The Liver Tumor Size Chart and How It Affects Staging

Liver tumor size is one of the most influential variables in cancer staging, but it never works alone. A tumor’s diameter in centimeters feeds into staging systems like the TNM classification and the Barcelona Clinic Liver Cancer (BCLC) framework, where it helps determine whether someone qualifies for transplant, surgery, ablation, or palliative care. The tricky part is that the same size tumor can land in different stages depending on how many tumors are present, whether blood vessels are involved, and how well the liver itself is functioning. Understanding how the numbers on a scan translate into a stage, and then into a treatment plan, gives you a much clearer picture of what the medical team is actually deciding.

How Size Feeds Into TNM Staging for Liver Cancer

The most widely used pathological staging framework is the AJCC TNM system, now in its 8th edition. For hepatocellular carcinoma (HCC), the “T” category is built around tumor size, number of tumors, and the presence of vascular invasion. A solitary tumor without vascular invasion is T1a if it is 2 cm or smaller and T1b if it exceeds 2 cm. Once you have a solitary tumor with vascular invasion, or multiple tumors none larger than 5 cm, the classification moves to T2. Larger or more numerous tumors push toward T3 and T4, with T4 specifically reserved for tumors that invade the portal vein or hepatic veins.

These groupings are meant to predict survival, but they are imperfect. A large analysis of the SEER database found that patients with a solitary tumor larger than 2 cm with vascular invasion actually survived longer than patients with multiple tumors under 5 cm, even though both groups are classified as T2. Among patients with multifocal tumors under 5 cm, those with vascular invasion fared worse than those without, yet the staging system does not distinguish between them.¹ This is a recurring theme in liver cancer staging: size alone does not capture the full biological story.

The BCLC System and the 5-Centimeter Divide

While TNM staging describes tumor anatomy in detail, clinicians more often rely on the Barcelona Clinic Liver Cancer (BCLC) system for treatment decisions, because it also accounts for liver function and the patient’s overall physical condition. Under the standard BCLC framework, stage A (early stage) includes a single tumor of any size or up to three tumors each 3 cm or smaller, provided the patient’s liver function is preserved. Stage B (intermediate) covers larger or multifocal tumors without vascular invasion or spread beyond the liver.

Where it gets complicated is the 5 cm threshold. Officially, a single tumor larger than 5 cm still qualifies as BCLC stage A if it is solitary and there is no vascular invasion or extrahepatic spread. But research has shown that patients with a single tumor over 5 cm behave more like stage B patients than like their stage A counterparts with smaller tumors. One study of over 1,000 HCC patients found that those with a solitary tumor exceeding 5 cm had a median survival of about 31 months, significantly worse than the roughly 43 months seen in patients with smaller stage A tumors, and nearly identical to the 34-month median of stage B patients.² Some researchers have argued that a single large HCC should be reclassified as stage B rather than stage A to more accurately reflect prognosis.³

That said, outcomes for single large tumors vary substantially depending on whether the patient undergoes resection and what the underlying liver looks like. In an Asian cohort where most patients received surgery, five-year overall survival for single large HCCs reclassified as BCLC stage A1 was about 77%, and there was no meaningful survival difference between these patients and those with standard stage A disease after adjusting for other factors.⁴ The takeaway is that a tumor slightly above 5 cm is not automatically a death sentence; it depends heavily on what treatment is offered and how the liver is functioning.

Transplant Eligibility and Size Cutoffs

Liver transplantation offers the best chance of long-term cure for HCC because it removes both the cancer and the diseased liver. But donor organs are scarce, so strict size criteria determine who gets listed. The Milan criteria, established in the 1990s, remain the global benchmark: a single tumor 5 cm or smaller, or up to three tumors each 3 cm or smaller, with no vascular invasion and no extrahepatic spread.⁵ These thresholds were chosen because patients meeting them showed five-year survival rates that rivaled those of transplant recipients without cancer.

The concern has always been that the Milan criteria are too restrictive. The UCSF criteria expand the window to a solitary tumor up to 6.5 cm, or up to three tumors with the largest 4.5 cm or smaller and a total tumor diameter of 8 cm or less. In a study of 70 transplant recipients, patients who exceeded the Milan criteria but met the UCSF criteria had a two-year survival of about 86%, well above the 70% threshold generally considered acceptable.⁶ Some centers have also reported that survival for patients transplanted beyond the Milan criteria is comparable to survival for those transplanted within them.⁷

For patients whose tumors exceed even the expanded criteria, downstaging through locoregional therapy is an option. The idea is to shrink the tumor with treatments like transarterial chemoembolization or ablation until it meets Milan or other transplant criteria. A multicenter analysis found that roughly 83% of patients achieved successful downstaging after a median of about two and a half months from the first procedure.⁸ Still, success rates decline as tumor burden increases, and not every patient can be safely downstaged. A structured protocol with clear eligibility criteria, endpoints, and a minimum observation period after downstaging is considered essential.⁹

Why Bigger Tumors Behave Worse

Size is not just a measurement; it correlates with biological aggressiveness. As a tumor grows, the odds that it has invaded small blood vessels (microvascular invasion) climb steeply. In one surgical series, microscopic vascular invasion was found in about 25% of tumors 3 cm or smaller, 40% of tumors between 3 and 5 cm, 55% of tumors between 5 and 6.5 cm, and 63% of tumors above 6.5 cm.¹⁰ Vascular invasion matters because it is one of the strongest predictors of recurrence after surgery or transplant. A separate study confirmed that microvascular invasion correlates with tumor size, histological grade, and the presence of tiny satellite tumors within the liver.¹¹

Alpha-fetoprotein (AFP), a blood marker often tracked in HCC patients, also rises with tumor size. AFP levels independently predict tumor size, and tumors 5 cm or smaller are significantly more common among patients who are AFP-negative than among those with elevated AFP.¹² This means that when a blood test shows a rising AFP and imaging shows a growing mass, the two findings are telling a consistent story about tumor behavior. Increasingly, clinicians are combining molecular markers like AFP with conventional staging to build more individualized prognostic models.¹³

Size Thresholds for Ablation Versus Surgery

For tumors small enough to treat locally, the choice between radiofrequency ablation (RFA) and surgical resection depends heavily on size. The evidence consistently points to about 3 cm as the line where ablation and surgery produce similar survival outcomes. A large propensity-matched analysis found that for solitary HCC tumors under 3 cm, patients who received ablation had survival outcomes comparable to those who underwent surgery. For tumors between 3 and 5 cm, however, surgery provided a clear survival advantage.¹⁴ Guidelines generally favor ablation for patients with tumors below 3 cm and low AFP levels, especially when surgery carries higher risk due to liver dysfunction or other medical conditions.¹⁵

There is also a practical issue with ablation: it becomes harder to completely destroy larger tumors. When tumors reach 4 cm or above, local recurrence at the ablation site increases significantly. One study found that a largest ablated tumor size of 4 cm or greater was independently associated with worse ablation-site recurrence-free survival.¹⁶ The physics behind this are straightforward: ablation heats and kills tissue in a roughly spherical zone, and ensuring complete destruction of a large, irregularly shaped tumor with adequate safety margins becomes exponentially more difficult as the tumor grows.

For intrahepatic cholangiocarcinoma, a different type of primary liver cancer, the size threshold for treatment choice is similar. In tumors smaller than 3 cm, ablation and resection produce comparable outcomes. Once the tumor hits 3 cm or above, surgical resection is associated with much better survival, with five-year rates of about 38% after surgery versus roughly 8% after ablation.¹⁷

When Tumors Are Large but Still Resectable

Surgery does not stop being an option just because a tumor is big. For single large HCCs in patients with reasonably preserved liver function, resection can still offer meaningful survival. A nationwide registry analysis of patients with single large HCCs found five-year overall survival rates above 50% after resection. The cutoff that mattered most was 10 cm: tumors at or above that diameter were independently associated with worse overall and progression-free survival.¹⁸

What predicts recurrence after resecting a big tumor is not just its diameter but the biological context around it. One study that built a scoring system for recurrence risk after surgery for single large HCCs found that the strongest drivers of recurrence were elevated AFP before surgery, portal vein invasion, and underlying cirrhosis. Patients with none of those factors had a median disease-free survival of about 65 months. Those with two or more risk factors saw that drop to under 9 months.¹⁹ Size matters, but vascular invasion and tumor biology matter at least as much once the decision to operate has been made.

Measuring Tumor Size Is Not as Simple as It Sounds

Staging depends on accurate measurement, and there are real discrepancies between what imaging shows and what the pathologist measures after surgery. A study of over 800 HCC patients found that while imaging and pathological tumor sizes correlated well overall, about 13% of patients had a meaningful size disparity between the two. In those patients, imaging tended to underestimate the tumor’s true size. Patients with size disparity had significantly worse five-year cancer-specific survival compared to those whose imaging and pathology measurements agreed.²⁰

The choice of imaging modality also matters. MRI has higher sensitivity and accuracy than CT for detecting and characterizing liver tumors, with one study reporting MRI sensitivity of about 79% compared to about 62% for CT.²¹ MRI also outperforms CT in detecting small metastases and in patients with fatty liver.²² Despite this, CT remains the more commonly used tool in many settings due to speed, availability, and lower cost. The practical implication is that a tumor measured on CT could appear slightly different in size on MRI, which can occasionally shift a patient across a staging boundary.

How Tumor Response Is Measured During Treatment

Once treatment begins, clinicians need to track whether tumors are shrinking, stable, or growing. The standard approach is RECIST 1.1, which measures the longest diameter of target lesions on imaging. But liver tumors often respond to treatment in ways that RECIST 1.1 does not capture well. A treated tumor might maintain its overall size while the interior dies off, leaving just a rim of viable tissue. By RECIST rules, that tumor looks unchanged or even larger. By modified RECIST (mRECIST), which measures only the enhancing, viable portion of the tumor on contrast imaging, it may count as a response.

This distinction has real clinical consequences. A systematic review found that the overall response rate using mRECIST was about 16%, roughly double the 8% response rate measured by RECIST 1.1.²³ For treatments like radioembolization, mRECIST responders showed significantly better survival than non-responders, while RECIST 1.1 could not meaningfully distinguish the two groups.²⁴ With newer immunotherapy drugs, this gets even more nuanced: a tumor treated with a PD-1 inhibitor might swell initially due to immune cell infiltration before it starts shrinking, making early size measurements misleading regardless of which RECIST version you use.²⁵

Size Thresholds in Liver Metastases

When cancer spreads to the liver from another organ, staging systems work differently. Colorectal liver metastases, the most common type, are not staged using TNM categories specific to the liver. Instead, treatment decisions revolve around whether the lesions can be completely removed by surgery or ablation and whether enough functional liver will remain afterward. A multidisciplinary consensus framework divides colorectal liver metastases into resectable/ablatable (stage IVa), requiring major surgery (IVb), needing downsizing first (IVc), and permanently unresectable if downsizing fails (IVd).²⁶

Size still matters for prognosis in this setting. Patients with colorectal liver metastases whose largest lesion was 21 mm or larger had significantly worse disease-free survival and disease-specific survival compared to those with smaller lesions, and this held up even after adjusting for other prognostic factors in the neoadjuvant treatment context.²⁷ A composite score combining the maximum diameter and number of liver metastases has also been proposed as an independent prognostic tool.²⁸

Benign Liver Tumors and the Size Question

Not every liver mass is cancer, and size thresholds guide management of benign tumors too. Hepatic adenomas, which are benign growths most commonly seen in women taking oral contraceptives, are typically monitored without surgery if they are under 5 cm and not causing symptoms. The 5 cm threshold exists because the risk of rupture and bleeding rises with size. In a multicenter analysis, ruptured adenomas were on average about 10.5 cm, and no tumor smaller than 5 cm ruptured. Larger tumors and recent hormone use were independently associated with rupture risk, leading some experts to recommend surgery for adenomas approaching 4 cm in patients who need to continue hormonal therapy.²⁹

More recent evidence suggests that even larger adenomas can sometimes be managed without surgery if the patient stops hormonal contraceptives, loses weight (in obesity-related cases), and is followed with regular imaging. The subtype of adenoma also matters: some subtypes carry a risk of malignant transformation that others do not, making the decision about surgery more about biology and less about centimeters alone.³⁰

Pediatric Liver Tumors Use a Different System Entirely

Children with primary liver tumors are staged using PRETEXT (Pre-Treatment Extent of Disease), a system developed by the International Childhood Liver Tumor Strategy Group. Rather than using tumor size in centimeters as the primary variable, PRETEXT divides the liver into four sections and stages based on how many sections are involved by tumor. A PRETEXT I tumor is confined to one section with three adjoining sections free; PRETEXT IV involves all four sections. The system also records vascular involvement, extrahepatic spread, and tumor rupture as annotation factors. This approach reflects the reality that in pediatric liver surgery, what matters most is not the diameter of the mass but whether enough healthy liver can be preserved after resection.³¹

The PRETEXT system is a reminder that tumor size is not universally the most important staging variable. In adult HCC, size drives treatment eligibility and carries prognostic weight, but it does so alongside vascular invasion, liver function, and tumor biology. In children, the anatomic distribution of tumor within the liver takes precedence. In metastatic disease, the number and location of lesions often outweigh the diameter of any single one. Every staging system reflects a compromise between simplicity and accuracy, and no single number on a chart captures the full clinical picture.