Back to 2026 Abstracts
Anatomical and Histologic Response to Neoadjuvant Therapy: Impact on Recurrence Free Survival and Early Recurrence After Liver Resection for Colorectal Liver Metastases
*Andrea Baldo
1, *Miho Akabane
1, *Jun Kawashima
1, *Odysseas P. Chatzipanagiotou
1, *Gaya Spolverato
2, *Andrea Ruzzenente
3, George Poultsides
4, *Kazunari Sasaki
4, *Itaru Endo
5, *Minoru Kitago
6, *Federico Aucejo
7, Irinel Popescu
8, *Tom Hugh
9, *Matthew Weiss
10, *Jin He
11,
Timothy M. Pawlik11Ohio State University Wexner Medical Cen, Worthington, Ohio; 2Department of Surgery, Oncology and Gastroenterology, University of Padua, Padua, PD; 3Division of General and Hepatobiliary Surgery, , University of Verona, Verona, VR; 4Department of Surgery, Stanford University, Stanford, California; 5Department of Gastroenterological Surgery, Yokohama City University School of Medicine, Yokohama; 6Department of Surgery, , Keio University, Tokyo; 7Department of Hepato-pancreato-biliary & Liver Transplant Surgery, Cleveland Clinic Foundation, Cleveland, Ohio; 8Department of Surgery, Fundeni Clinical Institute, Bucharest; 9Department of Surgery, The University of Sydney, Sydney, New South Wales; 10Department of Surgery, Northwell Health Cancer Institute, New York, New York; 11Division of Hepato-Pancreato-Biliary Surgery, Johns Hopkins School of Medicine, Baltimore, Maryland
Background: Tumor Regression Grade (TRG) and the Tumor Burden Score ratio (TBSr, postoperative/pre-neoadjuvant therapy) capture distinct biological dimensions of colorectal liver metastases (CRLM). TRG reflects histologic response to neoadjuvant therapy (NAT), whereas TBSr quantifies the anatomical response. Although both TBS and TRG may have prognostic relevance, their combined effect and the interplay between biological and anatomical response remain poorly defined. We sought to characterize recurrence-free survival (RFS) and early recurrence following neoadjuvant therapy (NAT) and resection of CRLM relative to TRG and TBSr.
Methods: Patients undergoing hepatic resection for CRLM after NAT were identified from a multi-institutional international database. TRG and TBSr were initially evaluated as continuous variables in multivariable Cox and logistic regression analyses. TRG was also categorized as good (1-2-3) versus poor (4-5) response. TBSr categories were derived using maximally selected rank statistics (<1.24 good, 1.24-1.60 mild, >1.60 poor response). Various ways of collapsing the combinations of these categories into three prognostic groups were tested, and the configuration maximizing the log-rank χ^2 statistic was selected for Kaplan-Meier analysis.
Results: Among 487 patients who underwent resection of CRLM following NAT, median age was 63 (IQR 55-70.8) years and 60.2% (n=293) were male; 68.8% (n=335) of the cohort had synchronous metastases, and 27.3%(n=133) underwent a major hepatectomy. Overall median RFS was 19.4 months (95% CI 16.2-25.4). On multivariable analysis, both TBSr (HR 1.23, 95% CI 1.03-1.46, p=0.019) and TRG (HR 1.32, 95% CI 1.20-1.46, p<0.001) were independently associated with shorter RFS, as well as early recurrence (TBSr OR 1.42, 95% CI 1.04-1.98, p=0.03; TRG OR 1.48, 95% CI 1.25-1.74, p<0.001). Good responders had a median RFS that was not reached, mild responders had a median RFS of 13.8 months (95% CI 11.6-16.9, p<0.001), and poor responders had a median RFS of 7.5 months (95% CI 6.3-15.0, p<0.001)
(Figure 1A). Contour modeling of the continuous TRG-TBSr relationship demonstrated a smooth and synergistic gradient of increasing risk of early-recurrence that exceeded 80% among patients with a combination of high TRG and high TBSr
(Figure 1B).Conclusions: TRG and TBSr capture distinct biological aspects of treatment response. Patients exhibiting both a poor histologic (TRG) and anatomical (TBSr) response represented a distinct high-risk phenotype characterized by substantially reduced RFS and a markedly increased likelihood of early recurrence.
Back to 2026 Abstracts
