How Do We Predict Anti-PD-1 Therapy Response in Melanoma?

July 30, 2018
Dave Levitan
Dave Levitan

A set of reproducible biomarkers are associated with better response rates to anti–PD-1 therapy in patients with advanced melanoma.

A set of reproducible biomarkers including the presence of metachronous metastases, CD163+ histiocytes at advancing edges, and NRAS mutations, among others, are associated with better response rates to anti–programmed death 1 (PD-1) therapy, and in some cases, with improved progression-free survival in patients with advanced melanoma, according to a new study. Using these biomarkers may help guide therapeutic decision making.

“Melanomas are highly immunogenic cancers, whose prognoses have been dramatically improved by immunotherapy,” wrote study authors led by Béatrice Vergier, MD, of the Hôpital Haut-Lévêque in France. Programmed death ligand 1 (PD-L1) expression is associated with better outcomes, but perhaps due to technical challenges in its evaluation, studies have consistently found some responses to anti–PD-1 therapy in PD-L1–negative tumors. The authors hypothesized that PD-L1 expression could be combined with other clinical, histological, and/or molecular criteria to better predict responses to the therapy.

They conducted a retrospective study on 70 metastatic melanoma patients (with a total of 118 melanoma samples) who received anti–PD-1 therapy. The cohort was 58.6% male, and 61.4% were over 60 years old; most patients received pembrolizumab (72.9%), with the remainder receiving nivolumab (27.1%). The results of the analysis were published in the British Journal of Cancer.

Patients who had metachronous metastases had a better 6-month objective response rate (ORR), at 35.3%, than those with synchronous metastases (P = .04). They also had a better 6-month overall survival rate, at 82% vs 36.8% (P = .002).

PD-L1 expression of greater than 5% was also associated with better ORR, at 35.7% vs 5% for < 5% expression (P = .02). Patients who had more than 10% CD163+ cells at advancing edges had a 6-month ORR of 34.3%, compared with 0% in those with < 10% (P = .009). Those who were NRAS mutation–positive had a 6-month ORR of 53.3%, compared with 19.6% in mutation-negative patients (P = .019). The researchers created a decision-tree algorithm using the factors with significant associations with ORR; it had a 40% sensitivity and a 94% specificity.

“According to our results, PD-L1 status alone cannot be used as a reliable biomarker of therapeutic response to anti–PD-1 immunotherapy but could be combined with other criteria,” the authors concluded. “Our findings highlighted the potential biomarker role of combining CD163+ histiocytes in melanomas and their metastases with other variables but need to be validated with a larger prospective cohort.”