Adjuvant 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), which were presented by Masatoshi Kudo, MD, PhD, (Kindai University, Japan) at the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting.1 Physician’s Weekly interviews Dr. Kudo to learn more.

The standard of care for curative intent therapy for early-stage HCC includes surgical resection and thermal ablation. However, 5 years after resection or ablation, the risk of recurrence is about 70% to 80%. There is an unmet need in patients with HCC for effective adjuvant approaches.

The randomized controlled phase 3 IMbrave050 trial2 researched the efficacy of an adjuvant combination treatment of the checkpoint inhibitor atezolizumab and the targeted therapy bevacizumab in delaying or preventing recurrence compared with active surveillance, which is the current standard of care for patients who have undergone surgical resection or ablation.

The trial enrolled patients with HCC who were at high risk for recurrence following tumor resection or ablation based on criteria like size and number of tumors, presence of cancer cells within the lumen of blood and/or lymphatic vessels, and tumor grade. Study participants were randomly assigned 1:1 to receive atezolizumab plus bevacizumab every 3 weeks for 1 year (or 17 cycles) or undergo active surveillance for 1 year. Patients in the control arm were eligible to switch to the experimental arm in case of recurrence. The primary endpoint was independent review facility-assessed recurrence-free survival (RFS).

After a median follow-up of 17.4 months, the trial met its primary endpoint, with the combination of atezolizumab and bevacizumab significantly increasing RFS when used as adjuvant therapy following surgical resection or ablation. Patients who received the combination treatment had their risk of recurrence or death reduced by 28% compared with patients in the active surveillance arm. The median RFS was not reached for either arm at this time-point. Because the primary endpoint was met earlier than expected, the data for the trial’s secondary endpoints, including overall survival, are not yet sufficiently mature. Subsequent analyses will supply these data. Physician’s Weekly spoke with Dr Kudo:

PW: Why was this trial important?

MK: There is simply no standard care for the adjuvant setting in HCC after resection or ablation. Several pervious trials have failed to meet the endpoint to suppress the tumor occurrence after curative treatment. This is very important because this is a first positive trial so far.

What was the rationale behind combining bevacizumab and atezolizumab? 

The IMbrave 150 trial (NCT03434379) for unresectable advanced stage HCC showed that atezolizumab combined with bevacizumab resulted in better overall and progression-free survival outcomes than sorafenib.3 The rationale is that combining a checkpoint inhibitor PD-L1 antibody plus anti-VEGF antibody bevacizumab, which improves the immune microenvironment, will shift the environment from immune-suppressive to immune-permissive. The response rate was 30% in advanced stage HCC, with a median survival of 19.2 months, which is the best outcome available for unresectable HCC. It follows that atezolizumab plus bevacizumab is the first choice for frontline treatment of unresectable HCC. The next obvious question is whether we could employ this combination in earlier-stage disease; in other words, does this combination improve the recurrence after resection or ablation, even in the absence of evident disease? It is very well known that the recurrence rate in this situation is around 60% to 80%, even after curative resection, because at the time of resection, micro-metastases have already been seeded in the remaining liver. It will take some time after resection or ablation, that those metastatic tumors become visible by CT or MRI imaging.

In that sense, the adjuvant therapy suppressing the recurrence is very important. And, it is generally believed that a single checkpoint inhibitor monotherapy is not strong enough to suppress recurrence. That is based on the understanding that the micro-metastases will have the same immune microenvironment as the primary resected tumor; thus PD-L1 antibody monotherapy will likely only achieve an approximately 50% response rate. Combining PD-L1 checkpoint inhibition with a targeted therapy like bevacizumab is more portent; this combination should suppress or kill the remaining microtumors.

What were the patient-reported outcome (PRO) results?

We use the IL42 tool, which is a version of  the EORTC QQL C30 questionnaire, with five domains: 1) HRQOL, 2) physical functioning, 3) role functioning, 4) emotional functioning, and 5) social functioning. Patients answered 15 questions at baseline and thereafter at every other visit.

The response rate was greater than 93%, which we were very pleased with. Treatment was continued until recurrence or unacceptable toxicity for 1 year, or about 17 cycles. At the time of the seventeenth cycle, or 1 year, more than 94% were under active surveillance, 96% of whom answered those questions.

In terms of baseline scores, the questionnaire was designed to measure the change on a scale of 0-100 points. There were no differences at baseline between the atezolizumab plus bevacizumab arm and the active surveillance arm. There were very high scores at baseline in both arms, around 80-90 for HRQOL, 81 points for physical function, and so on. We compared the intervention arm with the general population baseline scores, using published data from a general healthy population. We found that the intervention arm reported very similar scores to those scores: 71-89. We concluded that there was no deterioration in QOL at the outset of the adjuvant therapy after resection or ablation.

The pre-specified deterioration definition was a drop by -10 points, which we determined would be a significant deterioration drop due to adjuvant therapy. But compared with baseline scores, the treatment arm was very stable and showed no evidence of deterioration up to and including cycle 17. The intervention arm PRO scores also entirely overlapped the active surveillance in terms of a 95% CI. The line is almost similar, comparable throughout cycle 17. We could therefore conclude from this PRO-based HRQOL tool and functioning score that they did not deteriorate at all.’

Will this change guidelines?

Of course, this is a practice-changing result; Yes, absolutely, all international guidelines will be changed.

Any Safety concerns? 

The treatment-related grade 3-4 adverse event rate was around 35% in the atezolizumab plus bevacizumab group. Treatment-related serious adverse events occurred in 13% of patients. Treatment-related grade 5 occurred in 0.6%. Two patients died, but one case in the active surveillance group also died because of bleeding. The groups did not differ very much, and we felt that this combination was safe and tolerable.

What are the next steps? 

We have to extend the follow-up to accumulate more data, including on overall survival. That will be exciting to see!