A highly sensitive circulating tumor DNA (ctDNA) assay directed at KRAS mutations can detect residual disease in substantially more patients with localized pancreatic cancer than standard next-generation sequencing (NGS), a prospective study showed.

Researchers found that digital droplet polymerase chain reaction (ddPCR) targeting three common KRAS mutations identified ctDNA in nearly four times as many patients as standard NGS at diagnosis. And it continued to detect residual disease after chemotherapy and surgery when NGS did not.

The findings suggested that ddPCR can more accurately risk-stratify patients and, potentially, help guide the use of emerging KRAS-targeted therapies. 

“We showed that a highly sensitive KRAS-directed assay identifies a substantial group of patients with biologically aggressive disease who would have been considered ctDNA-negative by standard NGS alone,” said Akhil Chawla, MD, surgical oncologist at the Robert H. Lurie Comprehensive Cancer Center of Northwestern University in Chicago.

“Today, that information can help inform multidisciplinary discussions, surveillance, and clinical trial selection,” Chawla told Medscape Medical News. “As biomarker-driven perioperative trials and KRAS-targeted therapies emerge, accurately identifying these patients will become increasingly important.”

The study was published online in Clinical Cancer Research.

Spotting What NGS Misses

Pancreatic ductal adenocarcinoma (PDAC) remains one of the deadliest malignancies, with poor long-term survival even among patients undergoing surgery and multimodality treatment.

ctDNA has emerged as a promising marker of minimal residual disease and recurrence risk. While NGS enables broad mutational profiling, ddPCR offers higher sensitivity for specific targets, including the KRAS mutations that are present in most pancreatic cancers. 

For their study, Chawla and his colleagues evaluated the prognostic significance of KRAS-mutant ctDNA in 106 patients with localized PDAC and tested whether ddPCR revealed clinically relevant prognostic information missed by NGS.

Blood samples were collected at diagnosis, after neoadjuvant chemotherapy, and following surgical resection. The researchers compared tumor-agnostic NGS assays with ddPCR targeting the KRAS G12D, G12V, and G12R mutations, which account for most KRAS alterations in pancreatic cancer.

At diagnosis, ddPCR detected KRAS ctDNA in 65% of patients compared with 17% using NGS. After chemotherapy, detection with NGS dropped to 4.5% but remained high with ddPCR at 60%. And following surgery, ddPCR continued to detect KRAS ctDNA in 56% of patients vs 8.5% with NGS.

Patients who were positive on both tests had the poorest outcomes, with a median overall survival of about 11 months. That compared with a median overall survival of nearly 41 months among patients who were negative on both assays.

Patients missed by NGS but detected by ddPCR represented an intermediate-risk group, with a median survival of about 27 months.

“In other words, a negative NGS result may not be fully reassuring in localized pancreatic cancer,” said Mark Ashamalla, MD, chief of radiation oncology at Episcopal Health Services, in Far Rockaway, New York.

Implications for KRAS-Targeted Therapy

Ashamalla, who was not involved in the study, said the clinical implications are unclear at this point.

“Greater sensitivity for minimal residual disease detection absolutely has the potential to change management,” he told Medscape Medical News. “But I would say it’s still too early to call this routine practice-changing.”

For now, Ashamalla sees the greatest value in improving risk stratification, tailoring surveillance intensity, and informing the design of clinical trials evaluating KRAS-directed therapies.

The timing is notable. Last month, the phase 3 RASolute 302 trial showed that daraxonrasib, a pan-RAS inhibitor, doubled overall survival compared with chemotherapy for patients with previously treated metastatic pancreatic cancer.

But because the new findings come from a prognostic study, not an interventional one, it remains to be seen whether ddPCR is a useful way to guide therapy, Ashamalla said.

“We don’t yet know that changing therapy based on ddPCR positivity improves survival,” he stressed. “I would not change standard treatment solely on the basis of this assay today.”

Still, Ashamalla said the study points toward where the field is going: molecularly guided perioperative therapy.

According to the researchers, integrating ddPCR with standard NGS measures might provide additional prognostic information. But Chawla emphasized that prospective validation is the next critical step.

“Our study establishes that higher-sensitivity ctDNA testing improves risk stratification,” he said. “A critical question is whether acting on that information improves patient outcomes. The next step is a prospective biomarker-directed clinical trial.”

Chawla reported being supported by the Elsa U. Pardee Foundation and a grant from the National Institutes of Health. Ashamalla reported having no relevant disclosures.