Abstract: PD1 expression in CD4+ and CD8+ T cells is increased after treatment in multiple myeloma patients with persistent disease. The GEM-Pembresid trial analyzed the efficacy and safety of pembrolizumab as consolidation in patients achieving at least very good partial response but with persistent measurable disease after first- or second-line treatment. Moreover, the characteristics of the immune system were investigated to identify potential biomarkers of response to pembrolizumab. One out of the 17 evaluable patients showed a decrease in the amount of M-protein, although a potential late effect of high-dose melphalan could not be ruled out. Fourteen adverse events were considered related to pembrolizumab, two of which (G3 diarrhea and G2 pneumonitis) prompted treatment discontinuation and all resolving without sequelae. Interestingly, pembrolizumab induced a decrease in the percentage of NK cells at cycle 3, due to the reduction of the circulating and adaptive subsets (0.615 vs. 0.43, p = 0.007; 1.12 vs. 0.86, p = 0.02). In the early progressors, a significantly lower expression of PD1 in CD8+ effector memory T cells (MFI 1327 vs. 926, p = 0.03) was observed. In conclusion, pembrolizumab used as consolidation monotherapy shows an acceptable toxicity profile but did not improve responses in this MM patient population. The trial was registered at clinicaltrials.gov with identifier NCT02636010 and with EUDRACT number 2015-003359-23.