Supplementary Material for: A real-world evaluation of brain imaging in patients with advanced NSCLC — insights from a German claims data analysis

Introduction Brain metastases are a common risk in non-small-cell lung cancer (NSCLC) patients. Generally, brain metastases are associated with reduced survival rates, and brain imaging is recommended during tumour staging. Here, we assess the incidence and survival among patients with non-curative advanced NSCLC (aNSCLC) with brain metastases, as well as the use and timing of brain imaging based on German claims data. Methods Our study was based on claims data from about 4.5 million patients in Germany who are insured via statutory health insurance. 28,578 lung cancer patients were identified via ICD-10 code C34 during the study period from 2015 to 2021. Among them, patients with aNSCLC were determined using specific ICD-10 codes in combination with NSCLC disease- and stage-specific treatments. Brain metastases were identified via ICD-10 codes, while brain imaging was detected via OPS (operation and procedure classification system) and EBM (German Uniform Assessment Standard) codes. Results The study population comprised 8,111 patients with aNSCLC, of which 7,010 patients (86.4%) had metastatic NSCLC and 1,101 patients (13.6%) locally advanced NSCLC not amenable to curative-intended local therapy. The median overall survival (OS) of the entire study population was 9.2 months (95% CI: 8.9–9.6 months), of which 20.4% presented with brain metastases at initial diagnosis. Patients with brain metastases at diagnosis had a median OS of 6.1 months (5.7–6.6 months), as opposed to patients without brain metastases at diagnosis with a median OS of 9.5 months (9.0–9.9 months). Of the whole study population, 5,394 patients (66.5%) had received brain imaging at initial NSCLC diagnosis. Brain imaging was more common among patients with mNSCLC compared to those with locally advanced NSCLC (69.2% vs 49.5%). The diagnostic setting (inpatient vs. outpatient, rural vs. urban location) had no apparent impact on the usage of brain imaging. Conclusion Our study provides valuable insights into the routine practice of brain imaging in NSCLC patients in Germany. Our findings align with published evidence regarding incidence rates and mortality, again emphasising the prognostic relevance of brain metastases. Therefore, brain imaging at diagnosis should be used more commonly to detect brain metastases early on, which appeared insufficiently used in routine practice.