Liver Tumors and Transplantation: Hepatocellular Carcinoma End to End

0:00Welcome to Board Pearls. This is episode two of four of the Liver Tumors and Transplant chapter, in the Liver Disease module. This episode is hepatocellular carcinoma, from who gets surveilled through how a surveillance finding becomes a diagnosis and how that diagnosis picks a treatment.

0:15HCC begins with surveillance for one reason: the only way to deliver a curative treatment is to find the tumor while the patient still qualifies for one. The current recommendation is surveillance in every patient with Child-Pugh A or B cirrhosis whatever the cause, including patients whose hepatitis C was cured, because the cirrhotic liver stays at risk even after the virus is gone. Child-Pugh C cirrhosis gets surveilled only when the patient is a transplant candidate, and that cutoff is pure logic: if liver function is too poor for any HCC treatment short of transplant, then finding a tumor in a non-transplant candidate changes nothing, so you only surveil where a finding can change what you do.

0:53Non-cirrhotic chronic hepatitis B gets surveilled when it hits one of several risk triggers, because hepatitis B causes cancer directly through viral oncogenesis even without cirrhosis, and those triggers are Asian men over forty, Asian women over fifty, people of African ancestry at a younger age, a family history of HCC, a viral load over two hundred thousand international units per milliliter, HBeAg-positive status, prior HCC, or a PAGE-B score of ten or higher on long-term antiviral therapy, with hepatitis D co-infection surveilled regardless because it compounds the cancer risk.

1:23The modality is a right-upper-quadrant ultrasound with alpha-fetoprotein every six months, and the six-month interval is set to the tumor's doubling time, short enough to catch lesions while they're still within the size cutoffs that allow a curative treatment, which is the entire point. A yearly interval would let too many tumors outgrow those cutoffs between scans.

1:46The role of the AFP is worth understanding because it shifted from optional to a recommended add-on: adding it to ultrasound catches meaningfully more cancers than ultrasound alone, and that gain is worth the false positives when it's paired with imaging. But AFP is never a stand-alone test, because both its sensitivity and specificity are weak, a normal value doesn't exclude HCC since some tumors don't secrete it, and a mildly high value doesn't establish it since hepatitis flares and regeneration also raise it. The one setting where AFP carries diagnostic weight on its own is a value over two hundred nanograms per milliliter together with arterial enhancement on imaging, because concentrations that high are uncommon in benign disease, so that combination strongly supports HCC even when the imaging is otherwise equivocal.

2:35Two practical wrinkles: severe obesity or heavy steatosis can make the ultrasound useless because sound doesn't penetrate, and then you substitute MRI or CT every six months. And the override that runs through the whole section is that a new liver lesion in any cirrhotic or any chronic hepatitis B patient is HCC until proven otherwise, even when the imaging features would look benign in a normal liver, because the pre-test probability is high enough that reading a small lesion as a regenerative nodule or focal fat is the wrong move. The right move is to enter the diagnostic algorithm, which is LI-RADS.

3:14LI-RADS is what turns a surveillance finding into a decision. It grades cirrhotic liver lesions from one, definitely benign, up to five, definitely HCC, with three as indeterminate in between, plus two special categories: one for a mass that's malignant but not typical of HCC, and one for tumor growing inside a vein. These categories aren't arbitrary; they track the imaging features that reflect HCC biology. The five category is the diagnostic one, and it requires a lesion one centimeter or larger with arterial-phase hyperenhancement plus at least one major feature, where the major features are washout in the delayed or portal-venous phase, an enhancing capsule, or threshold growth defined as a fifty percent or greater increase within six months. When those are met in a cirrhotic liver, that's a diagnosis of HCC without biopsy, because the picture is essentially specific and a biopsy would only add tumor-seeding risk.

4:10Each of those features has a mechanism worth carrying. The arterial hyperenhancement is the tumor building its own arterial blood supply through angiogenesis, so it lights up brightly in the arterial phase while the normal liver, still fed mostly by the portal vein, hasn't enhanced yet. The washout is the lesion emptying quickly through leaky abnormal tumor vessels while the surrounding liver fills in during the portal phase, so the lesion goes relatively dark against a brightening background. And the enhancing capsule is the fibrous rim of compressed liver that a growing HCC pushes against, which takes up contrast.

4:45The four category is probably-HCC missing one feature, managed with short-interval imaging or biopsy depending on the patient. The three category is indeterminate and rescanned in three to six months, because lesions here can regress or slowly declare themselves, and the wait lets them.

5:03The malignant-but-not-HCC category exists to force a different path: rim arterial enhancement, peripheral washout, and a targetoid look point toward intrahepatic cholangiocarcinoma or a mixed tumor rather than HCC, and you biopsy these because management diverges sharply, since intrahepatic cholangiocarcinoma over two centimeters is a transplant contraindication and its drug therapy differs. So a cirrhotic with a lesion showing rim enhancement and peripheral washout is the classic case where reading it as HCC is exactly wrong, and the rim pattern is the tip-off.

5:36The tumor-in-vein category is almost always HCC that has invaded the portal vein, identified by an arterially enhancing thrombus, and that enhancement is what separates tumor thrombus from bland clot, because bland clot has no blood supply and doesn't enhance while tumor thrombus carries its own arterial feeders and lights up. The practical consequence is that tumor thrombus is metastatic disease for transplant purposes, it takes the patient out of the transplant criteria, and you don't anticoagulate it. Biopsy carries a small but real risk of seeding the needle tract, which is why you avoid it when the imaging is already diagnostic, but you do biopsy the malignant-but-not-HCC lesions, cases where you need tissue for molecular testing before systemic therapy, and genuinely ambiguous lesions.

6:21Once you have a diagnosis, the treatment decision runs through the BCLC staging system, which combines tumor burden, liver function, and performance status, and each stage points to the treatment the underlying physiology will actually permit. The earliest stages are the curative ones, a single small lesion or up to three lesions each at most three centimeters with preserved function and good performance status, and here resection, ablation, and transplant are all possible and the choice depends on the substrate.

6:50Resection is straightforward in a non-cirrhotic liver, including the hepatitis-B cancers that arise without cirrhosis and are often resectable even when large, but resection in a cirrhotic is narrow because taking out liver mass can tip a portal-hypertensive patient into failure, so you require a normal bilirubin and a hepatic venous pressure gradient under ten, since a gradient over ten signals the clinically significant portal hypertension that sets up postoperative decompensation. When resection is off the table, thermal ablation with radiofrequency or microwave substitutes for tumors typically under three centimeters, and the size limit matters because ablation only kills reliably within a defined zone, so larger tumors risk an incompletely treated margin, and proximity to a vessel causes a heat-sink effect where flowing blood carries the thermal energy away and leaves viable tumor at the edge, so a small lesion in a bad surgical location or in a patient with marginal reserve is an ablation case rather than a resection case.

7:44Transplant is the treatment of choice for early HCC in a cirrhotic, because it removes both the tumor and the cirrhotic liver that keeps generating new tumors, and eligibility is defined by the Milan criteria: a single lesion at most five centimeters, or up to three lesions each at most three centimeters, with no spread outside the liver and no macrovascular invasion, and that last exclusion follows directly from the tumor-in-vein teaching, because tumor thrombus is metastatic disease. Within Milan, long-term survival after transplant is strong.

8:14The UCSF criteria widen the door a bit, to a single lesion up to six and a half centimeters, or two to three lesions with the largest up to four and a half centimeters and a total tumor diameter up to eight centimeters, and a patient beyond Milan can be down-staged into Milan with locoregional therapy followed by three to six months of imaging stability before listing, and that waiting period is deliberate biological selection, because a tumor that responds and stays quiet has biology unlikely to recur after transplant.

8:40The intermediate stage is multifocal disease beyond Milan without vascular invasion or spread, and it's treated with locoregional therapy, either chemoembolization or selective radioembolization delivered through the artery, both of which exploit the fact that HCC feeds preferentially off the hepatic artery, so catheter-directed delivery hits the tumor hard while sparing the portal-fed normal liver, repeated as needed. Response is graded by looking at how much tumor still enhances rather than by total size, because successful therapy produces necrosis without immediately shrinking the mass, so viable enhancing tumor is the real readout.

9:17The advanced stage is macrovascular invasion or metastases with preserved liver function, and it's systemic, with first-line therapy being atezolizumab plus bevacizumab, a PD-L1 inhibitor paired with an anti-VEGF antibody that outperformed the old tyrosine kinase inhibitor on survival. The catch is bevacizumab's bleeding risk: because it blocks the VEGF-driven angiogenesis that also maintains vascular healing, it raises the risk of variceal bleeding in a cirrhotic, so the rule is a mandatory upper endoscopy within six months before starting it, with any high-risk varices treated first, and that endoscopy requirement is the direct practical consequence of the drug's mechanism and a high-yield testing pattern. When bevacizumab is contraindicated by uncontrolled varices, recent bleeding, or recent surgery, the alternative is the checkpoint combination of durvalumab plus tremelimumab, and lenvatinib or sorafenib remain options when checkpoint inhibitors themselves are contraindicated, for instance by autoimmune disease, with cabozantinib in the second line and ramucirumab for patients with an AFP over four hundred.

10:23The terminal stage, poor performance status or Child-Pugh C without a transplant option, is best supportive care, because the liver function and performance status rule out any tumor-directed treatment, and recognizing it protects the patient from futile therapy, so the patient with confirmed HCC, severe weight loss, poor performance status, and Child-Pugh C with no transplant option is a hospice referral, not a chemoembolization.

10:46So the HCC arc holds together. Surveillance goes to every Child-Pugh A or B cirrhotic and to high-risk hepatitis B, by ultrasound and AFP every six months timed to doubling time and curative-size cutoffs. LI-RADS turns a surveillance finding into a diagnosis, with the definite category diagnostic by imaging alone and the two special categories forcing biopsy or excluding transplant. And the staging system turns the diagnosis into a treatment, from resection and ablation through transplant within Milan to locoregional therapy for intermediate disease and atezolizumab plus bevacizumab for advanced disease. Running under all of it is the single override: a new lesion in a cirrhotic or hepatitis-B liver is HCC until proven otherwise, and the whole algorithm exists to prove or disprove it efficiently.

11:36The next episode turns to the other primary liver cancer, cholangiocarcinoma, where the organizing question is how the anatomy drives management. Intrahepatic disease behaves like a parenchymal tumor with a desmoplastic imaging look that distinguishes it from HCC, perihilar disease at the duct confluence, the Klatskin tumors, is the one place transplant enters the picture through a neoadjuvant chemoradiation protocol, and distal disease behaves more like a pancreatic head tumor, with systemic therapy built on gemcitabine, cisplatin, and durvalumab plus targeted agents for the molecular subsets.

12:09For the full chapter, the practice vignettes, and the topic-tagged question bank, head to board pearls dot com. You'll find the rest of the series on Apple Podcasts, Spotify, or wherever you listen to podcasts. That brings us to the end of episode two of four of chapter twenty-four, and I'll see you in the next one.