Multi-omic biomarkers of neoadjuvant treatment response in rectal cancer: A narrative review.
Authors
Affiliations (6)
Affiliations (6)
- Department of Radiation Oncology, University Hospitals Leuven, Leuven, 3000, Belgium; Department of Oncology, Laboratory of Experimental Radiotherapy, KU Leuven, Leuven, 3000, Belgium.
- Department of Abdominal Surgery, University Hospitals Leuven, Leuven, 3000, Belgium.
- Department of Pathology, University Hospitals Leuven, Leuven, 3000, Belgium.
- Department of Radiology, University Hospitals Leuven, Leuven, 3000, Belgium.
- Department of Digestive Oncology, University Hospitals Leuven, Leuven, 3000, Belgium.
- Department of Radiation Oncology, University Hospitals Leuven, Leuven, 3000, Belgium; Department of Oncology, Laboratory of Experimental Radiotherapy, KU Leuven, Leuven, 3000, Belgium. Electronic address: [email protected].
Abstract
Neoadjuvant treatment response in rectal cancer is highly heterogeneous, complicating patient selection for organ-preservation strategies. Robust biomarkers capable of accurately predicting treatment response are needed to improve personalized treatment decisions. We conducted a narrative review of studies published since 2015 evaluating predictors of response to neoadjuvant therapy in rectal cancer. A comprehensive PubMed/MEDLINE search identified evidence across six domains: (1) genomic and molecular biomarkers, (2) imaging-based biomarkers, (3) histopathological and digital pathology biomarkers, (4) liquid biopsy biomarkers (cfDNA and ctDNA), (5) patient-derived tumor models and (6) microbiome-associated biomarkers. Treatment response in rectal cancer reflects a complex interplay between tumor-intrinsic, immune microenvironment and stromal features. Immune-enriched tumors, characterized by high intratumoral CD8<sup>+</sup> T-cell infiltration, CMS1/iCMS3 subtype and high Immunoscore, consistently demonstrate higher rates of pathological and clinical complete response. Conversely, KRAS, TP53, BRAF and SMAD4 mutations, fibroblast activation, TGFβ signaling, inflammatory cancer-associated fibroblasts and epithelial-mesenchymal transition programs are associated with treatment resistance. Artificial intelligence applied to MRI, endoscopy and digital pathology enables accurate response prediction, particularly when incorporating longitudinal features. Emerging technologies including ctDNA monitoring, patient-derived tumor models and microbiome profiling provide additional insight into treatment sensitivity and show promise for predicting treatment response. Neoadjuvant treatment response in rectal cancer is dependent on genomic alterations, immune activation and stromal interactions. AI-driven biomarkers hold promise for personalized treatment and organ-preservation. Prospective, multicenter validation is essential to enable further clinical implementation.