1332P - Effects of immune checkpoint inhibitors for EGFR-wild/ALK-negative NSCLC patients with untreated brain metastases

Background

Significant survival benefit of immune checkpoint inhibitors (ICIs) had been reported for NSCLC patients regardless of the presence or absence of brain metastases (BMs)(NEJM2018). However, the detailed efficacy of ICIs on BMs remains unclear. The aim of this study was to clarify the effect of ICI on CNS control.

Methods

The clinical course of consecutive patients with BMs derived from NSCLC without sensitive EGFR nor ALK mutations treated in our hospital were retrospectively evaluated. The patients were divided into two groups depending on whether ICI was administrated as first-line treatment for BMs. The survival benefit of ICI was verified through competing risk analysis using CNS death, non-CNS death, and other death as competing risks. For verification of the effect of ICI on CNS control, CNS progression and death were used as competing risks. The predictive value of PD-L1 TPS was also evaluated.

Results

Of the 677 patients with BMs, 331 exhibited EGFR mutations and 21 had ALK fusion gene. Among the remaining 325 patients, 57 were treated with ICI as first-line treatment for BMs: pembrolizumab in 43, atezolizumab in 7, nivolumab in 6, and durvalumab in one case. ICIs improved the overall survival of patients (HR:0.44,95%CI:0.31-0.65, P<0.0001) and prolonged the time to non-CNS death (HR:0.44,95%CI:0.29-0.68, P<0.0001). In contrast, ICI did not prolong the time to CNS progression nor time to CNS death. PD-L1 TPS had measured in 135 cases. In the subgroup with TPS ≥50%, ICI significantly prolonged the time to CNS progression (HR:0.47, 95%CI:0.23-0.96, P=0.039) and no CNS deaths were observed in this group after ICI treatment. These effects of ICI were not observed in the subgroup with TPS <1% nor in the subgroup with TPS of 1-49%.

Conclusions

ICI was effective in controlling intracranial lesions in some patient populations. PD-L1 TPS predicted the efficacy of ICI in controlling intracranial lesions, but the threshold was estimated to be higher than that for extracranial lesions.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

JSPS KAKENHI Grant (grant no. 22K09272).

Disclosure

All authors have declared no conflicts of interest.