Genomic test can predict resistance to chemotherapy in prostate cancer

A team of investigators have developed a genomic test using chromosomal instability signature biomarkers that can predict resistance to chemotherapy across multiple tumor types, including metastatic prostate cancer.¹

The HR in predicting resistance to chemotherapy was 5.47.

“It was important to us to create a test that could be easily adopted in clinic, using material we already collect during diagnosis and well-established genomic sequencing methods,” said co-author Ania M. Piskorz, PhD, Head of Genomics at the Cancer Research UK Cambridge Institute, in a news release on the findings.² “The test is based on the full DNA sequence that we get from these methods, and we can adapt it to work alongside other genomic sequencing methods that are commonly used to personalize treatment for cancer.”

For the study, the investigators assessed the ability of the test to predict resistance to platinum-, taxane-, and anthracycline-based chemotherapies across 840 patients with metastatic prostate cancer, ovarian cancer, sarcoma, or metastatic breast cancer.

The metastatic prostate cancer cohort consisted of 238 adult patients. Among these, the test identified 91 patients at risk of chemotherapy resistance and 147 patients as sensitive to chemotherapy.

Within both the “chemotherapy resistant” and “chemotherapy sensitive” cohorts, patients were assigned to the experimental arm (taxane) if they had received at least 3 cycles of taxane-based chemotherapy or to the control arm (standard of care) if they had received any of the 5 most common non-taxane therapies in the cohort (abiraterone acetate [Zytiga], enzalutamide [Xtandi], olaparib [Lynparza], pembrolizumab [Keytruda], or radium-223 [Xofigo]). The treatment cohorts were then compared in each arm to assess the time to treatment failure, thus mimicking a randomized control trial.

Overall, among those predicted to be resistant to taxane-based chemotherapy, patients in the experimental arm demonstrated a significantly elevated risk of treatment failure vs those in the standard of care arm (HR, 5.47; 95% CI, 2.19 to 13.63; P = .0003). However, there was no statistical difference observed between the treatment arms in patients who were predicted to be sensitive to chemotherapy (HR, 0.98; P = .950).

The authors also noted, “Similar results were observed when limiting the analyses to single-agent taxane treatment.”

The test also demonstrated the ability to predict chemotherapy resistance in the other tumor types assessed. Patients who were identified as “chemotherapy resistant” demonstrated an increased risk of treatment failure with taxane-based (HR, 7.44) and anthracycline-based (HR, 1.88) therapy for ovarian cancer and taxane- (HR, 3.98) and anthracycline-based (HR, 3.69) therapy for metastatic breast cancer.

Additionally, “Nonrandomized emulations showed predictive capacity for platinum resistance in ovarian (HR, 1.46) and anthracycline in sarcoma (HR, 3.59),” the authors wrote.

Based on these findings, the authors suggest that chromosomal instability signatures may serve as predictive biomarkers for chemotherapy benefit, thus offering an opportunity to personalize therapy regimens in these patients.

“Chemotherapy is a mainstay of cancer treatment and saves many lives. Yet in many cases, it has been administered the same way for over 40 years. Sadly, there are too many cases where cancer is resistant to chemotherapy treatment – meaning unpleasant side-effects for the patient with limited benefit to them,” said co-author Professor James D. Brenton of the Cancer Research UK Cambridge Institute, in the news release.² “With genomic sequencing now more widely available, we can make some of the most well-established chemotherapies work better. By understanding who is most likely to respond to it, chemotherapy could become a more tailored treatment across different types of cancer.”

REFERENCES

1. Thompson JS, Madrid L, Hernando B, et al. Predicting resistance to chemotherapy using chromosomal instability signatures. Nat Genet. 2025 Jun 23. doi:10.1038/s41588-025-02233-y

2. Scientists create test to predict chemotherapy resistance in patients. News release. Cancer Research UK. June 23, 2025. Accessed June 26, 2025. https://www.eurekalert.org/news-releases/1088321