Introduction

The objective of this study is to evaluate the persistence patterns and discontinuation rates from ruxolitinib in clinical practice. Ruxolitinib was the first Janus kinase (JAK) inhibitor approved for treating intermediate-risk or high-risk myelofibrosis, including primary myelofibrosis, post‒polycythemia vera myelofibrosis, and post‒essential thrombocythemia myelofibrosis. Ruxolitinib was investigated in 2 randomized, controlled, phase 3 efficacy trials in patients diagnosed with myelofibrosis (COMFORT-I and COMFORT-II).

Methods

A retrospective database inception cohort study is developed from 2 large US claims databases, the IMS Health® database containing National Council for Prescription Drug Programs pharmacy-dispensed prescription claims and the MarketScan® database containing patient-level data from approximately 100 health plans (data from active employees, early retirees, COBRA continues, and dependents) and from Medicare Supplemental data. We identify a cohort of myelofibrosis patients who were prescribed ruxolitinib between November 1, 2011 and August 31, 2012 in the IMS database and between November 1, 2011 and June 30, 2012 from the MarketScan database. Patients are followed from the date of the first ruxolitinib claim, defined as the index date, until the last enrollment date or date of last claim recorded if greater than the last enrollment date in the MarketScan database or for a fixed 3- and 6-month period in the IMS database. When analyzing treatment discontinuation, patients with primary or secondary myelofibrosis are required to have ≥1 ruxolitinib dispensed prescription(s) and to have ≥3 months of pre-index data and ≥3 or 6 months of follow-up post-index data. Discontinuation of ruxolitinib was defined as a >30-day gap from the last day of supply from the prior prescription fill. Demographic and ruxolitinib discontinuation rates at 3 and 6 months post-index date for patients in each database are reported in this analysis with the Clopper-Pearson confidence interval.

Results

In patients prescribed ruxolitinib in the IMS database, mean (SD) age at the index date was 63.4 (13.2) years for patients followed for 3 months (n=183) and 63.3 (13.6) years for patients followed for 6 months (n=115). The MarketScan database inception cohort comprised 161 patients with mean (SD) age of 66.8 (12.2) years. Age was comparable and not substantially different across the 2 databases. Discontinuation rates in the IMS database were 60.9% and 73.0% at 3 and 6 months follow-up, respectively; in the MarketScan database, discontinuation rates were 41.1% and 48.4% for patients with a minimum of 3 and 6 months follow-up, respectively.

Conclusions

These analyses indicate that the discontinuation rates for patients receiving treatment with ruxolitinib for myelofibrosis are not low and are most pronounced within the first 3 months of treatment. Results suggest the magnitude of unmet need in myelofibrosis and for additional medications that can improve treatment outcomes.

Limitations of this study pertain to a traditional claims database inquiry. As patients were identified from 2 distinct databases, there may be some data overlap. Diagnostic information on the basis of claims databases is not always accurate. Future research using rich clinical data from chart reviews is needed to identify clinical determinants of persistence/discontinuation. The discontinuation rates observed from these 2 large representative databases are substantial, reflecting inherent variability of point estimates and differences in data collection for the 2 data sources. Future studies should examine the reasons for discontinuation so that findings may be incorporated into disease/medication adherence management programs.

Table. Discontinuation Rates From Ruxolitinib

DatabaseFollow-up Time% [95% CI]
IMS® 3 months 60.9 [53.2-67.8] 
6 months 73.0 [64.0-80.9] 
MarketScan® ≥3 monthsa 41.1 [32.9-49.7] 
≥6 monthsb 48.4 [37.9-59.0] 
DatabaseFollow-up Time% [95% CI]
IMS® 3 months 60.9 [53.2-67.8] 
6 months 73.0 [64.0-80.9] 
MarketScan® ≥3 monthsa 41.1 [32.9-49.7] 
≥6 monthsb 48.4 [37.9-59.0] 

Median follow-up time is a207 days and b241 days.

This study was sponsored by Sanofi.

Disclosures:

Fonseca: IMS Health - a company that received funds from Sanofi to conduct the study: Employment. Silver:Sanofi : Consultancy, Research Funding. Kazis:Sanofi and IMS: Consultancy. Iqbal:Sanofi: Employment, Equity Ownership. Rose:Sanofi: Employment. Khan:IMS Health: Employment.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution