The combination of atezolizumab plus carboplatin/etoposide continued to demonstrate an improvement in overall survival versus chemotherapy alone as a frontline treatment for patients with extensive-stage small cell lung cancer.
The combination of atezolizumab (Tecentriq) plus carboplatin/etoposide continued to demonstrate an improvement in overall survival (OS) versus chemotherapy alone as a frontline treatment for patients with extensive-stage small cell lung cancer (ES-SCLC), regardless of PD-L1 and blood tumor mutational burden (bTMB) status, according to an exploratory analysis of the phase IMpower133 trial that was presented during the 2020 AACR Virtual Annual Meeting II.1
Results showed that the median OS was similar across PD-L1 subgroups with cutoffs of 1% and 5%, suggesting that bTMB and PD-L1 status are not predictive of survival outcomes, and should not be used for patient selection for the immunotherapy/chemotherapy combination in this setting.
“When we looked at the impact of PD-L1 expression on survival, our results suggest that PD-L1 does not appear to be a predictive marker for checkpoint inhibitor therapy,” said Leora Horn, MD, MSc, Ingram Associate Professor of Cancer Research, associate professor of medicine (hematology and oncology), assistant director of the Educator Development Program, and clinical director of the Thoracic Oncology Program at Vanderbilt-Ingram Cancer Center during a poster presentation of the findings. “Our exploratory analysis indicates that neither PD-L1 expression or bTMB are predictive biomarkers for treatment with atezolizumab and carboplatin/etoposide. With longer follow-up, this updated analysis further supports first-line atezolizumab with platinum/etoposide for patients with untreated ES-SCLC in an all-comer population.”
The international, double-blind, randomized, placebo-controlled phase III IMpower133 trial evaluated the efficacy and safety of frontline atezolizumab added to the standard of care, combination carboplatin and etoposide, in 403 treatment-naïve patients with ES-SCLC.
All patients received four 21-day cycles of carboplatin area under the curve 5 mg/mL/min intravenously (IV) on day 1 and 100 mg/m2 etoposide IV on days 1 through 3. Patients were also randomized 1:1 to receive either concurrent atezolizumab at 1200 mg IV on day 1 (n = 201) or placebo (n = 202) during the induction phase. Treatment was followed by maintenance therapy with atezolizumab or placebo, according to the previous random assignment, every 3 weeks until progressive disease or loss of clinical benefit.
Investigator-assessed progression-free survival (PFS) and OS in the intent-to-treat population (ITT) served as the primary end points. Secondary end points included objective response rate (ORR), duration of response (DOR), and safety.
The median age was 64 years (range, 26-90) in both the atezolizumab and placebo groups, and the majority were male (64% vs 65%, respectively), white (81% vs 79%), and former smokers (58.7% vs 61.4%). The atezolizumab arm included 17 patients (8%) with brain metastases and 77 (38%) with liver metastases; while the placebo group consisted of 9% and 36%, respectively.
In March 2019, the FDA approved the combination of atezolizumab with carboplatin and etoposide for the frontline treatment of patients with ES-SCLC, based on earlier findings of the IMpower133 trial.
The preliminary data showed that after a median follow-up of 13.9 months, the median OS in IMpower133 was 12.3 months (95% CI, 10.8-15.9) in the atezolizumab arm compared with 10.3 months (95% CI, 9.3-11.3) in the carboplatin/etoposide and placebo arm, leading to a 30% reduction in the risk of death (HR, 0.70; 95% CI, 0.54-0.91; P = .0069).2 The OS benefit was observed in patients with both bTMB-high and bTMB-low status, using prespecified cutoffs of bTMB 10 and bTMB 16.
Moreover, the median PFS was 5.2 months (95% CI, 4.4-5.6) in the atezolizumab arm compared with 4.3 months (95% CI, 4.2-4.5) in the placebo group (HR, 0.77; 95% CI, 0.62-0.96; P = .017).
Updated findings, which were presented at the 2019 ESMO Congress, showed that at a median follow-up of 22.9 months, the median OS was maintained at 12.3 months in the atezolizumab and 10.3 months in the placebo group (HR, 0.76; 95% CI, 0.60-0.95; P = .0154).3
Exploratory biomarker analyses were conducted in the biomarker-evaluable population, according to PD-L1 expression levels on both tumor cells (TC) and immune cells (IC); prespecified secondary analyses were also conducted relating to bTMB status. Investigators also evaluated patterns of progressive disease (PD) and sites of disease progression in the ITT population, as well as in those in whom progression occurred. Safety analyses also included patients who had received at least 1 dose of the combination.
The survival benefit was observed across all prespecified subgroups, but was less apparent in patients younger than 65 years (HR, 0.94; 95% CI, 0.68-1.28) and in those with brain metastases (HR, 0.96; 95% CI, 0.46-2.01).
Updated ORR and DOR data were also included in the results presented at the 2020 AACR Virtual Annual Meeting II. The confirmed ORR was 60.2% (95% CI, 53.1%-67.0%) in the atezolizumab/chemotherapy arm versus 64.4% with the placebo/chemotherapy arm. Moreover, the median DOR was 4.2 months and 3.9 months, respectively (HR, 0.67; 95% CI, 0.51-0.88). Eleven patients on atezolizumab versus 3 patients on the placebo arm had ongoing responses.
In the exploratory biomarker analysis, the PD-L1 immunohistochemistry (IHC) biomarker-evaluable population comprised 34% of the ITT population, and VENTENA SP263 was utilized to determine PD-L1 status. Investigators also noted that PD-L1 expression was mainly observed on tumor-infiltrating ICs, and limited expression was found on TCs. Efficacy analyses were conducted through PD-L1 expression cutoffs of 1% and 5%.
Of the biomarker evaluable population (BEP; n = 126), 36 patients (28.6%) were bTMB-high (10 or higher) and 30 patients (23.8%) were PD-L1–positive; 38 patients (30.2%) were both bTMB-high and PD-L1–positive.
Results showed that the atezolizumab regimen was found to have an OS benefit across PD-L1–positive subgroups compared with the placebo arm, suggesting that PD-L1 expression does not serve as a predictive factor to use checkpoint inhibitors added to chemotherapy in the frontline setting.
Specifically, data showed that the median OS in the ITT-BEP (n = 137) was 9.9 months and 8.9 months with the atezolizumab regimen versus placebo/chemotherapy, respectively (HR, 0.70; 95% CI, 0.48-1.02). In the non-BEP (n = 266), the median OS was 14.6 months and 11.2 months, respectively (HR, 0.81; 95% CI, 0.61-1.08).
When stratified by PD-L1 expression at a 1% cutoff, the median OS was 10.2 months and 8.3 months for patients with less than 1% PD-L1 expression on TC or IC (n = 65) on atezolizumab and placebo, respectively (HR, 0.51; 95% CI, 0.30-0.89). For patients whose PD-L1 expression was 1% or higher (n = 72), the median OS was 9.7 months and 10.6 months, respectively (HR, 0.87; 95% CI, 0.51-1.49).
Patients were also stratified by having PD-L1 expression on 5% of TC or IC. For those with less than 5% PD-L1 expression (n = 108), the median OS with atezolizumab was 9.2 months and was 8.9 months in the chemotherapy/placebo arm (HR, 0.77; 95% CI, 0.51-1.17). In the 5% or greater PD-L1 expression group, the median OS was 21.6 months with atezolizumab versus 9.2 months with the placebo arm (HR, 0.60; 95% CI, 0.25-1.46).
Additionally, RECIST-defined progressive disease was observed in 181 patients (90.0%) who received atezolizumab plus carboplatin/etoposide and in 194 patients (96.0%) on the placebo arm. Patterns of PD were reported as existing lesions (57.7% with atezolizumab vs 64.9% with placebo), new lesions (42.8% vs 49.0%, respectively), and both new and existing lesions (20.9% vs 28.2%).
The patterns of disease progression in specific organs, and the incidence of new lesions, were similar between arms; the most common sites for new lesions in 10% or more of patients were central nervous system, lung, lymph nodes, and the liver.
Fewer patients on the atezolizumab regimen received at least 1 line of subsequent therapy (54.7%) versus those on the chemotherapy/placebo arm (61.9%). On the atezolizumab arm, 18.4% of patients received third-line therapy and 7.0% of patients received fourth-line treatment. On the placebo arm, 24.3% and 8.9% of patients received subsequent third- and fourth-line therapy, respectively.
The most common subsequent treatment in both arms was non-anthracycline chemotherapy (42.8% with atezolizumab vs 49.0% with placebo), followed by anthracycline chemotherapy (18.9% vs 25.2%, respectively), immunotherapy (3.5% vs 8.4%), other (1.5% each), and targeted therapy (1.0% vs 0.5%).
Regarding safety, these AEs were comparable with what was observed in prior analyses. Grade 3/4 AEs occurred in 67.7% and 63.3% of patients on atezolizumab and placebo, respectively. Treatment-related AEs (TRAEs) occurred in 94.9% and 92.3%, respectively, and serious AEs occurred in 38.9% of patients on atezolizumab and placebo, respectively.
Immune-related AEs were higher with atezolizumab (41.4%) versus placebo (24.5%); 20.2% and 5.6% in each arm were treated with steroids or hormone replacement therapy. AEs that led to treatment discontinuation occurred in 11.6% of those on atezolizumab and 2.6% of patients on placebo. No grade 5 immune-related AEs (irAEs) occurred in either arm, but 3 patients on each arm experienced grade 5 TRAEs.
All-grade irAEs that were also AEs of special interest, in the atezolizumab versus placebo arms, included rash (18.2% vs 10.7%, respectively), hepatitis (6.1% vs 4.6%), hypothyroidism (12.6% vs 0.5%), hyperthyroidism (5.6% vs 2.6%), infusion-related reaction (3.5% vs 4.6%), pneumonitis (2.0% vs 1.5%), colitis (0.5% vs 0%), and adrenal insufficiency (0% vs 1.5%). Grade 3/4 irAEs included rash (2.0% vs 0%), hepatitis (1.5% vs 0%), infusion-related reaction (2.0% vs 0.5%), pneumonitis (0.5% vs 1.0%), and colitis (1.0% vs 0%).
The investigators concluded that further studies are needed to evaluate potential biomarkers and their association with survival outcomes to atezolizumab plus carboplatin/etoposide in the frontline setting. Horn also noted that the PD-L1 analysis was based on a limited data set, which can be partially attributed to the poor quality of tissue sample collection methods that could have damaged tissue architecture and limited IHC analyses.