Long-read pan-cancer transcriptomics unravel distinct alteration trends between gene and isoform expression in tumorigenesis

This study leveraged Nanopore‑based long‑read RNA sequencing (lrRNA‑seq) to generate full‑length transcriptomic profiles for 144 paired tumor‑normal tissue sets spanning ten human organs, building a large‑scale bulk pan‑cancer resource to disentangle gene‑level and isoform‑level transcriptomic perturbations during tumorigenesis. After stringent quality filtering to remove outlier specimens, the final dataset demonstrated high technical reliability: transcript read‑length distribution, 5’‑end CAGE‑peak compatibility, 3’‑end poly‑A motif occupancy, and cross‑platform correlation against matched short‑read RNA‑seq collectively validated the integrity of full‑length transcript reconstruction. In total, nearly 300,000 isoforms were detected, among which a substantial proportion represented previously unannotated entities including novel splice variants and tumor‑associated fusion transcripts. Pan‑cancer comparative analysis exposed two uncoupled layers of transcriptomic dysregulation. At the gene expression level, tumors originating from diverse organs displayed convergent transcriptional signatures tightly linked to oncogenic dedifferentiation, a cellular reversion toward stem‑like phenotypic states. Many convergently altered genes, commonly down‑regulated across multiple tumor types, participated in tissue‑specific physiological metabolism and ion homeostasis. By contrast, isoform‑level remodeling behaved in a highly divergent, organ‑restricted manner. Cross‑organ comparisons indicated that tumor isoform‑ratio correlations resembled those observed in normal tissues, contradicting the heightened inter‑tumor correlation seen for gene expression. This key observation illustrated that widespread isoform shifts occur largely independently of bulk messenger‑RNA abundance changes, a phenomenon routinely missed by conventional short‑read workflows that mostly quantify aggregated gene‑level counts rather than individual full‑length transcripts. Thousands of tumor‑specific isoforms covering all seven major classes of alternative‑splicing events were recovered across cancer types, alongside hundreds of recurrent fusion transcripts with partial PCR‑Sanger experimental validation supporting their authenticity.

Splicing‑factor dysregulation constitutes one major driving force behind the extensive isoform reprogramming observed in malignant tissues. Using a random‑forest regression model trained on paired tumor‑normal expression profiles, this work quantified the predictive contribution of over four hundred splicing‑factor genes toward genome‑wide isoform‑composition shifts, pinpointing a small subset of high‑impact regulators such as HNRNPC and RBFOX2 that are known to govern alternative‑splicing circuits in cancer. Importantly, a considerable subset of genes exhibited robust isoform‑ratio alterations without any statistically significant difference in overall gene‑level expression. Genes such as DUT exemplified this pattern: pronounced isoform compositional remodeling took place in liver tumors while total gene transcript abundance remained stable. Such findings underscore a critical limitation of short‑read‑oriented pan‑cancer investigations: many oncogenic transcriptomic events manifest exclusively through altered transcript‑isoform proportions and structural variants, invisible when analytical pipelines collapse all reads onto aggregated gene models. It should be emphasized that all bulk lrRNA‑seq measurements represent population‑averaged signals derived from heterogeneous tissue biopsies containing malignant cells, stromal cells and immune infiltrates. Consequently, detected isoform alterations and fusion transcripts cannot be definitively mapped solely to cancer cells; shifts may partly reflect variations in tumor purity and cellular composition within specimens. Even though sensitivity analyses using deconvolution approaches confirmed that core isoform‑ratio trends persisted irrespective of purity fluctuations, the exact cellular origin of most detected transcript‑level aberrations awaits future single‑cell long‑read RNA‑seq experimentation to resolve intratumoral heterogeneity and assign molecular events to defined cell subpopulations.

To integrate these multi‑layer transcriptomic dimensions and prioritise biologically meaningful candidate genes, the present study constructed a custom pan‑cancer scoring framework that merges gene‑level differential expression, convergent‑divergent variance patterns, tumor‑specific isoform occurrence, fusion‑transcript evidence and isoform‑ratio perturbations, with supplementary weighted evidence retrieved from TCGA and the Cancer Gene Census database. Each gene received an organ‑specific score capped at one, which was then aggregated across all ten organs to compute a final pan‑cancer susceptibility score. More than ten thousand genes were scored and stratified into low‑, medium‑ and high‑score groups. High‑score entries included well‑characterized canonical cancer drivers such as TP53, RB1 and FAT4. More notably, the scoring system successfully unearthed under‑appreciated susceptibility genes whose dysregulation is dominated by isoform‑level defects rather than total gene‑expression change. Representative examples are IBSP, involved in extracellular‑matrix organization and metastasis, and FANCL, which participates in DNA‑damage‑repair pathways; both display recurrent tumor‑specific isoforms across multiple cancer types without prominent gene‑level dysregulation. Kaplan‑Meier survival analysis demonstrated that selected high‑score genes carried significant prognostic implications, whereas near‑zero‑score genes showed no survival association, confirming the framework’s capacity to nominate prognosis‑relevant molecules. This scoring resource is not intended for early‑cancer screening or treatment‑response prediction given the cohort’s lack of complete clinical follow‑up metadata. Collectively, this bulk long‑read pan‑cancer atlas expands our understanding of transcriptomic complexity in human tumorigenesis, furnishes a public resource for isoform‑centered cancer research, and supplies actionable candidates for downstream functional validation and translational biomarker or therapeutic‑target development.