AIM

To investigate the impact of PD-L1 expression status on consolidative durvalumab efficacy and safety in stage III NSCLC patients.

METHODS

This retrospective, ethics board approved, multi-centre study included all patients with histologically and/or cytologically confirmed unresectable stage III NSCLC, EGFR/ALK wild-type patients who completed concurrent chemoradiation therapy (cCRT) from January 2018 to August 2020 at the Cancer Centre of Southeastern Ontario and The Ottawa Hospital Cancer Centre. PD-L1 status was grouped as ≥50% vs. 1-49% vs. <1%. Primary study endpoint was overall survival (OS).

RESULTS

Of the total 63 patients, 27 (43%), 16 (25%), 8 (13%), and 12 (19%) patients in the PD-L1 ≥50%, PD-L1 1-49%, PD-L1 <1%, and PD-L1 unknown groups (reported separately), respectively. With the median follow-up of 17.0 months, our multivariable Cox analysis suggested PD-L1≥50% was independently associated with improved OS compared to PD-L1<1% group (HR 0.18, 95%CI 0.04-0.86, P = 0.03). There were no significant differences in OS outcomes between PD-L1 1-49% and PD-L1 <1% group (HR 0.36, 95%CI 0.08-1.63, P = 0.18). Any grade treatment-related pneumonitis was observed in 46% of patients. Sixty-two percent, 38%, and 18% of patients required systemic corticosteroid use, hospitalization, and permanent treatment discontinuation due to pneumonitis, respectively.

CONCLUSIONS

Our multi-centre, real-world study supported the use of consolidative durvalumab in inoperable stage III PD-L1 expressing NSCLC. Pneumonitis was significant, and higher than expected. Benefit appeared greater in high PD-L1 expressing patients, consistent with the subgroup analysis from the landmark PACIFIC trial. Our results need to be interpreted with cautions due to small sample size and a relatively short follow-up duration.