Real-World Study of Aumolertinib in Patients With Advanced EGFR T790M-Positive NSCLC

Lung cancer is one of the leading causes of cancer burden worldwide, and non-small cell lung cancer accounts for most lung cancer cases. Sensitizing EGFR mutations are common oncogenic drivers among Chinese patients with advanced non-small cell lung cancer. First- and second-generation EGFR tyrosine kinase inhibitors provide substantial clinical benefit, but acquired resistance is nearly inevitable. EGFR T790M is an important resistance mechanism after progression on these agents. Third-generation EGFR tyrosine kinase inhibitors that selectively inhibit sensitizing EGFR mutations and T790M have therefore become an important subsequent treatment option for this population.

Aumolertinib is a third-generation EGFR tyrosine kinase inhibitor. The single-arm phase II APOLLO registrational study enrolled 244 patients with EGFR T790M-positive locally advanced or metastatic non-small cell lung cancer after progression on a prior EGFR tyrosine kinase inhibitor. The study reported an objective response rate of 68.9%, a disease control rate of 93.4%, and a median progression-free survival of 12.4 months. With longer follow-up, median overall survival was 30.2 months and the 24-month overall survival rate was 57.5%. The safety profile was considered acceptable in the trial setting. These findings supported conditional marketing authorization for this indication in China.

Although APOLLO provided pivotal evidence of clinical benefit, its single-arm design and restrictive eligibility criteria limit direct generalization to routine practice. Older patients and those with poorer performance status, comorbidities, organ dysfunction, complex metastatic patterns, or different prior treatment pathways are more frequently encountered in the real world. In addition, the frequency of imaging assessment, dose modification, treatment interruption, subsequent therapy, and completeness of follow-up may differ from those in a registrational trial. Real-world outcomes may therefore differ from trial results. A large real-world study with long-term follow-up can provide further evidence on the effectiveness and safety of aumolertinib and can also inform assessment of health care utilization, medical costs, and the value of health insurance reimbursement.

This is a population-based, retrospective, observational, single-cohort study using health care data from Jiangsu Province, China. The China Health and Medical Big Data Center (East) covers outpatient and inpatient care for approximately 80 million urban and rural residents and includes demographic characteristics, diagnoses, imaging and laboratory findings, clinical and pathology notes, prescriptions and medication orders, health care institutions, medical costs, and health insurance reimbursement. Broad screening will identify patients with a lung cancer-related code (C34 or D02.200) and a record of aumolertinib between January 1, 2020, and December 31, 2024. Final eligibility will be confirmed from pathology or cytology, disease stage, prior treatment, disease progression, and molecular testing records. Within a secure environment, patients will be linked through a unique card number to the Jiangsu Province Resident Death Registration Database to determine vital status through January 1, 2025. The study will not assign treatment, contact patients, or alter clinical care.

Time zero is the first confirmed aumolertinib prescription, executed medication order, or administration. Eligibility assessment, treatment initiation, and the start of outcome follow-up will be aligned at time zero. The primary outcome is overall survival, defined as the time from time zero to death from any cause. Secondary outcomes include real-world progression-free survival and adverse events. Real-world disease progression will be ascertained from treating clinician documentation, radiology reports, pathology findings, and longitudinal clinical records. An adverse event will be described as treatment-related only when causality is explicitly documented in the medical record.

Using target trial emulation principles, the study will prespecify and align eligibility criteria, treatment strategy, time zero, follow-up, outcomes, and analysis. An indication-aligned new-user cohort and a stricter trial-aligned cohort will be analyzed separately. Overall survival and real-world progression-free survival will be described using Kaplan-Meier methods, survival probabilities at fixed time points, and restricted mean survival time. For the trial-aligned cohort, where common variable definitions and data quality permit, an unanchored matching-adjusted indirect comparison will align measured characteristics with published APOLLO baseline distributions. Weight distributions, effective sample size, and residual imbalance will be reported. The comparison with APOLLO is a trial-to-practice benchmark intended to assess consistency and transportability of outcomes; it is not a randomized causal comparison of aumolertinib versus another treatment.