# First‐Line Immune‐Combination Therapy for Driver Gene‐Negative NSCLC With Brain Metastases: Real‐World Outcomes

## TL;DR

This study examines the effectiveness of immune-based therapies for lung cancer with brain metastases, finding that PD-L1 levels predict better outcomes.

## Contribution

The study identifies PD-L1 tumor proportion score as a predictive biomarker for intracranial efficacy in driver gene-negative NSCLC with brain metastases.

## Key findings

- Adding bevacizumab to ICI-based chemotherapy did not improve survival outcomes.
- Local brain metastasis therapy did not enhance survival when combined with first-line ICI-based systemic therapy.
- Extracranial PD-L1 TPS ≥ 50% was associated with significantly better intracranial and overall survival.

## Abstract

Optimal treatment for driver gene‐negative non‐small cell lung cancer (NSCLC) with brain metastases (BM) remains unclear, particularly regarding immune checkpoint inhibitor (ICI)‐based combinations and local BM therapy. Predictive biomarkers for intracranial efficacy are also undefined.

This retrospective study analyzed driver gene‐negative NSCLC patients with BM treated with first‐line ICI‐based systemic therapy (ICI plus chemotherapy [ICI + CT] or ICI + CT plus bevacizumab [ICI + CT + Bev]) from June 2019 to June 2024. The intracranial progression‐free survival (icPFS), progression‐free survival (PFS), and overall survival (OS) were compared between treatment groups and by BM local therapy. The PD‐L1 tumor proportion score (TPS) expression was evaluated for correlation with intracranial efficacy.

A total of 36 patients were enrolled in the study. The intracranial objective response rate (icORR) was 70.6% (ICI + CT) versus 78.6% (ICI + CT + Bev) (p = 0.689), with no significant differences in icPFS, PFS, or OS between the two different first‐line systemic regimens (all p > 0.05). Local BM therapy (n = 18) did not improve icPFS and OS (all p > 0.05). Extracranial PD‐L1 (TPS ≥ 50%, n = 13) correlated with superior icPFS, PFS, and OS (all p < 0.05) versus PD‐L1 TPS < 50%. Multivariate analysis confirmed PD‐L1 ≥ 50% as an independent prognostic factor (HR = 0.155; 95% CI, 0.025–0.939; p = 0.042).

Adding bevacizumab to first‐line ICI‐chemotherapy did not enhance survival outcomes. Local treatment for BM did not provide additional survival advantages when combined with first‐line ICI‐based systemic therapy. Extracranial PD‐L1 TPS ≥ 50% predicted improved intracranial efficacy.

The key question of this study is “what is the optimal first‐line therapy for driver gene‐negative NSCLC with brain metastases (BM), and can PD‐L1 predict intracranial efficacy?” This is a retrospective analysis of 36 patients treated with immune checkpoint inhibitor (ICI) + chemotherapy (CT) (n = 21) and ICI + CT + bevacizumab (Bev) (n = 15). This study evaluated icPFS, PFS, OS, and the impact of local BM therapy (n = 18). PD‐L1 TPS was assessed for correlation with outcomes. No significant difference was seen in icORR (70.6% vs. 78.6%, p = 0.689), icPFS, PFS, or OS between ICI + CT and ICI + CT + Bev. Local BM therapy did not improve survival (p > 0.05). Biomarkers used are PD‐L1 TPS ≥ 50% (n = 13) associated with longer icPFS, PFS, and OS (all p < 0.05) and independent prognostic factor (HR = 0.155, p = 0.042). Adding bevacizumab to ICI + CT did not improve survival. Local BM therapy offered no additional benefit with systemic ICI‐based treatment. PD‐L1 TPS ≥ 50% predicts superior intracranial efficacy.