Atezolizumab Plus Bevacizumab Shows OS Benefit in Albumin-Bilirubin Grade 1 HCC

Masatoshi Kudo, MD, PhD

Atezolizumab (Tecentriq) given in combination with bevacizumab (Avastin) led to an improvement in overall survival (OS) compared with sorafenib (Nexavar) in patients with albumin-bilirubin (ALBI) grade 1 hepatocellular carcinoma (HCC), findings from a subgroup analysis of the phase 3 IMbrave150 trial (NCT03434379) show.

Results presented during the 2021 International Liver Cancer Association Conference show that at a median follow-up of 15.6 months, the median OS was not estimable (NE; 95% CI, 23.7-NE) with atezolizumab/bevacizumab vs 15.4 months (95% CI, 11.7-20.8) with sorafenib in patients with ALBI grade 1 HCC (HR, 0.50; 95% CI, 0.35-0.72). In patients with ALBI grade 2 disease, the median OS was 11.7 months (95% CI, 9.1-16.1) vs 12.2 months (95% CI, 7.2-16.1), respectively (HR, 0.92; 95% CI, 0.66-1.29).

At a median follow-up of 15.6 months, the median OS was not estimable (NE; 95% CI, 23.7-NE) with atezolizumab/bevacizumab vs 15.4 months (95% CI, 11.7-20.8) with sorafenib in patients with ALBI grade 1 HCC (HR, 0.50; 95% CI, 0.35-0.72). In patients with ALBI grade 2 disease, the median OS was 11.7 months (95% CI, 9.1-16.1) vs 12.2 months (95% CI, 7.2-16.1), respectively (HR, 0.92; 95% CI, 0.66-1.29).

Further subtyping of patients with grade 2 disease showed the median OS was 14.1 months (95% CI, 9.3-18.0) with atezolizumab/bevacizumab vs 12.4 months (95% CI, 6.9-20.4) with sorafenib in patients with modified ALBI (mALBI) grade 2a HCC (HR, 0.97; 95% CI, 0.59-1.59). The median OS was 10.5 months (95% CI, 7.1-15.5) vs 10.4 months (95% CI, 5.8-15.7), respectively, in patients with mALBI grade 2b HCC (HR, 0.85; 95% CI, 0.54-1.34).

“Atezolizumab plus bevacizumab provided the longest OS seen in a frontline phase 3 study in advanced HCC, confirming this combination as the standard of care for [patients with] previously untreated, unresectable HCC,” said Masatoshi Kudo, MD, PhD, professor and chairman of the faculty of medicine in the Department of Medicine and member in the Department of Gastroenterology and Hepatology at Kindai University in Osaka-Sayama, Japan. “In this exploratory analysis, patients with mALBI or ALBI grade 1 [disease] derived a greater OS benefit with atezolizumab plus bevacizumab than with sorafenib.”

The combination of atezolizumab and bevacizumab has been approved in over 70 countries for the treatment of patients with unresectable HCC who have not received prior systemic therapy, Kudo explained. The regulatory decisions were based on findings from the primary analysis of the IMbrave150 study, which demonstrated clinically meaningful improvement in the coprimary end points of OS and independent review facility (IRF)-assessed progression-free survival (PFS) per RECIST v1.1 criteria.

The exploratory analysis evaluated the possible effects of ALBI grade at baseline on outcome in patients treated with atezolizumab/bevacizumab or sorafenib.

“The ALBI grade is an evidence-based method of assessing liver function that relies only on serum bilirubin and albumin levels and has demonstrated similar performance to the Child-Pugh grade,” Kudo said.

In IMbrave150, patients were randomized 2:1 to receive 1200 mg of intravenous (IV) atezolizumab plus 15 mg/kg of IV bevacizumab every 3 weeks or 400 mg of oral sorafenib twice daily until loss of clinical benefit or unacceptable toxicity.

Patients had previously untreated, locally advanced or metastatic and/or unresectable HCC. Stratification was based on region, ECOG performance status, macrovascular invasion and/or extrahepatic spread, and baseline alpha-fetoprotein.

ALBI score equates to log10 bilirubin in µmol/L x 0.66 plus albumin in g/L x -0.085. For this analysis, ALBI grade definitions stated that grade 1 was at least -2.60, grade 2 was between -2.60 and -1.39, and grade 3 was more than -1.39. mALBI grading defined grade 2a as more than -2.60 and -2.270 or lower and grade 2b as more than -2.270 and -1.39 or lower.

Patient characteristics were well balanced between mALBI grades. Overall, 191, 72, and 72 patients with mALBI grade 1, grade 2a, and grade 2b disease were treated with atezolizumab plus bevacizumab, respectively. Sorafenib was given to 87, 37, and 71 patients, respectively.

Further results of the analysis revealed that the median PFS per IRF RECIST v1.1 criteria was 8.8 months (95% CI, 6.7-10.0) with atezolizumab/bevacizumab compared with 4.4 months (95% CI, 3.9-5.7) with sorafenib in patients with ALBI grade 1 HCC (HR, 0.61; 95% CI, 0.45-0.82). The median PFS was 5.6 months (95% CI, 5.3-7.0) vs 4.2 months (95% CI, 2.7-5.8), respectively, in patients with ALBI grade 2 disease (HR, 0.70; 95% CI, 0.51-0.95).

By mALBI grade, patients with grade 2a disease treated with atezolizumab plus bevacizumab had a median PFS of 5.6 months (95% CI, 4.3-8.6) compared with 4.7 months (95% CI, 2.5-6.7) in patients treated with sorafenib (HR, 0.68; 95% CI, 0.43-1.07). Patients with mALBI grade 2b disease had a median PFS of 6.5 months (95% CI, 4.5-7.1) vs 4.2 months (95% CI, 2.6-6.2), respectively (HR, 0.74; 95% CI, 0.49-1.14).

The confirmed overall response rates (ORRs) with atezolizumab/bevacizumab were 32% (n = 58) in patients with mALBI grade 1 HCC, 30% (n = 21) in patients with mALBI grade 2a HCC, and 25% (n = 18) in patients with mALBI grade 2b disease. The ORRs with sorafenib were 6% (n = 5), 11% (n = 4), and 23% (n = 9), respectively.

Complete responses (CRs) with atezolizumab/bevacizumab were observed in 10% (n = 18) of responders with mALBI grade 1 disease, 6% (n = 4) of responders with mALBI grade 2a disease, and 4% (n = 3) of responders with mALBI grade 2b disease. One patient with mALBI grade 2b disease derived a CR with sorafenib.

The disease control rate (DCR) in patients with mALBI grade 1 disease was 78% with atezolizumab/bevacizumab vs 59% with sorafenib. In patients with mALBI grade 2a disease, the DCRs were 73% vs 53%, respectively. In patients with mALBI grade 2b disease, the DCRs were 65% vs 48%, respectively.

The median time to deterioration in liver function in the intention-to-treat patient population was 10.2 months (95% CI, 8.0-11.0) with atezolizumab/bevacizumab (n = 336) compared with 8.6 months (95% CI, 6.2-11.8) with sorafenib (n = 165; HR, 0.82; 95% CI, 0.65-1.03).

“The safety and tolerability profile of atezolizumab plus bevacizumab was consistent with the known safety profiles of each individual drug and with the underlying disease, regardless of mALBI grade,” Kudo said.

The median treatment duration in patients with mALBI grade 1 HCC was 10.4 months (range, 0-28) with atezolizumab, 9.6 months (range, 0-28) with bevacizumab, and 2.9 months (range, 0-25) with sorafenib. In patients with mALBI grade 2a disease, the median treatment durations were 6.9 months (range, 0-26), 5.1 months (range, 0-25), and 1.9 months (range, 0-21), respectively. In patients with mALBI grade 2b disease, the median treatment durations were 4.3 months (range, 0-24), 4.7 months (range, 0-24), and 2.8 months (range, 0-21), respectively.

Serious adverse effects (AEs) were observed in 42% of patients with mALBI grade 1 HCC treated with atezolizumab/bevacizumab (n = 189) vs 27% of patients treated with sorafenib (n = 81). Treatment-related serious AEs were observed in 22% vs 12% of patients, respectively. In patients with mALBI grade 2a disease, 54% of patients experienced serious AEs with atezolizumab/bevacizumab (n = 71) vs 32% with sorafenib (n = 37). Treatment-related serious AEs were observed in 17% vs 14% of patients, respectively. In patients with mALBI grade 2b disease, 62% of patients experienced serious AEs with atezolizumab/bevacizumab (n = 69) vs 45% with sorafenib (n = 38). Treatment-related serious AEs were observed in 33% vs 26% of patients, respectively.

Grade 5 AEs were observed in 4% of patients with mALBI grade 1 HCC treated with atezolizumab/bevacizumab; 4% of patients with mALBI grade 1 HCC treated with sorafenib; 7% of patients with grade 2a HCC treated with atezolizumab/bevacizumab; 5% of patients with grade 2a HCC treated with sorafenib; 16% of patients with grade 2b disease treated with atezolizumab/bevacizumab; and 11% of patients with grade 2b disease treated with sorafenib. Treatment-related grade 5 AEs occurred in 1%, 0%, 1%, 0%, 4%, and 3% of patients, respectively.

Additionally, 17%, 9%, 25%, 14%, 30%, and 16% of AEs led to patient withdrawal from any component of treatment, respectively. The rates of dose interruptions because of AEs as a result of any study treatment were 65%, 35%, 52%, 51%, 52%, and 55%, respectively.

Finally, 37% of patients with mALBI grade 1 disease, 38% of patients with mALBI grade 2a disease, and 37% of patients with mALBI grade 2b disease experienced AEs that led to dose modification of sorafenib.

In concluding remarks, Kudo noted that because of “the exploratory nature of this analysis, the limited number of patients within each mALBI subgroup, and imbalances between baseline characteristics between subgroups, results should be interpreted with caution.”

_Reference:

_{Kudo M, Finn RS, Cheng AL, et al. IMbrave150: albumin-bilirubin grade analyses in phase III study of atezolizumab + bevacizumab vs sorafenib in patients with unresectable hepatocellular carcinoma. Presented at: 2021 International Liver Cancer Association Conference; September 2-5, 2021; virtual.}