No safety signals were raised, no major toxicity and no surgery impairment were reported following one cycle of neoadjuvant atezolizumab therapy in patients with resectable non-small cell lung cancer (NSCLC) according to phase II trial PRINCEPS findings presented by Dr Benjamin Besse of the Department of Cancer Medicine, Institut Gustave Roussy in Villejuif, France at the ESMO Virtual Congress 2020.
The study primary endpoint was 2-month tolerance rate defined as rate of patients without major toxicities or morbidities during the period from start of treatment and 1 month after surgery. Major toxicities or morbidities included treatment toxicity leading to a delay of at least 15 days of the surgery, grade ≥3 toxicity occurring within 2 months after atezolizumab infusion, major postoperative morbidities, any death related to the experimental treatment and occurring in the period from the day of injection of atezolizumab to the 30 postoperative days, and patients that did not have surgery because of early progression.
Secondary objectives were toxicities according NCI-CTC v4.0 and postoperative morbidities according Clavien-Dindo classification, the objective response rate evaluated by CT using RECIST v1.1 criteria, mid-term (2-years) and long-term (5-years) overall survival rates, mid-term(2-years) and long-term (5-years) disease-free survival and pattern of recurrence, pathological complete response, pathological findings (response pattern, immune infiltrate), and response with FDG-PET before surgery.
Exploratory objectives were PD-L1 tissue expression by immunohistochemistry, adaptive immune response in the microenvironment in post-surgical fresh tissue and blood, rate of circulating immunomarkers by circulating tumour cells (CTC) and ctCTC, molecular profiling at resection and, if available, prior to treatment and at disease progression, changes in oncogenic mutations profiles and mutational load at resection and at disease progression, if available.
From December 2016 to February 2020, the study enrolled 30 surgery-eligible patients with clinical stage IA (≥ 2 cm)-IIIA, ECOG performance status 0-1, non-N2 NSCLC. The patients’ mean age was 64 years, 50% were female, just 7% were never smokers, and 83% of patients had adenocarcinoma. Fifteen (50%) patients had pathological stage I, two (20%) patients were stage II, and 9 (30%) patients had stage III disease.
Mutation status was reported in 24 patients for BRAF (2), EGFR (3), KRAS (8), TP53 (13), PI3KCA (1), STK11 (1), ESR1 (1), WT (1) and baseline PD-L1 status was reported in 29 patients as following: <1% in 18 patients (62%), ≥1% in 6 patients (21%) and ≥50% in 5 patients (17%).
The patients were treated with one i.v. injection of atezolizumab at 1200 mg on study day 1, which was followed by surgery between day 21 to day 28.
The primary endpoint was the rate of patients without major toxicities or morbidities from day 1 until 1 month post-surgery. Response by RECIST v1.1 and major pathological response (MPR; ≤10% viable tumour cells) were also assessed.
Fresh tumour samples were obtained during surgery and analysed within 4 hours.
Most tumour samples showed histological features of response
All patients underwent the planned surgery and no surgery was delayed more than 15 days. Following surgery, 29 had R0 resection, and one had R1.
However, no complete response was observed, MPR was reported in 4 patients (14%) and pathological response ≥50 (less than 50% residual tumour cells) was reported in 12 patients (41%).
Metabolic response (18F-FDG PET/CT, variation of SUVmax) was reported in 28 patients as following: +20% or more in 7 patients (25%), stable (between +20% and -20%) in 18 patients (64%) and -20% or more in 3 patients (11%).
Correlation between pathological response and response by RECIST was done in 28 patients with Pearson correlation 0.24 (p = 0.2); no correlation observed between percentage of pathological regression and variation of SDM at week 3 post atezolizumab.
Correlation between pathological response and metabolic response was done in 27 patients with Pearson correlation of 0.12 (p = 0.55); no correlation observed between percentage of pathological regression and variation of SUVmax at week 3 post atezolizumab.
Correlation between pathological response and PD-L1 expression at baseline was done in 23 patients with Pearson correlation of 0.45 (p = 0.0311); Dr Besse summarised that an increased level of PD-L1 tumour cells status before treatment is associated to a larger pathological regression. When PD-L1 was assessed on tumour cells and considered in 4 classes (<1%, 1-5%, 5-50% and ≥50%), Spearman correlation was -0.50560 (p = 0.0051); Dr Besse summarised that quantitative results confirmed the correlation between pathological response and PD-L1 expression.
Three patients experienced surgical complications that included one patient with respiratory distress grade 3 plus sepsis grade 4, one heart block atrioventricular, and one patient had paresthesia grade 1.
No grade 5 toxicity was observed.
Treatment-related adverse events (TRAEs) were uncommon with only one grade 1 TRAE of parietal pain, which related to surgery.
Based upon these data, the authors concluded that surgery after one infusion of atezolizumab was safe. MPR was observed in 4 patients. Pathological responses were not correlated with RECIST v1.1 response rate, as well as with metabolic variations, but they were correlated with high PD-L1 expression.
Recruitment of a control group of 30 untreated tumours is ongoing. Final analysis of the immune contexture in blood and fresh tissue will be performed when the control group will be fully recruited.
Prof. Lyudmila Bazhenova of the Moores Cancer Center, University of California San Diego in San Diego, USD who discussed the study findings said that surgery after immune checkpoint inhibitor is safe, but complexity increases. Valuable translational correlates will provide insights into potential biomarkers for predicting response to immune checkpoint inhibitor therapy.
Questions to answer with future investigation are will the immune checkpoint inhibitors be used according to biomarkers or for all-comers, does the tumour pathological response rate or MRP translate into disease-free and overall survival, how many induction cycles, best time for surgery after induction, will the MRP rates be affected by surgical timing, should immune checkpoint inhibitors be continued in the adjuvant setting for all patients or for responders only, are neoadjuvant immune checkpoint inhibitors more efficacious given as monotherapy or dual therapy or in combination with chemotherapy or radiation.
This study was funded by Roche-Genentech.
1215O – Besse B, Adam J, Cozic N, et al. Neoadjuvant atezolizumab (A) for resectable non-small cell lung cancer (NSCLC): results from the phase II PRINCEPS trial. ESMO Virtual Congress 2020.