Characterization of Time-to-Treatment-Discontinuation (TTTD) as a Proxy for Progression-Free Survival (PFS) in Real-World Evidence (RWE) Based Analyses of Relapsed-Refractory Multiple Myeloma (RRMM)
Author(s)
O'Reilly J1, Perkins A2, Tunaru F3, Wallis J3, Carpenter L3
1Arcturis Data, Oxford, OXF, UK, 2Arcturis Data, Kidlington, UK, 3Arcturis Data, Oxford, UK
Presentation Documents
OBJECTIVES: RWE is playing an increasingly important role in assessing the comparative effectiveness and cost effectiveness of novel therapeutics in multiple myeloma (MM). The International Myeloma Working Group (IMWG) define disease progression based on longitudinal changes in laboratory markers, however many RWE databases that have been used as part of regulatory filings have relied on approximate measures of progression-free survival (PFS), such as Time to Treatment Discontinuation (TTTD), due to the lack of laboratory data. The Arcturis Data Platform contains comprehensive therapy and laboratory data, allowing for characterisation of the difference in real-world PFS (rwPFS) and TTTD.
METHODS: This retrospective study used de-identified electronic health records for MM patients diagnosed between 2000 and 2023 from UK NHS partners collated as part of the Arcturis Data Platform. TTTD and rwPFS were used to calculate the delay between progression and treatment discontinuation for each patient. Time-heterogeneity of the accuracy of TTTD was investigated by obtaining distributions of TTTD percentage error stratified by quartiles of progression time.
RESULTS: From a total of 6,749 MM patients, a cohort of 184 RWE RRMM patients treated with either daratumumab or pomalidomide was constructed that matched patients from the GEN501 and SIRIUS trials. The median delay between progression and TTTD is 22 days (-14, 91), with the percentage error of TTTD being greatest amongst patients with the shortest time to progression. This effect decreased in both magnitude and variance amongst patients with longer times to progression.
CONCLUSIONS: TTTD has been used as a proxy for PFS in cost and comparative-effectiveness analyses as part of regulatory assessment. This study shows that in RRMM, the delay between TTTD and rwPFS can be large, and the relationship is not time-homogenous.
Conference/Value in Health Info
Value in Health, Volume 26, Issue 11, S2 (December 2023)
Code
RWD18
Topic
Clinical Outcomes, Methodological & Statistical Research, Study Approaches
Topic Subcategory
Comparative Effectiveness or Efficacy, Electronic Medical & Health Records, Meta-Analysis & Indirect Comparisons, Missing Data
Disease
Oncology, Rare & Orphan Diseases