Safety and efficacy of atezolizumab plus bevacizumab for unresectable hepatocellular carcinoma

Prof Richard Finn - Geffen School of Medicine at UCLA, Los Angeles, USA

Hi there, this is Dr Richard Finn from The Geffen School of Medicine at UCLA. Thank you for your interest in our data presented at AACR 2021. This was a subgroup analysis from the IMbrave150 study looking at patients with high risk tumour features. By way of background, IMbrave150 was a global randomised phase III study looking at the PD-L1 inhibitor atezolizumab in combination with the VEGF antibody bevacizumab, so atezo-bev, given every three weeks versus standard of care sorafenib for patients with advanced liver cancer. This includes patients who were both intermediate, Barcelona B and Barcelona C, those would be patients who had progressed on local regional treatment or were not candidates and were first line treatment for advanced disease.

This study was a hallmark study and it was the first time that a new regimen was superior to sorafenib. In the initial dataset, the primary analysis, we had not reached the median overall survival but the hazard ratio was significant. Then at the updated analysis that we presented at ASCO GI a few weeks ago we saw that the median overall survival with atezo-bev was 19.2 months compared to sorafenib at 13.4 months. This was supported by an improvement in PFS with a hazard ratio of 0.65 comparable to the OS data of 0.66. Importantly, there was an objective response rate of 30% which included eight complete responses. The median duration of response was over 18 months.

So a very new active regimen in advanced liver cancer and IMbrave150, unlike some other phase III studies, included patients with very high risk features. This included large bulky tumours in the liver, over 50% of the liver could be included; patients could have invasion into the main portal vein or the Vp4 branch or bile duct invasion. Again, these criteria, all of them or some of which have been excluded from modern phase III studies. This was not an insignificant proportion of patients in this cohort. We had about 64 patients in the atezo-bev arm and 37 patients in the sorafenib arm.

We saw that even in this group of patients that was high risk we had a significant improvement in overall survival with a hazard ratio of 0.62 which compares very favourably to the intent to treat population. That was a median survival from 5.5 months with sorafenib to 7.6 months with atezo-bev.

Interestingly and not surprisingly, when we extracted those patients from the intent to treat population, the so-called non-high risk patients, survival in that group actually went up to just under 23 months – 22.8 months with atezo-bev.

This pattern of improving benefit in these high risk patients was maintained with an improvement of PFS as well as response rate. Again, this is a group of patients who probably needs a response because they have high tumour burden. We saw that the objective response rate in this group was 25%, pretty much in line with the intent to treat population, including 8% complete responses.

Atezo-bev is very well tolerated in both the high risk and intent to treat population but the one thing that did turn up is that the few grade 5 bleeding events that we saw with atezo-bev, despite patients having upper endoscopies, were those that had main portal vein invasion. This physiologically makes some sense given that they have a large tumour burden in their portal vein that this could lead to more portal hypertension. Therefore, even though these are single digit events, something to pay a little more close attention to in these high risk patients and also gives us some reassurance that in the majority of patients who are not high risk that they incidence of high grade bleeding events are even less common.

So, in summary, we presented a dataset that supports the overall findings of IMbrave150 in that it is an active regimen for advanced liver cancer, setting really a new standard of care in the front-line setting based on it is the first time since the approval of sorafenib in 2008 that we have a regimen that improves survival versus an active control.

Thank you very much for your attention and I encourage you to view the presentation for yourself at AACR. Thank you. I’d like to thank all the patients, families, as well as the research staff that all contributed to the success of IMbrave150. Thank you very much.