Hepatocellular Carcinoma Staging and the BCLC System

Hepatocellular carcinoma (HCC) staging is unusually complex because it has to capture two diseases at once: the cancer itself and the damaged liver it grows in. The Barcelona Clinic Liver Cancer (BCLC) system, first proposed in 1999, has become the dominant staging framework in Western practice and links each stage directly to a recommended treatment strategy.¹ But BCLC is far from the only system in use, and the way a tumor gets classified can vary depending on the country, the scoring method, and how well the liver is still functioning.

The BCLC Framework

Most Western guidelines, including those from the American and European liver disease associations, use the BCLC system to match patients with appropriate treatments. It divides HCC into five categories (very early through terminal) based on three pillars: tumor characteristics such as size and number of nodules, liver function typically measured by the Child-Pugh score, and the patient’s overall physical condition measured by performance status. Since 1999, it has been updated several times but has retained its fundamental structure as both a prognostic tool and a treatment algorithm.²

At the earliest stages (BCLC 0 and A), patients have small, asymptomatic tumors and preserved liver function, making them candidates for curative treatments like surgical removal, liver transplantation, or tumor ablation. Intermediate stage (B) covers patients with multiple tumor nodules who still feel well and have reasonable liver function. Advanced stage (C) applies when the tumor has invaded blood vessels, spread beyond the liver, or the patient’s physical condition has declined. Terminal stage (D) describes patients with severely impaired liver function or very poor physical status, where the expected survival is under three months and only supportive care is appropriate.³ ⁴

Anatomical Staging and Its Limits

The AJCC/UICC TNM system, used broadly across cancers, classifies HCC by tumor size, number, vascular invasion, lymph node involvement, and distant spread. The 8th edition introduced a split for small tumors: a solitary tumor 2 cm or smaller without vascular invasion is classified T1a, while a solitary tumor larger than 2 cm without vascular invasion is T1b. This matters after surgery because even among those small T1a tumors, the presence of microscopic blood vessel invasion dramatically worsens outcomes.⁵

The TNM system’s main weakness for HCC is that it ignores liver function entirely. Two patients with identically sized tumors can have radically different prognoses if one has a healthy liver and the other has advanced cirrhosis. A large study of over 13,000 patients in Japan found that the Japanese TNM system, which uses three simple binary criteria (vascular invasion, tumor size above or below 2 cm, and single versus multiple tumors) had stronger prognostic discrimination than the AJCC version. Five-year survival ranged from 70% in the best group down to 24% in the worst, with clear separation at each level.⁶

Another analysis using the U.S. SEER database found that the AJCC 8th edition had a concordance index of just 0.60, meaning its ability to correctly rank patients by survival was only modestly better than a coin flip. The study also found that a solitary tumor with vascular invasion actually had better survival than multiple tumors under 5 cm, even though the staging system lumps them into the same T2 category.⁷ These kinds of mismatches are why purely anatomical staging has never been the whole story for liver cancer.

Measuring Liver Function Alongside Tumor Burden

The Child-Pugh score has been the default tool for grading liver function in HCC for decades. It uses five clinical measures: bilirubin, albumin, clotting time, ascites, and encephalopathy, producing a grade of A (best), B, or C (worst). The BCLC system leans heavily on it to determine which treatments a patient can tolerate. But the score is coarse. A patient graded Child-Pugh A5 (the best possible within class A) can have a very different prognosis from one graded A6 (the lowest tier within the same class).

The Albumin-Bilirubin (ALBI) grade was developed to offer a more objective, continuous measure of liver function using only two blood tests: albumin and bilirubin. In patients who had liver resection, the ALBI grade showed better ability to predict long-term survival than the Child-Pugh score.⁸ A separate study in patients treated with the targeted drug sorafenib found that median survival was about 11 months for ALBI grade 1, roughly 7 months for grade 2, and 3 months for grade 3. The survival gap between Child-Pugh A5 and A6 patients was similarly stark, at about 10 versus 7 months.⁹ Both tools have their place, but the ALBI grade is increasingly used to refine decisions within Child-Pugh A, where the traditional score lumps together patients who actually face quite different odds.

How Imaging Determines the Stage

HCC is one of the few cancers that can be diagnosed by imaging alone, without a biopsy, if the tumor shows a characteristic pattern of taking up and then washing out contrast dye on CT or MRI. The Liver Imaging Reporting and Data System (LI-RADS), endorsed by the American College of Radiology, standardizes how radiologists interpret and report these scans.¹⁰ It assigns each liver observation a category from LR-1 (definitely benign) through LR-5 (definitely HCC), providing a standardized vocabulary that feeds directly into staging.¹¹

The LI-RADS category does not replace the staging system, but it determines the inputs. Whether a nodule is counted as HCC, how large it measures, and whether vascular invasion is present all come from imaging interpretation. Errors or ambiguity at this step ripple through the entire staging and treatment process, which is why standardized reporting has become so important.

Biomarkers and What They Add

Alpha-fetoprotein (AFP) remains the most widely used blood marker for HCC, but its sensitivity for early-stage disease is poor. Many small tumors produce little or no AFP, and some non-cancerous liver conditions can elevate it.¹² Two additional markers, AFP-L3 (a specific sugar-modified fraction of AFP) and DCP (des-gamma-carboxy prothrombin), help fill the gaps. The GALAD score, which combines AFP, AFP-L3, and DCP with age and sex, has shown better sensitivity for catching early-stage HCC than AFP alone. Both AFP-L3 and DCP also correlate with aggressive tumor features like microvascular invasion, and they are finding a role in monitoring treatment response and predicting recurrence after transplant.¹³

Biomarkers do not currently have their own slot in most staging systems, but they increasingly influence treatment decisions at the margins. For example, AFP level can determine eligibility for a second-line drug, and high AFP-L3 or DCP readings after transplant can signal recurrence before imaging shows anything visible.

Transplant Eligibility and the Milan Criteria

Liver transplantation is the only treatment that addresses both the tumor and the underlying liver disease simultaneously. The Milan criteria, established over two decades ago, set the eligibility bar: a single tumor no larger than 5 cm, or up to three tumors each no larger than 3 cm, with no vascular invasion or spread beyond the liver. These criteria remain the international benchmark.¹⁴

The problem is that the Milan criteria are conservative. Some patients with slightly larger or more numerous tumors can still do well after transplant. UNOS Region 4 in the United States expanded its criteria to allow a single tumor up to 6 cm or up to three tumors with the largest no more than 5 cm and total diameter no more than 9 cm. A review of over 2,000 patients transplanted under these rules showed no meaningful difference in ten-year survival compared with patients who met the traditional Milan criteria.¹⁵ A living-donor transplant study using similar expanded limits (largest tumor up to 6 cm, total tumor diameter under 10 cm) found a five-year survival of about 77%, comparable to Milan-criteria patients, while expanding the eligible pool by roughly a third.¹⁶

Downstaging to Reach Transplant

Patients whose tumors initially exceed transplant criteria can sometimes be “downstaged” through locoregional treatments like transarterial chemoembolization or ablation, shrinking the cancer enough to qualify. A ten-year study found that about 83% of patients within the standard downstaging protocol were successfully reduced to within Milan criteria. After transplant, their five-year recurrence-free survival was roughly 64%, compared with about 47% for patients who went ahead with transplant despite remaining beyond Milan criteria.¹⁷

Patients with more advanced disease at the start face a harder path. A prospective multiregional study found that patients in the “all-comers” group beyond standard downstaging limits had a successful downstaging rate of about 65%, compared with 83% in the standard group. They also needed more treatment sessions and a longer time period to reach the goal.¹⁸ Downstaging essentially uses treatment response as a biological test: tumors that shrink and stay down are likely less aggressive than those that resist or recur quickly.

Why Intermediate Stage Is So Difficult to Manage

BCLC stage B, the intermediate category, is arguably the most frustrating in clinical practice. It covers an enormous range of patients: someone with four small nodules and excellent liver function sits in the same category as someone with twenty nodules of varying sizes and borderline liver function. Yet the traditional BCLC recommendation for this entire group is a single treatment, transarterial chemoembolization.¹⁹

To address this, several groups have proposed subclassifications. The Kinki criteria, for example, split stage B into three substages using two factors: liver function (Child-Pugh 5–7 versus 8–9) and whether the tumor burden falls within or beyond the “up-to-seven” rule (where the number of tumors plus the diameter of the largest tumor does not exceed seven). This simple split helps identify which intermediate-stage patients might benefit from more aggressive approaches and which should be directed toward supportive care earlier. Recent BCLC updates have themselves begun incorporating some of this heterogeneity, but the fundamental tension between simplicity and accuracy remains.

Systemic Therapy for Advanced Disease

Advanced-stage HCC (BCLC C) has seen a genuine revolution in treatment over the past several years. For more than a decade, the targeted drug sorafenib was the only systemic option. The landscape shifted when the combination of atezolizumab (an immune checkpoint inhibitor) and bevacizumab (which targets blood vessel growth) showed improved survival in a large trial and became the new first-line standard for patients with preserved liver function and adequate physical status.²⁰ The combination of tremelimumab and durvalumab has also shown an overall survival benefit over sorafenib, offering another immunotherapy-based frontline option.²¹

For patients who cannot receive immunotherapy, whether because of autoimmune conditions, prior organ transplants, or other contraindications, the tyrosine kinase inhibitors sorafenib and lenvatinib remain available as first-line options. After first-line treatment fails, second-line choices include cabozantinib, regorafenib (only for patients who previously tolerated sorafenib), and ramucirumab (specifically for patients with AFP levels at or above 400 ng/mL).²² Stage classification matters here because all of these therapies are studied in patients with Child-Pugh A liver function. For patients with worse liver function, the evidence is thin and the risks of treatment are higher.

Early-Stage Treatment Choices

For small, early-stage tumors, three options compete: surgical resection, transplantation, and radiofrequency ablation. When liver function is well preserved, resection is generally preferred regardless of tumor size, because it is not constrained by the same size and number limits that apply to ablation and transplantation.²³ For tumors under about 3 cm with low AFP levels, ablation has shown survival rates comparable to surgery in some studies, making it a reasonable alternative, especially in patients who are not ideal surgical candidates.²⁴

A practical staging consideration here is the “salvage transplant” strategy: resect first, and if the cancer comes back, the patient may still be eligible for transplantation. This only works if the recurrence is caught early and the patient’s liver function has not deteriorated too far, which underscores the importance of continued surveillance after any treatment.

Microvascular Invasion Changes Everything

One feature that doesn’t appear in BCLC staging but has enormous prognostic weight is microvascular invasion (MVI): the presence of tumor cells within small blood vessels around the main tumor. This can only be confirmed by examining tissue under a microscope, so it is typically a post-surgical finding rather than a pre-treatment staging input. Patients whose tumors show severe MVI (invasion of five or more vessels or involvement of large clusters of tumor cells) have roughly three times the risk of death compared with those without it.²⁵

This creates a frustrating gap: two patients can be staged identically before surgery, receive the same operation, and then have wildly different outcomes based on what the pathologist finds afterward. Nomograms that incorporate MVI alongside other pathological features are being developed to predict both survival and early recurrence more precisely than stage alone.²⁶ Biomarkers like AFP-L3 and DCP, mentioned earlier, show promise in predicting MVI before surgery, which could eventually feed back into pre-treatment staging decisions.

HCC Without Cirrhosis

Most staging systems were built around the assumption that HCC develops in a cirrhotic liver, but a meaningful minority of cases arise in patients without cirrhosis. These patients tend to present very differently. In one study, tumors in non-cirrhotic livers were diagnosed at a more advanced stage, with nearly half classified T3 or T4 compared with about a third in cirrhotic patients. The tumors were also substantially larger, with a median size of 7.5 cm versus 3.5 cm, and metastatic disease at diagnosis was more than twice as common.²⁷

The reason is straightforward: patients with cirrhosis are often already in surveillance programs that catch tumors when they are small, while patients without cirrhosis typically have no reason to be screened. Their cancer is found incidentally or when symptoms appear, by which point it is usually much more advanced.²⁸ On the other hand, these patients often have better underlying liver function, which can make them better surgical candidates if the tumor is resectable.

How the Underlying Liver Disease Affects Staging and Outcomes

The cause of liver disease behind the cancer also shapes the staging picture. HCC driven by non-alcoholic steatohepatitis (NASH, now often called metabolic dysfunction-associated steatohepatitis) presents differently from HCC linked to hepatitis B, hepatitis C, or alcohol. NASH-related HCC patients tend to be older, more often female, and less likely to have full-blown cirrhosis at diagnosis.²⁹ That last point is significant: because many do not have cirrhosis, they are not enrolled in surveillance, and their tumors are more likely to be found late. Yet when NASH-related HCC is caught early enough for curative treatment, overall survival can be longer than in viral hepatitis patients, likely because their underlying liver function is better preserved.

Paradoxically, one study found that despite similar BCLC stage at diagnosis, NASH patients had worse overall survival compared with those whose HCC arose from alcohol-related liver disease, though outcomes were similar to viral hepatitis patients.³⁰ The metabolic syndrome, obesity, and cardiovascular disease that accompany NASH can independently limit treatment options and survival, introducing prognostic factors that no current staging system captures.

Regional Differences in Staging Systems

There is no single global staging system for HCC. While Western guidelines rally around the BCLC system, much of Asia uses different frameworks. Korea and Japan favor modified UICC-based TNM staging. Hong Kong developed its own system (the Hong Kong Liver Cancer staging system) using data from nearly 4,000 local patients, and it tends to be more aggressive in recommending active treatment for intermediate and advanced cases than BCLC would. China established its own staging system in 2017, incorporating tumor characteristics, liver function, and performance status in a structure that resembles BCLC but with distinct treatment allocations.³¹

These differences are not just academic turf wars. HCC’s biology varies by region. In East Asia, hepatitis B is the dominant driver, and patients often present at a younger age with different tumor behavior than in the West, where hepatitis C and metabolic liver disease predominate. Organ availability for transplantation also varies enormously, and so does access to expensive systemic therapies. A staging system designed around the assumption that transplant is available, as the BCLC partly is, becomes less useful in settings where living-donor transplant is the primary option or where transplant is simply not accessible.³²

Competing Integrated Scores

Beyond BCLC and TNM, several systems attempt to combine tumor and liver information into a single score. The Cancer of the Liver Italian Program (CLIP) score was one of the earliest integrated systems and has been widely validated. However, it performs poorly in early-stage patients, which increasingly make up a larger share of diagnoses thanks to screening. The Japan Integrated Staging (JIS) score addresses this by combining the Japanese TNM classification with the Child-Pugh grade. In head-to-head comparisons, the JIS score discriminated between stages more effectively than the CLIP score, and its best prognostic group had a ten-year survival of 65% versus 23% for the best CLIP group.³³ Multivariate analyses have confirmed that the JIS and CLIP systems outperform pure TNM systems at predicting survival, though each has its strengths in different populations and disease stages.³⁴

The proliferation of staging systems reflects an underlying reality: no single scheme perfectly captures HCC’s complexity. Each system was developed in a specific patient population, validated against particular endpoints, and optimized for certain treatment algorithms. The challenge, as several researchers have noted, is balancing the desire for precision (adding more variables) against the need for simplicity in daily clinical use.³⁵ Staging HCC is a problem that has been solved many times over, and each solution highlights different trade-offs.

1
PubMed. Prognosis of hepatocellular carcinoma: the BCLC staging classification ↩︎
2
PubMed. Merits and boundaries of the BCLC staging and treatment algorithm for hepatocellular carcinoma ↩︎
3
PubMed. Prognosis of hepatocellular carcinoma: the BCLC staging classification ↩︎
4
PubMed Central. Palliative Care for People with Hepatocellular Carcinoma and Specific Benefits to Older Adults ↩︎
5
PubMed Central. Evaluation of the American Joint Committee on Cancer AJCC 8th Edition Staging System for Hepatocellular Carcinoma in 1008 Patients with Curative Resection ↩︎
6
PubMed Central. Staging of Hepatocellular Carcinoma: Assessment of the Japanese TNM and AJCC/UICC TNM Systems in a Cohort of 13,772 Patients in Japan ↩︎
7
Journal of Surgical Oncology. Critical evaluation of the American Joint Commission on Cancer (AJCC) 8th edition staging system for patients with Hepatocellular Carcinoma (HCC): A Surveillance, Epidemiology, End Results (SEER) analysis ↩︎
8
PubMed Central. Comparison between Child-Pugh score and Albumin-Bilirubin grade in the prognosis of patients with HCC after liver resection using time-dependent ROC ↩︎
9
Liver International. A multicentre comparison between Child Pugh and Albumin‐Bilirubin scores in patients treated with sorafenib for Hepatocellular Carcinoma ↩︎
10
PubMed Central. CT-MRI LI-RADS v2017: A Comprehensive Guide for Beginners ↩︎
11
PubMed Central. Up-to-Date Role of CT/MRI LI-RADS in Hepatocellular Carcinoma ↩︎
12
Journal of Digestive Cancer Research. Serum Biomarkers for Hepatocellular Carcinoma: A Comparative Review of Traditional and Emerging Markers ↩︎
13
Current Hepatology Reports. The Role of AFP-L3 and DCP Biomarkers in the Diagnosis and Management of Hepatocellular Carcinoma ↩︎
14
PubMed Central. Liver Transplantation in Patients with Hepatocellular Carcinoma beyond the Milan Criteria: A Comprehensive Review ↩︎
15
PubMed. A long-term experience with expansion of Milan criteria for liver transplant recipients ↩︎
16
BMC Surgery. Living donor liver transplantation for hepatocellular carcinoma beyond the Milan criteria: outcome of expanded criteria in tumor size ↩︎
17
JAMA Surgery. Ten-Year Outcomes of Liver Transplant and Downstaging for Hepatocellular Carcinoma ↩︎
18
American Journal of Transplantation. Down-staging outcomes and liver transplantation for hepatocellular carcinoma beyond conventional criteria ↩︎
19
Digestive Diseases. Subclassification of BCLC B Stage Hepatocellular Carcinoma and Treatment Strategies: Proposal of Modified Bolondi’s Subclassification (Kinki Criteria) ↩︎
20
PubMed Central. Systemic Therapy for Advanced Hepatocellular Carcinoma: Current Stand and Perspectives ↩︎
21
PubMed Central. Systemic Therapy in Advanced Hepatocellular Carcinoma: Patient Selection and Key Considerations ↩︎
22
Journal of Clinical Oncology. Systemic Therapy for Advanced Hepatocellular Carcinoma: ASCO Guideline ↩︎
23
PubMed Central. Management of small hepatocellular carcinoma: a review of transplantation, resection, and ablation ↩︎
24
PubMed Central. Appropriate treatment modality for solitary small hepatocellular carcinoma: Radiofrequency ablation vs. resection vs. transplantation? ↩︎
25
PubMed Central. Classification of microvascular invasion of hepatocellular carcinoma: correlation with prognosis and magnetic resonance imaging ↩︎
26
PubMed Central. Novel microvascular invasion-based prognostic nomograms to predict survival outcomes in patients after R0 resection for hepatocellular carcinoma ↩︎
27
Scandinavian Journal of Surgery. Hepatocellular carcinoma in cirrhotic versus non-cirrhotic liver: Treatment and survival differences in a nationwide cohort ↩︎
28
PubMed Central. Hepatocellular Carcinoma in Patients Without Cirrhosis: The Fibrosis Stage Distribution, Characteristics and Survival ↩︎
29
Hepatology. Outcomes of curative treatment for hepatocellular cancer in nonalcoholic steatohepatitis versus hepatitis C and alcoholic liver disease ↩︎
30
PubMed. Comparative Analysis of Nonalcoholic Steatohepatitis- Versus Viral Hepatitis- and Alcohol-Related Liver Disease-Related Hepatocellular Carcinoma ↩︎
31
Clinical and Molecular Hepatology. Overview of Asian clinical practice guidelines for the management of hepatocellular carcinoma: An Asian perspective comparison ↩︎
32
Cancer. The clinical management of hepatocellular carcinoma in the United States, Europe, and Asia: A comprehensive and evidence‐based comparison and review ↩︎
33
PubMed. Prognostic staging system for hepatocellular carcinoma (CLIP score): its value and limitations, and a proposal for a new staging system, the Japan Integrated Staging Score (JIS score) ↩︎
34
PubMed. Comparative study of survival of patients with hepatocellular carcinoma predicted by different staging systems using multivariate analysis ↩︎
35
The Oncologist. The Challenge of Prognosis and Staging for Hepatocellular Carcinoma ↩︎