Antitumour Activity of the Poly(Adp-Ribose) Polymerase (Parp) Inhibitor Olaparib in Unselected Sporadic Castration Resistant Prostate Cancer (Crpc) in the Toparp Trial

Background: Next generation sequencing (NGS) has identified genomic aberrations causing homologous recombination (HR) DNA repair defects in sporadic, metastatic CRPC. These aberrations sensitize tumours to PARP inhibitors (PARPi) and platinum treatment. To date very little is known about these cancers. We hypothesize that single agent olaparib would be active in a subset of unselected metastatic CRPC patients and that NGS could discover putative biomarkers of response. Methods: We designed TOPARP, an open-label, investigator-initiated adaptive biomarker phase II study (CRUK/11/029) with a two-stage design to assess antitumour activity of olaparib (po = 0.05; p1 = 0.20; α = 0.02; β = 0.10; n1 = 30, n2 = 15). TOPARP-A involved recruitment of unselected sporadic, metastatic CRPC patients with retrospective evaluation of putative biomarkers of response and TOPARP-B will involve the prospective validation of predictive biomarkers to enable co-development of therapeutics and predictive biomarkers. The primary endpoint was response rate, defined as either objective response by RECIST 1.1 and/or PSA decline ≥50% and/or confirmed circulating tumour cell (CTC) count falls from ≥5 to <5/7.5 ml blood. Secondary endpoints included progression-free survival (PFS), overall survival (OS), safety and tolerability. Olaparib 400 mg twice daily was administered continuously to patients with metastatic CRPC progressing after 1–2 lines of taxane-based chemotherapy, with no prior exposure to PARPi/platinum, good organ function and ECOG-PS 0–2. Correlative biomarkers were generated from exome and transcriptome studies from paired, mandatory fresh CRPC biopsies. Functional markers of DNA damage and repair including γH2AX and RAD51 foci formation were also evaluated in pre and post-treatment tumour biopsies. The role of putative biomarkers of response identified in the “test set” (TOPARP-A) will be later validated in TOPARP-B. Results: 50 patients from 7 UK centers were enrolled. These patients were heavily pretreated. All patients had prior docetaxel, 48 (96%) had prior abiraterone and 29 (58%) receivedprior cabazitaxel. Response rate was 16/49 evaluable patients (32.7%, 95% CI: 20.0 to 47.5), with confirmed PSA falls >50%, prolonged CTC count falls and with confirmed partial response per RECIST in 5/33 patients with measurable disease. Notably, 3 responders remained on treatment for >1 year. NGS fresh biopsy analyses have identified aberrations in DNA repair genes among the responding patients including BRCA2 and ATM loss. Sequencing results, correlation with response to PARPi as well as multicolour immunofluorescence pharmacodynamics studies evaluating RAD51, 53BP1 and γH2AX will be presented. Consistent with prior studies of olaparib, anemia (9/50, 18%) and fatigue(5/50, 10%) were the most common grade >3 adverse events, with 13 (26%) patients requiring a dose reduction. Conclusions: Olaparib has antitumour activity in heavily pre-treated patients with sporadic metastatic CRPC with a 32.7% response rate and prolonged responses lasting >6 months. Several patients remain on treatment beyond 1 year. The predefined criteria for seeking a biomarker definedsubgroup were met. Exome and transcriptome studies indicate that defects in DNA repair genes including BRCA2 and ATM loss associate with olaparib sensitivity in sporadic metastatic CRPC. The identified biomarkers are being prospectively evaluated in a validation cohort (TOPARP-B).