Neural regulation of cancer therapy resistance is established unevenly across tumor types, signal sources, and treatments. This review proposes an evidence-calibrated framework that evaluates each candidate mechanism through its source, signal or activity pattern, receptor, target cell, and treatment context. We organize candidate neural contributions into four author-defined, overlapping analytical programs: survival and damage tolerance; cell-state plasticity and drug-tolerant persistence; neuroimmune suppression; and metabolic, vascular, and stromal adaptation. These categories facilitate comparison rather than denote equally established mechanisms: survival and selected neuroimmune routes have more direct treatment-linked support than many vascular and stromal hypotheses. Local innervation is distinguished from systemic neuroendocrine signaling and non-neuronal ligand production, while circuit integration in brain tumors is separated from predominantly paracrine and perineural interactions in peripheral cancers. An explicit hierarchy differentiates prospective biomarker-supported clinical benefit (Level 1), early interventional pharmacodynamic evidence (Level 2), treatment-specific preclinical intervention evidence (Level 3), and association (Level 4). Source resolution and rescue are assessed separately within each study. Therapy-associated sprouting, receptor adaptation, and cancer-induced nerve injury are evaluated as distinct processes; current evidence does not establish treatment-induced neural rewiring as a universal resistance driver. Study-level comparisons, model-selection criteria, and operational biomarker examples identify where neural targeting is supported, where only pathway-level dependence is demonstrated, and where additional causal validation is required. The framework is intended to improve selection and interpretation of mechanism-matched combinations, rather than infer oncologic efficacy from the availability of neuromodulatory drugs.
Background Patients with pancreatic ductal adenocarcinoma (PDAC) and concurrent diabetes mellitus frequently exhibit poor outcomes. This study aims to elucidate the molecular mechanism linking diabetes with therapeutic failure. Methods We integrated clinical analysis, diabetic-KPC mouse models, multi-omics approaches including scRNA-seq and eCLIP-seq analysis, and functional validations to investigate the mechanisms of therapy resistance in PDAC associated with diabetes. Results It was found that pancreatic cancer with concurrent diabetes is prone to chemoresistance, which is accompanied by Serine/arginine-rich splicing factor 1(SRSF1)-driven altered splicing and expression of Dihydrolipoamide S-Acetyltransferase(DLAT). Our results showed that lactate accumulation in the diabetic tumor microenvironment induces K48-lactylation and stabilization of SRSF1. Crucially, stabilized SRSF1 directly binds to DLAT pre-mRNA, promoting the retention of Exon 5. This enhanced exon retention prevents nonsense-mediated decay (NMD) of DLAT, thereby elevating DLAT protein levels.The resulting SRSF1/DLAT axis drives mitochondrial hyperfusion, conferring chemoresistance by inhibiting senescence and apoptosis. Furthermore, the hyperfusion of mitochondria facilitates immune evasion by promoting the acetylation and stabilization of PD-L1. Notably, pharmacological intervention with Berberine effectively suppressed this SRSF1-DLAT axis, thereby re-sensitizing diabetes-associated PDAC tumors to combined chemo- and immunotherapy in vivo. Conclusions Our findings reveal that the SRSF1-DLAT-mitochondrial fusion axis drives therapeutic resistance in diabetes-associated PDAC. Targeting this metabolic-immune crosstalk offers a promising strategy for PDAC treatment.
P-glycoprotein (P-gp, also known as ABCB1) is a member of the ATP-binding cassette (ABC) transporter family that protects tissues from diverse xenobiotics and contributes to multidrug resistance in cancer. Despite extensive structural, biochemical, and biophysical studies, the molecular basis of its remarkable substrate polyspecificity remains incompletely understood. Recent structures of P-gp and its homologs, including zebrafish Abcb4, have revealed extensive conformational heterogeneity and provided unprecedented views of substrate and inhibitor recognition. In this review, we integrate structural and functional studies to examine how conformational variability of the drug-binding site contributes to P-gp polyspecificity. We highlight the continuum of inward-facing conformations and the coupling between local changes in drug-binding site architecture, including transmembrane-helix rotation and tilting, and the global opening and closing of the transporter’s two pseudo-symmetric halves. We propose a dynamic conformational sampling model in which diverse substrates can recognize and possibly stabilize different conformational states of P-gp and, in some cases, adopt distinct binding poses. We further discuss how this dynamic view provides a framework for understanding the complex pharmacological behavior of P-gp and may inform the development of more effective inhibitors.
