Atezolizumab±Bevacizuman Combo Yields Modest Benefit in Advanced Biliary Tract Cancer

Atezolizumab (Tecentriq) plus cisplatin/gemcitabine with or without bevacizumab (Avastin) resulted in a modest clinical benefit in a population diagnosed with advanced biliary tract cancer (BTC), according to data from the dual experimental arm of the phase 2 IMbrave 151 trial (NCT04677504).¹

Data presented during the 2023 Gastrointestinal Cancers Symposium showed that treatment-naïve patients who received bevacizumab, atezolizumab, and cisplatin/gemcitabine (n = 79) had a median progression-free survival (PFS) of 8.3 months (95% CI, 6.8-10.0) vs 7.9 months (95% CI, 6.2-8.4) for those who received placebo plus atezolizumab and cisplatin/gemcitabine (n = 83; HR, 0.76; 95% CI, 0.51-1.14). The 6-month PFS rates were 78.2% (95% CI, 68.8%-87.7%) vs 63.1% (95% CI, 52.6%-73.6%), respectively.1

“IMbrave 151 is the first randomized study to evaluate concurrent PD-L1/VEGF blockade plus chemotherapy as first-line treatment for advanced BTC,” said Anthony B. El-Khoueiry, MD, associate professor of clinical medicine in the Division of Medical Oncology at the Keck School of Medicine of the University of Southern California in Los Angeles, California, in a presentation of the data. “The purpose of this study was hypothesizing and estimation of the impact of atezolizumab plus bevacizumab plus [gemcitabine/cisplatin] and atezolizumab plus placebo plus gemcitabine/cisplatin on PFS and it was designed to randomize 2 experimental arms and there was no formal hypothesis testing and therefore no P value.”

In a discussion of the abstract, Imane H. El Dika, MD, noted, “Although PFS is a valuable [end point] in a phase 2 study, I wonder if we can rely on this with an immunotherapy mechanism of action [that sets in] later with more action being observed at the tail end of the curve.” El Dika, who is an assistant attending physician at Memorial Sloan Kettering Cancer Center in New York New York, added, “With the limitation of the small sample size, lack of statistical power, both arms did not compare favorably to historic controls…there was no clear signal to move either arm forward. Gemcitabine, cisplatin, durvalumab [Imfinzi] remains the standard frontline treatment [for this patient population.]”

A key secondary end point was objective response rate (ORR). The confirmed ORR was 24.1% (95% CI, 15.1%-35.0%) with the addition of bevacizumab vs 25.3% (95% CI, 16.4%-36.0%) without the VEGF inhibitor. Complete responses were observed in 1 patient in each arm. More than half of patients in each arm (63.3% and 54.2%, respectively) had stable disease. The disease control rate with bevacizumab was 78.5% vs 75.9% without.

The duration of response (DOR) was prolonged in the bevacizumab arm with 88.5% (95% CI, 73.6%-100.0%) of responders having a response lasting at least 6 months compared with 47.4% (95% CI, 23.0%-71.8%) of responders in the placebo arm. The median duration of response was not estimable (NE; 95% CI, 6.4-NE) in the bevacizumab arm compared with 5.8 months (95% CI, 4.3-6.7).

Overall survival (OS) data were immature at the time of analysis. The median OS was NE (95% CI, 11.0-NE) in the bevacizumab arm vs 11.4 months (95% CI, 10.6-NE) in the placebo arm (HR, 0.74; 95% CI, 0.43-1.27). The 6-month OS rates were 92.0% (95% CI, 85.8%-98.1%) vs 80.5% (95% CI, 72.0%-89.1%), respectively.

IMbrave 151 evaluated the safety and efficacy the PD-L1 inhibitor atezolizumab plus chemotherapy backbone of blockade cisplatin/gemcitabine with or without the anti-VEGF monoclonal antibody bevacizumab. Investigators hypothesized the inhibiting effects of anti-VEGF agents added to anti–PD-L1 with chemotherapy may result in an immune-permissive tumor microenvironment in patients with BTC.2

Patients with histologically confirmed intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma, or gallbladder cancer with measurable disease per RECIST v1.1 criteria were eligible for enrollment. “The eligibility criteria were relatively standard for a study of advanced biliary tract cancers,” El- Khoueiry noted. This included patients having an ECOG performance status of 0 or 1; unresectable, recurrent, or metastatic disease; and no prior treatment. He added that those with recurrent disease within 6 months after surgery or adjuvant therapy, and patients with bleeding from esophageal or gastric varices within 6 months prior to treatment were excluded. “Patients at high-risk for varices were required to have endoscopy and treatment of varices by local standard within the 6 months prior to treatment,” he said.1

Patients had a median age of 63 years (range, 36-79) with 42.6% of patients located in Asia vs the rest of the world. Most patients had metastatic disease (81.9%) and intrahepatic cholangiocarcinoma (54.3%) or gallbladder cancer (27.2%). One-third of patients had prior surgery for BTC. Among 121 patients, tumor area positive score for PD-L1 of less than 1% was noted in 56.2% of patients.1

Patients were randomly assigned to receive intravenous (IV) atezolizumab 1200 mg IV once every 3 weeks in combination with gemcitabine (1000 mg/m2 on days 1 and 8) and cisplatin (25 mg/m2 IV on days 1 and 8) of each 21-day cycle with or without bevacizumab 15 mg/kg IV once every 3 weeks. The chemotherapy regimen was administered for 8 cycles, with patients continuing bevacizumab or placebo and atezolizumab maintenance for cycle 9 and beyond until disease progression or unacceptable toxicity.1,2 No crossover was allowed.

The primary end point was PFS per RECIST 1.1 criteria. Key secondary end points included ORR, DOR, OS, safety, and patient-reported quality-of-life outcomes. Exploratory end points of the study include landmark PFS and OS analysis and patient-reported outcomes using Common Terminology Criteria for Adverse Events.1

Patients were stratified by anatomical location of primary tumor, metastatic disease, and geographic region.1 El-Khoueiry said that although PFS across subgroups marginally favored the addition of bevacizumab, the most notable benefit was among those without metastatic disease (HR, 0.51), women (HR, 0.56) and patient residing outside of Asia (HR, 0.60).1

Results from an exploratory analysis of PFS among responders and those with stable disease was reported. Among the 19 responders with either a complete or partial response in the bevacizumab arm the median PFS was NE (95% CI, 9.7-NE) compared with 8.3 months (95% CI, 8.1-NE) among 21 responders in the placebo arm (HR, 0.51; 95% CI, 0.18-1.49). For the 50 patients with stable disease in the bevacizumab arm, the median PFS was 8.3 months (95% CI, 6.3-10.0) vs 8.4 months (95% CI, 6.4-10.3) in the placebo arm (n = 45; HR, 1.00; 95% CI, 0.59-1.71).1

The safety analysis was assessed during the chemotherapy phase, the maintenance phase, and among all treated patients. During the chemotherapy phase, grade 3/4 any-cause adverse effects (AEs) were experienced by 69.2% of patients in the bevacizumab arm (n = 78) and 70.4% of patients in the placebo arm (n = 81). AEs led to withdrawal from any treatment in 5 and 7 patients, respectively. An AE of special interest was whether patients required systemic treatment with corticosteroids with atezolizumab; 9 patients in the bevacizumab were treated with corticosteroids in the bevacizumab arm vs 4 patients in the placebo arm.

Treatment-related AEs (TRAEs) linked to atezolizumab were experienced by 47.4% vs 38.3% of patients in the bevacizumab and placebo arms, respectively. TRAEs related to bevacizumab or placebo were experienced by 56.4% vs 38.3%, respectively.

In the maintenance phase, among patients in the bevacizumab arm (n = 54), the rate of grade 3.4 any-cause AEs, AEs leading to withdrawal of any treatment, and AEs requiring corticosteroids were 31.5%, 7.4%, and 1.9%, respectively. These rates in the placebo arm (n = 53) were 24.5%, 1.9%, and 1.9%, respectively. “In regard to frequency and nature of adverse events they were comparable across both arms and there were no new safety signals,” El-Khoueiry said. “Of note, hypertension was more common in the bevacizumab arm, which is no surprise. Not shown here because it was at a frequency of less than 20%, grade 3 and grade 4 bleeding events were comparable across both arms in 3.7% of patients on the placebo arm and 2.6% of patients on the bevacizumab arm. All-grade bleeding events were 19% in the bevacizumab arm and 10% in the placebo arm.”

The most common AEs overall were anemia (50.0% vs 65.4%, respectively), neutrophil count decrease (48.7% vs 39.5%), nausea (41.0% vs 40.7%), and hypertension (38.5% vs 18.5%).

El Dika concluded her discussion by saying, “I urge us all to continue to do molecular testing at diagnosis early on for [patients with] advanced BTC as integration of targeted therapies and immune therapies is underway in the frontline setting to expand our treatment options for our patients.”

References

1. El-Khoueiry AB, Ren Z, Chon H, et al. IMbrave151: a phase 2, randomized, double-blind, placebo-controlled study of atezolizumab with or without bevacizumab in combination with cisplatin plus gemcitabine in patients with untreated, advanced biliary tract cancer. J Clin Oncol. 2023;41(suppl 4):491. doi:10.1200/JCO.2023.41.3_suppl.491
2. Hack SP, Verret W, Mulla S, et al. IMbrave 151: a randomized phase II trial of atezolizumab combined with bevacizumab and chemotherapy in patients with advanced biliary tract cancer. Ther Adv Med Oncol. 2021;13:17588359211036544. doi:10.1177/17588359211036544