THE ASSOCIATION BETWEEN EVENT-FREE SURVIVAL (EFS) AND OVERALL SURVIVAL (OS) IN HIGH-RISK NON-MUSCLE-INVASIVE BLADDER CANCER (NMIBC) IN THE NETHERLANDS: AN OBSERVATIONAL STUDY OF ELECTRONIC HEALTH RECORDS
Author(s)
Leo Chen, MPH1, Joseph C. Cappelleri, PhD, MPH2, Jane Chang, MPH3, Julia Brinkmann, MD, MBA4, Hollie Pilkington, MSc5, Sarah Smith, MSc5, Louise Lemmens, MSc6, Sarah Bailey, PhD6, Jetty Overbeek, PhD6, Patricia Silva, PhD7, Meena Venkatachalam, BSc, MSc7, Anthony Eccleston, MSc8.
1Pfizer, Bothell, WA, USA, 2Pfizer, Groton, CT, USA, 3Pfizer, New York, NY, USA, 4Pfizer, Berlin, Germany, 5Lumanity, Manchester, United Kingdom, 6PHARMO Institute for Drug Outcomes Research, Utrecht, Netherlands, 7Lumanity, London, United Kingdom, 8Pfizer, Edinburgh, United Kingdom.
1Pfizer, Bothell, WA, USA, 2Pfizer, Groton, CT, USA, 3Pfizer, New York, NY, USA, 4Pfizer, Berlin, Germany, 5Lumanity, Manchester, United Kingdom, 6PHARMO Institute for Drug Outcomes Research, Utrecht, Netherlands, 7Lumanity, London, United Kingdom, 8Pfizer, Edinburgh, United Kingdom.
OBJECTIVES: 5-year cancer-specific survival in high-risk non-muscle invasive bladder cancer (NMIBC) is relatively high (71 to 96%), so recent trials have used event-free survival (EFS) as the primary endpoint. However, the relationship between EFS and overall survival (OS) in high-risk NMIBC is not well established. This study aimed to leverage observational data to assess the validity of EFS as a surrogate outcome for OS in these patients.
METHODS: An observational study of patients diagnosed with high-risk NMIBC in the Netherlands from 01/01/2016-31/12/2022 was conducted. Data were obtained from the PHARMO Data Network linked to national cancer and pathology registry. High-risk NMIBC was defined according to the European Association of Urology 2019 criteria. EFS and OS were estimated using the Kaplan-Meier (KM) method. The correlation between EFS and OS at the individual level was assessed using a three-state illness-death model and an inverse probability of censoring weights (IPCW) approach to estimate Kendall’s Tau.
RESULTS: A total of 1753 patients were included (mean age 72.5 (SD 9.8) years, 82% male) with a mean follow up of 35.6 (SD 23.6) months. There were 756 EFS events (43%) and 322 deaths (18%). Median EFS was 48.4 (95% CI 39.6 to 58.2) months, while median OS was not reached. The three-state illness-death model yielded a Tau of 0.50 (SE 0.04), and the IPCW approach produced a Tau of 0.56 (SE 0.02).
CONCLUSIONS: Results were consistent across the two analytical approaches and demonstrated a moderate and tangible correlation between EFS and OS in high-risk NMIBC. Patients who remained event-free tended to have longer survival, suggesting that EFS is a modest yet an informative indicator of OS. Further validation with datasets containing longer follow-up is warranted.
METHODS: An observational study of patients diagnosed with high-risk NMIBC in the Netherlands from 01/01/2016-31/12/2022 was conducted. Data were obtained from the PHARMO Data Network linked to national cancer and pathology registry. High-risk NMIBC was defined according to the European Association of Urology 2019 criteria. EFS and OS were estimated using the Kaplan-Meier (KM) method. The correlation between EFS and OS at the individual level was assessed using a three-state illness-death model and an inverse probability of censoring weights (IPCW) approach to estimate Kendall’s Tau.
RESULTS: A total of 1753 patients were included (mean age 72.5 (SD 9.8) years, 82% male) with a mean follow up of 35.6 (SD 23.6) months. There were 756 EFS events (43%) and 322 deaths (18%). Median EFS was 48.4 (95% CI 39.6 to 58.2) months, while median OS was not reached. The three-state illness-death model yielded a Tau of 0.50 (SE 0.04), and the IPCW approach produced a Tau of 0.56 (SE 0.02).
CONCLUSIONS: Results were consistent across the two analytical approaches and demonstrated a moderate and tangible correlation between EFS and OS in high-risk NMIBC. Patients who remained event-free tended to have longer survival, suggesting that EFS is a modest yet an informative indicator of OS. Further validation with datasets containing longer follow-up is warranted.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
CO207
Topic
Clinical Outcomes, Methodological & Statistical Research, Real World Data & Information Systems
Topic Subcategory
Relating Intermediate to Long-term Outcomes
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, Oncology