Durvalumab Versus Chemotherapy as First-line Treatment for Metastatic NSCLC With Tumor PD-L1 Expression of 25% or Higher: Results From the Randomized Phase 3 PEARL Study.

Background: PEARL (NCT03003962) is an open-label, phase 3 study comparing first-line durvalumab monotherapy with chemotherapy in patients with metastatic NSCLC (mNSCLC [EGFR/ALK wild type]) with programmed cell death ligand 1 (PD-L1) tumor cell (TC) membrane expression status of 25% or higher. We report the final analysis of PEARL.

Methods: Adults (N = 669) with previously untreated stage IV mNSCLC were randomized (1:1) to durvalumab 20 mg/kg every four weeks or chemotherapy every three weeks for four to six cycles. The dual primary endpoints were overall survival (OS) in the population with PD-L1 TC of 25% or higher and OS in the population at low risk of early mortality (LREM) with PD-L1 TC of 25% or higher.

Results: Durvalumab was associated with a numerical reduction in the risk of death versus chemotherapy in the 25% and higher PD-L1 TC population (OS hazard ratio [HR] = 0.84, 95% confidence interval [CI]: 0.71-0.99, p = 0.037; median OS 14.6 months, 95% CI: 12.2-16.9 versus 12.8 months, 95% CI: 10.1-14.7, respectively). In the 25% and higher PD-L1 TC low risk of early mortality population the OS hazard ratio for durvalumab versus chemotherapy was 0.96 (95% CI: 0.79-1.15, p = 0.628); median OS 14.6 months (95% CI: 12.6-17.2) versus 15.0 months (95% CI: 13.1-16.8), respectively. In the safety population, the incidence of grade 3 or 4 treatment-related adverse events was 15.5% (durvalumab) and 45.9% (chemotherapy).

Conclusions: Durvalumab did not statistically significantly improve OS versus chemotherapy as first-line treatment in patients with mNSCLC and 25% and higher PD-L1 TC. The numerical improvement in OS was consistent with previous studies of first-line immune checkpoint inhibitor monotherapy in patients with mNSCLC.