Efficacy, safety, and quality of life data from IMbrave150 have changed the standard of care for unresectable hepatocellular carcinoma.
Richard S. Finn, MD: One of the most striking outcomes from the study is the response data. We have known for a long time that we could improve overall survival in both frontline and second-line liver cancer without significantly improving response rates, essentially slowing progression. This is an important observation because the most important thing we do for our patients is improve overall survival. However, we now have data with atezolizumab/bevacizumab showing that an objective response rate by the RECIST criteria in the study was over 27%, and this includes about 5.5% complete responses and about 22% partial responses. With sorafenib, we saw about 12% objective responses, all of these are partial responses, but clearly the duration of response and the depth of the response that we see with immunotherapy agents differs from what we see with the TKIs [tyrosine kinase inhibitors].
These efficacy data need to be balanced against the [adverse] effect profile. However, if we look at overall adverse events, they are similar between both arms of the study. Grade 3/4 events were similar between both arms of the study: 56% and 55%, respectively, between atezolizumab/bevacizumab and sorafenib.
If we look at grade 5 events, it is also similar. If anything, it is a little higher with sorafenib, where we saw with atezolizumab/bevacizumab about 4.5%, and with sorafenib, it was just under 6%. If we look at the requirement to withdraw study drugs, it was 10% with sorafenib. To stop the regimen of atezolizumab/bevacizumab completely, it was only 7%.
The adverse event profile was predictable. When we look at atezolizumab/bevacizumab, we see that the most frequent things were those that were related to bevacizumab, such as hypertension and proteinuria. These were of higher frequency and grade compared to sorafenib, but if we look at those that were more frequent with sorafenib, such as diarrhea and hand-foot-skin reaction syndrome, those did not occur frequently with atezolizumab/bevacizumab. If they did, such as with GI [gastrointestinal] toxicity, it was generally low grade. Other things like fatigue were generally similar between both arms, but they were higher grade with sorafenib. There were other toxicities that we saw. The incidence of autoimmune hepatitis was uncommon: only about 12% of patients in the atezolizumab/bevacizumab arm required steroids for the treatment of an autoimmune event.
If we look at significant adverse events, it is important to note that there were some grade 5 bleeding events that occurred more frequently with atezolizumab/bevacizumab. There were a total of 4, or about 1.2%, fatal events, related to GI hemorrhage, 1%, and esophageal varices, 0.3%. Overall, the bleeding events that occurred with atezolizumab/bevacizumab were more common than with sorafenib, but this was mostly driven by low-grade events, such as epistaxis or low-grade hematuria.
What is really supportive of making this the standard of care is the patient-reported outcome data. Several key quality of life measurements were done, and we consistently saw a delay in the deterioration of quality of life, role functioning, or physical well-being with atezolizumab/bevacizumab. If we look at time to deterioration of quality of life, one of the readouts, that occurred at 3.6 months with sorafenib and over 11 months with atezolizumab/bevacizumab.
Based on these data, this regimen was approved in the United States in June of 2020, and it has now become the standard of care for most patients with advanced liver cancer. We need to find a reason not to offer patients this regimen. They should have some contraindication, such as severely decompensated liver disease, recent variceal bleeding, or high-risk varices seen on endoscopy, or some other medical contraindication to either bevacizumab or atezolizumab. If they do, then sorafenib or lenvatinib are certainly appropriate options, and I would still argue that a clinical trial is the best option for patients.
Transcript edited for clarity.