Richard D. Kim, MD

With atezolizumab plus bevacizumab being the standard first-line treatment, there are other phase 3 studies underway in the first line. Most of the studies are comparing to the control arm of a TKI [tyrosine kinase inhibitor], whether it’s lenvatinib or sorafenib. There are no trials right now against atezolizumab plus bevacizumab. There are combinations of a TKI plus immunotherapy [IO]. There’s the COSMIC-312 study, where we are combining cabozantinib plus atezolizumab, a checkpoint inhibitor. There’s the LEAP-002 study, where we’re combining lenvatinib with pembrolizumab, another checkpoint inhibitor. The rationale for combining those 2 classes is that the TKI can change the microenvironment. All those TKIs have anti-VEGF properties as well. You could change the microenvironment so that it could be synergistic with the checkpoint inhibitors. Now, a lot of early data show a very promising outcome with that combination. Recently, regarding the COSMIC-313 data, there was a press release telling us that the combination of cabozantinib plus nivolumab showed an improvement in PFS [progression-free survival] but no improvement in overall survival. This was just a press release. We need to get more data in terms of what that number really means. There’s another trial, the LEAP-002 study, which is a combination of lenvatinib plus pembrolizumab in the phase 3 setting. That’s a trial that’s already finished accrual. We’ll wait to hear what the results of that trial show.

Another combination that’s been studied in the first line is a combination of 2 immunotherapy agents, a checkpoint inhibitor with a CTLA-4 inhibitor. There’s the CheckMate 9DW study combining nivolumab plus ipilimumab, and there’s the HIMALAYA study, which is another trial combining 2 immunotherapy agents. The HIMALAYA study is combining durvalumab plus tremelimumab. That trial has already finished accrual. We’re eagerly awaiting the results of those studies. When those trials read out, then you may have a couple of combination therapies available in the first line: not only with atezolizumab plus bevacizumab, but with an option of using a TKI plus IO, or using an IO plus IO. We’re eagerly awaiting the results of those phase 3 studies.

For the patient’s sake, you may have more options available if those trials are positive. If those trials are positive, as clinicians, we will have to decide which we want to use. I think the devil will be in the details in terms of which regimen to use. I think efficacy does matter, in terms of how effective they are. I also think the safety profile matters a lot because unfortunately, we’re not curing any of these patients. We are mostly prolonging their lives. But their quality of life is very important. The safety profile will be important as well. We’re eagerly anticipating the results of all those phase 3 studies underway. Hopefully, there are some other studies that will be positive, which will lead to more indications in this setting.

In the treatment of hepatocellular carcinoma, we’ve come a very long way. In the past, as medical oncologists, we didn’t have much to offer except sorafenib. It was mostly a disease treated by interventional radiologists and hepatologists. But with the evolvement of all these new drugs—TKIs and immunotherapy, checkpoint inhibitors now being approved or being studied—as medical oncologists, we should get more heavily involved earlier in treating these patients. One of the things a community oncologist should do is, as I mentioned, the multidisciplinary approach, which is very important. It’s a must. We have to get everybody at the table, to go with the same plan. But with all these drugs available, we should treat them a lot earlier. In the past, there were a lot of cases where local therapy failed. Despite that, the patients were getting more local therapies, for example. But now, if they fail local therapy or if they’re refractory to local therapy, we should treat those patients more aggressively with all the combinations available. Based on the IMbrave150 study, the median survival is hovering close to 20 months. This is the longest overall survival we’ve seen with advanced disease. As the treatment paradigm evolves in hepatocellular carcinoma, it is very important for us, medical oncologists, to get involved earlier in conjunction with interventional radiologists and offer patients treatment that could potentially give them a good quality of life and prolong their overall survival.

Videos

Considerations for ongoing clinical trials that may shape the future treatment landscape of unresectable hepatocellular carcinoma. 

Looking Toward the Future Management of Unresectable HCC 

The pivotal study that got atezolizumab and bevacizumab approved is IMbrave150. This was a randomized phase 3 study comparing the control arm, which is sorafenib, to atezolizumab plus bevacizumab, based on the phase 1 trial that had very promising results. This trial had a dual primary end point of PFS [progression-free survival] and overall survival [OS]. The PFS data were presented last year; and this year at ASCO GI [American Society of Clinical Oncology Gastrointestinal Cancers Symposium], the follow-up on overall survival was presented. Basically what the trial showed was that the combination of atezolizumab and bevacizumab was superior to sorafenib in terms of the response rate, which was hovering around 30%. There was improvement in PFS, which was first reported last year, by a little over 2 months. This year at ASCO GI, there were overall survival data presented showing 19 months with atezolizumab plus bevacizumab compared to about 13 to 14 months of OS with sorafenib.

Nineteen months in advanced hepatocellular carcinoma is the longest overall survival we’ve seen in any of the studies. Clearly, these data show that the combination of atezolizumab plus bevacizumab is superior to sorafenib, and not only in terms of response rate and PFS. There was definitely a trend toward overall survival even though the 

value was not significant. In terms of adverse events in the study, there was not much difference between the 2 arms. We saw a little more hypertension in the bevacizumab arm because it is a more potent VEGF inhibitor, and we saw more hand-foot skin reactions with sorafenib, which is expected. But the arm of atezolizumab and bevacizumab was very well tolerated. There were few grade 3 or 4 events. Most of the adverse events we saw were grade 1 and 2, and the highest grade 3 and 4 adverse event seen in the trial was hypertension, which we know how to manage. Overall, the trial concluded that atezolizumab and bevacizumab were very well tolerated. It showed superior efficacy, response rate, PFS, and a trend toward overall survival.

Based on that pivotal study, the combination was approved by the FDA and is in the NCCN [National Comprehensive Cancer Network] guidelines as well. In our practice, that was a practice-changing trial. A lot of patients with advanced hepatocellular carcinoma with a Child-Pugh A score, as that study was done in the Child-Pugh A setting, will get this combination. One of the caveats to this is that in my practice, I use endoscopy just like in the trial to avoid any unwanted bleeding. We perform an endoscopy to make sure there’s no variceal bleeding. In the era of COVID-19, if you’re having a tough time performing endoscopy, what happens in my practice is that I will start with atezolizumab first. If the endoscopy’s done later, and if there’s no evidence of variceal bleeding, then I add Avastin [bevacizumab]. That’s how I do it in the era of COVID-19.

Richard Kim, MD, shares insight on positive results from IMbrave150, which tested the frontline combination of atezolizumab and bevacizumab in patients with unresectable HCC. 

IMbrave150: Frontline Atezolizumab + Bevacizumab in Unresectable HCC

In terms of molecular profiling in HCC [hepatocellular carcinoma], in all other GI [gastrointestinal] tumors, molecular profiling is a must for all patients. In HCC, it’s a little less clear because we don’t know exactly what kind of mutations or aberrations we’re looking for at this time. At least in some practices, we are sending out tissue for profiling purposes, but I don’t think it’s used as much compared to other diseases because we don’t know exactly what to do with the profiling. In some cases in HCC, we don’t have tissue to profile. We could radiographically diagnose patients with HCC using their backgrounds and by doing an MRI or a CT scan that is consistent with HCC with arterial enhancement and portal venous washout. For those patients, I don’t have tissue to profile, so those cases would be an exception. Generally speaking, even though we should test for molecular profiling purposes to see what comes out of it and see if there are actionable aberrations, at this time I don’t think that’s being done commonly compared to the other GI tumors.

One of the biomarkers that would sometimes be used in other diseases that can predict or enrich a patient population that can potentially benefit from immunotherapy is PD-L1 status. Sometimes, if PD-L1–positive, those patients tend to benefit more from immunotherapy, for example, in GE [gastroesophageal] junction tumors. But in HCC, there’s no correlation between expression of PD-L1 and response to atezolizumab. Therefore, in our practice, we don’t do PD-L1 testing as standard testing for patients with advanced hepatocellular carcinoma.

As the treatment of hepatocellular carcinoma evolves with atezolizumab and bevacizumab in the first line, the question is, what do you do in the second line? All the second- and third-line trials so far are in patients who failed sorafenib. None of the trials were done in patients who failed immunotherapy. There are a lack of data in terms of what to do in the second line if atezolizumab and bevacizumab fail. However, most of us would prefer to use a TKI [tyrosine kinase inhibitor] in the second-line setting without much data available. This is an area where a lot of studies are being done. The question that comes up is, if you fail atezolizumab plus bevacizumab, which TKI should you use? I mentioned 4 or 5 TKIs. There’s lenvatinib [Lenvima], sorafenib, regorafenib, and cabozantinib. They’re all available. Most of them were studied in patients who failed sorafenib and not patients who failed immunotherapy, so it is unclear what to do in the second line. Without much data out there, we do tend to use one of the TKIs.

In my practice, I would probably use Lenvima, or you could use cabozantinib as well. There are some data for cabozantinib that with its mechanism of action, it blocks the AXL pathway and MET pathway, and those are some of the resistance mechanisms to VEGF inhibition. If you’re getting VEGF-targeted therapy first line and become resistant, then we may give you a TKI that blocks that AXL pathway or MET pathway and may resensitize, but it’s unclear. But in the second-line setting, a TKI will be used. In the third line, you would use a TKI that you didn’t use in the second line.

A brief synopsis on optimal sequencing strategies for patients who have been diagnosed with unresectable hepatocellular carcinoma. 

Advice on Molecular Testing and Sequencing in Unresectable HCC

In terms of the management of first-line treatment for advanced hepatocellular carcinoma [HCC], in the past, sorafenib was the only drug available. Sorafenib is a TKI [tyrosine kinase inhibitor] that blocks many pathways and has been around for over 10 years. A couple of years ago, lenvatinib came onboard. That was based on a study showing that lenvatinib was not inferior to sorafenib. However, it did have a better response rate, longer PFS [progression-free survival], and a different adverse event profile. When those data came out, many of us moved on from sorafenib to lenvatinib. Recently, in the IMbrave150 study, the combination of atezolizumab, which is an immune checkpoint inhibitor, plus bevacizumab, which is an anti-VEGF inhibitor, was shown to be superior compared to TKI sorafenib. Based on that, our standard of first-line treatment has changed to using atezolizumab plus bevacizumab in patients who are eligible to get those agents.

There are other TKIs available in the second or third line of therapy. For example, in the second-line setting, other TKIs called cabozantinib and regorafenib are approved. In the past, immune checkpoint inhibitors, such as pembrolizumab and nivolumab, were approved as single-agent therapy in the second-line setting as well. With all the data emerging, the main question remains how to sequence them. In terms of the mechanism of action, there is a rationale for combining atezolizumab and bevacizumab because most cases of HCC are hypervascular. They overexpress VEGF, and anti-VEGF therapy tends to mitigate the VEGF-mediated immunosuppression. Therefore, in combining that with a checkpoint inhibitor, there’s some synergistic activity with those 2 agents. That was the rationale for doing the IMbrave150 study, which had a positive result showing benefit over sorafenib.

In the first-line setting, based on the Imbrave150 study, the standard should be atezolizumab plus bevacizumab. There are some patients who are not eligible for bevacizumab or atezolizumab: for example, if the patient has a history of bleeding from varices. In the trial, everybody was required to get an EGD [esophagogastroduodenoscopy], and yet the varices prohibited it. Therefore, if at high risk for bleeding from varices, a patient may not be a good candidate for bevacizumab. The other potential patients who are not candidates for that combination have autoimmune disease: if you have Crohn disease or active ulcerative colitis, for example. Giving a checkpoint inhibitor to those patients may not be a good idea. There also is a lack of data in patients with Child-Pugh B disease. If you have a patient with a Child-Pugh score of B8 or B9, can you safely give the combination of atezolizumab and bevacizumab? That’s unclear. I’m sure there will be studies done, but at this moment, there are no safety data on that. That’s another portion of patients who may not be eligible for the combination of atezolizumab plus bevacizumab. If there’s any contraindication to the combination, at least in our practice, we’ll go with a TKI.

Regarding a first-line TKI, there are 2 choices available, sorafenib and lenvatinib. They are very similar because they are both TKIs. Efficacy-wise, even though the OS [overall survival] showed lenvatinib was noninferior, due to the high response rate, the longer PFS, and a slightly better adverse effect profile—meaning that for lenvatinib, you’re dealing with more hypertension and high blood pressure compared to sorafenib, where you deal more with rash and hand-foot skin rashes, which can be cumbersome—many of us have adapted to use more lenvatinib in the first line than sorafenib. In patients who are not a candidate for atezolizumab plus bevacizumab, based on some of the factors I mentioned, a single agent TKI, such as lenvatinib, may be reasonable.

One of the difficulties we see when we treat patients with hepatocellular carcinoma is that unfortunately, many of these patients don’t have a Child-Pugh A score. Most of the studies were conducted with Child-Pugh A patients. If you see a patient with Child-Pugh B score, whether it’s B7, B8, or B9, what do you do in that setting? Atezolizumab plus bevacizumab in the IMbrave150 study was tested in Child-Pugh A patients. Are there data in Child-Pugh B patients? Not that I’m aware of. Therefore, if you want to use that combination in a patient with Child-Pugh B7, there will be a risk to benefit ratio. You’d have to discuss that with the patient and decide what to do.

However, other TKIs such as sorafenib and lenvatinib, or nivolumab as a single agent, have been studied in Child-Pugh B patients. Clearly, anybody Child-Pugh B won’t be similar to patients with Child-Pugh A because they’re going to perform worse no matter what, due to their underlying liver condition. But the safety data show that you can safely use sorafenib, lenvatinib, or nivolumab—the latter based on the CheckMate 040 study—in patients with Child-Pugh score B. Therefore, if you have patients with Child-Pugh score B7, B8 or B9, there are data to support giving TKIs in that setting such as lenvatinib or sorafenib; or possibly using a checkpoint inhibitor as a single agent, such as nivolumab, based on the CheckMate 040 study. There are some data with that. Moving forward, there will be more data collected using the combination of atezolizumab and bevacizumab in patients with Child-Pugh B. But to date, I don’t have any data to support that. However, in those subsets of patient who are not candidates for atezolizumab plus bevacizumab because of Child-Pugh B status, giving a TKI such as sorafenib or lenvatinib, or a single-agent checkpoint inhibitor, is definitely reasonable.

A comprehensive overview of the frontline treatment landscape for patients with unresectable HCC, including those who present with contraindications to standard-of-care therapy.

Unresectable HCC: Selecting the Optimal Frontline Therapy

As most of us know, HCC [hepatocellular carcinoma] is an aggressive tumor that typically starts in the setting of liver cirrhosis. In the United States, the incidence is around 42,000 cases per year. If you look at the data, it’s lumped in with the primary liver cancers along with intrahepatic cholangiocarcinoma, understanding that the bulk of it is primary liver cancer. For most of the patients who develop HCC, there are certain risk factors. For example, if you look at who’s most commonly affected, males tend to be more affected than females at a ratio of 3:1. It tends to be more prevalent in the Asian population and the African American population, most likely due to some of the risk factors they carry.

The risk factors most commonly known that can cause liver cirrhosis—most HCC occurs in the setting of liver cirrhosis—the common risk factors, at least in the United States, include hepatitis C. They include what we call nonalcoholic fatty liver, which is one of the most common risk factors evolving. Diabetes and obesity are 2 other common risk factors. In Asia, the most common risk factor we see is hepatitis B, which can potentially lead to HCC. Overall, the prognosis for liver cancer is not very good, and if you look at the data, the incidence and the mortality rates are very similar.

Obviously treatment will depend upon early detection. If you have any of those risk factors I mentioned, you definitely should be seen by your primary care physician or hepatologist so you can undergo screening. Hepatologists have their own guidelines, if you have hepatitis B or hepatitis C, on how often you should undergo screening with ultrasound or with laboratory tests. I think that’s a very important point.

One of the challenges in diagnosing and managing HCC is that depending on where you practice, you may not see a lot of hepatocellular carcinoma. I think that’s one of the challenges. If you don’t see a lot of the patient population, then it may be difficult to manage. Having said that, unfortunately, the bulk of patients will present with advanced-stage disease. Only 20% to 30% of patients will present in early stages. Now, if you look at the overview of how we manage hepatocellular carcinoma, there are only 2 ways of curing the disease: one is liver resection, and the other is liver transplant. Once again, that only accounts for about 20% to 30% of the patients we see.

The rest of the patients, unfortunately, are not curable. This is where we truly need a multidisciplinary approach. At our tertiary medical center, we have a tumor board, for example. We have a lot of specialties involved in caring for those patients, including oncologists, surgeons, interventional radiologists, pathologists, and sometimes even radiation oncologists. I think one of the hurdles we face in the community is that there is not a team approach, there’s not a multidisciplinary approach in some cases. I think this is where it’s important, if you see anybody who you suspect has or was diagnosed with hepatocellular carcinoma to refer those patients to a tertiary center where they can truly be treated with a multidisciplinary approach.

Regarding the unmet needs at this time, in terms of the systemic treatment perspective, we’ve come a very long way in terms of treating a patient with advanced disease. Sorafenib used to be the only TKI [tyrosine kinase inhibitor] that was available. However, with the recent approval of atezolizumab and bevacizumab based on the IMbrave150 study, that has taken over. Now, we don’t know what to do afterward. In the case that atezolizumab and bevacizumab fail, what do you use in the second line? That’s a truly unmet need. The other area of unmet need is, can we move this combination of atezolizumab and bevacizumab into an earlier setting, for example, in a patient with BCLC [Barcelona Clinic Liver Cancer] stage B? These are patients with a multifocal hepatocellular carcinoma, just within the liver. Can you move atezolizumab and bevacizumab up to see if they could be used in combination with local therapy, or by themselves in this setting, and get the same efficacy? Those are 2 of the unmet needs at this time.

Expert Richard Kim, MD, reflects on the nature of HCC, particularly the challenges inherent in diagnosis and management.

Richard D. Kim, MD | Authors

Articles

Looking Toward the Future Management of Unresectable HCC

IMbrave150: Frontline Atezolizumab + Bevacizumab in Unresectable HCC

Advice on Molecular Testing and Sequencing in Unresectable HCC

Unresectable HCC: Selecting the Optimal Frontline Therapy

Approaching the Identification and Management of HCC