medwireNews: The PEARLS/KEYNOTE-091 trial has demonstrated a significant improvement in disease-free survival (DFS) with the use of adjuvant pembrolizumab versus placebo for patients with resectable non-small-cell lung cancer (NSCLC) regardless of PD-L1 expression.
Luis Paz-Ares (Hospital Universitario 12 de Octubre, Madrid, Spain) presented the results for the phase 3 study at the ESMO Virtual Plenary, saying that “pembrolizumab has the potential to be a new adjuvant treatment option for patients with stage IB (T≥4 cm) to IIIA NSCLC following complete resection and adjuvant chemotherapy when recommended.”
He explained that while pembrolizumab and concurrent radiotherapy has shown efficacy in patients with unresectable stage III NSCLC, regardless of PD-L1 expression or histology, there has been no evidence for the earlier use of pembrolizumab as an adjuvant treatment.
To investigate, the team randomly assigned patients with complete surgical resection and negative margins to receive up to 18 cycles of pembrolizumab 200 mg every 3 weeks (n=580) or placebo (n=581). The patients were stratified by receipt of up to four cycles of adjuvant platinum-based chemotherapy, disease stage, PD-L1 tumor proportion score (TPS), and geographic region.
Of note, the majority of both the pembrolizumab and control arms had nonsquamous histology (67.5 vs 61.8%), stage II disease (55.8 vs 57.6%), and received adjuvant chemotherapy (85.8 vs 85.9%). The PD-L1 TPS distribution was also similar for <1% (39.5 vs 39.5%), 1–49% (32.0 vs 32.4%), and ≥50% (28.5 vs 28.1%).
The pembrolizumab and control arms received a median of 17 and 18 doses of treatment, respectively, and were followed-up for a median of 35.6 months, reported Paz-Ares.
The coprimary endpoint of DFS in the overall population was significantly longer with pembrolizumab than placebo, at a median of 53.6 versus 42.0 months and a hazard ratio (HR) of 0.76, and the 18-month DFS rates were 73.4% versus 64.3%.
However, the coprimary endpoint of DFS in patients with a PD-L1 TPS of at least 50% was comparable in the pembrolizumab and placebo arms, with the median duration unreached in both and a nonsignificant HR of 0.82. The 18-month DFS rates were 71.7% versus 70.2%.
Paz-Ares noted that subgroup analysis of the overall population indicated that pembrolizumab was beneficial across most subgroups but this was not true for patients who did not receive adjuvant chemotherapy and those with squamous histology, albeit the confidence intervals for these subgroups were “wide.”
Interim analysis of overall survival was immature and did not show a significant difference between the pembrolizumab and placebo arms, with median durations unreached in both and 18-month rates of 91.7% and 91.3%, respectively.
The presenter said the safety profile for pembrolizumab was “as expected,” with grade 3–5 adverse events (AEs) occurring in 34.1% of pembrolizumab-treated patients and 25.8% of controls. Treatment-related deaths occurred in a corresponding 0.7%, including two cases of myocarditis, and 0.0% of the treatment arms.
Pembrolizumab was also associated with a higher rate of serious AEs (24.5 vs 15.5%) and AEs leading to discontinuation (19.8 vs 5.9%) and interruption (38.1 vs 25.0%). The presenter highlighted the higher rate of immune-mediated AEs and infusion reactions with pembrolizumab, mostly at grade 1–2, including hypothyroidism (20.7 vs 4.6%), hyperthyroidism (10.7 vs 2.9%), and pneumonitis (6.9 vs 2.9%).
“Pembrolizumab provided statistically significant, clinically meaningful DFS improvement versus placebo in the overall population,” concluded Paz-Ares.
He added that both DFS in the PD-L1 defined patient groups and OS will be “tested at future analyses according to the analysis plan.”
Session discussant Martin Reck, from the LungenClinic Grosshansdorf in Germany, noted that the DFS outcomes of the PEARLS trial were “not so different” from those of the IMpower010 trial of adjuvant atezolizumab for resectable NSCLC.
However, he noted that more data are required to determine the impact of PD-L1 expression for adjuvant checkpoint inhibitor use, as well as to assess the importance of EGFR mutation status, the use of adjuvant chemotherapy and neoadjuvant therapies, and the application of ctDNA status as a biomarker.
Finally, Reck said that further research is required to confirm if there is a correlation between DFS and OS for adjuvant immuno- and targeted treatments.
medwireNews is an independent medical news service provided by Springer Healthcare Ltd. © 2022 Springer Healthcare Ltd, part of the Springer Nature Group
ESMO Virtual Plenary; 17–18 March 2022
Ann Oncol 2022; doi: 10.1016/j.annonc.2022.02.224