PROpel shows high ctDNA concordance

Prostate cancer genomics is running into a very practical problem — the biology is sophisticated, but the sample often is not. In metastatic castration-resistant prostate cancer, treatment decisions can hinge on whether a patient carries a homologous recombination repair, or HRR, mutation, especially BRCA1 or BRCA2. But tissue from bone-heavy metastatic disease is often old, scant, or just unusable. The new PROpel biomarker analysis matters because it shows blood-based ctDNA testing can recover a lot of that missing information, and do it with pretty strong agreement against tissue. ### What was actually analyzed? PROpel was the phase 3 first-line mCRPC trial of olaparib plus abiraterone versus placebo plus abiraterone. Patients were enrolled regardless of HRR mutation status, but the study prospectively collected tumor tissue and plasma for genomic testing using FoundationOne CDx for tissue and FoundationOne Liquid CDx for ctDNA. The new analysis focused on how often those two methods agreed and what happened when the trial used both together. (ascopubs.org) ### Why is ctDNA even a big deal here? Because prostate cancer is unusually annoying to genotype from tissue. Many men with mCRPC have metastases in bone, and bone biopsies are technically hard, can yield low tumor content, and may not leave enough DNA for a clean next-generation sequencing readout. A blood draw is easier, faster, and repeatable — but only if the tumor is shedding enough DNA into the bloodstream to be detectable. (clinicaltrials.gov) ### So how close was the match? Pretty close. In 491 matched tumor-tissue and ctDNA samples, overall percent agreement was 85.1% for HRR mutation status and 93.9% for BRCA mutation status when tissue was used as the reference. Negative predictive value was 93.7% for HRR and 97.3% for BRCA, which is the reassuring part if you are worried about ctDNA missing too many true positives. ### What were the actual testing success rates? (ascopubs.org) This is where the story gets more useful. HRR status was obtained for 778 of 796 randomized patients — about 98% — when tissue and ctDNA results were aggregated. The paper says that combining the two methods identified 226 patients with HRR mutations, including 85 with BRCA mutations. The point is simple: using both tests solved more cases than leaning on one alone. ### Where did the mismatches come from? Mostly from low ctDNA fraction. That means the blood sample did not contain much tumor DNA, so the assay had less signal to work with. This is the catch with liquid biopsy in prostate cancer — a negative plasma result can mean “no mutation,” but sometimes it means “not enough tumor DNA in circulation today.” That is why tissue still matters. ### Does this mean blood can replace tissue? (ascopubs.org) Not really. It means blood can rescue the workflow. Tissue remains the reference standard in the analysis, and tissue can capture alterations that plasma misses when shedding is low. But ctDNA adds speed, accessibility, and a second shot at finding actionable mutations when archival tissue is exhausted or a fresh biopsy is unrealistic. ### Why does BRCA matter more than the broader HRR bucket? (ascopubs.org) Because BRCA mutations are the clearest predictive biomarker for PARP inhibitor benefit in prostate cancer. PROpel itself enrolled all comers, but later analyses have made the BRCA subgroup especially important for interpreting benefit. So the 93.9% BRCA agreement number is the one clinicians will probably remember. ### Bottom line? The clean read on PROpel is that ctDNA is not a replacement for tissue in mCRPC, but it is no longer just a backup plan. (ascopubs.org) In this dataset, it tracked tissue well enough — especially for BRCA — to make combined testing look like the practical way to minimize missed patients and reduce dead-end specimen workups.