For patients whose colorectal cancer returns in the liver years after the original tumor has been removed, surgery can still be curative — but only for some. A new study in Genome Medicine reports that specific patterns of chromosome-arm loss in these late-appearing metastases track with how quickly the disease comes back after metastatic surgery, addressing the current lack of genomic markers to personalize surgical decisions in this setting.
Researchers assembled a deeply annotated cohort of 41 colorectal cancer patients who underwent resection of metachronous liver metastases, meaning tumors that appeared after the primary tumor had been removed. The team performed whole-exome and transcriptome sequencing on the metastases and, where available, on matched primary tumors. The genomic data were paired with structured annotations of each patient’s longitudinal clinical course, including a median follow-up of 66.4 months, relapse after metastatic surgery, adjuvant therapies, radiologic assessments, and blood-based markers.
The analysis revealed a clear split. Early relapse, defined as recurrence within 12 months of metastatic surgery, correlated with a larger number of metachronous liver metastases and with a characteristic copy-number pattern: loss of the p arm of chromosome 3 and loss of the q arm of chromosome 5. In contrast, metastases from patients with late or no relapse were characterized by multi-hit APC mutations, in which the APC gene is damaged repeatedly. These genomic features also correlated with significant differences in relapse-free survival after metastatic surgery, and the associations held whether or not patients had received adjuvant therapy for their primary or metastatic tumors.
Notably, the early-relapsing metastases did not typically carry just one of these changes. Instead, they harbored an accumulation of multiple relapse-associated genomic alterations, suggesting that the burden of specific chromosomal events, rather than any single event alone, distinguishes tumors associated with rapid recurrence.
The study also compared metachronous metastases with synchronous ones — metastases present at the time of initial diagnosis — using an independent cohort. The metachronous tumors differed genomically from their synchronous counterparts, showing a higher frequency of 5q loss and a higher burden of copy-number alterations relative to their matched primary tumors. According to the authors, this pattern suggests continued evolution of the cancer after primary-tumor resection.
Transcriptomic analysis added a functional layer to the copy-number findings. The relapse-associated copy-number changes correlated with gene-dosage effects on the affected chromosome arms, meaning that losing genetic material changed the expression levels of genes located there. These changes were linked to differential expression of genes involved in proliferation and metabolism. In two independent datasets, the same relapse-associated copy-number signatures also correlated with advanced disease and shorter survival.
The researchers dissected the relapse-associated genomic features across five independent datasets in total, examining whether the signals observed in their own 41-patient cohort could be reproduced elsewhere. The consistency of the 3p and 5q loss associations across adjuvant-therapy subgroups and external data strengthens the case that these are genuine correlates of relapse behavior rather than artifacts of a single small sample.
The clinical context makes the findings consequential. As cancer screening expands, colorectal tumors are increasingly diagnosed before they have spread, yet up to 25 percent of patients with early-stage disease later develop metachronous metastases, often years after primary tumor resection. When these metastases are confined to the liver, they remain potentially curable by surgery, but outcomes vary substantially: some patients stay durably disease-free while others relapse within months. Until now, clinicians have lacked genomic markers to personalize surgical decisions for this population.
The study characterizes the clinico-genomic basis of early relapse rather than delivering a ready-made clinical test. The cohort, while richly annotated, is modest in size, and the associations described are correlational. Still, the researchers report that these features could ultimately help identify patients with a high propensity to relapse after surgery and personalize the management of colorectal cancer patients with metachronous liver metastases.
Subject of Research: Clinico-genomic determinants of relapse in metachronous colorectal cancer liver metastases
Article Title: Clinico-genomic determinants of relapse in metachronous colorectal cancer metastases
Article References: Holý, P., Möhrmann, L., Zhou, Y., Hlaváč, V., Rao, B. H., Červenková, L., Heczko, L., Liška, V., Vyčítal, O., Mírka, H., Fiala, O., Souček, P., & Dietlein, F. (2026). Clinico-genomic determinants of relapse in metachronous colorectal cancer metastases. Genome Medicine. https://doi.org/10.1186/s13073-026-01818-4
Image Credits: AI Generated
DOI: 10.1186/s13073-026-01818-4
Keywords: colorectal cancer, liver metastases, metachronous metastases, copy-number alteration, whole-exome sequencing, transcriptome, relapse-free survival, APC mutations, 5q loss, 3p loss, cancer genomics, Genome Medicine
News Source: Nathaniel Bowman. (October 24, 2026). Genomic arm losses flag early relapse after colorectal liver metastasis surgery. Scienmag.



