Impact of IDO1 expression and inhibition on the sensitivity of anti-PD-1.

Abstract

e14094 Background: Concurrent inhibition of IDO1 and PD-1 may have enhanced anti-tumor effects. However, inhibition of IDO1 failed to enhance efficacy of anti-PD-1 in late phase clinical trials. In this study, computational models were established to identify factors involved in interactions between these two types of agents. Methods: Abstracts published on IDO1, PD-1, PD-L1, anti-PD1/PD-L1 were downloaded from PubMed, and analyzed by text mining. The information on interactions among gene, compound/therapy, cell/animal model, pathway/disease, and clinical study was extracted. Networks were constructed to simulate effects of IDO1 inhibition on anti-PD-1. Results: The IDO1/anti-PD-1 network consists of 79 nodes, 227 interactions with 1277 possible regulatory paths from IDO1 to anti-PD-1. 675 paths are associated with agonism and 602 with antagonism between IDO1 inhibition and anti-PD-1. Up-regulation of [TNF-alpha, IFN-gamma, type I immunity & CD8+ T cells], and down-regulation of [Tregs, MDSC, JAK/STAT pathway, IL-6 and PD-L1] are associated with enhancement of anti-PD-1 by IDO1 inhibition. Besides, treatments with PARP inhibitor, EGFR inhibitor or radiation may enhance the combination efficacy of IDO1 and PD-1 inhibitors. Up-regulation of TIM-3 and NK cells are associated with reduced activities of anti-PD-1 upon IDO1 inhibition. PD-L1 is a predictive biomarker for anti-PD-1 response and its expression can be regulated by IDO1. The IDO1/PD-L1 network consists of 83 nodes, 204 interactions and 318 paths. Up-regulation of IL-6, AhR, MDSC, JAK/STAT, HIF1A, IL10, NF-kappaB & MYC was associated with induction of PD-L1 by IDO1. Conclusions: As the network analyses revealed, inhibition of IDO1 could either up- or down-regulate PD-L1, and enhance or reduce efficacy of anti-PD-1. Multiple factors, other than PD-L1, play roles in determining the efficacy of combining IDO1 and PD-1 inhibitors.