Longer OS in Asian than non-Asian patients given atezolizumab plus chemotherapy for extensive-stage SCLC

Before matching, more never-smokers and fewer liver metastases in the Asian cohort

The 363 Asian patients also included more men and more patients with an Eastern Cooperative Oncology Group performance status of 1, and had a lower body mass index than the 344 non-Asian patients. All had received at least one dose of atezolizumab with carboplatin and etoposide in IMpower133, SKYSCRAPER-02, SKYSCRAPER-02C or the fourth pooled trial.

In this unmatched population, median OS was 15.3 months for Asian patients and 12.2 months for non-Asian patients (HR 0.80; p = 0.008). These baseline differences can affect survival, so the comparison was repeated after propensity score matching by race.

A median OS difference of 5.1 months after matching

With 169 patients in each group, median OS was 16.4 months in Asian patients and 11.3 months in non-Asian patients, for a hazard ratio of 0.70 (p = 0.004).

After matching, the hazard ratio was 0.70 against 0.80 before, and the difference in median OS was 5.1 months against 3.1 months, so matching did not remove the survival difference. The matched groups are subsets of the original cohorts, and the abstract does not report how well the matching balanced them.

Progression-free survival of about five months in both groups

Median PFS after matching was 4.8 months for Asian patients and 5.1 months for non-Asian patients, so the longer survival in the Asian group was not accompanied by longer PFS on first-line treatment.

The abstract does not describe what treatment patients received after progression. It therefore cannot show whether later lines of therapy, or differences in how they were used across regions, contributed to the OS difference seen after a similar time to progression.

More grade 3–4 adverse events in Asian patients, more grade 5 events in non-Asian patients

Almost every patient had at least one adverse event, 98.8% of Asian and 100% of non-Asian patients. Grade 3–4 events occurred in 82.8% of the Asian group against 60.9% of the non-Asian group.

The pattern reversed for fatal events: grade 5 adverse events were reported in 0.6% of Asian and 4.7% of non-Asian patients. The abstract does not give the causes of these deaths.

“This analysis demonstrates the utility of atezolizumab plus chemotherapy for ES-SCLC and longer OS for Asian patients.”Zhang et al., Lung Cancer (2026)

An observational comparison drawn from randomised trials

Every patient in the analysis was treated within a phase 3 trial, but race was not a randomised factor in any of them. The comparison between Asian and non-Asian patients is therefore observational, and propensity score matching can balance only the characteristics that were measured.

The authors’ rationale is that differences in tumour biology, the tumour microenvironment and immunology can make race affect outcomes. The abstract reports no direct measure of these, so the analysis describes an association between race and survival on atezolizumab plus chemotherapy, and it does not identify the mechanism behind it.