Author response

Re: Xie J, Diener M, De G, et al. Budget impact analysis of everolimus for the treatment of hormone receptor positive, human epidermal growth factor receptor-2 negative (HER2-) advanced breast cancer in the United States. J Med Econ 2013;16(2):278–88

Dear Editor,

We thank Dr Diaby and his colleagues for their comments on our budget impact analysis of everolimus for the treatment of hormone receptor positive (HR+), human epidermal growth factor receptor-2 negative (HER2−) advanced breast cancer (ABC)1. Their comments focus on the sensitivity analysis where we include the impact of adverse events (AEs) on a payer’s budget following the introduction of everolimus. This response will address the rationale for our analytic decisions regarding AEs and demonstrate the effect of these decisions on the budget impact.

The AE sensitivity analysis focused on the common and severe AEs. We selected grade 3/4 AEs if their incidence was greater than or equal to 5% in either the everolimus + exemestane arm or the placebo + exemestane arm in the BOLERO-2 trial. To ensure a fair comparison, once the AEs were selected, we extracted the actual grade 3/4 AE rates from the trial for both treatment arms instead of arbitrarily assigning 0% if the rate was below 5% in a specific arm. Costs of AEs for each arm were estimated by multiplying the rates of selected grade 3/4 AEs and the cost associated with treating the AE. We assumed that costs of AEs for tamoxifen and fulvestrant were the same as those for exemestane. The rationale for considering the costs of AEs with an incidence of more than 5% in either treatment arm at a grade 3/4 level (i.e., either severe or life threatening) is that AEs less than grade 3 normally do not incur substantial resource utilization. Similarly, AEs that occur at a rate of less than 5% generally have a small impact on health plan costs due to the low rates of the events. Such assumptions are common in economic evaluations2,3.

While we agree with Dr Diaby and his colleagues that the selection of AEs in the sensitivity analysis may under-estimate the impact of adverse events, we would like to point out that the assumption had minimal impact on model results. While the base case analysis resulted in a budget impact of $0.0435 per member per month (PMPM), the budget impact only increased to $0.0442 PMPM when grade 3/4 AEs with ≥5% in either arm were included (resulting in a negligible budgetary difference of $0.0007 PMPM). Even after doubling the costs of adverse events for everolimus + exemestane treatment, the budget impact increase was still negligible, changing from $0.0435 to $0.0450 PMPM (a difference of $0.0015); therefore, we do not think that including the additional adverse events that are less severe or prevalent would have a meaningful effect on the budget impact results from a payer’s perspective.

In summary, AEs have a minimal effect on the overall budget impact of adding everolimus to a health plan’s formulary for the treatment of HR+, HER2− ABC. We appreciate the opportunity to respond to the letter to the editor and hope that this discussion clarifies our assumption and helps readers assess the impact of AEs on a payer’s budget following the introduction of everolimus.

Sincerely,

Jipan Xie

10 Rockefeller Plaza, 15th Floor, New York, NY 10020 USA. Tel: [phone]; Fax: [phone]; [email protected]

Transparency

Declaration of funding

There was no funding for this letter.

Declaration of financial/other relationships

Jipan Xie is responding on behalf of all the authors and has disclosed that she is employed by Analysis Group, Inc., a company that received consulting fees for this study discussed in this letter.