The PD-1 inhibitors nivolumab and pembrolizumab are currently widely used as treatment of advanced melanoma. While being tolerated very well by some patients, they cause severe and even fatal immune related adverse events (irAEs) in others. Factors associated with increased risk of toxicity have barely been identified. With the perspective of long-term survival in stage IV disease, and increasing numbers of patients receiving adjuvant immunotherapy in stage III melanoma, being able to predict severe (≥grade 3) toxicity becomes more and more important. In this study, we aimed to assess factors associated with severe irAEs in clinical data and routine peripheral blood tests.
Patients with advanced (stage III-IV) melanoma, treated with nivolumab or pembrolizumab with at least 3 months follow-up were included. Demographical parameters together with lactate dehydrogenases (LDH), C-reactive protein (CRP), leukocytes, eosinophils, neutrophils and lymphocytes at baseline and at last measurement before toxicity were retrieved retrospectively. IrAEs were graded according to CTCAE 4.03. Non parametric tests were used to assess the difference between groups.
60 patients treated with anti-PD1 monotherapy were included, of whom 7 developed ≥grade 3 irAEs. Median CRP at baseline was significantly higher in patients that experienced severe irAEs (25mg/L) compared to patients who did not (3 mg/L; p = 0,038). Furthermore, in patients with severe irAEs, median eosinophil level at last measurement before toxicity (0,27x109/L) was significantly increased compared to baseline (0,15x109/L; p = 0,047).
Our data indicate that baseline CRP is a potential predictor of severe anti-PD1 induced toxicity. If confirmed in a larger study, CRP could play a role in clinical decision making at the start of treatment. Furthermore, our observation that increase of eosinophils precedes ≥grade 3 toxicity underpins its possible role in the pathogenesis of irAEs and could help identify and treat irAEs at an early stage.
Clinical trial identification
Legal entity responsible for the study
University Medical Center Utrecht.
Has not received any funding.
K. Suijkerbuijk: Consulting/advisory relationship: Bristol-Myers Squibb, MSD; Honoraria (paid to institution): Novartis, Roche. All other authors have declared no conflicts of interest.