CHARACTERISTICS, TREATMENT PATTERNS AND CLINICAL OUTCOMES OF ADVANCED NSCLC PATIENTS RECEIVING SECOND-LINE PLUS TREATMENT IN GERMANY: FIRST EXPERIENCES WITH THE HEALTH DATA LAB
Author(s)
Elena E. Möhrmann, MD1, Gina Rüter, MD1, Julia Lovera, MSc2, Peter Gray, MSc2, Maciej Rowinski, MSc2, Agnes Kisser, PhD1, Ariane Höer, MD3, Frank Griesinger, Prof. Dr. Med4, Esther Denecke, PhD1, Heike Aichinger, PhD1, Caoimhe Cawley, PhD1, Eveline Otte im Kampe, MSc5.
1Pfizer Pharma GmbH, Berlin, Germany, 2IGES GmbH, Berlin, Germany, 3IGES Institut, Berlin, Germany, 4Pius Hospital, Oldenburg, Germany, 5Pfizer, Berlin, Germany.
1Pfizer Pharma GmbH, Berlin, Germany, 2IGES GmbH, Berlin, Germany, 3IGES Institut, Berlin, Germany, 4Pius Hospital, Oldenburg, Germany, 5Pfizer, Berlin, Germany.
OBJECTIVES: The objective of this study is to describe the characteristics, treatment patterns and clinical outcomes of patients with advanced non-small cell lung cancer (aNSCLC) receiving second-line plus (2L+) therapy, using the German Health Data Lab (HDL) as a novel healthcare database.
METHODS: As of October 2025, the HDL provides access to claims data covering ~73 million insured individuals (~90% of the German population). Adults (≥18 years) continuously insured from 2019-2023 with incident aNSCLC initiating 2L+ therapy between 2020-2023 were identified via proxies, using ICD-10 German Modification diagnosis codes and treatment-based algorithms. Patient characteristics (e.g. age, sex, location of metastases, histology), first-line to third-line treatment patterns (proportions receiving immunotherapy, chemotherapy with/without immunotherapy, targeted therapies), and overall survival from 2L therapy initiation are described for the study population as a whole, as well as stratified by type of 2L therapy (targeted versus non-targeted). In this submission we describe first experiences with accessing and analysing HDL data.
RESULTS: Analysis scripts were developed within the Secure Data Processing Environment, based on synthetic use files (SUF) provided by the HDL. Preliminary results indicate that the study population can be identified via the defined proxies. However, working with the SUF has been associated with challenges, including unstable access, data quality issues, and that not all relationships in the original data are reproduced in the synthetic dataset. Analysis scripts will be further refined and run on the real-world pseudonymised datasets to extract final results.
CONCLUSIONS: This analysis will provide population-level RWE on patient characteristics, treatment patterns and outcomes in aNSCLC 2L+ therapy in Germany. The HDL is a promising new data source, however, initial experience shows that additional time must be planned in for such projects.
METHODS: As of October 2025, the HDL provides access to claims data covering ~73 million insured individuals (~90% of the German population). Adults (≥18 years) continuously insured from 2019-2023 with incident aNSCLC initiating 2L+ therapy between 2020-2023 were identified via proxies, using ICD-10 German Modification diagnosis codes and treatment-based algorithms. Patient characteristics (e.g. age, sex, location of metastases, histology), first-line to third-line treatment patterns (proportions receiving immunotherapy, chemotherapy with/without immunotherapy, targeted therapies), and overall survival from 2L therapy initiation are described for the study population as a whole, as well as stratified by type of 2L therapy (targeted versus non-targeted). In this submission we describe first experiences with accessing and analysing HDL data.
RESULTS: Analysis scripts were developed within the Secure Data Processing Environment, based on synthetic use files (SUF) provided by the HDL. Preliminary results indicate that the study population can be identified via the defined proxies. However, working with the SUF has been associated with challenges, including unstable access, data quality issues, and that not all relationships in the original data are reproduced in the synthetic dataset. Analysis scripts will be further refined and run on the real-world pseudonymised datasets to extract final results.
CONCLUSIONS: This analysis will provide population-level RWE on patient characteristics, treatment patterns and outcomes in aNSCLC 2L+ therapy in Germany. The HDL is a promising new data source, however, initial experience shows that additional time must be planned in for such projects.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
RWD53
Topic
Clinical Outcomes, Real World Data & Information Systems, Study Approaches
Topic Subcategory
Health & Insurance Records Systems
Disease
Oncology