Summary
In a randomized phase 2 trial of 54 men with high-volume metastatic hormone-naive prostate cancer, talazoparib plus enzalutamide produced a high PSA response rate but no clear advantage over enzalutamide alone. Severe adverse events were more frequent with the combination, and two patients developed fatal acute myeloid leukaemia.
A randomized phase 2 trial in 54 men with high-volume metastatic hormone-naive prostate cancer found no clear additional clinical benefit from adding talazoparib to enzalutamide, while severe adverse events were more common with the combination. The ZZ-FIRST study, published in Nature Medicine, also used tumour biopsies to examine how prostate cancer cells respond to androgen-receptor inhibition.
Talazoparib is a PARP inhibitor, a drug that interferes with enzymes involved in DNA repair. Enzalutamide blocks signalling through the androgen receptor, which prostate cancer cells can use to grow. Both groups also received androgen-deprivation therapy (ADT), which lowers the effects of male sex hormones.
A high response rate, but no clear advantage between groups
The multicentre, open-label trial enrolled patients at eight sites in Spain. All began ADT and enzalutamide; after two 28-day cycles, they were randomly assigned to continue that treatment or add talazoparib. The study included 37 patients in the combination group and 17 in the enzalutamide-only group.
The trial’s primary endpoint was a confirmed prostate-specific antigen (PSA) level below 0.2 ng ml−1 at 12 months. This response occurred in 27 of 37 patients (73.0%) receiving talazoparib plus enzalutamide, exceeding the study’s prespecified 40% efficacy threshold. In the control group, 11 of 17 patients (64.7%) reached the same endpoint. An exploratory comparison between the groups found no statistically significant difference (P = 0.54).
The results therefore show activity for the combination, but the primary endpoint does not demonstrate that it worked better than enzalutamide alone. A secondary measure, median radiographic progression-free survival, was 45.3 months with the combination and 31.1 months with enzalutamide alone. The estimated hazard ratio was 0.64, with a 95% confidence interval of 0.28–1.42 (P = 0.265). The interval and comparison do not establish a clear difference between treatments.
Severe (grade 3 or higher) treatment-emergent adverse events occurred in 25 of 37 patients (67.6%) in the combination group and 6 of 17 (35.3%) in the control group. Severe anaemia was the most common such event with the combination, affecting 15 patients (40.5%). Two patients receiving talazoparib and enzalutamide developed acute myeloid leukaemia and both died. One case was considered related to talazoparib; the other was considered possibly related. Baseline testing identified a DNMT3A mutation in one patient and a germline FLT3 mutation in the other, both associated with susceptibility to myeloid malignancies.
Biopsies point to cell-cycle changes, not clear loss of repair function
The researchers collected tumour biopsies before treatment and after about four weeks of enzalutamide, before talazoparib was introduced. Spatial transcriptomic analysis—a method that measures gene activity while retaining information about where it occurs in tissue—was possible in 34 biopsy samples, including matched before-and-after samples from 12 patients.
After androgen-receptor inhibition, tumour samples showed reduced activity in gene programmes associated with cell division and proliferation. Gene signatures related to homologous recombination repair (HRR), a process cells use to repair certain forms of DNA damage, also fell. The researchers found that this reduction largely tracked the suppression of cell-cycle activity.
Functional tests of DNA damage and HRR activity did not show a clear reduction in RAD51 foci after hormonal treatment. These tests included a small number of on-treatment samples from ZZ-FIRST and a separate group of prostate cancer samples collected during a neoadjuvant study. The authors interpret the combined findings as evidence that lower HRR gene expression after androgen-receptor inhibition mainly reflects reduced proliferation, rather than a newly induced functional HRR-deficient state.
That distinction matters because PARP inhibitors are best established for tumours with particular DNA-repair gene alterations. In this trial, the combination did not establish clear extra benefit in a biomarker-unselected population, and treatment was associated with greater toxicity. The investigators note that the small phase 2 study limits the precision of treatment comparisons; its subgroup findings are exploratory. Larger studies would be needed to determine which patient groups, if any, gain a meaningful additional benefit.