Immune checkpoint inhibitors targeting the programmed death receptor 1 (PD-1) axis are current standards of care for metastatic melanoma as single agents or in combination with the CTLA-4 inhibitor ipilimumab, but more than half of patients do not respond to immunotherapy. Unfortunately, there are no reliable biomarkers available in melanoma to accurately predict for success of immunotherapy treatment. However, circulating tumor DNA (ctDNA) is indicative of tumor burden in patients with melanoma, and elevated ctDNA is typically associated with decreased response and shortened progression-free survival (PFS).
A longitudinal cohort study sought to evaluate the predictive value of ctDNA for response, PFS, and overall survival (OS) in 76 patients with metastatic melanoma treated with PD-1 inhibitors, including either pembrolizumab or nivolumab monotherapy or nivolumab/ipilimumab combination. Efficacy following treatment with PD-1 inhibitor was evaluated in patients with undetectable ctDNA at baseline and during therapy (Group A), patients with elevated ctDNA at baseline that became undetectable within 12 weeks of therapy (Group B), and patients with elevated levels of ctDNA at baseline that remained elevated despite therapy (Group C).
While pretreatment levels of ctDNA did not have an impact on patient outcomes on PD-1 inhibitors, elevated ctDNA during treatment was a negative prognostic factor. The objective response rates (ORR) were 72% in Group A and 77% in Group B, compared to 6% in Group C. Only patients in Group A achieved a complete response (n = 3). Likewise, the median PFS and overall survival (OS) were improved in patients with low or undetectable levels of ctDNA compared to those with elevated levels during treatment. Median PFS was not reached in groups A and B, but was only 2.7 months for patients in Group C (Group A vs C: HR 0.08, P<.001; Group B vs C: HR 0.15, P<.001). Similarly, median OS was not reached in Groups A and B, and was 9.8 months for patients in Group C (Group A vs C: HR 0.11, P<.001; Group B vs C: HR 0.12, P<.001). Results were consistent in multivariate analysis when adjusted for LDH, performance status, tumor stage, and disease volume. Of interest, ctDNA levels in patients with predominant brain metastases were not accurate predictors of response, supporting observations from other studies that the blood-brain barrier may prevent ctDNA from entering the circulation.
The investigators concluded that ctDNA at baseline and during therapy is an accurate marker of anti-PD-1 therapy efficacy in metastatic melanoma. Persistent elevation of ctDNA during therapy is associated with a poor prognosis, and this may be useful in guiding sequencing and combination of therapies. For patients with detectable baseline ctDNA, they suggest repeating ctDNA 5-8 weeks after first dose of anti-PD-1 therapy to identify those who may benefit from treatment; however, they noted that these results must be validated in a larger cohort study.