For patients facing one of the most surgically demanding abdominal cancers, a reliable blood-based monitoring tool would be transformative. Pseudomyxoma peritonei (PMP) already demands radical cytoreductive surgery combined with heated intraperitoneal chemotherapy, and current postoperative surveillance — relying on imaging and tumor markers like CA19-9 — is notoriously imprecise. The question of whether circulating tumor DNA (ctDNA) could fill that gap is therefore clinically urgent, even if the answer here is disappointing.
In a prospective cohort of 95 PMP patients known to harbor KRAS and/or GNAS tumor mutations — the most common oncogenic drivers in this disease — researchers applied droplet digital PCR to plasma samples across multiple timepoints. The detection rate was strikingly low: ctDNA was identified in just 8 of 95 patients (8.4%), with a median mutated allele frequency of only 0.1%, near the technical floor of reliable detection. Of those eight positives, four occurred at baseline, three at confirmed recurrence, and one during routine follow-up. Independently, high-grade peritoneal disease burden, appendiceal tumor histology, and elevated baseline CA19-9 emerged as predictors of inferior disease-free survival on Cox regression analysis.
The core problem appears biological rather than methodological. PMP characteristically grows as slow-moving mucinous deposits confined to the peritoneal cavity, shedding far less DNA into the bloodstream than epithelial tumors with hematogenous spread. This anatomical containment — the very feature that makes locoregional surgery potentially curative — also renders blood-based liquid biopsy nearly blind to the disease. The broader ctDNA literature has demonstrated excellent sensitivity in colorectal and lung cancers, but peritoneal-confined malignancies consistently underperform. This study adds rigorous prospective data confirming that KRAS/GNAS-targeted plasma ctDNA, at current sensitivity thresholds, cannot substitute for conventional surveillance in PMP. Future directions may include peritoneal fluid ctDNA or broader methylation-based approaches, but those remain investigational. For now, imaging and CA19-9 retain their imperfect but irreplaceable role.