Combined PARP and HSP90 inhibition: preclinical and Phase 1 evaluation in patients with advanced solid tumours

PURPOSE

PARP inhibitor resistance may be overcome by combinatorial strategies with agents that disrupt homologous recombination repair (HRR). Multiple HRR pathway components are HSP90 clients, so that HSP90 inhibition leads to abrogation of HRR and sensitisation to PARP inhibition. We performed in vivo preclinical studies of the HSP90 inhibitor onalespib with olaparib and conducted a Phase 1 combination study.

PATIENTS AND METHODS

Tolerability and efficacy studies were performed in patient-derived xenograft(PDX) models of ovarian cancer. Clinical safety, tolerability, steady-state pharmacokinetics and preliminary efficacy of olaparib and onalespib were evaluated using a standard 3 + 3 dose-escalation design.

RESULTS

Olaparib/onalespib exhibited anti-tumour activity against BRCA1-mutated PDX models with acquired PARPi resistance and PDX models with RB-pathway alterations(CDKN2A loss and CCNE1 overexpression). Phase 1 evaluation revealed that dose levels up to olaparib 300 mg/onalespib 40 mg and olaparib 200 mg/onalespib 80 mg were safe without dose-limiting toxicities. Coadministration of olaparib and onalespib did not appear to affect the steady-state pharmacokinetics of either agent. There were no objective responses, but disease stabilisation ≥24 weeks was observed in 7/22 (32%) evaluable patients including patients with BRCA-mutated ovarian cancers and acquired PARPi resistance and patients with tumours harbouring RB-pathway alterations.

CONCLUSIONS

Combining onalespib and olaparib was feasible and demonstrated preliminary evidence of anti-tumour activity.

Phase 2, two-group, two-stage study of avelumab in patients (pts) with microsatellite stable (MSS), microsatellite instable (MSI), and polymerase epsilon (POLE) mutated recurrent/persistent endometrial cancer (EC)

5502

Background: This non-randomized phase 2 study evaluated the PD-L1 inhibitor avelumab in two cohorts of EC: i) MSI/ POLE cohort including ECs with immunohistochemical (IHC) loss of expression of at least one of the mismatch repair (MMR) proteins and/or documented mutation in the exonuclease domain of POLE and ii) MSS cohort including ECs with normal IHC expression of all MMR proteins. Methods: Eligibility criteria included measurable disease, unlimited prior therapies, and any EC histology. Co-primary endpoints were confirmed objective response (OR) and progression-free survival rate at 6 months (PFS6). Avelumab 10 mg/kg IV was given every 2 weeks until progression or unacceptable toxicity. In the 1st stage, 16 pts were enrolled in each cohort; if there were ≥2 ORs or ≥2 PFS6 responses, accrual would continue to the 2nd stage with enrollment of 19 additional pts. Overall, if there are ≥4 ORs or ≥8 PFS6 responses, avelumab would be considered worthy of further study in each cohort. Results: As of 12/2018, 33 pts were enrolled. The MSS cohort was closed at the 1st stage due to futility; of 16 pts in the MSS cohort, only 1 pt exhibited an OR and PFS6 response [ORR and PFS6 rate 6.25% (95% CI 0.16%-30.2%)]. Conversely, the MSI/POLE cohort reached the primary endpoint of 4 ORs after accrual of only 17 pts. Two pts in the MSI/POLE cohort did not initiate protocol therapy and were excluded from all analyses. Of 15 pts in the MSI/POLE cohort, 4 pts exhibited OR [1CR+3PRs, OR rate (ORR) 26.7% (95% CI 7.8%-55.1%)] and 6 pts (including the 4 pts with OR) exhibited PFS6 responses [PFS6 rate 40.0% (95% CI 16.3%-66.7%)], 4 ongoing and 3 approaching 2 yrs. Twenty-two pts (71%) reported treatment related toxicities, 6 patients (19%) G3 toxicities; there were no treatment-related G4 and G5 toxicities. In the MSI/POLE cohort, 5 of 6 PFS6 responses were observed in pts with ≥3 lines of prior therapy (p = 0.011) and in tumors who were PD-L1 negative by IHC. Further correlative work will be reported at the meeting. Conclusions: In EC pts stratified by MSI/POLE status, MSI vs MSS status appears to be correlated with avelumab response even in PD-L1 negative tumors. Responses in the MSI/POLE cohort were more frequent in more heavily pretreated patients, a finding that warrants further investigation. Clinical trial information: NCT02912572.