RESUMO
When evaluating the real-world treatment effect, the analysis based on randomized clinical trials (RCTs) often introduces generalizability bias due to the difference in risk factors between the trial participants and the real-world patient population. This problem of lack of generalizability associated with the RCT-only analysis can be addressed by leveraging observational studies with large sample sizes that are representative of the real-world population. A set of novel statistical methods, termed "genRCT", for improving the generalizability of the trial has been developed using calibration weighting, which enforces the covariates balance between the RCT and observational study. This paper aims to review statistical methods for generalizing the RCT findings by harnessing information from large observational studies that represent real-world patients. Specifically, we discuss the choices of data sources and variables to meet key theoretical assumptions and principles. We introduce and compare estimation methods for continuous, binary, and survival endpoints. We showcase the use of the R package genRCT through a case study that estimates the average treatment effect of adjuvant chemotherapy for the stage 1B non-small cell lung patients represented by a large cancer registry.
RESUMO
BACKGROUND: The selection of later-line treatment for older patients with AJCC (version 7) stage IV non-small cell lung cancer (NSCLC) remains controversial. Nanoparticle albumin-bound (nab)-paclitaxel is approved with carboplatin for the first-line treatment of patients with NSCLC and subgroup analysis of phase 3 data has suggested superior survival in older patients. METHODS: The authors conducted a phase 2 study of nab-paclitaxel in 42 patients aged ≥70 years who had been treated previously with a platinum doublet regimen; patients also could have received a PD-1 inhibitor. The primary endpoint of the current study was grade 3 to 5 toxicity (according to the National Cancer Institute Common Terminology Criteria for Adverse Events [version 4.0]). In addition to response rate, progression-free survival (PFS), and overall survival (OS), geriatric assessments also were performed before and during treatment, associations between baseline sarcopenia and outcomes were explored, and changes in T lymphocyte p16 before and during treatment were measured. The authors also performed a retrospective subgroup analysis of 19 older patients who were treated with nab-paclitaxel as part of a larger, randomized, phase 2 study; data were not combined. RESULTS: The rate of grade 3 to 5 toxicities was 33.7%. The most common grade 3 to 5 toxicities were decreased white blood cell count (11.9%), neutropenia (9.5%), and fatigue (11.9%). The response rate was 34.2% (2.6% complete response rate and 31.6% partial response rate). The median PFS was 5.2 months and the median OS was 9.3 months. Adverse prognostic factors were common: 42% of patients were frail and 39% of patients were prefrail, whereas 21% had an Eastern Cooperative Oncology Group performance status of 2 and 27% were sarcopenic. Only frailty was found to be predictive of inferior survival. A subgroup analysis of 19 older patients treated with nab-paclitaxel alone in a prior trial demonstrated a response rate of 15.8%, a PFS of 4.2 months, and an OS of 13.6 months. CONCLUSIONS: Fit and prefrail older patients with stage IV NSCLC should be considered for treatment with nab-paclitaxel after disease progression with doublet chemotherapy.