Therapy with atezolizumab and bevacizumab improved recurrence-free survival in patients with hepatocellular carcinoma (HCC) following surgical resection or ablation, according to results from the phase 3 IMbrave050 clinical trial (NCT04102098),2 presented at the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting.1 As this combination therapy is finding its way into practice, Dr. Cha Len Lee (University of Toronto) performed a study to assess the efficacy, safety, and value of esophagogastroduodenoscopy with regard to this combination therapy in the real world.4 Physician’s Weekly spoke with her to discuss the findings.

PW: What was the unmet need in patients with hepatocellular carcinoma? Why is it important to have real-world data on the combination of atezolizumab and bevacizumab

CLL: Hepatocellular carcinoma (HCC) is a significant global health concern, currently ranking the fourth leading cause of cancer-related death worldwide. Since 2008, there has been a huge search for novel treatments targeting this disease, particularly in the area of immunotherapy. When the IMbrave150 trial showed the superiority of atezolizumab with bevacizumab in terms of overall survival, progression-free survival, and response rate against sorafenib, this combination treatment was recognized as the first-line treatment option.3 Furthermore, the IMbrave150 trial reported that the treatment-related gastrointestinal bleeding rate was 7%, compared with 4.5% in the sorafenib group. As a result, the trial recommended that all the patients should have endoscopic esophagogastroduodenoscopy (EGD) within 6 months prior to the treatment with atezolizumab plus bevacizumab in order to identify any varices, which may increase the likelihood of bleeding when undergoing this treatment. Since then, 3 years have passed and it is evident that this treatment has significantly impacted the landscape of HCC. However, the value of performing pre-treatment EGD in all patients remains uncertain. Besides this, the efficacy and the bleeding risk of atezolizumab with bevacizumab in the real-world setting are still unclear. With all those unanswered questions, we decided to perform a retrospective analysis on this topic.

Can you explain the methodologies you used?

In Canada, we have an HCC Cancer Health Outcome Research Database. It is a consortium that is existing across 10 cancer centers in different provinces. For this specific retrospective analysis, we managed to include collaborators from five cancer centers in the provinces of Ontario, Alberta, and Manitoba. That’s how we managed to get all the retrospective data for this project.

What were the results? 

We focused on the efficacy, safety, and uptake of EGD. We recruited 112 patients who have been treated from July 1, 2020 to August 31, 2022. Overall, patient baseline characteristics were similar to those of the patients enrolled in the IMbrave150 trial, although, we did treat more Barcelona Clinic Liver Cancer (BCLC) stage B patients and patients who previously underwent local therapy compared with those who have been reported in the IMbrave150 trial. All our patients received atezolizumab with bevacizumab outside the clinical trial setting: 90% received treatment as a first-line therapy, 9% received it as a second-line treatment, and 1% received this combination therapy as a third-line treatment. After a median follow-up of 10.4 months, the reported overall survival for our study population was 20.3 months, and the median progression-free survival was 9.6 months. Moreover, the disease control rate achieved with the treatment was 76.8 months. Altogether, these numbers are comparable to those reported in the IMbrave150 trial.

The second part of the study focused on the EGD uptake and treatment-related bleeding. We discovered that 70% of the patients completed EGD within 6 months before starting treatment, while the other 30% proceeded with the treatment without EGD or any invasive determinations of portal hypertension or compensated advanced chronic liver disease. For those who had undergone EGD, 41% had evidence of varices seen on their endoscopic investigations, and 20% of those required treatment with either ligation or beta blockers. Those numbers were two times higher than the rates reported in the IMbrave150 trial. This is not surprising; these results are reflecting the real-world, high-risk population that we are treating in this study.

Overall, the treatment-related bleeding rate observed in this study was 15%. If you break it down between the EGD group and non-EGD group, the rate in the EGD group was higher, with 18%, compared to the non-EGD group, with 9%. Again, this is not surprising when you think about it; the EGD group is probably the higher risk group. The doctor sent these high-risk patients to have some investigation before they received their treatment. However, when we further analyzed the data, 10% of the bleeding complications were actually non-gastrointestinal bleedings, like bruises and gum bleedings. The other 5% of bleedings were specific gastrointestinal bleedings. When we compared patients who had gastrointestinal bleedings with those who had non-gastrointestinal bleedings, there wasn’t a meaningful overall survival or progression-free survival difference between them.

What are the implications of these results, especially with regard to bleeding?

It’s still a very small study, and I can’t make any firm conclusion based on that. However, our study does show potential for the selective use of EGD in patients with low risk of hepatic decompensation. In addition, our study showed that the approach of selective EGD, based on patient characteristics and risk factors, is feasible in certain patients in order to allow patients to start his or her treatment sooner, without having to wait for EGD. However, until we find out more about this, we need to lean on the cautious side in regard to not investigating all patients with pre-treatment EGD before they go on atezolizumab with bevacizumab, as this combination therapy involves a high dose of bevacizumab.

What are the next steps?

At the moment, the sample size of our study is definitely small, but we are excited about the results. Therefore, we are in the process of publishing our findings. In addition, the community needs to have real-world data with regard to this combination therapy. It does have value to get our results published. Meanwhile, we want to expand the sample size of this retrospective analysis with more collaborators from our other Canadian cancer centers. In the long run, we are considering to conduct a prospective study with our multidisciplinary team, including hepatologists, to create a better guideline with respect to stratifying patients with HCC around the area of variceal bleeding related to cirrhosis and atezolizumab with bevacizumab.