Study: Overall Patient Survival Estimates for NSCLC Are Transferable From US to Canada

Patient overall survival (OS) estimates for advanced non–small cell lung cancer (NSCLC) from US-based clinical practices could validate decision making for physicians in Canada, according to the results of a study published in JAMA Network Open.

The results showed that, under specific circumstances and criteria, OS estimated from clinical practices in the United States can be adjusted, using baseline clinical characteristics, to approximate OS in selected populations of patients in Canada who have advanced NSCLC.

Investigators used transportability analysis to assess if the adjustment for pretreatment characteristics of eligible patient cohorts could approximate OS from patients in the United States to those in Canada. They found data on 17,432 individuals who were diagnosed with advanced NSCLC on or after January 1, 2011, who were followed up with until September 30, 2020.

Ethnicity and race were available in the United States database but not in the Canadian database, so these characteristics were not analyzed.

Among the 17,432 individuals, 15,669 were from the United States and 1763 were from Canada. Of those, 11,863 were included in the subset of individuals with complete data for baseline covariates.

Investigators reported that 13,532 individuals from the United States received first-line chemotherapy and 2137 received first-line pembrolizumab monotherapy. Approximately 62.4% of individuals in the chemotherapy group and 77.3% in the pembrolizumab group had complete data on baseline covariates for the outcome estimations.

Additionally, 1476 individuals from Canada received first-line chemotherapy, and 287 received first-line pembrolizumab monotherapy and were identified from the target population.

Investigators used standardization to baseline patient covariates in the Canadian cohorts. They found that with the OS estimates there was less than a 5% mean absolutely difference from the observed OS in the target population, with 0.56% over 60 months of follow-up for the first-line chemotherapy group and 4.54% over 30 months of follow up in the first-line pembrolizumab group.

The negative control analysis, which used a mismatched outcome model, showed a 6.64% discrepancy and an incompatible survival curve shape, according to investigators.

The results showed robust assumptions of random missingness for baseline covariates, as well as unadjusted differences in baseline comorbidities and metastases. Additionally, the results showed differences in the standard of care between individuals in Canada and the United States, which were related to administration of second-line anti-programmed cell death 1 ligand 1 immunotherapy for individuals who had first-line chemotherapy.

Investigators reported that the study had limitations, which included that model misspecification could be a bias for parametric outcome models. Additionally, the analyses were conducted among a selected group of individuals, so the results might not be generalizable to individuals with recurrent disease or other disease sites, drug indications, or geographic regions.

Other unmeasured factors, such as PF-L1 expression for the first-line pembrolizumab cohort, could not be ruled out, investigators said.

References

Ramagopalan SV, Popat S, Gupta A, et al. Transportability of overall survival estimates from US to Canadian patients with advanced non–small cell lung cancer with implications for regulatory and health technology assessment. JAMA Netw Open. 2022;5(11):e2239874. doi:10.1001/jamanetworkopen.2022.39874