TY - JOUR
T1 - Distinct patterns of late recurrence in long-term hepatocellular carcinoma survivors
AU - Akabane, Miho
AU - Kawashima, Jun
AU - Woldesenbet, Selamawit
AU - Lee, Ghee Rye
AU - Cauchy, François
AU - Aucejo, Federico
AU - Popescu, Irinel
AU - Kitago, Minoru
AU - Martel, Guillaume
AU - Ratti, Francesca
AU - Aldrighetti, Luca
AU - Poultsides, George A.
AU - Imaoka, Yuki
AU - Ruzzenente, Andrea
AU - Endo, Itaru
AU - Gleisner, Ana
AU - Marques, Hugo P.
AU - Lam, Vincent
AU - Hugh, Tom
AU - Bhimani, Nazim
AU - Shen, Feng
AU - Pawlik, Timothy M.
PY - 2025/9
Y1 - 2025/9
N2 - Background: Among patients with hepatocellular carcinoma (HCC), late recurrence – defined as recurrence occurring ≥2 years after treatment – has often been treated as a singular, uniform event, despite being inherently heterogeneous and driven by diverse biologic mechanisms. This study aimed to identify prognostic factors associated with recurrence among long-term survivors of HCC after treatment, with particular emphasis on the role of underlying liver fibrosis and intrinsic tumor aggressiveness. Methods: Patients who underwent hepatectomy for HCC between 2000 and 2021 were identified from an international database. The prognostic factors for recurrence-free survival (RFS) were evaluated using multivariate Cox regression. The recurrence timing patterns were assessed using kernel density plots. Results: Among 769 patients, 166 (21.6%) developed late recurrence. Compared with patients who did not experience late recurrence, individuals who experienced late recurrence had a higher fibrosis-4 (FIB-4) index (median: 2.09 vs 2.31, respectively; P =.002) and tended to have more frequent microvascular invasion (13.6% vs 19.3%, respectively; P =.089). A high FIB-4 index (hazard ratio [HR], 1.090 [95% CI, 1.011–1.174]; P =.024) and the presence of microvascular invasion (HR, 2.064 [95% CI, 1.260–3.383]; P =.004) were independently associated with worse RFS. Patients were stratified into low-, intermediate-, and high-risk groups based on these factors relative to RFS (P =.027). The hazards of recurrence at 5 years were 2-fold higher among high-risk patients (HR, 2.07 [95% CI, 1.20–3.59]) and 34% higher among intermediate-risk patients (HR, 1.34 [95% CI, 0.93–1.95]) (both P <.05). Kernel density plots demonstrated that microvascular invasion was associated with a peak in recurrence risk at approximately 3 years and that a high FIB-4 index was associated with a more gradual and sustained risk, peaking at approximately 4 years that persisted beyond 5 years. Conclusion: A high FIB-4 index and microvascular invasion were independent predictors of late recurrence. Distinct temporal risk patterns emphasize the need for tailored, risk-based postoperative surveillance to enhance detection and early intervention of HCC recurrence.
AB - Background: Among patients with hepatocellular carcinoma (HCC), late recurrence – defined as recurrence occurring ≥2 years after treatment – has often been treated as a singular, uniform event, despite being inherently heterogeneous and driven by diverse biologic mechanisms. This study aimed to identify prognostic factors associated with recurrence among long-term survivors of HCC after treatment, with particular emphasis on the role of underlying liver fibrosis and intrinsic tumor aggressiveness. Methods: Patients who underwent hepatectomy for HCC between 2000 and 2021 were identified from an international database. The prognostic factors for recurrence-free survival (RFS) were evaluated using multivariate Cox regression. The recurrence timing patterns were assessed using kernel density plots. Results: Among 769 patients, 166 (21.6%) developed late recurrence. Compared with patients who did not experience late recurrence, individuals who experienced late recurrence had a higher fibrosis-4 (FIB-4) index (median: 2.09 vs 2.31, respectively; P =.002) and tended to have more frequent microvascular invasion (13.6% vs 19.3%, respectively; P =.089). A high FIB-4 index (hazard ratio [HR], 1.090 [95% CI, 1.011–1.174]; P =.024) and the presence of microvascular invasion (HR, 2.064 [95% CI, 1.260–3.383]; P =.004) were independently associated with worse RFS. Patients were stratified into low-, intermediate-, and high-risk groups based on these factors relative to RFS (P =.027). The hazards of recurrence at 5 years were 2-fold higher among high-risk patients (HR, 2.07 [95% CI, 1.20–3.59]) and 34% higher among intermediate-risk patients (HR, 1.34 [95% CI, 0.93–1.95]) (both P <.05). Kernel density plots demonstrated that microvascular invasion was associated with a peak in recurrence risk at approximately 3 years and that a high FIB-4 index was associated with a more gradual and sustained risk, peaking at approximately 4 years that persisted beyond 5 years. Conclusion: A high FIB-4 index and microvascular invasion were independent predictors of late recurrence. Distinct temporal risk patterns emphasize the need for tailored, risk-based postoperative surveillance to enhance detection and early intervention of HCC recurrence.
KW - Fibrosis-4 index
KW - Hepatocellular carcinoma
KW - Late recurrence
KW - Microvascular invasion
KW - Recurrence-free survival
KW - Resection
UR - http://www.scopus.com/inward/record.url?scp=105010425970&partnerID=8YFLogxK
U2 - 10.1016/j.gassur.2025.102135
DO - 10.1016/j.gassur.2025.102135
M3 - Article
C2 - 40578426
AN - SCOPUS:105010425970
SN - 1091-255X
VL - 29
SP - 1
EP - 7
JO - Journal of Gastrointestinal Surgery
JF - Journal of Gastrointestinal Surgery
IS - 9
M1 - 102135
ER -