Alternative splicing in cancer drug resistance: Mechanisms and therapeutic prospects (Review).

Alternative splicing (AS) is one of the principal mechanisms of post‑transcriptional regulation that confers transcriptomic plasticity and proteomic diversity in cancer, thereby enabling tumor adaptation to therapeutic pressure. However, two obstacles impede the translation of these findings into clinical benefit: The absence of systematic functional annotation of the numerous splice variants associated with drug resistance and the paucity of biomarkers capable of distinguishing from acquired splice‑mediated resistance. In the present review, the current mechanistic understanding of AS‑driven drug resistance was briefly synthesized, and it was evaluated how existing strategies address these challenges. It was also described how knowledge of dysregulated splicing networks, due to mutations in cis‑regulatory elements such as ESS, overexpression of trans‑acting factors such as , as well as mechanisms such as alternative trans‑splicing, in which the spliceosome interacts with splice sites on two distinct RNA molecules and which can be driven by complementary sequences or other trans‑acting factors, could be used to more accurately identify tumors dependent on aberrant splicing for survival. In addition, it was outlined how targeting aberrant splice variants to overcome therapeutic resistance can be achieved, such as through spliceosome inhibition (for example, H3B‑8800) or antisense oligonucleotides directed to a specific exon or splice junction (for example, targeting exon 2 of , which is implicated in cis‑regulated AS isoforms, or alternatively spliced isoforms of , and ). However, therapeutic strategies to target adaptive resistance mechanisms such as AS remain limited, as intratumoral heterogeneity may facilitate the emergence of resistant subpopulations, and as most spliceosome inhibitors are not spliceosome‑specific, they exhibit off‑target effects. Importantly, it was also discussed how pan‑cancer splicing databases and single‑cell isoform expression profiling can be integrated with deep‑learning models, thereby informing the design of therapeutic strategies to overcome splicing‑mediated adaptive drug resistance. Notably, such integration will enable the rational design of isoform‑specific combination regimens to dismantle drug‑resistance circuits. It is anticipated that the present review will assist the scientific community, including both basic and translational researchers, in translating these findings into interventions that mitigate therapeutic failure in recalcitrant cancers.