NCI 9938: Phase I clinical trial of ATR inhibitor berzosertib (M6620, VX-970) in combination with irinotecan in patients with advanced solid tumors.

Abstract

3012 Background: Ataxia telangiectasia and Rad3 related (ATR) is activated in response to replication stress from topoisomerase 1 inhibitors. Selective ATR inhibition with berzosertib potentiates the efficacy of irinotecan in colorectal mouse xenograft models. We hypothesized that berzosertib in combination with irinotecan is well tolerated, modulates the DNA damage repair response to irinotecan, and the combination is associated with clinical activity. Methods: This phase I study utilized a modified Storer’s up and down dose escalation design. Dose Levels (DLs) combined berzosertib 60 to 270 mg/m2 with irinotecan 180 mg/m2, every 2 weeks in a 4-week cycle. The primary endpoint was identification of the maximum tolerated dose (MTD) and recommended phase II dose (RP2D). Activity, pharmacokinetics (PK), and pharmacodynamics (PD) were secondary endpoints. The identification of molecular subpopulations sensitized to the combination was exploratory. Results: Between July 2016 and July 2021, 51 patients (pts) enrolled, of whom 50 received treatment. Pts most commonly had colorectal cancer (CRC, 39%), pancreatic cancer (24%), small cell lung cancer (SCLC, 6%) and non-small cell lung cancer (6%). The median number of prior lines of therapy was 4 (range, 2 to 11). In Stage I, 1 of 3 evaluable pts experienced dose-limiting toxicity (DLT) of grade 3 lung infection at DL3 (berzosertib 180 mg/m2 - irinotecan 180 mg/m2), and Stage II was initiated enrolling cohorts of 5 pts. In Stage II, 4 of the first 11 pts treated at DL4 (berzosertib 270 mg/m2 - irinotecan 180 mg/m2) were unable to complete the DLT evaluation period due to clinically significant toxicity not meeting DLT criteria: grade 2 diarrhea (1 pt), grade 3 diarrhea (1 pt), and grade 3 neutrophil decrease (2 pts). The protocol was amended to limit dose escalation beyond DL4. At DL4, 1 of 21 evaluable pts experienced DLT (grade 4 febrile neutropenia). Most common treatment-related grade ≥ 3 toxicities were neutrophil decrease (34%), lymphocyte decrease (30%), WBC decrease (28%), anemia (20%), diarrhea (16%), fatigue (8%) and hypokalemia (8%). 2 partial responses were observed, occurring in pts with pancreatic cancer and ATM alterations: 32% decrease in an ATM E11828/ATM K1109* tumor lasting 15.3 months and 68% decrease in an ATM R3008H/germline ATM R1882* tumor ongoing at 11 months. An additional pt with ATM S214fs*40 mutant colorectal cancer (CRC) experienced a 26% decrease lasting 7.5 months. Conclusions: Berzosertib 270 mg/m2 - irinotecan 180 mg/m2 was declared the RP2D. The combination is associated with manageable side effects and promising disease activity in ATM mutant solid tumors. PK and PD studies are in process. Tumor biopsy studies are planned in a 15 pt dose expansion cohort at DL4, enrolling pts with CRC, pancreatic cancer, SCLC and DNA damage repair deficient tumors. Clinical trial information: NCT02595931.