Relationship between PD-L1 Expression and Outcomes of Salvage Treatment Following Durvalumab Consolidation

Disease progression during or after consolidation treatment with durvalumab, an immune-oncology agent (IO) for non-small cell lung cancer, confers a poor prognosis. Systemic therapy is the mainstay. Currently, standard first-line systemic therapy includes IO; however, in the post-durvalumab setting, it remains unclear if IO should be used again.

We performed an analysis of a nationwide, deidentified database of patients with stage III non-small cell lung cancer who received concurrent chemoradiation, durvalumab consolidation, and salvage systemic therapy. Overall survival, measured from salvage treatment, was compared between salvage IO or IO-based regimens (ie IO regimens) and non-IO regimens. Predictive factor of PD-L1 level was demonstrated by effect modification.

Analyses included 104 patients: 49 patients (47%) received salvage IO regimens and 55 (53%) received non-IO regimens. The median overall survival was 12.7 months: 13.4 months with IO regimens versus 9.5 months with non-IO regimens, P = .27. Among patients with PD-L1 < 1% (n = 41), survival was numerically worse with IO regimens than with non-IO regimens: hazard ratio (HR) 1.88 (95% CI, 0.81-4.39, P = .14). However, among those with PD-L1 ≥ 1% (n = 63), survival was better with IO regimens than non-IO regimens: HR 0.48 (95% CI, 0.25-0.93, P = .03). The association between regimens and survival differed by PD-L1 level, P-interaction =.007. Poor performance status and lower socioeconomic index were significant adverse prognostic factors.

In post-durvalumab setting, this analysis suggests that salvage IO regimens are superior to non-IO regimens, but only among patients with PD-L1 ≥ 1%.