A NOVEL APPROACH FOR GENERATING ROBUST POPULATION ESTIMATES FOR USE IN HTA AND REGULATORY SUBMISSIONS: DEMONSTRATION IN CLL IN GERMANY
Author(s)
Kai Strobel, MSc, Zuzana Dostalova, MSc, Patricia Linke, MSc, Stefanie Spirin, MSc, Ewelina Kubietz, MSc, Maria Friese, BSc, Franziska Haug, MSc, Amanda Moore, PharmD, PhD, Jeffrey Brown, PhD, Markus Rückert, PhD.
TriNetX, LLC, Cambridge, MA, USA.
TriNetX, LLC, Cambridge, MA, USA.
OBJECTIVES: The goal was to validate a two-step methodology for generating representative real-world epidemiological estimates in oncology, demonstrated using chronic lymphocytic leukemia (CLL) in Germany. These estimates are critical for interpreting study findings by providing robust epidemiological estimates and to evaluate real-world data (RWD) representativeness.
METHODS: We implemented a 2-step process using public epidemiologic data to describe CLL treatment landscape and targeted epidemiologic surveys to assess national treated prevalence and incidence. Centers were grouped into A/B/C categories, ranking them by patient volume. The top two quartiles (A/B; ~80% of patients) were defined as care relevant. To determine healthcare segment contribution to CLL treatment, institutions were categorized into university hospitals (UH), non-university hospitals (NUH), and office-based practices (OBP). Targeted epidemiological surveys were conducted at a representative sample of care-relevant centers. Data were proportionally sampled and weighted by facility type to extrapolate national treated prevalence and incidence.
RESULTS: The analysis identified 321 CLL-treating centers: 18 UHs (6%), 43 NUHs (13%), and 260 OBPs (81%). ABC analysis indicated that 51% of hospitals treated 80% of hospitalized patients. Surveys from 61 centers (1,631 patients) yielded a national treated prevalence of 11,382-12,189 patients (13.66-14.69 per 100,000) and incidence of 3,388-3,657 patients (4.06-4.39 per 100,000). Distribution was UH 6%, NUH 10%, OBP 84%. Representative findings were validated via chart review (n=682). Hospital distributions aligned with German Federal Joint Committee data. Incidence aligned with local projections, which reflect diagnosed cases, while our approach estimates treated prevalence and incidence that drive resource allocation, cost-effectiveness modeling, and RWD interpretation.
CONCLUSIONS: This healthcare structure-based methodology provides robust, reproducible epidemiological estimates suitable for HTA and regulatory submissions. Systematic characterization and proportional sampling ensure representativeness, and multi-source validation supports accuracy. This method is scalable across oncology indications and geographies, offering a standardized framework for fit-for-purpose real-world evidence generation.
METHODS: We implemented a 2-step process using public epidemiologic data to describe CLL treatment landscape and targeted epidemiologic surveys to assess national treated prevalence and incidence. Centers were grouped into A/B/C categories, ranking them by patient volume. The top two quartiles (A/B; ~80% of patients) were defined as care relevant. To determine healthcare segment contribution to CLL treatment, institutions were categorized into university hospitals (UH), non-university hospitals (NUH), and office-based practices (OBP). Targeted epidemiological surveys were conducted at a representative sample of care-relevant centers. Data were proportionally sampled and weighted by facility type to extrapolate national treated prevalence and incidence.
RESULTS: The analysis identified 321 CLL-treating centers: 18 UHs (6%), 43 NUHs (13%), and 260 OBPs (81%). ABC analysis indicated that 51% of hospitals treated 80% of hospitalized patients. Surveys from 61 centers (1,631 patients) yielded a national treated prevalence of 11,382-12,189 patients (13.66-14.69 per 100,000) and incidence of 3,388-3,657 patients (4.06-4.39 per 100,000). Distribution was UH 6%, NUH 10%, OBP 84%. Representative findings were validated via chart review (n=682). Hospital distributions aligned with German Federal Joint Committee data. Incidence aligned with local projections, which reflect diagnosed cases, while our approach estimates treated prevalence and incidence that drive resource allocation, cost-effectiveness modeling, and RWD interpretation.
CONCLUSIONS: This healthcare structure-based methodology provides robust, reproducible epidemiological estimates suitable for HTA and regulatory submissions. Systematic characterization and proportional sampling ensure representativeness, and multi-source validation supports accuracy. This method is scalable across oncology indications and geographies, offering a standardized framework for fit-for-purpose real-world evidence generation.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
RWD55
Topic
Real World Data & Information Systems
Topic Subcategory
Reproducibility & Replicability
Disease
SDC: Oncology