Ian Jhemes Oliveira Sousa
Kerolayne de Melo Nogueira
Ester Miranda Pereira- 1Northeast Biotechnology Network, Federal University of Piauí, Teresina, Brazil
- 2Postgraduate Program in Pharmacology, Universidade Federal do Ceara, Fortaleza, Brazil
- 3Laboratory of Immunogenetics and Molecular Biology, Universidade Federal do Piaui, Teresina, Brazil
1 Introduction: from morphology to functional precision in oncology
Modern oncology has witnessed a paradigm shift, moving from therapeutic approaches based on histopathological and anatomical characteristics to personalized strategies defined as Precision Medicine (1). This evolution is driven by the increasing ability to characterize tumors molecularly, identifying the genetic and epigenetic alterations that drive disease growth and progression. Although DNA sequencing (genomics) has provided crucial insights into cancer-associated mutations, a comprehensive understanding of tumor biology increasingly depends on functional molecular analyses, particularly transcriptome profiling (2).
The transcriptome, which represents the complete set of RNA molecules in a cell at a given time, reflects real-time genetic activity and, consequently, the functional state of a tumor (1). Although most advances in transcriptomics and proteomics are associated with the study of cancer, these approaches have also been widely applied to elucidate pathophysiological mechanisms in other complex diseases. Investigations in this direction demonstrate how the characterization of molecular profiles can reveal crucial early events in pathogenesis, providing support for the development of more precise diagnostic and therapeutic strategies (3).
High-throughput RNA sequencing (RNA-seq) technologies have enabled an unprecedented exploration of this molecular universe, revealing not only the expression of protein-coding genes, but also non-coding RNAs, splicing isoforms and gene fusions, which play critical roles in oncogenesis and treatment response (2). Total tumor transcriptome analysis, therefore, offers a comprehensive and functional view of the molecular identity of a cancer, overcoming the limitations of static genomic analysis by capturing the dynamics of cellular adaptation. This integration makes it possible not only to identify molecular signatures with diagnostic and prognostic value, but also to delineate more specific therapeutic targets, consolidating itself as an essential element for the development of the next generation of personalized cancer strategies (4).
Extracting biologically relevant patterns from these high-dimensional datasets will enable the discovery of more sensitive and specific biomarkers, the refinement of patient stratification processes, and the design of more effective and personalized therapeutic interventions. Thus, the most immediate impact of transcriptomics lies in its transformative potential on cancer diagnosis (5). While conventional histopathology, although fundamental, has limitations in capturing the complexity and heterogeneity of tumor behavior, transcriptome analysis paves the way for more refined molecular characterization (6).
2 Clinical applications of whole tumor transcriptomics in precision oncology
This potential extends to early detection, where the analysis of cell-free RNA (cfRNA) in liquid biopsies can allow the identification of tumors even before the onset of clinical symptoms (7). Specific transcriptomic signatures of cfRNA can indicate the presence of tumor cells and their tissue of origin, facilitating faster interventions. In addition, many cancers, such as breast and lung cancer, are collections of diseases with distinct molecular profiles. The transcriptome allows the identification of molecular subtypes, such as Luminal A, Luminal B, HER2-enriched, and basal subtypes in breast cancer, that are not discernible by histopathology but are essential to guide treatment selection (8).
This molecular precision is also invaluable for cancers of unknown primary, which account for 3%−5% of all cancer cases. For these challenging diagnoses, metastasis transcriptome analysis can reveal the tumor's tissue of origin, allowing for more targeted and effective therapy (9). However, the complexity of RNA-seq data requires sophisticated computational methods, such as machine learning algorithms like Support Vector Machines and neural networks, to build accurate classifiers (10). At this stage, it is essential to clearly distinguish exploratory models that demonstrate strong performance in experimental cohorts from computational tools that have achieved clinical validation and meet regulatory requirements for routine diagnostic use.
Beyond this distinction, important challenges persist, including the confounding effects of tumor heterogeneity, where the presence of non-cancerous cells can obscure tumor-specific signals (11). Furthermore, variations in laboratory protocols require robust data normalization methods to ensure comparability and reproducibility (12). In addition to refining diagnosis, the transcriptome provides an equally powerful lens for predicting disease trajectory. Prognosis, which underpins decisions about treatment intensity and follow-up, is traditionally based on clinicopathological factors that often fail to capture the biological aggressiveness of a tumor. Transcriptomic analysis can identify gene expression signatures that stratify patients into high- and low-risk groups for recurrence or disease progression, thus personalizing treatment intensity and avoiding overtreatment or undertreatment (13).
This is particularly critical for predicting cancer recurrence, a major challenge in oncology. Transcriptomic signatures can predict the likelihood of recurrence, allowing for more rigorous monitoring and the proactive use of adjuvant therapies in high-risk patients (14). In fact, this concept is already in clinical practice with tests such as Oncotype DX® and Prosigna® for breast cancer, which use gene expression signatures to predict the risk of recurrence and guide chemotherapy decisions (15).
The computational approaches behind these tools involve machine learning algorithms trained on large patient cohorts. More advanced methods, such as gene co-expression network analysis, provide additional prognostic insights by identifying co-regulated gene modules linked to specific biological processes, such as cell proliferation or invasion (16). Integrating this molecular data with established clinical factors in multivariate models can further increase prognostic accuracy (17).
However, the development and validation of these models require large, well-annotated patient cohorts with long-term follow-up, and any model must be externally validated to ensure its generalizability. Furthermore, improving the interpretability of these often “black box” computational models is crucial for building clinical confidence and understanding the underlying biology. Ultimately, improved diagnoses and prognoses should translate into more effective treatments, which is the central promise of Precision Medicine. The whole tumor transcriptome is fundamental in this endeavor, providing detailed information to guide the selection of therapies that maximize efficacy and minimize toxicity (18).
Its most direct application is the identification of actionable therapeutic targets. The transcriptome can reveal the overexpression of genes that encode drug targets, such as growth factor receptors, or identify unique molecular vulnerabilities, such as targetable gene fusions and aberrant splicing variants that are invisible to genomic analysis alone (19).
Furthermore, transcriptomic signatures can predict a tumor's sensitivity or resistance to specific treatments. For example, while PD-L1 protein expression is a common biomarker for immunotherapy, a broader transcriptomic profile of the tumor immune microenvironment can offer a more comprehensive prediction of response to PD-1 blockade (20).
This dynamic monitoring capability is also critical for tracking treatment resistance. As tumors evolve under therapeutic pressure, serial transcriptomic analysis, especially from non-invasive liquid biopsies, can detect the emergence of resistance mechanisms, allowing physicians to modify treatment strategies in real time (21).
3 Barriers, limitations, and future directions
Despite these advances, the translation of these vast transcriptomic data into clinical decisions is computationally hard, relying on pathway and network analyses to understand drug effects and on machine learning models trained on large drug response datasets to predict sensitivity in silico (22). The main challenges in this domain are the complex and multifactorial nature of drug resistance and the long and high risk of developing new drugs against newly identified targets. The path to the widespread clinical implementation of transcriptome-guided oncology requires overcoming several significant practical obstacles. The main one is the need for rigorous standardization of sample handling, sequencing, and data analysis protocols to ensure that results are reproducible and comparable across different centers (13, 23).
The enormous volume of data generated also requires a robust and accessible computational infrastructure for storage and processing, which remains a bottleneck for many institutions (24). To achieve a truly holistic understanding of cancer, it will be essential to integrate transcriptomic data with other “omics” layers such as: genomics, proteomics, and metabolomics, a task that represents a major computational and statistical challenge (25).
As we deal with this deeply personal patient data, we need to navigate the associated ethical and privacy issues, ensuring the existence of robust regulatory frameworks to protect patients while enabling research (26). However, the future is rich in opportunities as continuous advances in artificial intelligence and deep learning promise to uncover even more complex patterns in transcriptomic data, and technologies such as single-cell RNA sequencing (scRNA-seq) are already providing unprecedented results. Finally, standardization of analyses and the application of best practices remain fundamental to the consolidation of transcriptome-based precision oncology (27–29).
In conclusion, whole-tumor transcriptome analysis represents an indispensable cornerstone for the evolution of precision oncology. By capturing functional dynamics that transcend static genomic information, transcriptome-guided approaches offer unprecedented resolution for refined diagnosis, definition of a prognosis based on cellular status, and personalized therapeutic selection, which can in itself overcome the challenges of tumor unpredictability and contribute to the next generation of personalized cancer therapies.
Author contributions
IS: Conceptualization, Writing – original draft, Writing – review & editing, Formal analysis, Methodology, Validation. KN: Writing – review & editing, Writing – original draft. EP: Writing – original draft, Writing – review & editing.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1. Ergin S, Kherad N, Alagoz M. RNA sequencing and its applications in cancer and rare diseases. Mol Biol Rep. (2022) 49:2325–33. doi: 10.1007/s11033-021-06963-0
2. Hong M, Tao S, Zhang L, Diao LT, Huang X, Huang S, et al. RNA sequencing: new technologies and applications in cancer research. J Hematol Oncol. (2020) 13:166. doi: 10.1186/s13045-020-01005-x
3. Hasin Y, Seldin M, Lusis A. Multi-omics approaches to disease. Genome Biol. (2017) 18:83. doi: 10.1186/s13059-017-1215-1
4. Neto JTM, Kirsztajn GM, Pereira EM, Andrade HM, Oliveira IHR, Labilloy A, et al. Proteomic profiling of engineered human immortalized podocyte cell model of Fabry disease. Mol Genet Metab. (2019) 126:S106–7. doi: 10.1016/j.ymgme.2018.12.269
5. Kourou K, Exarchos TP, Exarchos KP, Karamouzis MV, Fotiadis DI. Machine learning applications in cancer prognosis and prediction. Comput Struct Biotechnol J. (2015) 13:8–17. doi: 10.1016/j.csbj.2014.11.005
6. Conesa A, Madrigal P, Tarazona S, Gomez-Cabrero D, Cervera A, McPherson A, et al. A survey of best practices for RNA-seq data analysis. Genome Biol. (2016) 17:13. doi: 10.1186/s13059-016-0881-8
7. Ignatiadis M, Sledge GW, Jeffrey SS. Liquid biopsy: current challenges and future directions. Nat Rev Clin Oncol. (2021) 18:189–201. doi: 10.1038/s41571-020-00457-x
8. Perou CM, Sørlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, et al. Molecular portraits of human breast tumours. Nature. (2000) 406:747–52. doi: 10.1038/35021093
9. Ching T, Himmelstein DS, Beaulieu-Jones BK, Kalinin AA, Do BT, Way GP, et al. Opportunities and obstacles for deep learning in biology and medicine. J R Soc Interface. (2018) 15:20170387. doi: 10.1098/rsif.2017.0387
10. Dagogo-Jack I, Shaw AT. Tumour heterogeneity and resistance to cancer therapies. Nat Rev Clin Oncol. (2018) 15:215–31. doi: 10.1038/nrclinonc.2017.166
11. Crolley VE, Marashi H, Rawther S, Sirohi B, Parton M, Graham J, et al. The impact of Oncotype DX breast cancer assay results on clinical practice: a UK experience. Breast Cancer Res Treat. (2020) 180:809–17. doi: 10.1007/s10549-020-05578-6
12. Zhang S, Xiao X, Yi Y, Wang X, Zhu L, Shen Y, et al. Tumor initiation and early tumorigenesis: molecular mechanisms and interventional targets. Signal Transduct Target Ther. (2024) 9:149. doi: 10.1038/s41392-024-01848-7
13. Benkirane H, Vakalopoulou M, Planchard D, Adam J, Olaussen K, Michiels S, et al. Multimodal CustOmics: a unified and interpretable multi-task deep learning framework for multimodal integrative data analysis in oncology. PLoS Comput Biol. (2025) 21:e1013012. doi: 10.1371/journal.pcbi.1013012
14. Qian Y, Daza J, Itzel T, Betge J, Zhan T, Marmé F, et al. Prognostic cancer gene expression signatures: current status and challenges. Cells. (2021) 10:648. doi: 10.3390/cells10030648
15. Tsakiroglou M, Evans A, Pirmohamed M. Leveraging transcriptomics for precision diagnosis: lessons learned from cancer and sepsis. Front Genet. (2023) 14:1100352. doi: 10.3389/fgene.2023.1100352
16. Sparano JA, Gray RJ, Makower DF, Pritchard KI, Albain KS, Hayes DF, et al. Adjuvant chemotherapy guided by a 21-gene expression assay in breast cancer. N Engl J Med. (2018) 379:111–21. doi: 10.1056/NEJMoa1804710
17. Barabási AL, Gulbahce N, Loscalzo J. Network medicine: a network-based approach to human disease. Nat Rev Genet. (2011) 12:56–68. doi: 10.1038/nrg2918
18. Simon R. Clinical trials for predictive medicine. Nat Rev Clin Oncol. (2010) 7:516–24. doi: 10.1177/1740774510366454
19. Elzein A, Iyer G, Solit DB. Lessons from the Study of Exceptional Responders. Cancer Cell. (2021) 39:11–3. doi: 10.1016/j.ccell.2020.11.008
20. Tang Q, Chen Y, Li X, Long S, Shi Y, Yu Y, et al. The role of PD-1/PD-L1 and application of immune-checkpoint inhibitors in human cancers. Front Immunol. (2022) 13:964442. doi: 10.3389/fimmu.2022.964442
21. Galon J, Bruni D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nat Rev Drug Discov. (2019) 18:197–218. doi: 10.1038/s41573-018-0007-y
22. Alix-Panabières C, Pantel K. Clinical applications of circulating tumor cells and circulating tumor DNA as liquid biopsy. Cancer Discov. (2016) 6:825–41. doi: 10.1158/2159-8290.CD-15-1483
23. Kuenzi BM, Park J, Fong SH, Sanchez KS, Lee J, Kreisberg JF, et al. Predicting drug response and synergy using a deep learning model of human cancer cells. Cancer Cell. (2020) 38:672–684.e6. doi: 10.1016/j.ccell.2020.09.014
24. Lamb J, Crawford ED, Peck D, Modell JW, Blat IC, Wrobel MJ, et al. The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and diseases. Science. (2006) 313:1929–35. doi: 10.1126/science.1132939
25. Li MM, Datto M, Duncavage EJ, Kulkarni S, Lindeman NI, Roy S, et al. Standards and guidelines for the interpretation and reporting of sequence variants in cancer: a joint consensus recommendation of the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists. J Mol Diagn. (2017) 19:4–23. doi: 10.1016/j.jmoldx.2016.10.002
27. Vayena E, Tasioulas J. The ethics of big data in public health. Annu Rev Public Health. (2016) 37:357–71. doi: 10.1146/annurev-publhealth-032315-021448
28. Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. (2019) 25:44–56. doi: 10.1038/s41591-018-0300-7
Keywords: computational biology, personalized therapy, precision oncology, RNA-Seq, tumor transcriptome
Citation: Sousa IJO, Nogueira KdM and Pereira EM (2026) Whole tumor transcriptome: a possibility for diagnosis, prognosis, and therapy in precision oncology using computational analysis. Front. Med. 13:1759083. doi: 10.3389/fmed.2026.1759083
Received: 02 December 2025; Revised: 03 January 2026;
Accepted: 05 January 2026; Published: 20 January 2026.
Edited by:
S. Balasubramaniam, Digital University Kerala, IndiaReviewed by:
Anto Lourdu Xavier Raj Arockia Selvarathinam, Grand Valley State University, United StatesCopyright © 2026 Sousa, Nogueira and Pereira. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Ian Jhemes Oliveira Sousa, aWFuamhlbWVzQHVmcGkuZWR1LmJy