the placebo group the most common adverse events were fatigue (25/38; 66%), coughing (22/38; 58%), dyspnea (15/38; 39%) and nausea (11/38; 29%). There were no treatment-related deaths.

PFS was longer in the olaparib arm, but this did not reach statistical significance. When the PFS HR was adjusted for smoking status and histology, a significant difference at the one-sided 0.2 level was observed, suggesting that tumour control may be achieved for chemosensitive NSCLC treated with PARP monotherapy. We speculate that this signal may be driven by a molecular subgroup harbouring HRD.

This study was funded between AstraZeneca CRUK, National Cancer Research Institute, and Cancer Research UK Feasibility Study Committee.