Abstract 4736: Dose optimization of olaparib plus temozolomide in small cell lung cancer (SCLC) patient-derived xenograft (PDX) models for clinical translation

Abstract

Abstract Introduction: SCLC is an aggressive high-grade neuroendocrine malignancy in which targeting DNA-damage repair pathways has shown efficacy in pre-clinical and clinical contexts. We previously conducted a phase I/II trial combining the PARP inhibitor olaparib (O) tablets with the alkylating agent temozolomide (T) in patients with relapsed SCLC, in which O and T were both given days (d) 1-7 of each 21 d cycle. The recommended phase 2 dose (RP2D) was O 200 mg PO BID and T 75 mg/m2 PO daily, and the response rate among the first 29 evaluable patients was 41.4%, with the primary toxicities being hematologic (Farago et al., ASCO 2018). Continuous PARP inhibition with O may improve efficacy, though it is unknown how best to incorporate T onto this backbone. Simultaneous clinical testing of multiple dosing schedules for concurrent OT would be impractical. Here we present an optimization of the OT regimen using PDX models generated from patients treated with OT. Methods: SCLC PDX models derived from two patients prior to their durable responses to OT, MGH1518-1B and MGH1528-1, were used for dose optimization. Based on a human-murine comparison of serum Cmax for OT, the estimated murine-equivalents for the RP2D doses were O 25 mg/kg BID and T 6.25 mg/kg/d. The RP2D regimen was then modulated in both dose intensity and duration in a 9-arm PDX trial. The arms included 3 schedule categories, with doses varied within each category: discontinuous regimens (OT d 1-7); continuous O regimens (O d1-21 + T d 1-7); and rapid- or slow-alternating regimens administering O and T on non-overlapping days. Tumor bearing mice were treated when subcutaneous tumors reached a volume of 400-800 cc, enabling measurement of tumor regression (PDX response) and time to volume doubling (PDX TTP). Findings: The RP2D regimen resulted in near-complete response in both models. This degree of regression was recapitulated with nearly all dosing schedules, and was not improved by increasing T to 2x-RP2D. However, the PDX TTP varied. In both models, the longest TTP was seen with continuous O at full- or half-RP2D and d 1-7 T at half-RP2D. This regimen prolonged TTP by 12% in MGH1518-1B and 20% in MGH1528-1, versus discontinuous OT at RP2D. Based on these results, we initiated clinical treatment in a new cohort of patients on OT, with a phase 1 dose escalation starting with O 50 mg BID d 1-21 and T 50 mg/m2 d 1-7 of each 21 d cycle (NCT02446704). At this first dose level, a partial response by RECIST 1.1 criteria was observed in a heavily pre-treated patient, indicating clinical activity of this combination with this new dosing strategy. Conclusions: In PDX models sensitive to OT at our previously determined RP2D, we find that continuous O with intermittent T permits dose reduction of both O and T without loss of efficacy. A clinical trial is ongoing to determine whether this strategy may enable longer-term dosing and improve efficacy in patients. Citation Format: Benjamin J. Drapkin, Sarah Phat, David T. Myers, Jun Zhong, Beow Y. Yeap, Elisabetta Leo, Anna Stansiszewska, Aaron Smith, Elaine Cadogan, Maria C. Pietanza, Nicholas J. Dyson, Anna Farago. Dose optimization of olaparib plus temozolomide in small cell lung cancer (SCLC) patient-derived xenograft (PDX) models for clinical translation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4736.