Abstract
Alternative Lengthening of Telomeres (ALT) allows a substantial minority of human cancers to maintain their telomeres without reactivating telomerase, using a recombination-based mechanism that is mechanistically inseparable from the genomic instability it also generates. This review traces ALT from its structural hallmarks, ALT-associated PML bodies, C-circles, and G-quadruplexes through the regulatory circuit now understood to govern it: transcription of the telomeric long non-coding RNA TERRA, its conversion via RAD51AP1 from a transcription-associated R-loop into a recombination-competent D-loop, and the demonstration that this cycle behaves as a homeostat rather than a switch, self-destructive if left unrestrained. It then asks whether this dependence on strict RNA governance is unique to ALT or reflects a deeper constraint shared with telomerase, by comparing the decay machinery that polices TERC on the telomerase side (PARN, the nuclear RNA exosome) against the machinery that polices TERRA on the ALT side (XRN2, RNaseH1): structurally distinct systems converging on the same functional requirement, consistent with convergent evolution on a shared design problem rather than one pathway doing double duty. The review goes on to examine ALT’s genomic consequences, its divergent, tissue-dependent prognostic associations, its emerging sensitivity to ATR inhibition, and a candidate explanation for that prognostic paradox rooted in cGAS-STING signalling and the tumour microenvironment. It closes by identifying the central gap between cell-autonomous molecular mechanism and tumour-level clinical behaviour, and proposing specific experiments that would close it.