Long-Term Survival by Number of Immune Checkpoint Inhibitors in PD-L1–Negative Metastatic NSCLC: A Systematic Review and Meta-Analysis
A partir d'une revue systématique de six essais cliniques randomisés incluant au total 1 684 patients atteints d'un cancer du poumon non à petites cellules n'exprimant pas PD-L1 et de stade métastatique, cette méta-analyse évalue, en fonction du sous-type histologique de la tumeur, l'efficacité d'une chimiothérapie en combinaison avec 1 ou 2 inhibiteurs de point de contrôle immunitaire
Introduction : Immune checkpoint inhibitors (ICI) with or without chemotherapy is the current standard of care first-line treatment for patients with metastatic non–small cell lung cancer (NSCLC) without actionable variants. However, patients with negative programmed death-ligand 1 (PD-L1) status are known to experience worse survival outcomes compared with those with positive PD-L1 status despite the addition of ICI to chemotherapy.
With multiple front-line ICI treatment regimens available, the optimal treatment regimen for PD-L1–negative patients has yet to be defined. It is also unclear whether cytotoxic T-lymphocyte–associated protein inhibition, in addition to targeting PD-1/PD-L1 axis, could yield further survival benefits. With the availability of long-term survival data from several phase 3 studies evaluating ICI plus chemotherapy in patients with advanced NSCLC, we studied whether there was a difference between the different regimens in treating patients with PD-L1–negative NSCLC.
Methods : This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline, with the study review and selection process described in eFigure 1 in Supplement 1. We included phase 3 randomized clinical trials (RCTs) from 2022 to 2024 that reported 5-year overall survival (OS) or progression-free survival (PFS) of patients with advanced or metastatic PD-L1–negative NSCLC who received an ICI-containing regimen compared with chemotherapy alone. Patients with indeterminate PD-L1 status were excluded from the analysis. Pooled hazard ratios (HR) for OS and PFS and 95% CIs were calculated using the generic inverse variance method. Subgroup analysis based on histological subtype for OS and PFS was further performed if at least 3 eligible studies were found. Cochran Q test was used to determine statistical heterogeneity, with 2-sided ; values less than .05 indicating statistical significance for heterogeneity. R version 4.3.2 (R Project for Statistical Computing) and meta package version 7.0-0 were used for analysis.
Results : Six RCTs consisting of 1684 patients met the criteria and were included in the meta-analysis.1-6 A total of 1480 and 1228 events were observed for OS and PFS, respectively. The study characteristics, along with the study quality assessment, are summarized in the Table. Patients who received ICIs exhibited superior OS and PFS compared with those who did not (pooled HR, 0.75; 95% CI, 0.66-0.85; I2 = 41% and HR, 0.72; 95% CI, 0.64-0.81; I2 = 0%, respectively). Doublet ICI had a lower HR for OS than single-agent ICI, but the results were not statistically significant (pooled HR, 0.69; 95% CI, 0.60-0.79 vs HR, 0.80; 95% CI, 0.66-0.95; P for subgroup differences = .21) (Figure, A and B). Interestingly, among studies evaluating single-agent ICI plus chemotherapy, KEYNOTE 189 had better OS compared with other studies (HR 0.55; 95% CI, 0.39-0.77). Leave-one-out analysis excluding KEYNOTE-189 resulted in reduced OS benefits (pooled HR 0.86; 95% CI, 0.75-0.99; I2 = 0%).
PFS was not different between the doublet ICI and the single ICI subgroup (pooled HR, 0.73; 95% CI, 0.61-0.88 vs HR, 0.71; 95% CI, 0.61-0.83; P for subgroup differences = .81). Egger tests did not show the presence of publication bias. The Grading of Recommendations, Assessment, Development, and Evaluations assessment showed moderate certainty of evidence.
Stratifying by histological subtype, we analyzed 3 RCTs on patients with squamous cell lung cancer. We found that doublet ICI (pooled HR, 0.52; 95% CI, 0.33-0.81), but not singlet ICI plus chemotherapy (pooled HR, 0.85; 95% CI, 0.66-1.10), was associated with improved OS.3,5 In contrast, patients with nonsquamous histological subtypes attained similar OS benefits from either doublet or singlet ICI (Figure, C and D).3,4,6
Discussion : There is limited data on the outcomes of patients with negative PD-L1 expression and, to our knowledge, this is the first meta-analysis to address this question. Our analysis showed that the OS benefit from KEYNOTE-189 in patients with PD-L1–negative NSCLC was an outlier compared with similar studies. This is possibly due to differences in patient characteristics, such as inclusion of only nonsquamous histology, more patients with brain metastasis, and having less patient crossover to the ICI group upon disease progression. Other limitations of this meta-analysis included the small number of studies and the heterogeneity of the treatment regimen, particularly the difference in the number of chemotherapy cycles and the use of pemetrexed maintenance therapy in patients with nonsquamous histology. Also, doublet ICI regimen had better OS outcomes in patients with PD-L1–negative squamous cell lung cancer. Our findings suggest that doublet ICI regimens may be beneficial in patients with PD-L1–negative squamous cancer, while singlet ICI regimens may be preferred in nonsquamous histology. Further RCTs are needed to confirm the potential OS benefit of doublet ICI in the squamous histology subpopulation.
JAMA Network Open 2024