Hypoxia, defined as low oxygen availability, represents an early environmental stress in the evolution of solid tumors. Far from being a mere epiphenomenon, it fuels tumor progression by promoting plasticity, heterogeneity, stem-like properties, and genomic instability in cancer cells. By reprogramming the metabolism of the tumor microenvironment, it exacerbates already unfavorable conditions—an acidic, nutrient-depleted, and poorly vascularized milieu—that impair the function of immune cells. This review aims to clarify the ambivalent relationship between hypoxia and the antitumor response, and to examine how it might be exploited to predict the efficacy of immunotherapy.
The authors note that the effects of hypoxia operate through mechanisms that are both dependent on and independent of hypoxia-inducible factors (HIF). These proteins are heterodimers combining a constitutive β subunit and an inducible α subunit (HIF-1α, HIF-2α, or HIF-3α). Under normal oxygenation conditions, the α subunits are hydroxylated by prolyl hydroxylase (PHD), leading to their ubiquitination by the Von Hippel-Lindau (VHL) tumor suppressor protein and subsequent degradation by the proteasome. Under hypoxic conditions, in contrast, these proteins are stabilized and regulate the transcription of downstream genes, shaping a microenvironment hostile to immune cells and resistant to immunotherapy. Through its impact on the fundamental hallmarks of cancer and by creating a physical barrier conducive to tumor survival, hypoxia promotes the escape of cancer cells from the ongoing immune response.
The crosstalk between tumor cells and immune cells within a hypoxic microenvironment tips the balance toward a so-called "cold," immunosuppressed milieu that is refractory to immune checkpoint inhibitors (ICIs). The authors nonetheless emphasize that emerging data could turn hypoxia into an asset for improving the response to these treatments. This ambivalence—hypoxia being, in turn, an ally and an adversary of the antitumor response—is the guiding thread of the review.
The analysis of the immune context of tumors has progressively shifted toward computational approaches. A growing body of work draws on bioinformatics to deconvolve the cellular and non-cellular components of the tumor microenvironment, often combining these analyses with gene signatures serving as surrogate indicators of hypoxia. The review thus surveys numerous hypoxia-related signatures developed for various cancers—hepatocellular carcinomas, bladder, colorectal, lung, ovarian, pancreatic, and kidney cancers, among others—aimed at predicting prognosis and characterizing the immune microenvironment. The authors highlight the mechanisms by which hypoxia influences immune functions and how this understanding could, in the era of machine learning and computational biology, help anticipate the response to immunotherapy.