Highlights
Among 298 adult transplant recipients with HCC, recurrence occurred in 19% during a median follow-up of 64 months, with a median time to recurrence of 31 months.
The RETREAT score retained strong prognostic value for hepatocellular carcinoma recurrence after liver transplantation in a real-life, single-center cohort spanning more than two decades.
Five-year recurrence-free survival differed substantially by RETREAT risk group: 93% in the low-risk group, 78% in the medium-risk group, and 58% in the high-risk group.
Competing-risk analysis showed a stepwise increase in recurrence hazard, with hazard ratios of 2.3 for the medium-risk group and 6.4 for the high-risk group versus the low-risk group.
Background
Liver transplantation remains one of the most effective curative strategies for selected patients with hepatocellular carcinoma, because it removes both the tumor and the cirrhotic liver in which new tumors can arise. Even so, post-transplant HCC recurrence continues to limit long-term benefit. Recurrence is clinically consequential not only because it reduces survival, but also because it complicates surveillance planning, counseling, and the potential design of adjuvant or risk-adapted approaches.
Selection of transplant candidates with HCC has historically relied on morphologic criteria, most notably the Milan criteria, which are based on tumor size and number. Although these criteria have been highly influential, they incompletely capture tumor biology. Two patients with similar radiologic tumor burden may have very different recurrence risks after transplantation depending on serum alpha-fetoprotein levels, microvascular invasion, and residual viable tumor after locoregional therapy.
The RETREAT score, or Risk Estimation of Tumor Recurrence After Transplant, was developed to address this gap. It incorporates three post-transplant variables: alpha-fetoprotein at transplantation, microvascular invasion on explant pathology, and the sum of the largest viable tumor diameter plus the number of viable tumors on explant. Because it combines pretransplant biology with explant-based pathologic information, it is intended as a pragmatic tool for estimating recurrence risk and tailoring surveillance after transplant.
The current study by Neri and colleagues is clinically relevant because it evaluates whether the RETREAT score performs well outside the original derivation setting and over prolonged follow-up. Such validation is important before a prediction model can be broadly embedded into routine post-transplant care.
Study Design
Design and setting
This was a retrospective, single-center study including all adult patients who consecutively underwent liver transplantation for hepatocellular carcinoma from January 2000 through July 2022. The investigators assessed the ability of the RETREAT score to predict post-transplant HCC recurrence and overall survival in a real-world cohort.
Population
The analysis included 298 adult patients transplanted for HCC. Patients were grouped according to RETREAT score into three clinically practical categories: low risk with a score of 0 or 1, medium risk with a score of 2 or 3, and high risk with a score of 4 to 6. In this cohort, 54% were in the low-risk group, 36% in the medium-risk group, and 10% in the high-risk group.
Outcomes
The main outcomes were HCC recurrence, recurrence-free survival, and overall survival. Because death without recurrence can preclude the observation of recurrent HCC, the authors also used competing-risk analysis to better estimate the association between RETREAT group and recurrence.
Why the design matters
Validation studies are often less dramatic than derivation studies, but they are more informative for practice. A score that performs only in the population in which it was created is of limited value. The long inclusion period and median follow-up of 64 months make this study especially useful for assessing late recurrence, an issue that shorter series may underestimate.
Key Findings
Overall recurrence burden
During follow-up, 56 of 298 patients, or 19%, developed recurrent HCC. Median time to recurrence was 31 months, underscoring that recurrence after transplantation is not exclusively an early event. The cumulative recurrence rates were 4% at 1 year, 16% at 5 years, and 23% at 10 years.
These numbers are clinically important for two reasons. First, they confirm that recurrence remains a substantial long-term problem even in transplanted patients. Second, the 10-year recurrence estimate supports the concept that surveillance strategies should not always end after the first few years, particularly in patients with biologically higher-risk disease.
Risk discrimination by RETREAT category
The RETREAT score separated patients into groups with markedly different recurrence-free survival. Five-year recurrence-free survival was 93% in the low-risk group, 78% in the medium-risk group, and 58% in the high-risk group, with a statistically significant overall difference of p less than 0.001.
This degree of separation is clinically meaningful. A patient with a low RETREAT score had an excellent 5-year recurrence-free outcome, whereas a patient in the highest risk category had a substantially reduced probability of remaining recurrence-free. In practical terms, this supports the use of RETREAT as more than a purely academic index; it can help determine who might benefit from intensive imaging surveillance, closer biomarker monitoring, or consideration for future trials of adjuvant strategies.
Competing-risk analysis
In the competing-risk framework, the medium-risk RETREAT group had a hazard ratio of 2.3 for HCC recurrence compared with the low-risk group, with p = 0.017. The high-risk group had a hazard ratio of 6.4 versus the low-risk group, with p less than 0.001.
The use of competing-risk analysis is methodologically important in transplantation research. Standard survival methods can overestimate recurrence probability when non-recurrence death is common. By accounting for this issue, the study strengthens the conclusion that the RETREAT score is genuinely associated with recurrence rather than merely reflecting global frailty or mortality risk.
Overall survival
A total of 119 patients died during follow-up. Of these deaths, 32, or 27%, were attributed to HCC recurrence. Overall 5-year survival for the whole cohort was 74%. When stratified by RETREAT category, 5-year survival was 95% in the low-risk group, 86% in the medium-risk group, and 61% in the high-risk group.
In multivariable analysis, the RETREAT score was associated with overall survival only for the highest risk class. Compared with the lowest risk categories, the highest risk class had a hazard ratio for death of 2.5, with p less than 0.001.
This pattern is noteworthy. The score was more strongly linked to recurrence than to all-cause mortality, especially outside the highest risk category. That makes clinical sense. Overall survival after liver transplantation is influenced by multiple competing determinants, including graft function, cardiovascular disease, infection, renal impairment, and other malignancies. A recurrence-specific score would not necessarily be expected to perform equally well for all-cause mortality.
Long-term relevance
One of the most useful contributions of this study is its long observation window. Many prior studies of recurrence prediction focus on early events occurring within the first 2 to 5 years. Here, recurrence rates continued to rise through 10 years. That finding reinforces the value of biologic risk stratification for determining the duration of follow-up rather than relying on a uniform surveillance cutoff for all transplant recipients.
Clinical Interpretation
The main message is straightforward: the RETREAT score works in routine practice. It identifies patients at low, intermediate, and high long-term risk of HCC recurrence after liver transplantation, and its prognostic gradient remains clear over prolonged follow-up.
For clinicians, the most immediate implication is surveillance tailoring. A low RETREAT score may justify less intensive imaging schedules after the early post-transplant years, especially when balanced against cost, radiation exposure from repeated CT, patient burden, and competing medical issues. By contrast, a high RETREAT score identifies a subgroup in whom recurrence risk remains high enough to support more structured and prolonged surveillance.
The study also supports a broader shift in transplant oncology: from simple morphologic eligibility rules toward integrated models that include tumor biology and treatment response. The RETREAT score is particularly attractive because it uses variables already collected in standard practice and does not depend on advanced molecular testing. That practicality increases its likelihood of adoption.
At the same time, clinicians should remember that RETREAT is a post-transplant score. Its strongest use is after transplantation for recurrence prediction, not candidate selection before transplantation. For pretransplant decision-making, other models and clinical factors remain necessary.
Context Within the Literature
The present findings are consistent with the original RETREAT derivation and validation work led by Mehta and colleagues, which demonstrated that the score predicts HCC recurrence after liver transplantation and can guide surveillance intensity. In that seminal study, recurrence risk increased substantially with higher scores, and very low recurrence was observed among patients with a score of 0, supporting de-escalated surveillance in selected individuals.
Subsequent studies have generally confirmed that post-transplant recurrence is driven by a combination of tumor burden, alpha-fetoprotein biology, and explant pathology, especially microvascular invasion and residual viable tumor. This biologic framework also aligns with current transplant oncology practice, in which response to locoregional therapy and AFP kinetics are increasingly used as surrogates for tumor aggressiveness.
Current society guidance, including expert recommendations from the American Association for the Study of Liver Diseases and transplant-focused literature, recognizes the value of risk-based surveillance after liver transplantation, although standardized surveillance schedules remain heterogeneous across centers. A validated score such as RETREAT helps move the field toward a more consistent and evidence-informed approach.
Strengths of the Study
This analysis has several strengths. First, it represents a real-world consecutive cohort, which improves clinical relevance. Second, the long enrollment period and long follow-up allow assessment of both early and late recurrence. Third, the authors evaluated not only recurrence-free survival but also recurrence using competing-risk methods, which is methodologically appropriate in a transplant population. Fourth, the stratification into low-, medium-, and high-risk groups is simple enough to be used in clinical workflow.
Another strength is that the study validates a score rather than proposing yet another center-specific model. External and pragmatic validation are often more useful than repeated creation of novel tools that lack portability.
Limitations and Cautions
Several limitations should temper interpretation. The study was retrospective and conducted at a single institution, which raises the possibility of center-specific selection practices, locoregional therapy protocols, pathology interpretation, and surveillance patterns. These factors may influence both measured recurrence risk and the apparent performance of the score.
The transplant period spanned from 2000 to 2022, during which HCC management changed substantially. Advances in imaging, bridging therapy, downstaging, immunosuppression strategies, and candidate selection may have introduced temporal heterogeneity. While this breadth reflects real life, it may also blur the effect of contemporary practice patterns.
Because RETREAT incorporates explant pathology, it cannot inform all decisions before transplantation. In addition, the study summary does not provide discrimination metrics such as C-statistics or calibration plots, which would have further clarified how accurately predicted risk matched observed risk across score values.
Finally, although the score identified patients at higher recurrence risk, the study does not establish that changing surveillance or treatment based on RETREAT improves outcomes. Prognostic utility does not automatically translate into interventional benefit, and this remains an important area for future research.
Implications for Practice and Research
In current practice, the RETREAT score can reasonably be used to structure post-transplant counseling and follow-up planning. Low-risk patients may be reassured that long-term recurrence risk is relatively low, whereas medium- and high-risk patients can be informed that recurrence risk remains clinically meaningful and may justify more intensive monitoring.
The score may also help standardize eligibility for clinical trials testing adjuvant strategies after liver transplantation. Historically, such trials have been challenging because recurrence events are relatively infrequent in unselected transplant populations. Enriching enrollment with patients in higher RETREAT strata could improve feasibility and statistical efficiency.
Future work should focus on multicenter prospective validation, calibration across geographic regions, and integration with emerging biomarkers such as dynamic AFP kinetics, inflammatory markers, radiomics, and circulating tumor DNA. Whether such additions materially improve prediction beyond RETREAT alone remains uncertain, but the current score provides a strong and practical benchmark.
Another important research question is surveillance optimization. A score is most useful when linked to an actionable care pathway. Studies should test whether RETREAT-guided surveillance can reduce unnecessary imaging in low-risk patients while improving earlier detection or salvage options in high-risk patients.
Funding and Trial Registration
The provided study summary and citation do not specify a ClinicalTrials.gov registration number. No funding details are available in the source material provided here. Readers should consult the full article for complete disclosures, funding statements, and any institutional support.
Conclusion
This single-center real-life validation adds meaningful evidence that the RETREAT score is a robust tool for predicting long-term HCC recurrence after liver transplantation. In 298 transplant recipients followed for a median of 64 months, the score clearly separated low-, intermediate-, and high-risk groups for recurrence-free survival and showed a strong graded association with recurrence in competing-risk analysis. Its association with overall survival was most evident in the highest-risk category, which is biologically plausible given the many non-cancer determinants of death after transplantation.
For clinicians, the practical implication is that RETREAT can support risk-adapted post-transplant surveillance and more precise patient counseling. For researchers, it provides a validated framework for trial enrichment and for studying how recurrence prediction can be translated into better outcomes. The next step is not simply to confirm that RETREAT predicts recurrence, but to determine how best to act on that information.
RETREAT Score：
https://medxy.ai/calculators/retreat-score-for-hcc-recurrence-after-liver-transplant
References
Neri F, Viganò M, Camagni S, Zambelli MF, Loglio A, De Giorgio M, Muglia R, Licini L, Francavilla A, Marra P, Sironi S, Erba PA, Gotuzzo I, Gianatti A, Fagiuoli S, Pinelli D. RETREAT Score Accurately Predicts the Long-Term Risk of HCC Recurrence After Liver Transplantation: A Single-Center Real-Life Validation. Cancers (Basel). 2026 Feb 9;18(4):556. doi: 10.3390/cancers18040556. PMID: 41749810; PMCID: PMC12938944.
Mehta N, Heimbach J, Harnois DM, Sapisochin G, Dodge JL, Lee D, Burns JM, Sanchez W, Grewal HP, Yao FY. Validation of a Risk Estimation of Tumor Recurrence After Transplant (RETREAT) Score for Hepatocellular Carcinoma Recurrence After Liver Transplant. JAMA Oncol. 2017 Apr 1;3(4):493-500. doi: 10.1001/jamaoncol.2016.5116.
Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, Montalto F, Ammatuna M, Morabito A, Gennari L. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996 Mar 14;334(11):693-699. doi: 10.1056/NEJM199603143341104.
Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, Zhu AX, Murad MH, Marrero JA. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology. 2018 Jan;67(1):358-380. doi: 10.1002/hep.29086.
