Could more complex molecular analysis aid biomarker-driven clinical management of poor-prognosis colorectal cancer?

Although cancer has a very complex biology, until recently relatively simple biomarkers have been used to inform therapy. “There is a need to bring a higher degree of sophistication to our analyses to match the intrinsic biological complexity of colorectal cancer (CRC), and this includes transcriptomic analyses in baseline tissue and circulating tumour DNA (ctDNA) for comprehensive molecular profiling,” urges Dr Manuel Salto-Tellez from Queen's University Belfast, and the Institute of Cancer Research and Royal Marsden Hospital, London, UK, welcoming research efforts presented at ESMO Gastrointestinal Cancers Congress 2026 (Munich, 1–4 July) that are moving the field towards more personalised treatment strategies.

A molecular analysis of the phase III BREAKWATER trial used whole exome, whole transcriptome sequencing and ctDNA profiling to determine molecular correlates of overall survival (OS) and treatment-specific genomic resistance patterns in patients treated for BRAF V600E-mutated metastatic CRC (Abstract 7RO). By comparing first-line encorafenib plus cetuximab with or without modified (m)FOLFOX6 with standard care (chemotherapy with or without bevacizumab), the study previously demonstrated that patients who received encorafenib plus cetuximab with mFOLFOX6 experienced the most favourable OS outcomes (N Engl J Med. 2025;392:2425–2437), and the second-generation BRAF inhibitor, in combination with cetuximab and chemotherapy, is now approved for BRAF V600E-mutated metastatic CRC by the U.S. Food and Drug Administration.

As discussed at the Congress, the OS benefit of encorafenib plus cetuximab with mFOLFOX6 versus control was consistent across common alterations (APC, TP53, PIK3CA, RNF43), consensus molecular subtypes and BRAF-mutant transcriptional subtypes. However, some transcriptomic data, such as higher baseline ultraviolet response pathway activity, were most prominent in patients who received encorafenib plus cetuximab with mFOLFOX6. Compared with the other arms, this subgroup also had fewer end-of-treatment acquired resistance alterations identified by ctDNA profiling (NRAS, KRAS, MAP2K1 and BRAF exon deletions).

“These results provide further evidence that complex molecular analysis may be of assistance in guiding treatment decisions in patients with poor-prognosis CRC,” comments Salto-Tellez.
In a separate study, ctDNA was used to identify the value of molecular residual disease (MRD) in predicting OS benefit from adjuvant chemotherapy following curative-intent resection of CRC liver metastases in the observational GALAXY study (Abstract 8O) (Nat Med. 2024;30:3272–3283). MRD status following ctDNA assessment during a 2–10-week window post-surgery was able to differentiate between patients who would or would not derive substantial benefit from adjuvant therapy. Patients with MRD-negative disease experienced prolonged OS without clear benefit from adjuvant chemotherapy. However, in MRD-positive patients, adjuvant chemotherapy significantly improved disease-free survival (adjusted hazard ratio [aHR] 0.07; 95% confidence interval [CI] 0.02–0.24; p<0.0001) and OS (aHR 0.27; 95% CI 0.08–0.88; p=0.03) compared with observation. MRD status did not impact survival outcomes following adjuvant chemotherapy in patients who also received neoadjuvant chemotherapy. “These data suggest a broader value of ctDNA detection in CRC. In addition to being a powerful prognostic marker in general, MRD detection may be predictive of subsequent response to chemotherapy and therefore of chemotherapy need in the context of curative-intent resection of CRC liver metastases,” notes Salto-Tellez.

He points out that comprehensive molecular profiling is currently used to guide targeted therapies for colorectal cancer, including immunotherapies for microsatellite instability-high/mismatch repair deficient disease, and targeted therapies for poor-prognosis, previously ‘undruggable’ molecular phenotypes, such as BRAF V600E and KRAS G12C-mutated CRC (Clin Colorectal Cancer. 2024;23:215–229; Nat Med. 2024;30:969–983). Liquid biopsy, utilising ctDNA testing, is also becoming a valuable tool to monitor for disease recurrence and to identify potential drug targets (Diagnostics (Basel). 2026;16:523). Currently, ESMO recommends the use of sensitive and validated ctDNA assays to identify actionable mutations to direct targeted therapy in patients with advanced cancers when rapid results are clinically important or when tissue biopsies are not possible (Ann Oncol. 2022;33:750–768). However, as existing ctDNA assays are limited by low sensitivity in detecting trace quantities of ctDNA – particularly in patients with early-stage disease or low-shedding tumours – and can produce false-negative results, the routine use of ctDNA profiling for MRD detection is not currently recommended.