Berzosertib Combo Shows No Response Benefit in Metastatic CRPC

The Ataxia telangiectasia and Rad3-related (ATR) inhibitor berzosertib combined with carboplatin produced no responses and more frequent severe toxicity than docetaxel plus carboplatin in patients with heavily pretreated metastatic castration-resistant prostate cancer (mCRPC), according to results from a randomized phase 2 trial (NCT03517969) published in Cancer Research Communications.¹

The overall response rate (ORR), defined as a PSA decline of 50% or more or a confirmed radiographic response, was 15% (n = 5/34) in the control arm (arm A) and 0% (n = 0/31) in the berzosertib arm (arm B). All responses in arm A occurred among the 26 patients who received docetaxel plus carboplatin, for an ORR of 19% in that group. None of the 8 patients who received carboplatin alone responded. Of the 5 responders, 2 had both a PSA and a radiographic response, and 3 had a PSA response only. None of the 14 patients who crossed over from arm A to arm B responded.

Enrollment was halted for futility at a planned interim analysis after 65 of the 130 planned patients had been treated.

The median radiographic progression-free survival (PFS) was 3.2 months (95% CI, 2.1-5.1) in arm A and 2.6 months (95% CI, 2.1-4.8) in arm B (HR, 1.14; 95% CI, 0.62-2.1). The median PFS by Prostate Cancer Working Group 3 (PCWG3) criteria was 2.5 months (95% CI, 2.1-4.2) and 2.4 months (95% CI, 2.1-3.5), respectively (HR, 1.05; 95% CI, 0.58-1.90). The median time to PSA progression was 2.1 months (95% CI, 1.4-3.6) vs 1.4 months (95% CI, 1.0-2.8; HR, 1.65; 95% CI, 0.88-3.09). The median overall survival (OS) was 8.4 months (95% CI, 4.5-12.8) in arm A and 5.7 months (95% CI, 4.6-8.1) in arm B after a median follow-up of 5.7 months (range, 1.3-26). The 6-month OS rates were 57% (95% CI, 39%-72%) and 44% (95% CI, 26%-61%), respectively.

All 65 treated patients experienced treatment-related adverse effects (TRAEs) of any grade. Grade 3/4 TRAEs occurred in 38% of patients (n = 13/34) in arm A and 65% (n = 20/31) in arm B. One patient in arm B had a grade 5 sepsis event attributed to study treatment. Grade 3/4 thrombocytopenia was more common with berzosertib (26% vs 9%). After crossover to arm B, 29% of patients (n = 4/14) had grade 3 TRAEs.

“Because there are no prior randomized trials of carboplatin plus docetaxel compared with carboplatin alone, this trial provides compelling evidence that the combination of carboplatin with docetaxel is favored clinically over carboplatin monotherapy in a biomarker-unselected population, a finding that has important implications for clinical practice,” lead author Atish D. Choudhury, MD, PhD, senior physician in the Lank Center for Genitourinary Oncology at Dana-Farber Cancer Institute and assistant professor of medicine at Harvard Medical School, wrote in the publication with study coinvestigators.¹

The randomized, open-label, multicenter phase 2 study was conducted across Experimental Therapeutics Clinical Trials Network sites. Eligible adults had an ECOG performance status of 0 to 2 and had previously received at least 1 androgen receptor pathway inhibitor and a taxane. They also needed castrate testosterone levels and disease evaluable by PSA or RECIST v1.1 criteria, and all underwent a mandatory pretreatment biopsy. There was no limit on prior lines of therapy, and prior PARP inhibitor treatment was permitted. Patients who had previously received carboplatin or an ATR inhibitor were excluded.

Patients were randomly assigned 1:1 to 1 of 2 regimens, each given every 21 days:

• Arm A: docetaxel at 60 mg/m² plus carboplatin AUC4 on day 1. Patients who were not docetaxel candidates received carboplatin AUC5 alone.
• Arm B: berzosertib at 90 mg/m² on days 2 and 9 plus carboplatin AUC5 on day 1.

Randomization was stratified by prior PARP inhibitor use and RECIST-evaluable disease, and patients in arm A could cross over at progression. Between June 2019 and July 2020, 73 patients were randomly assigned, and 65 received treatment. The median number of prior systemic therapies was 4 (range, 2-8).

The primary end point was ORR. Secondary end points included radiographic PFS, PFS per PCWG3, time to PSA progression, safety, and the relationship between homologous recombination deficiency (HRD) on baseline biopsy and clinical outcomes. OS and response after crossover were exploratory.

The investigators called the lack of synergy between berzosertib and carboplatin unexpected given preclinical data. They noted that the berzosertib dose was limited by overlapping toxicities with platinum. Because on-treatment biopsies were not performed, they could not confirm adequate ATR inhibition. Few enrolled patients had definitive HRD.

To the authors’ knowledge, this is the first randomized trial of an ATR inhibitor in prostate cancer. They noted that the ATR inhibitors ceralasertib, elimusertib, and camonsertib have shown only modest monotherapy activity in metastatic CRPC, even in biomarker-selected patients. The limited predictive value of HRR mutations for carboplatin benefit is consistent with the PRO-CARBO trial (NCT03652493), which was stopped early for limited carboplatin activity in heavily pretreated, HRR-deficient mCRPC.²

References

1. Choudhury AD, Zhong C, Xie W, et al. A phase II study of berzosertib in combination with carboplatin compared with docetaxel with carboplatin in metastatic castration-resistant prostate cancer. Cancer Res Commun. 2026;6(9):2206-2219. doi:10.1158/2767-9764.CRC-26-0488
2. Coquan E, Penel N, Lequesne J, et al. Carboplatin in metastatic castration-resistant prostate cancer patients with molecular alterations of the DNA damage repair pathway: the PRO-CARBO phase II trial. Ther Adv Urol. 2024;16:17562872241229876. doi:10.1177/17562872241229876