Zarxio Switching Fails to Meet Budget Impact Model Expectations

According to research presented at the AMCP 2017 Annual Meeting, Zarxio (filgrastim-sndz; Sandoz) cost savings, predicted using a budget impact model, failed to meet the expectations of researchers due to the drug’s higher price.

Anna Hung, PharmD, of the University of Maryland School of Pharmacy, sought to validate a previously developed budget impact model, which predicted the 1-year cost of Zarxio through a clinical management program.

“In March 2015, Zarxio became the first biosimilar to be approved in the United States,” Dr Hung wrote in her study. “In 2015, a large commercial insurer used a clinical management program to encourage providers to switch from Neupogen (filgrastim; Amgen) to Zarxio. A budget impact model was developed to predict 1-year cost savings from the clinical management program. One year post the clinical management program implementation, we want to compare predicted savings to realized costs.”

In order to study the budget impact model’s accuracy, Dr Hung conducted a study with three phases. The first phase of the study updated two of the budget model’s key assumptions: that Zarxio’s price would be 30% lower than its reference product, and that 30% of the reference product users would switch to Zarxio. However, these assumptions were significantly inaccurate. In real-world scenarios, the price of Zarxio was only discounted 15%, while switching utilization was only 9%. 

For the second phase of the study, Dr Hung built an additional component into the budget model that took into account patients who switched from Nuepogen to Granix (tbo-filgrastim; Teva). Cost saving associated with this switch were also added into the budget impact model.

The third phase of the study involved updating the assumed utilization and cost with real-world pricing and utilization data. Dr Hung used pharmacy claims data from September 2015 to September 2016. 

Study results showed that the assumed cost saving were greater than the actual cost savings generated by the retooled budget impact model. Cost savings in the original model were $56.94 per utilizing member per month, compared to $3.28 after phase 1 of the study, $20.74 after phase 2, and $29.45 after phase 3. 

“The original predicted cost savings was higher than what was realized primarily due to a smaller reduction in price for Zarxio versus Neupogen,” Dr Hung concluded. “Future budget impact models should be validated by updating key assumptions, building in additional components as needed, and comparing against claims data.” —David Costill