Unresectable Liver Cancer: The Standard Regimen's Own Bleeding Risk
The standard first-line regimen for this cancer includes a drug whose own defining risk is the exact complication this patient's cirrhosis has already put on the table.
K.T., a 66-year-old man, has run a small vineyard for twenty-five years and jokes that the cirrhosis that eventually caught up with him "wasn't from the tasting room, it was from thirty years before I owned the place" — hepatitis B acquired decades ago, well-controlled on antiviral therapy for the past ten years but not before it left him with Child-Pugh A cirrhosis. Surveillance imaging found a 6cm hepatocellular carcinoma with portal vein branch involvement, unresectable by the liver surgery team's assessment. Screening endoscopy, done as standard workup before starting any antiangiogenic-containing regimen, found grade 2 esophageal varices with red wale signs — a recognized high-risk stigma for future bleeding, not yet treated with banding.
Atezolizumab plus bevacizumab, established by IMbrave150 as standard first-line therapy for unresectable hepatocellular carcinoma with a real overall survival advantage over prior standards, carries a well-documented bleeding risk from bevacizumab's antiangiogenic mechanism — exactly why variceal screening is required before starting it, and exactly why active, high-risk varices are treated as a reason to delay or avoid it until banded. The STRIDE regimen — durvalumab plus a single priming dose of tremelimumab — was validated in HIMALAYA with comparable overall survival benefit and, critically, contains no antiangiogenic agent, sidestepping the variceal bleeding risk entirely. K.T.'s red-wale-sign varices sit squarely in the population bevacizumab's own screening protocol exists to flag, making the choice between these two, otherwise comparably effective, first-line regimens turn substantially on a risk one of them simply doesn't carry. His Child-Pugh A status and normal blood pressure are what make this a choice rather than an exclusion — he is well enough for either regimen, so the red wale signs are not deciding whether he can be treated, only which of two comparably effective regimens makes him carry a risk he does not have to carry.
Multidisciplinary liver tumor board
Atezolizumab plus bevacizumab remains the regimen with the most mature outcomes data in unresectable HCC — IMbrave150 is what most guidelines still lead with. I'd band his varices first, which typically takes a session or two, then start the standard regimen once that risk is addressed. It's a short delay for the option with the longest track record.
Banding is the right first step regardless of which regimen we choose, but I want to be clear about what it does and doesn't do — red-wale-sign varices carry meaningfully elevated bleeding risk, and banding reduces that risk without eliminating it. If we then add a drug whose defining mechanism is antiangiogenic, we're layering a known bleeding-risk agent onto a variceal bed that's improved, not resolved. STRIDE's HIMALAYA data (Abou-Alfa et al., NEJM Evidence 2022) showed comparable survival without that specific risk at all.
This isn't a claim that bevacizumab is unsafe in general — it's a claim that his specific variceal findings make the alternative's risk profile a better fit.
I'd take that a step further: band the varices as standard endoscopic care either way, but choose STRIDE regardless of banding outcome. The whole reason we screen before bevacizumab is that the risk is real and persistent in exactly this population. Once a comparably effective alternative exists that doesn't carry that mechanism at all, I don't think residual post-banding risk is something we need to accept just to use the more familiar regimen.
Agreed: esophageal variceal banding as standard endoscopic care, followed by STRIDE (durvalumab plus a single tremelimumab priming dose) as first-line systemic therapy, specifically to avoid layering an antiangiogenic bleeding risk onto a variceal bed that banding improves but does not fully normalize.
The medical oncologist's preference for the more extensively validated atezolizumab/bevacizumab regimen was heard and was a reasonable position given its longer track record, but was not adopted — the group judged the bleeding-risk argument decisive once a comparably effective alternative without that specific risk was on the table.