Soluble immune checkpoints and T-cell subsets in blood as biomarkers for resistance to immunotherapy in melanoma patients

Different mechanisms lead to immune checkpoint inhibitor (ICI) resistance. Identifying clinically useful biomarkers might improve drug selection and patients' therapy. We analyzed the soluble immune checkpoints sPD1, sPDL1, sLAG3, and sTIM3 using ELISA and their expression on circulating T cells using FACS in pre- and on-treatment blood samples of ICI treated melanoma patients. In addition, pre-treatment melanoma metastases were stained for TIM3 and LAG3 expression by IHC. Results were correlated with treatment response and progression-free survival (PFS). Resistance to anti-PD1 treatment (n = 48) was associated with high pre-treatment serum levels of sLAG3 (DCR: p = .009; PFS: p = .018; ROC cutoff >148 pg/ml) but not sPD1, sPDL1 or sTIM3. In contrast, resistance to ipilimumab plus nivolumab (n = 42) was associated with high levels of sPD1 (DCR: p = .019, PFS: p = .046; ROC cutoff >167 pg/ml) but not sPDL1, sLAG3 or sTIM3. Both treatment regimens shared a profound increase of sPD1 serum levels with treatment (p < .0001). FACS analysis revealed reduced frequencies of CD3+ CD8+ PD1 + T cells (p = .028) in anti-PD1-resistant patients, whereas increased frequencies of CD3+ CD4+ LAG3 + T cells characterized patients resistant to ipilimumab plus nivolumab (p = .033). Unlike anti-PD1 monotherapy, combination blockade significantly increased proliferating T cells (CD3+ CD8+ Ki67 + T cells; p < .0001) and eosinophils (p = .001). In melanoma metastases, an increased infiltration with TIM3+ or LAG3 + T cells in the tumor microenvironment correlated with a shorter PFS under anti-PD1 treatment (TIM3: p = .019, LAG3: p = .07). Different soluble immune checkpoints characterized checkpoint inhibitor-resistant melanoma. Measuring these serum markers may have the potential to be used in clinical routine.

Automated summary

Different mechanisms lead to resistance to immune checkpoint inhibitors in melanoma patients. Soluble immune checkpoints and T cell expression were analyzed, showing associations between certain markers and resistance to specific treatments, with potential implications for clinical use.