Document resource
Immune checkpoints maintain host integrity by enforcing graded levels of restraint across the immune system, operating not as binary regulators but as components of a continuous tolerance–resistance axis. This continuum is established by spatially and molecularly specialised microenvironments, including thymic and peripheral lymphoid priming niches, epithelial barrier surfaces and stromal–myeloid compartments. Physiological tolerance is maintained through transcriptional programmes (autoimmune regulator–forebrain embryonic zinc finger-like protein 2, NR4A, TOX, basic leucine zipper ATF-like [BATF] transcription factor), classical inhibitory pathways (PD-1, CTLA-4) and a wider network of non-canonical checkpoint regulators such as LAG-3, TIGIT, TIM-3, VISTA, B7-H3/B7-H4, Siglec-15 and HHLA2, each showing context-dependent dominance across tissue sites. Spatial transcriptomic and multiplex imaging studies now demonstrate how these circuits interact with metabolic and structural cues to enforce localised immune quiescence. Cancer subverts these physiological architectures, converting tolerance mechanisms into tumour-protective resistance. Tumours orchestrate exhaustion hierarchies, remodel lymphoid microanatomy and activate myeloid-centric suppressive networks, many of which remain unaffected by PD-1 or CTLA-4 blockade. These same processes underpin primary and acquired resistance to immune checkpoint inhibitors and are highly dependent on tissue context, stromal topology and myeloid composition. This review synthesises current knowledge across four domains: the molecular and spatial basis of physiological tolerance; how tumours hijack these programmes to generate resistance; emerging checkpoint targets beyond PD-1/CTLA-4, with emphasis on myeloid and barrier-site pathways; and clinical implications, including mechanistic insights into immune-related adverse events as manifestations of tolerance breach.