During a Targeted Oncology case-based roundtable event, Richard S. Finn, MD, discussed the frontline treatment options for a patient with hepatocellular carcinoma.
SNEDDEN: I think the history of Crohn disease is certainly something to consider. Yes, the LI-RADS score is 5, but [I would] want to make sure it’s not colorectal cancer.
So, that would be one of the main reasons I would consider the biopsy.
FINN: That’s a good thought, [because] chronic inflammatory disease of the colon is obviously associated with an increased risk of colon cancer.
JANA: Biopsy tends to have a high false-negative [result] and can cause considerable confusion in this setting when you have strong imaging confirmation and elevated AFP. That’s why I [would choose] no biopsy.
FINN: So does this patient need any further imaging? Or are you satisfied with [the CT scan of the] chest, abdomen, and pelvis?
GHABACH: [Chest, abdomen, and pelvis imaging is] sufficient unless the patient had any clinical signs that would suggest bone disease.
FINN: I would tend to agree with that; [I would feel] similarly [if the patient had signs that suggested] neurologic disease.
It’s very uncommon to see HCC metastasize to the central nervous system, but as patients are living longer, we do sometimes see that.
Do any of you send [for] Foundation Medicine testing on these patients? [Do you use] some other next-generation sequencing?
GHABACH: Not up front, no.
FINN: Yes. Other than, I think, NTRK [translocation], which is a tumor-agnostic indication that occurs very rarely, there’s no biomarker-driven approach for HCC based on the tumor. [However], the serum AFP level is linked to the approval of ramucirumab [Cyramza].1
FINN: [The disease classification is] stage IV [based on the] T and M components. [According to] the Barcelona system, which is a very common staging system for HCC, this patient would have stage C disease, because they’re well compensated and have extrahepatic spread.
FINN: As mentioned in the National Comprehensive Cancer Network [NCCN] guidelines, the combination of atezolizumab [Tecentriq] plus bevacizumab [Avastin] is [the preferred regimen], supported by level 1 evidence.2 We published that in the New England Journal of Medicine [in 2020].3 This combination showed a marked improvement in survival vs frontline sorafenib [Nexavar].
Sorafenib [had been] the standard of care since 2008.2 It had been shown to improve survival vs placebo by inducing stable disease [but it did] not have a very high response rate.4-6 [Since then,] no [other regimens] improved on that until [the advent of] the atezolizumab plus bevacizumab combination.
[Sorafenib] was the only approved drug until 2018, when lenvatinib [Lenvima], another recommended regimen,2 was approved.7 Yet, lenvatinib was only approved [because it was] noninferior, not superior, to sorafenib in terms of overall survival [OS], though it did have a higher response rate and higher progression-free survival [PFS].8
Listed as “useful in certain circumstances,” single-agent nivolumab [Opdivo]2 currently has no FDA approval. It did have accelerated approval for second-line use,9 but the phase 3 CheckMate 459 study [NCT02576509] of frontline nivolumab vs sorafenib was negative,10 so nivolumab lost its approval.11 I think this NCCN guideline probably came out before that.
Another regimen listed as useful in certain circumstances is FOLFOX, the combination of folinic acid [leucovorin], fluorouracil, and oxaliplatin [Eloxatin]. It is not used in the United States. It has a low level of evidence, [2B].2 It is more often used in China where they’ve studied it a little more, but [it] never was shown to be superior to sorafenib in terms of OS.
SNEDDEN: [What would make me hesitant is the patient’s history of] Crohn disease.
FINN: Yes. This patient has Crohn disease, and that’s a relative contraindication to using checkpoint inhibitors.12 [The patient] is on infliximab as well; you’d have to stop that, presumably. So, [given that] this patient obviously has bad Crohn disease, I would be hesitant to try [the atezolizumab plus bevacizumab combination]. However, I understand the urge to do it, because it’s the most active thing we have; objective response rates were 30%.3
The other thing that should give us pause, though, is that the patient also has a history of varices and was banded 2 months ago. Not that that is an absolute contraindication, but in the IMbrave150 study [NCT03434379], all patients [were required] to have had an upper endoscopy within 6 months of coming on study, and they had to have varices treated. [Approximately one-fourth] of the patients did have varices, and approximately 10% to 20% had them treated before coming on study.3
So there are 2 contraindications [to using atezolizumab plus bevacizumab]. Lenvatinib [Lenvima] or sorafenib would be very reasonable, and I think it is perceived as being more active because of the high response rate and improvements in PFS.8 It does have [an adverse] effect profile that is different from that of sorafenib.5,8
FINN: I think, for most of us, patients with advanced HCC are not usually getting tyrosine kinase inhibitor [TKI] monotherapy because presumably there is no contraindication [for atezolizumab plus bevacizumab]. I think [we would] estimate between 10% and 20% of patients have a reason why they can’t get atezolizumab or bevacizumab.
I think [contraindications for] atezolizumab are probably relatively uncommon; most patients don’t have autoimmune diseases. [Consideration of bevacizumab does raise] a concern about bleeding. However, even though bevacizumab is associated with bleeding, the incidence of very high-grade bleeding events was very uncommon—with a single-digit percentage of grade 4/5 events.3
FINN: [I think most of us consider] some efficacy measurement, either response rate or survival, to be the most important factor; [others primarily] consider safety and tolerability. Obviously, it’s a complicated question because all of those [factors need to be considered].
JANA: Typically, I wouldn’t take age or sex [into consideration].
FINN: Right, I think very few of us do that. Performance status [is worth considering], and age can be related to performance status, which goes together with symptom burden. What about etiology of cirrhosis? Is your selection of a treatment regimen influenced by whether the patient has hepatitis B, hepatitis C, or nonalcoholic steatohepatitis?
GHABACH: I don’t think that [cirrhosis etiology] is relevant.
JANA: [I agree that] it’s not relevant.
FINN: I think we do tend to approach the disease like that. There was a paper this year that suggested that [immunotherapy] is not as active in nonviral HCC [as in viral HCC]; however, that has not been proven, per se, in prospective studies.
Are there any toxicities in this patient, or in patients with HCC in general, that concern you?
GREENBERG: Yes. In this patient, specifically, I’d be concerned about bleeding. I think it’s safe to use atezolizumab plus bevacizumab, but this patient had varices that were banded, so that would just be something that [would have to be] monitored closely.
FINN: I would tend to agree with that if they didn’t have the autoimmune issue. Does it concern you if a patient has untreated hepatitis C or hepatitis B? Or HIV, for that matter?
GHABACH: If it’s autoimmune hepatitis, I might think twice about immunotherapy.
FINN: Sure, that’s very fair. There are 2 issues, I think, for patients with autoimmune disease. One issue is [that] if they’re on immunosuppression, it’s counterintuitive to use checkpoint inhibitors; it’s like pushing the brakes and the gas at the same time.
The other issue is the challenge of flare. If a patient has psoriasis that is not too disruptive, I give them checkpoint inhibitors without too much concern and if they flare up, I treat them. But something such as autoimmune hepatitis or bad inflammatory bowel disease would have to be [treated more cautiously]. In this case [the patient has had a] prior liver transplant, right? I think that is probably [another] firm contraindication to using checkpoint inhibitors.
How do the Child-Pugh score and liver function [influence your choice of frontline therapy]?
SNEDDEN: Well, if the Child-Pugh score were B, then we [would] use neither the atezolizumab plus bevacizumab combination nor lenvatinib, but rather sorafenib, so it does have an impact on treatment options.
FINN: That brings up an interesting point. The NCCN guidelines initially [listed] even nivolumab [as an option] for [disease with a] Child-Pugh [score of] B, based on a single-arm exploratory study that showed that it was relatively safe. Sorafenib has certainly been around the longest, and there are some safety data [regarding its use in disease with a] Child-Pugh [score of] B.
Survival outcomes are always inferior for patients with a Child-Pugh [score of] B. But we don’t have randomized data that [show whether] that is just the effect of the liver disease or [whether] the drug does improve outcomes for those patients. With the newer drugs such as lenvatinib or, now, atezolizumab plus bevacizumab, I think we need to consider the patients with a Child- Pugh [score of] B as a heterogenous group.
There are some patients who are very functional [in terms of] performance status and symptoms. They have minimal symptoms, they’re up and about, they can come to clinic. For those patients, I don’t have pause about possibly treating them with atezolizumab plus bevacizumab, as long as they don’t have big varices, because I don’t think we add any toxicity with this combination in [patients with a score of] B. Decompensation events are not high risk and, short of autoimmune liver disease, [these patients] should be able to tolerate some [adverse] effects.
The other thing to keep in mind is that some of these patients have Child-Pugh B physiology because they have a very large tumor burden or because they have vascular invasion. If you can induce a response, then their liver disease should get better, and that’s when you need to look at the scans. A patient who has a liver that is coarse on imaging but has a 12- or 15-cm tumor might have some ascites, and maybe their bilirubin would be a little high or their albumin would be low. These are patients who, if you can induce a response, will get better. In that case, atezolizumab plus bevacizumab probably gives them the best chance of a response.
[In contrast], there are some patients who have very shrunken and cirrhotic livers, and the tumor is not that big— perhaps only 5 cm—but the liver is so small [that the tumor] looks giant. In those patients, even if the tumor disappears, they’re still going to be very sick and will probably die around the same time [regardless of treatment], just because they’re decompensated. I think [it is important to remember] that all the phase 3 studies done with any HCC drug are always done in patients who have a Child-Pugh [score of] A.
Does patient preference, or your preference, for an oral regimen vs an intravenous [IV] regimen influence your choice of treatment?
HSU: No, I think only when it comes to [the fact that] some of the drugs may be more expensive.
FINN: So there may be [some influence] on the patient side?
FINN: What about COVID-19? Has that affected your use or management of oral vs IV regimens?
GHABACH: Initially, maybe, a year and a half ago, but not anymore.
FINN: Yes, not anymore. [Keep in mind that] the IV regimen, atezolizumab plus bevacizumab, [requires a visit] once every 3 weeks. But also, because the oral drugs have [adverse] effects, I typically see patients on the oral drugs at least once every 10 to 14 days for the first month or so until we know that the patient is stable. The last thing you want is someone to have diarrhea for 2 weeks and not call or not come in to be seen.
GREENBERG: You mentioned the flare phenomenon. Do you ever, in a patient with an autoimmune condition such as psoriasis, use flare as a surrogate for response?
FINN: No. I don’t think flare or the development of a new autoimmune disease has been correlated with activity of these drugs, or at least with anticancer activity. I think that [is true] across melanoma, lung cancer, all the diseases [in which it has been studied]. Those [adverse] effects seem a little idiosyncratic, and some, such as rash and hypothyroidism, are very common.
FINN: In my opinion, the most common challenges with lenvatinib are anorexia, weight loss, and fatigue. Hypertension may occur, but that generally can be managed. This patient had a partial response, and the response rate with lenvatinib [has been shown to be approximately] 19% by RECIST criteria.
DASGUPTA: It’s not a very common disease in my practice, so my range of experience is necessarily limited, but I have found better outcomes with lenvatinib. I’ve not had any problems [with tolerability]. I start at the 8-mg dose, which I’m leery about [increasing] the dose, and I’ve had great responses, but I guess I’ve been lucky. Anorexia [can be a problem so] I use dronabinol, and I’ve not had major problems with weight loss. The hypertension is easily treatable, but sorafenib has been terrible in my hands.
FINN: A friend of mine once said that TKIs [such as lenvatinib and sorafenib]5 are different from chemotherapy, [in that] if you give someone a taxane, you know what’s going to happen. Everybody gets bone marrow suppression, eventually they’ll get neuropathy, lose their hair, have nausea.
[In contrast,] with the TKIs some patients have no problems and some patients do. I’ve seen a patient get sorafenib and [then require] a wheelchair in a week because they have horrible blisters on their feet. It is very variable.
GUO: My experience [with lenvatinib] is like what you described. Some patients do fine; they’re just fatigued. Normally I would do the same thing—start on 8 mg and try to increase to 12 mg if the patient can tolerate it.
But I’ve also had other patients who are extremely sensitive. I, too, had a patient who developed sores within a week. [Some patients develop] oral mucositis, diarrhea, hand-foot syndrome, hypertension, fatigue, leg swelling, everything. So, yes, it’s variable. I normally start with a lower dosage and see how things go. If the patient can’t tolerate 8 mg, then of course we can’t [increase the dosage], but if they do fine, then I’ll increase the dosage gradually.
FINN: Is there any trick [with managing] fatigue? I struggle with that. [I do discuss] sleep hygiene and check patients’ thyroid-stimulating hormone level. A lot of these TKIs can cause hypothyroidism.5
GREENBERG: I often will use low-dose dextroamphetamine while monitoring blood pressure, obviously. You can have a little problem because amphetamines can be appetite suppressants, so [you have to find a good] balance, but I found that 2 mg of dextroamphetamine can give these patients some life. I’ve had more than 1 patient tell me that they have gone dancing for the first time in a long time because it made them feel a little better. [It improves their] quality of life.
FINN: I think most of us are fairly comfortable [with managing the adverse events associated with lenvatinib], using dose reductions and monitoring for [adverse] events.
[In addition to hypothyroidism], proteinuria [is something] you should watch for on a regular basis. Dysphonia is something that can happen as well, and I don’t think there’s any quick trick. If you do need dose reductions based on weight, [the recommended progression is] 8 mg given daily, then 4 mg daily, and then 4 mg every other day; if you start at the higher dose you have a little more leeway beginning with 12 mg given daily and then following the same progression.
After [our discussion], would [anyone here] change their recommended frontline therapy for this patient?
REDDY: Initially, I [would have chosen] the combination of atezolizumab plus bevacizumab; I’m not convinced [enough] to change my opinion.
FINN: With regard to upfront atezolizumab plus bevacizumab, I think the Crohn disease, controlled with infliximab, would make me [hesitant] to [use that combination]. I think you’d be better off using lenvatinib or sorafenib. Given, diarrhea can happen with these drugs, but it’s by a different mechanism. [Similarly, considering] the gastrointestinal bleeding, I think [both lenvatinib and sorafenib are preferrable to atezolizumab plus bevacizumab].
1. FDA approves ramucirumab for hepatocellular carcinoma. FDA. May 10, 2019. Accessed February 2, 2022. https://bit.ly/34yuCUw
2. NCCN. Clinical Practice Guidelines in Oncology. Hepatobiliary cancers, version 5.2021. Accessed January 31, 2022. https://bit.ly/3BooXg0
3. Finn RS, Qin S, Ikeda M, et al; IMbrave150 Investigators. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med. 2020;382(20):1894-1905. doi:10.1056/NEJMoa1915745
4. Lang L. FDA approves sorafenib for patients with inoperable liver cancer. Gastroenterology. 2008;134(2):379. doi:10.1053/j.gastro.2007.12.037
5. Llovet JM, Ricci S, Mazzaferro V, et al; SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359(4):378-390. doi:10.1056/NEJMoa0708857
6. FDA approves lenvatinib for unresectable hepatocellular carcinoma. FDA. August 16, 2018. Accessed February 2nd, 2022. https://bit.ly/3oJcVI
7. Kudo M, Finn RS, Qin S, et al. Lenvatinib vs sorafenib in first-line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non-inferiority trial. Lancet. 2018;391(10126):1163-1173. doi:10.1016/S0140-6736(18)30207-1
8. FDA grants accelerated approval to nivolumab for HCC previously treated with sorafenib. FDA. September 22, 2017. Updated September 25, 2017. Accessed February 2, 2022. https://bit.ly/3GJIAQC
9. Yau T, Park JW, Finn RS, et al. Nivolumab vs sorafenib in advanced hepatocellular carcinoma (CheckMate 459): a randomised, multicentre, open-label, phase 3 trial. Lancet Oncol. 2022;23(1):77-90. doi:10.1016/S1470-2045(21)00604-5
10. Bristol Myers Squibb statement on Opdivo (nivolumab) monotherapy post-sorafenib hepatocellular carcinoma U.S. indication. News release. Bristol Myers Squibb; July 23, 2021. Accessed February 2, 2022. https://bit.ly/3oHdItM