Combining regular liquid plasma biopsies using plasma cell-free DNA (cfDNA) and solid-tissue biopsies with a mathematical modeling of tumor evolution helped anticipate tumor progression in patients with metastatic colorectal cancer, according to the results of a phase II study published in Cancer Discovery.
These methods could inform clinicians about the “timing of clinical decisions and future treatment strategies,” according to researchers led by Nicola Valeri, MD, PhD, team leader in gastrointestinal cancer biology and genomics at the Institute of Cancer Research, London, and consultant medical oncologist at the Royal Marsden NHS Foundation Trust.
According to the study, although clinicians often use tumor biopsies for cancer genotyping, many tumors have intratumor heterogeneity that can drive treatment resistance. Therefore, multiple biopsies in time and space could help provide a better understanding on how tumors evolve to resist therapy.
Valeri and colleagues analyzed the results of the PROSPECT-C trial, a study looking at biomarkers of response and resistance to anti–epidermal growth factor receptor (EGFR) treatments in patients with KRAS/NRAS wild-type chemorefractory metastatic colorectal cancer. The trial included 47 patients enrolled from November 2012 – December 2016. Patients had tumor biopsies at regular time points at baseline, disease progression, and at partial response in some patients. Patients also provided plasma samples every 4 weeks until disease progression.
Analysis of baseline cfDNA showed that 50% of patients with tumors defined as RAS wild-type based on tissue analysis that harbored aberrations in the RAS pathway in pretreatment cfDNA. These aberrations could explain why patients were resistant to the EGFR inhibitor cetuximab. Patients with aberrations in the RAS pathways were associated with worse progression free (hazard ratio [HR] = 3.41) and overall survival (HR = 2.78) compared with wild-type disease.
Valeri and colleagues developed models that used both cfDNA and carcinoembryonic antigen levels from plasma to predict time to progression. They validated their findings using RECIST measurements taken from radiological imaging data. This model was applied to 6 patients to predict time to clinical progression; of the predictions, 3 were within 10% of progression time as measured by RECIST.
“Integration of novel monitoring technologies like cfDNA, in combination with mathematical modeling of tumor forecasting, may offer the opportunity to act early, stop therapy, or change treatment to stay one step ahead of the disease,” Valeri said in a press release. “Our method allows for a more accurate prediction as well as improved monitoring of response to therapy.”
The researchers acknowledged several limitations to the study including its small size and its focus solely on the RAS pathways. The ability of the model to predict time to progression will need to be validated in other prospective clinical trials, the researchers wrote.