Recurrence After Resection of Colorectal Liver Metastases and Its Impact on Survival: A Multistate Model-Based, Multi-institutional Observational Cohort Study

BACKGROUND: Conventional survival analyses cannot fully account for recurrence as an intermediate event that dynamically influences subsequent mortality. Multistate models (MSMs) enable explicit modeling of transitions between clinically relevant disease states; however, their application among patients undergoing resection for colorectal liver metastases (CRLM) remains limited. METHODS: An international multi-institutional database was queried to identify patients undergoing curative-intent hepatectomy for CRLM between 1998 and 2023. A four-state MSM was constructed comprising alive and recurrence-free, alive with recurrence, colorectal cancer (CRC)-related death, and non-CRC-related death. Transition probabilities and transition-specific prognostic factors were evaluated RESULTS: Among 3002 patients, 1624 (54.1%) developed recurrence and 911 (56.1%) died, predominantly from CRC-related causes (92.8%). The probability of recurrence decreased substantially the longer patients remained recurrence-free after surgery, whereas recurrence markedly worsened subsequent prognosis. High tumor burden score was independently associated with transition to recurrence (hazard ratio [HR] 1.06 per unit; 95% confidence interval 1.03-1.08; p < 0.001) but not post-recurrence mortality, suggesting that the adverse prognostic impact of tumor burden was mediated primarily through recurrence. Adjuvant chemotherapy reduced recurrence (HR 0.76; p = 0.005) and CRC-related death from the recurrence-free state (HR 0.39; p = 0.002). CONCLUSION: MSMs provide a dynamic framework to characterize disease evolution after CRLM resection and represent recurrence as a clinically meaningful time-dependent intermediate state rather than a simple binary endpoint. Recurrence was the dominant determinant of long-term prognosis, supporting the role of MSMs for individualized prognostication and surveillance after hepatectomy for CRLM.

Authors

Institutions

Publication Details

Journal
Annals of Surgical Oncology
Published
2026-10-13
DOI
https://doi.org/10.1245/s10434-026-20765-3
Primary Topic
Colorectal Cancer Treatments and Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Recurrence After Resection of Colorectal Liver Metastases and Its Impact on Survival: A Multistate Model-Based, Multi-institutional Observational Cohort Study

Minoru Kitago, Itaru Endo, Timothy Michael Pawlik, Gaya Spolverato et al.
Annals of Surgical Oncology
Colorectal Cancer Treatments and Studies
article

Recurrence After Resection of Colorectal Liver Metastases and Its Impact on Survival: A Multistate Model-Based, Multi-institutional Observational Cohort Study

Minoru Kitago, Itaru Endo, Timothy Michael Pawlik, Gaya Spolverato, George A. Poultsides, Miho Akabane, Abdulaziz Elemosho, Jin He, Matthew Weiss, Charalampos M. Charalampous, Andrea Ruzzenente, Irinel Popescu, Andrea Baldo, Jun Kawashima, Kazunari Sasaki, Tom Hugh, Federico Aucejo
article en

Abstract

BACKGROUND: Conventional survival analyses cannot fully account for recurrence as an intermediate event that dynamically influences subsequent mortality. Multistate models (MSMs) enable explicit modeling of transitions between clinically relevant disease states; however, their application among patients undergoing resection for colorectal liver metastases (CRLM) remains limited. METHODS: An international multi-institutional database was queried to identify patients undergoing curative-intent hepatectomy for CRLM between 1998 and 2023. A four-state MSM was constructed comprising alive and recurrence-free, alive with recurrence, colorectal cancer (CRC)-related death, and non-CRC-related death. Transition probabilities and transition-specific prognostic factors were evaluated RESULTS: Among 3002 patients, 1624 (54.1%) developed recurrence and 911 (56.1%) died, predominantly from CRC-related causes (92.8%). The probability of recurrence decreased substantially the longer patients remained recurrence-free after surgery, whereas recurrence markedly worsened subsequent prognosis. High tumor burden score was independently associated with transition to recurrence (hazard ratio [HR] 1.06 per unit; 95% confidence interval 1.03-1.08; p < 0.001) but not post-recurrence mortality, suggesting that the adverse prognostic impact of tumor burden was mediated primarily through recurrence. Adjuvant chemotherapy reduced recurrence (HR 0.76; p = 0.005) and CRC-related death from the recurrence-free state (HR 0.39; p = 0.002). CONCLUSION: MSMs provide a dynamic framework to characterize disease evolution after CRLM resection and represent recurrence as a clinically meaningful time-dependent intermediate state rather than a simple binary endpoint. Recurrence was the dominant determinant of long-term prognosis, supporting the role of MSMs for individualized prognostication and surveillance after hepatectomy for CRLM.

Annals of Surgical Oncology
University of Verona (IT), The University of Sydney (AU), Northwell Health (US), Cleveland Clinic (US), University of Padua (IT), Johns Hopkins University (US), Keio University (JP), Johns Hopkins Medicine (US), The Ohio State University Wexner Medical Center (US), Stanford Medicine (US), Johns Hopkins Hospital (US), Institutul Clinic Fundeni (RO), The Ohio State University (US), Yokohama City University (JP), Stanford University (US)
Openalex Percentile: Top 18%
Colorectal Cancer Treatments and Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.