Hepatocellular Carcinoma at a Center of Clinical Excellence in Colombia: A Real-World Cohort Study of Clinical Characteristics, Treatment Trajectories, and Survival

Abstract

Purpose

Hepatocellular carcinoma (HCC) is the sixth most common malignancy worldwide. In middle-income settings, fragmentation of care may delay diagnosis and limit access to timely treatment. This study aimed to describe the sociodemographic and clinical characteristics, treatment patterns, and overall survival of patients with HCC managed at a Center of Clinical Excellence (CCE).

Patients and Methods

We conducted a retrospective descriptive cohort study of adults (≥18 years) with HCC managed at a CCE between September 2022 and December 2025. Baseline variables included etiology, comorbidities, Child–Pugh class, Model for End-Stage Liver Disease (MELD) score, Barcelona Clinic Liver Cancer (BCLC) stage, and alpha-fetoprotein (AFP) levels. Treatment strategies and clinical course, including stage migration, liver transplantation, and death, were described. Overall survival was estimated using the Kaplan–Meier method.

Results

A total of 134 patients were included. Median age was 67.93 years (IQR 63.12–72.59) and 61.19% were men. The most common underlying etiology was metabolic dysfunction-associated steatotic liver disease (MASLD) (34.38%). Most patients were diagnosed at early BCLC stages 0/A (64.18%; n = 86). Ablation and transarterial chemoembolization were the most frequently used treatments, and liver transplantation was performed in 9.70% of patients. After a median follow-up of 21 months, median overall survival was not reached; estimated survival was 93.8% at 1 year and 80.0% at 3 years.

Conclusion

In this retrospective cohort managed at a CCE, MASLD was the most frequent underlying etiology, and patients were diagnosed at early stages BCLC stages 0/A. Treatment patterns were dominated by locoregional therapies, and median overall survival was not reached during the available follow-up period. These findings provide real-world data on HCC care in a middle-income setting and highlight the clinical relevance of describing treatment trajectories and survival within structured multidisciplinary care programs.

Graphical Abstract

Introduction

Hepatocellular carcinoma (HCC) accounts for nearly 90% of primary liver cancers and represents a major global health burden, with approximately 905,000 new cases diagnosed annually and a mortality-to-incidence ratio of 0.95, reflecting a five-year overall survival below 18%.^1,2 In low-incidence regions such as Latin America, age-adjusted incidence rates range from 2.9 to 6.6 cases per 100,000 population.³ The prognosis of HCC is determined not only by tumor burden but also by underlying hepatic functional reserve, which together shape both therapeutic eligibility and clinical outcomes, and often necessitate a multidisciplinary management approach.⁴

Describing clinical characteristics, treatment trajectories, and survival in patients with HCC is essential to understanding how the disease presents in real-world settings, how treatment decisions evolve across stages, and how patients progress over time. Clinical characterization informs the local epidemiological profile of the disease, including the distribution of underlying etiologies, tumor stage at diagnosis, and functional liver status. Treatment trajectories are particularly informative in HCC because management is dynamic and may involve sequential or simultaneous transitions between imaging surveillance, locoregional therapies, systemic treatment, transplant evaluation, and supportive or palliative care. These transitions are shaped by tumor burden, liver function, performance status, treatment response, and multidisciplinary decision-making, and constitute a critical dimension of HCC care that static cross-sectional analyses often fail to capture.⁵ Finally, survival estimates derived from real-world cohorts provide context for interpreting outcomes in settings where patient profiles, resource availability, and health system organization differ substantially from those of randomized controlled trials.

Such information is particularly relevant in middle-income settings (MICs), where the interplay of delayed diagnosis, heterogeneous access to locoregional, systemic, and transplant therapies, and fragmented care pathways may profoundly influence clinical trajectories and survival. In Latin America, adherence to BCLC guideline recommendations may be below 50%⁶ and more than 60% of patients may be diagnosed at advance stages, limiting access to potentially curative therapies to fewer than 20% of the population.^6,7 These figures underscore a persistent gap between evidence-based recommendations and clinical practice in this region and highlight the need for detailed documentation of real-world HCC care pathways in MICs.

Effective delivery of HCC care is frequently hindered in routine practice by fragmentation of health services. Studies have shown up to a 12% higher risk of death (aHR = 1.12; 95% CI 1.05–1.18) among patients with delays exceeding 90 days between diagnosis and treatment initiation.⁸ Beyond delays access, recent evidence-primarily from high-income countries-indicates that fragmented care is independently associated with worse overall survival, with adjusted hazard ratios ranging from 1.14 to 1.81 compared with centralized management.^9,10 Moreover, this excess risk appears to follow a dose–response pattern according to the number of institutions involved in care.⁹ In MICs, where care fragmentation remains structurally entrenched, these findings carry particular clinical and programmatic relevance.¹¹

As a structural response to care fragmentation, the Center of Clinical Excellence (CCE) model integrates HCC management through multidisciplinary boards, coordinated clinical teams, and standardized care protocols. International evidence suggests that this approach may be associated with improved care coordination and more consistent application of guideline-concordant therapies in patients with HCC,¹² though causal attribution remains difficult in the absence of comparative or prospective data. Evidence on CCE models in MICs remains limited, and findings from high-income settings may not be directly transferable to health systems characterized by different referral patterns, resource constraints, and administrative fragmentation.¹³

At the same time, it must be acknowledged that cohorts derived from CCE settings may not fully represent the broader population of patients with HCC in Colombia or in the region. Centers of clinical excellence typically attract patients who have been referred or identified through institutional networks, and who may therefore differ from the general HCC population in terms of performance status, underlying liver function, stage at presentation, or access to resources.¹³ This inherent selection bias is an important contextual factor in interpreting the clinical profile and outcomes described in CCE-based cohort and should be explicitly considered when generalizing findings to non-specialized care settings.

In this context, there is a need for detailed, real-world descriptions of how patients with HCC are managed within structured CCE setting in MICs-not to demonstrate superiority of the CCE model over alternative care arrangements, but to characterize the clinical profile, treatment patterns, longitudinal trajectories, and survival outcomes of patients managed within such programs. To our knowledge, no previous study has described the clinical characteristics, treatment trajectories, and survival of a cohort of patients with HCC managed within an internationally accredited CCE in Colombia.^14,15 Accordingly, this study aims to describe the sociodemographic and clinical characteristics, treatment patterns and trajectories, and overall survival of patients with HCC enrolled in a CCE at a referral institution in a middle-income setting.

Materials and Methods

We conducted an observational descriptive cohort study with retrospective secondary data collection based on institutional medical records. The study population comprised patients aged ≥18 years with a confirmed diagnosis of HCC who were enrolled in the CCE between September 2022 and December 2025; the last patient was enrolled in December 2025 (Figure 1). Survival status and clinical follow-up information were updated through April 2026, which was used as data cutoff for survival analysis. HCC diagnosis was established either by histopathology or by imaging criteria using contrast-enhanced computed tomography (CT) and/or magnetic resonance imaging (MRI), according to the Liver Imaging Reporting and Data System (LI-RADS).⁵

Figure 1.

Patient flow diagram of patients with hepatocellular carcinoma enrolled in the Center of Clinical Excellence.

The CCE is an institutional multidisciplinary care program for patients with HCC located at Fundación Cardioinfantil–Instituto de Cardiología in Bogotá, Colombia. The CCE is not a separate hospital but an organized care pathway within the institution that integrates multidisciplinary tumor board discussion, hepatology follow-up, specialized nursing coordination, imaging and laboratory surveillance, transplant evaluation, and treatment planning according to tumor stage, liver function, performance status, and multidisciplinary recommendations. International accreditation of this program was granted by the Joint Commission (JCI), which certifies adherence to standardized quality and safety benchmarks in clinical care delivery. The multidisciplinary tumor board convenes biweekly and comprises specialists in hepatology, radiology, transplant surgery, oncology, liaison medicine, and specialized nursing.

Enrollment into the CCE follows predefined institutional criteria. Patients are eligible for program admission if they have a confirmed diagnosis of HCC and can undergo periodic follow-up within the institutional multidisciplinary pathway, which may include clinical assessment, laboratory testing, imaging surveillance, and transplant evaluation according to tumor stage, liver function, performance status, and multidisciplinary board recommendations. Patients with non-HCC malignant hepatic lesions, extrahepatic malignancies requiring primary management elsewhere, or conditions precluding institutional follow-up are not admitted to the program. The present study includes all consecutive patients who fulfilled institutional admission criteria and were formally enrolled in the CCE during the study period.

A consecutive sampling strategy was used, including all eligible patients with confirmed HCC who were formally enrolled in the CCE during the study period. Because this was a retrospective descriptive cohort based on all available eligible cases, no formal a priori sample size calculation was performed. The final sample size was determined by the total number of patients meeting the inclusion criteria during the study period. This approach is consistent with descriptive cohort studies in which the objective is to characterize all available cases rather than to test a hypothesis.

Patients with suspected HCC are identified across different care settings, including outpatient clinics, emergency departments, inpatient wards, or diagnostic support services. All cases are discussed at the Multidisciplinary Mass Board, a biweekly meeting comprising specialists in hepatology, gastroenterology, radiology, transplant surgery, oncology, liaison medicine, and specialized nursing. At each meeting, clinical history, laboratory results, and imaging studies are reviewed to confirm or rule out HCC and to determine the need for liver biopsy. Enrollment is documented in the medical record upon diagnostic confirmation, and follow-up visits are scheduled every three months to assess clinical course, adherence to the treatment plan, and the need for management adjustments.

After diagnostic confirmation, disease is staged using the BCLC system,¹⁶ which integrates tumor burden, liver function, and performance status to individualize treatment allocation. Management options considered within the CCE include resection, ablation, liver transplantation, transarterial chemoembolization (TACE), and systemic therapy, allocated according to BCLC stage and multidisciplinary board recommendation. Selected patients are contacted by the CCE liaison nurse to coordinate an in-person hepatology visit, during which the diagnosis, treatment options, and treatment expectations are discussed with the patient and family.

We collected sociodemographic and clinical variables, including etiology, comorbidities, Child-Pugh class, MELD score, BCLC stage, LI-RADS classification, AFP levels, and treatment strategies. Progression was defined as migration to a higher BCLC stage or death. Overall survival was defined as the time from HCC diagnosis to death from any cause. We also collected information on disease progression, treatment trajectories, and vital status throughout the follow-up period.

Patients who discontinued follow-up within the CCE after enrollment for administrative reasons-primarily insurance-related barriers-were not excluded from the cohort. These patients were retained in baseline analyses according to their BCLC stage at enrollment and were included in the overall survival analysis, as vital status and date of death, when applicable, were retrospectively ascertained through the Colombian ADRES administrative database. However, treatment trajectories, transplant status, or BCLC stage migration occurring after discontinuation of CCE follow-up could not be systematically captured unless documented in institutional records; therefore, progression-related analyses were based exclusively on information available during program-based or institutional follow-up. The number of patients who discontinued CCE follow-up during the study period was 38, all for administrative reasons related to insurance coverage changes.

Clinical data were retrospectively extracted from institutional medical records from September 2022 onward, systematically reviewed, and entered into REDCap by the CCE liaison physician.¹⁷ Vital status and date of death, when applicable, were retrospectively ascertained through clinical history and the ADRES national health registry. The dataset was subsequently anonymized and exported for statistical analysis. Data were independently verified by two physicians, and standardized criteria for data collection and entry were applied throughout.

Analyses were performed in R.¹⁸ Categorical variables were summarized as frequencies and percentages. The distribution of continuous variables was assessed by visual inspection of histograms; normally distributed variables were reported as mean and standard deviation (SD), whereas non-normally distributed variables were reported as median and interquartile range (IQR). Overall survival was estimated using the Kaplan–Meier method, with follow-up time calculated from HCC diagnosis to death or censoring at liver transplant or end of study; the median follow-up time for the cohort is reported in the Results section. Missing data were reported descriptively, without imputation. Tables, transition matrices, and alluvial plots were used to illustrate stage progression and treatment trajectories.

Results

Sociodemographic Characteristics

A total of 134 patients with confirmed HCC enrolled in the CCE were included in the study. During follow-up, 38 patients discontinued CCE follow-up after enrollment for administrative or insurance-related reasons; these patients were retained in the baseline and overall survival analyses, as vital status was ascertained through ADRES. Among patients who discontinued CCE follow-up, 19/38 (50.0%) died during follow-up, compared with 11/96 (11.5%) among those who remained in program-based follow-up. Death was considered the event of interest; patients who were alive and had not undergone liver transplantation were censored at the last available follow-up or at the April 2026 data cutoff, whichever occurred first.

The median age of the cohort was 67.93 years (IQR 63.12–72.59), with a predominance of men (61.19%; n = 82). Regarding educational level, secondary education was most frequent (46.30%; n = 50), followed by primary education (26.85%; n = 29) and higher education (15.74%; n = 17). The most frequent underlying etiology of HCC was MASLD (34.38%; n = 44); other relevant etiologies included alcohol use (15.63%; n = 20), autoimmune hepatitis, and hepatitis C. The most prevalent comorbidities were hypertension (58.21%; n = 78) and type 2 diabetes mellitus (49.25%; n = 66); other frequent comorbidities included osteoporosis, dyslipidemia, hypothyroidism, and chronic kidney disease. Baseline sociodemographic characteristics, underlying etiologies, and comorbidities are summarized in Table 1.

Table 1.

Baseline Sociodemographic, Etiologic, and Clinical Characteristics of Patients with Hepatocellular Carcinoma Enrolled in the Center of Clinical Excellence

Variable	N	N = 1341
Sociodemographic Characteristics
Age, median (IQR)	134	67.93 (63.12–72.59)
Sex	134
Men		82 (61.19%)
Women		52 (38.81%)
Educational Attainment	108
No formal education		12 (11.11%)
Primary education		29 (26.85%)
Secondary education		50 (46.30%)
Higher education		17 (15.74%)
Unknown		26
Underlying Etiology of HCC
MASLD		44 (34.38%)
Alcohol		20 (15.63%)
Autoimmune		20 (15.63%)
Hepatitis C		17 (13.28%)
MetALD		11 (8.59%)
Cryptogenic		7 (5.47%)
Hemosiderosis		4 (3.13%)
Hepatitis B		4 (3.13%)
DILI		1 (0.78%)
Unknown		6
Baseline Comorbidities
Hypertension	134	78 (58.21%)
Type 2 diabetes mellitus		66 (49.25%)
Osteoporosis		38 (28.36%)
Dyslipidemia		24 (17.91%)
Hypothyroidism		23 (17.16%)
Chronic kidney disease		17 (12.69%)
Cerebrovascular disease		8 (5.97%)
Other		7 (5.22%)
Heart disease		6 (4.48%)
COPD		4 (2.99%)
Autoimmune disease		4 (2.99%)
Clinical Characteristics at Enrollment
Child–Pugh Class	129
A		84 (65.12%)
B		40 (31.01%)
C		5 (3.88%)
Not applicable (non-cirrhotic)		5
MELD Score, median (IQR)	134	10.00 (8.00–12.00)
BCLC Stage at Diagnosis	134
0/A		86 (64.18%)
B		35 (26.12%)
C		11 (8.21%)
D		2 (1.49%)
AFP Categorized	125
<20 ng/mL		75 (55.97%)
20–399 ng/mL		24 (17.91%)
400–999 ng/mL		8 (5.97%)
≥1000 ng/mL		18 (13.43%)
Unknown		9 (6.72%)
Cirrhosis-related Complications	134
Esophageal varices		62 (46.27%)
History of variceal bleeding		43 (32.09%)
Hepatic encephalopathy		39 (29.10%)
Ascites		31 (23.13%)

Notes: Continuous variables are presented as median and interquartile range (IQR); and categorical variables as absolute frequencies and percentages. Percentages were calculated based on the number of patients with available data for each variable. ¹ Median (IQR); n (%). Percentages calculated based on patients with available data for each variable.

Abbreviations: AFP, alpha-fetoprotein; BCLC, Barcelona Clinic Liver Cancer; COPD, chronic obstructive pulmonary disease; DILI, drug-induced liver injury; IQR, interquartile range; MASLD, metabolic dysfunction-associated steatotic liver disease; MELD, Model for End-Stage Liver Disease; MetALD, metabolic dysfunction-associated alcohol-related liver disease.

At program enrollment, among patients with cirrhosis, 65.12% (n = 84) were Child–Pugh class A, 31.01% (n = 40) were class B, and 3.88% (n = 5) were class C; Child-Pugh classification was not applicable to five non-cirrhotic patients. The median MELD score was 10 (IQR 8.0–12.0). According to the BCLC classification, most patients were stage 0/A (64.18%; n = 86), followed by stage B (26.12%; n = 35); smaller proportions were classified as stage C (8.21%; n = 11) and stage D (1.49%; n = 2). Regarding AFP levels at enrollment, 55.97% (n = 75) had values <20 ng/mL, while 13.43% (n = 18) had levels ≥1000 ng/mL. Among cirrhosis-related complications, esophageal varices were most frequent, followed by history of variceal bleeding, hepatic encephalopathy, and ascites. Clinical characteristics at enrollment are summarized in Table 1.

Initial Treatment Strategies

Among the 134 patients included in the cohort, the most frequent treatment strategy was ablation (23.88%; n = 32), followed by transarterial chemoembolization (23.13%; n = 31) and systemic therapy (19.40%; n = 26). Combination therapy was used in 15.67% of patients (n = 21). Regarding surgical management, liver transplantation was performed in 9.70% (n = 13) and hepatectomy in 2.99% (n = 4). Other treatment strategies were recorded in 5.22% of cases (n = 7). Initial treatment strategies are summarized in Table 2.

Table 2.

Initial Treatment Strategies in Patients with Hepatocellular Carcinoma Enrolled in the Center of Clinical Excellence (N=134)

Initial Treatment Strategy	N = 134 (%)
Ablation	32 (23.88%)
Transarterial chemoembolization (TACE)	31 (23.13%)
Systemic therapy	26 (19.40%)
Combined therapy	21 (15.67%)
Liver transplantation	13 (9.70%)
Other	7 (5.22%)
Hepatectomy	4 (2.99%)

Note: n (%): absolute frequency and percentage of total cohort (N = 134).

Abbreviation: TACE, transarterial chemoembolization.

BCLC Stage Transitions During Follow-Up

A total of 134 patients with confirmed HCC were followed longitudinally to assess disease progression. Clinical progression was evaluated using a BCLC stage transition matrix, defined as migration to a higher BCLC stage or death during follow-up. Overall transition patterns are illustrated in Figures 2 and 3.

Figure 2.

BCLC stage transition matrix from initial BCLC stage to follow-up status in patients with hepatocellular carcinoma.

Figure 3.

Sankey diagram of BCLC stage transition from initial stage to follow-up status in patients with hepatocellular carcinoma.

Most patients entered follow-up at early BCLC stage 0/A (n = 88). Among them, 51.1% (n = 45) remained in the same stage during follow-up, while 8.0% (n = 7) progressed to stage B, 3.4% (n = 3) to stage C, and 1.1% (n = 1) to stage D. Additionally, 14.8% (n = 13) underwent liver transplantation and 21.6% (n = 19) died during follow-up.

Among patients initially classified as BCLC stage B (n = 35), 54.3% (n = 19) remained stable in the same stage. Progression to stage C and D occurred in 8.6% (n = 3) of patients for each stage, while 8.6% (n = 3) underwent liver transplantation and 20.0% (n = 7) died during follow-up.

Among those with initial BCLC stage C (n = 9), 33.3% (n = 3) remained at the same stage and 66.7% (n = 6) died during follow-up. Both patients with initial BCLC stage D died during follow-up.

Treatment Trajectories According to Initial BCLC Stage

Treatment trajectories according to initial BCLC stage are illustrated in Figure 4. Among patients with initial BCLC stage 0/A, the most frequent treatment strategies were ablation, chemoembolization, and combined therapy; a smaller proportion underwent liver transplantation or hepatectomy. At the end of follow-up, most patients in this group remained at early stage, while a subset underwent transplantation or died during follow-up. Among patients initially classified as BCLC stage B, treatment strategies were more heterogeneous, with chemoembolization, systemic therapy, and combined approaches being the most frequent; a smaller proportion underwent liver transplantation, while others progressed to more advanced stages or died during follow-up. Among patients with initial BCLC stage C, systemic therapy was the predominant treatment strategy; at the end of follow-up, a proportion remained at stage C while the remainder died during follow-up.

Figure 4.

Sankey diagrams of treatment trajectories and follow-up status stratified by initial BCLC stage in patients with hepatocellular carcinoma. The three panels correspond to patients with initial BCLC 0-A, B, and C, respectively. The diagrams include patients with complete and clinically informative transition pathways. Among patients initially classified as BCLC 0-A, 85 of 88 were included; among those classified as BCLC B, 33 of 35 were included; and all 9 patients initially classified as BCLC C were included. Flows represent transitions from initial BCLC stage to treatment received and current clinical status, and the numbers displayed in each stratum indicate the number of patients in that category.

Survival

Overall survival was estimated using the Kaplan-Meier method in the full cohort of 134 patients. The median follow-up time for the overall cohort was 21 months. The maximum follow-up time reached 36 months. Survival probability showed a gradual decline over time, with estimated survival rates of 97.7% (95% CI: 95.1–100) at 6 months, 93.8% (95% CI: 89.3–98.4) at 12 months, 87.7% (95% CI: 81.3–94.6) at 18 months, and 83.2% (95% CI: 75.6–91.7) at 24 months. Survival remained relatively stable thereafter, with an estimated survival of 80.0% (95% CI:70.8–90.5) at both 30 and 36 months. Median overall survival was not reached during the available follow-up period. The Kaplan-Meier overall survival curve is shown in Figure 5.

Figure 5.

Kaplan-Meier overall survival curve for the study cohort.

When overall survival was stratified by baseline BCLC stage, significant differences were observed between groups according to the Log rank test (p < 0.001). Patients classified as BCLC 0/A and BCLC B showed the most preserved survival over time, whereas patients in BCLC C experienced an earlier decline in survival. The BCLC D group showed the steepest decline; however, this estimate should be interpreted with caution due to the very small number of patients in this category. Median follow-up was 21.2 months for BCLC 0/A, 17.5 months for BCLC B, 22.8 months for BCLC C, and 3.8 months for BCLC D. The Kaplan-Meier survival curves stratified by BCLC stage are shown in Figure 6. Detailed Kaplan-Meier estimates, including the number of patients at risk, cumulative events, cumulative censoring, and survival probabilities during follow-up, are presented in Supplementary Tables S1 and S2; Supplementary Table S1 shows the estimates for the overall cohort and Supplementary Table S2 presents the results stratified by baseline BCLC stage.

Figure 6.

Kaplan-Meier of BCLC stage survival curve for the study cohort.

Discussion

This study provides a real-world description of the clinical profile, treatment allocation, BCLC stage transitions, and survival estimates of patients with HCC managed within a structured multidisciplinary CCE in Colombia. Rather than establishing causal evidence regarding the effect of integrated care on outcomes, these findings should be interpreted as contextual and hypothesis-generating, contributing data from a middle-income Latin American setting where evidence on organized HCC care pathways remains limited. Three findings stand out: MASLD was the most frequent underlying etiology of HCC in this cohort; most patients were diagnosed at early BCLC stages 0/A, characterizing the case-mix of a specialized referral center; and despite this early-stage distribution, a substantial proportion of patients experienced BCLC stage migration or death during follow-up, underscoring the dynamic course of HCC and the need for sustained, coordinated care over time.

Globally, HCC is associated with a marked etiology transition,¹⁹ with MASLD increasingly contributing while viral hepatitis-related cases decline in settings with established vaccination programs and broader access to antiviral therapy.^20–22 This transition has also been described in Latin America. Piñero et al⁷ reported an increase in the prevalence of MASLD as an etiology of HCC in liver transplant recipients, from 2% in 2005 to 13% in 2013. In the Americas, Debes et al²³ reported HCV as the leading cause (48%), followed by alcohol (22%), HBV (14%), and MASLD (9%). Against this background, our cohort shows a more advanced shift toward metabolic disease, with MASLD as the most frequent etiology (34.38%), accompanied by a high prevalence of metabolic comorbidities, including hypertension (58.21%) and diabetes (49.25%), while HBV (3.13%) and HCV (13.28%) were comparatively less frequent. Colombian cohorts have similarly documented MASLD as the leading etiology, ranging from 21% to 35%,^24,25 supporting the notion of a continuing epidemiologic transition. This etiologic shift is consistent with the high prevalence of metabolic risk factors observed in our cohort, including hypertension (58.21%) and type 2 diabetes mellitus (49.25%), which are well-recognized drivers of MASLD-related HCC in the Latin American region.²³

Survival is a key outcome for evaluating HCC management. In our cohort, median overall survival was not reached after a median follow-up of 21 months, and estimate survival was 93.8% at 1 year and 83.2% at 2 years. Comparisons across studies should be interpreted cautiously, as survival depends on epidemiologic profile, stage at diagnosis, availability of locoregional, systemic, and transplant therapies, as well as health-system organization. In settings characterized by late presentation and limited access to specialized care, substantially shorter survivals have been reported, including a median survival of 2.5 months from clinical presentation in sub-Saharan Africa.²⁶ Similarly, a large single-center cohort from China reported a median survival of approximately 9 months.²⁷ In contrast, a hospital-based cohort from the United States reported a median overall survival of 35 months, with clear variation by BCLC stage (44.9 months for stage A and 21.3 months for stage B).¹¹ The survival estimates observed in our cohort are consistent with the high proportion of patients diagnosed at early BCLC stages and the availability of curative or locoregional therapies within a specialized multidisciplinary center; however, given the relatively short median follow-up of 21 months and the small number of patients at risk beyond 24 months, these estimates should be interpreted with caution.

Treatment trajectories also reflected the dynamic nature of HCC management and were consistent with BCLC-based treatment allocation.¹⁶ Among patients with initial BCLC stage 0/A, curative-intent or locoregional therapies, predominated, including ablation, chemoembolization, combined therapy, and liver transplantation. Among those initially classified as BCLC stage B, treatment strategies were more heterogeneous, reflecting the clinical complexity of intermediate-stage disease and the need for individualized multidisciplinary decision-making. Patients with initial BCLC stage C primarily received systemic therapy, consistent with guideline recommendations for advanced disease.⁵ These patterns highlight the importance of longitudinal multidisciplinary reassessment, since therapeutic decisions in HCC may change over time according to tumor burden, liver function, treatment response, and transplant eligibility. The stage transition data further illustrate that even among patients diagnosed at early stages, a proportion experienced disease progression or death during follow-up, underscoring the need for sustained coordinated care beyond initial treatment allocation.

The organizational structure of our CCE supports coordinated delivery of HCC therapies through multidisciplinary decision-making, in line with international recommendations aimed at improving adherence and facilitating selection of curative-intent strategies.²⁸ This integrated approach is particularly relevant in oncology care, where fragmented care has been associated with worse clinical outcomes. In the United States, Harrison et al⁹ reported lower overall survival among patients with fragmented care (HR 1.81; 95% CI 1.72–1.91), with similar findings reported by Hester et al¹⁰ In Colombia, patients treated within fragmented networks for colorectal cancer had a 35% higher risk of death at 4 years.¹¹ Nonetheless, system-level barriers may persist even in organized programs. In our cohort, 38 patients discontinued CCE follow-up after enrollment, primarily due to administrative reasons related to insurance coverage. Among these patients, 19/38 (50.0%) died during follow-up, compared with 11/96 (11.5%) among those who remained in programed-based follow-up. While this difference should be interpreted descriptively given the observational design and the absence of adjustment for confounders, it illustrates how administrative fragmentation may disrupt continuity of care and alter clinical trajectories even within structured multidisciplinary programs. Despite discontinuing program-based follow-up, these patients were retained in the overall survival analysis through vital status ascertainment via ADRES, the Colombian national administrative health registry.

Limitations

This study has several limitations that should be considered when interpreting its findings. First, the retrospective, single-center design may introduce selection bias and limits generalizability. As this cohort was derived from an internationally accredited CCE, enrolled patients may differ systematically from the broader Colombian HCC population in terms of performance status, stage at diagnosis, insurance coverage, and access to multidisciplinary evaluation; thus, the observed early-stage distribution and survival estimates may not be representative of population-level HCC outcomes in Colombia. Second, the relatively short median follow-up of 21 months and the limited number of patients at risk beyond 24 months preclude robust long-term survival conclusions, and the survival estimates reported here should be considered preliminary. Third, the purely descriptive design does not allow for quantification of associations between clinical variables and outcomes; the absence of multivariable analysis limits the ability to identify independent prognostic factors. Fourth, discontinuation of CCE follow-up in 38 patients for administrative reasons may have influenced progression and trajectory estimates, as post-discontinuation clinical data could not be systematically captured. These limitations highlight the need for prospective, multicenter studies to validate these findings and to evaluate the independent contribution of structured multidisciplinary care models to HCC outcomes in middle-income settings.

Conclusion

This study describes the clinical characteristics, treatment patterns, treatment trajectories, and survival estimates of patients with HCC managed within a Center of Clinical Excellence in Colombia. In this cohort, MASLD was the most frequent underlying etiology, and most patients were diagnosed at early BCLC stages 0/A. Treatment strategies were mainly locoregional, particularly ablation and transarterial chemoembolization, while treatment trajectories varied according to initial BCLC stage. Median overall survival was not reached during the available follow-up period, with estimated survival of 93.8% at 1 year and 80.0% at 3 years. Administrative or insurance-related barriers led to discontinuation of program-based CCE follow-up in 38 patients; however, these patients remained included in the overall survival analysis through vital status ascertainment via ADRES.

Funding Statement

This study did not report external funding for study conduct. AstraZeneca funded medical writing support and covered the article processing charge.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of Fundación Cardioinfantil, Instituto de Cardiología, Colombia (protocol code CEIC-004-2026; approved on 16 January 2026).

Use of Generative AI

During the preparation of this manuscript, the authors used ChatGPT (Open AI) to assist with language editing, text refinement, and organizational improvement of the manuscript.

Data Sharing Statement

The data presented in this study are available from the corresponding author upon request. The data are not publicly available due to institutional policies and Colombian regulations governing the protection of personal health data.

Informed Consent Statement

Patient consent was waived by the institutional ethics committee due to the retrospective design of the study and the use of anonymized clinical data, in accordance with national regulations and international ethical guidelines.

Author Contributions

Conceptualization, C.T.-V., A.V.-P., O.B.-G., and A.S.-J.; methodology, A.S.-J. and J.A.-N.; data curation, M.A.-D.L.H. and C.T.-V.; formal analysis, A.S.-J. and J.A.-N.; writing—original draft preparation, C.T.-V., A.S.-J., A.V.-P., and J.A.-N.; writing—review and editing, all authors.

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.