Response to: McBride A, Campbell K, Bikkina M, et al. Cost-efficiency analyses for the US of biosimilar filgrastim-sndz, reference filgrastim, pegfilgrastim, and pegfilgrastim with on-body injector in the prophylaxis of chemotherapy-induced (febrile) neutropenia. J Med Econ 2017;20:1083–93

McBride et al.¹ recently published what was referred to as “cost-efficiency analyses”. The analyses completed by the authors compared four alternative granulocyte colony-stimulating factor (G-CSF) options including: biosimilar filgrastim, reference filgrastim, pegfilgrastim, and pegfilgrastim with on-body injector; focusing specifically on the use of these options as prophylaxis for chemotherapy-induced febrile neutropenia (FN).

We acknowledge that evidence supporting the value of medicines is of substantial interest to decision-makers, including payers, providers, and patients. With this letter, we wish to highlight three major concerns we identified within the McBride et al.¹ analyses; concerns that we felt it was important to highlight to your readers.

Cost-efficiency is not a recognized form of pharmacoeconomic analysis

Our first concern is that use of the term cost-efficiency may lead readers to misinterpret the findings of the McBride et al.¹ analyses. Cost-efficiency is not a recognized form of pharmacoeconomic analysis^2–4. Established forms of pharmacoeconomic analysis include: cost of illness, cost-minimization, cost-benefit, cost-effectiveness, and cost-utility; each form having different and accepted methods, based primarily on the unit of effectiveness selected.

Given the stated assumption that all “prophylaxis options were considered to be similar to each other in efficacy”, it is clear that the authors have attempted to perform a cost-minimization analysis. Cost-minimization analysis can be an appropriate method for comparing all relevant costs and consequences of two or more therapeutic interventions, but only if the alternatives are all equivalent or equally efficacious, and the objective is to choose the least costly alternative⁴.

We think readers must, therefore, be cautioned; the findings of the McBride et al.¹ analyses do not provide evidence supporting “cost-efficiency”, as the title states. As a cost-minimization analysis, the findings of the McBride et al.¹ analyses can only provide information on the least costly alternative (if one accepts the assumption of equivalence or equal efficacy between, and the clinical appropriateness of, all of the alternatives being compared). However, having identified the pharmacoeconomic analysis actually being attempted leads us to our second major concern.

Inclusion of inappropriate alternatives based on available evidence

With approval by both the EMA and FDA of filgrastim-sndz as a biosimilar to the reference product filgrastim, it is appropriate to include these two alternatives in a cost-minimization analysis. With this we highlight: the situation is different for comparisons between either form of filgrastim and pegfilgrastim or pegfilgrastim-injector, as the equivalence or equal efficacy of those options has not been shown. Therefore, it is inappropriate to include these alternatives in a cost-minimization analysis.

To defend the assumption of similar efficacy, the authors state that they considered the evidence from the pivotal filgrastim vs pegfilgrastim non-inferiority studies^5,6. In other sections of their report, the authors cite, but disregard, evidence from a pooled analysis of the non-inferiority studies⁷, along with three meta-analyses^8–10. Each of these analyses provides evidence that filgrastim and pegfilgrastim are not equivalent. Instead the results of these analyses collectively provide evidence that pegfilgrastim provides significant benefit over filgrastim. The authors do not provide sufficient rationale, nor a sufficiently robust critical appraisal, to justify excluding this evidence.

We think readers must, therefore, be cautioned; while the assumption of similar efficacy between filgrastim-sndz and filgrastim is appropriate and supported by evidence, the extension of this assumption to the comparative efficacy of filgrastim and pegfilgrastim or pegfilgrastim-injector options is not supported by the currently available evidence and is, therefore, inappropriate. In addition, the McBride et al.¹ analyses included 1–14 doses of filgrastim and filgrastim-sndz. Use of inappropriately low doses of filgrastim in real-world clinical practice has been shown to increase the risk of febrile neutropenia^11–14.

Identification of limitations in the evidence supporting the assumption of similar efficacy between filgrastim and pegfilgrastim, as well as limitations in the range of figrastim/filgrastim-sndz doses included in the McBride et al.¹ analyses, leads us to our third major concern.

Inappropriate exclusion of relevant costs, based on available evidence

Our third major concern relates to the exclusion of costs “related to the management of (febrile) neutropenia, clinician fees, administrative fees, evaluation-and-management fees, and (actual or opportunity) costs of associated reductions, delays, or cancellations of chemotherapy administrations”. Excluding these FN-related costs is inappropriate, as it fails to capture the comprehensive financial consequences of this serious event. As a recently published study by members of the CDC’s division of cancer prevention and control group identified, the cost of cancer-related neutropenia or fever hospitalizations in the US in 2012 was, on average, $24,770 per stay (2017 $28,612) for a total cost of $2.7 billion (2017 $3.2 billion)¹⁵.¹

We think readers must, therefore, be cautioned; with insufficient evidence of similarity between filgrastim and pegfilgrastim, and limitations in filgrastim/filgrastim-sndz dosing, the exclusion of costs beyond “reductions in the cost of prophylaxis” biases the findings of the McBride et al.¹ analysis against options having a lower incidence of FN.

Although we have additional concerns: the use of a mix of data sources and perspectives that don’t immediately appear to align, and the exclusion of any mention of the potential benefits of the pegfilgrastim-injector option over real-world administration of pegfilgrastim on the same-day as chemotherapy (which is associated with an increase in the odds of febrile neutropenia)^16,17; we feel the three highlighted above are of most concern.

In summary, evidence supporting the value of medicines is critical for payers, providers, and patients. However, the evidence that is generated must be based on the rigorous application of established methods, supported with appropriate evidence, and inclusive of relevant information. The analyses presented by McBride et al.¹ are limited by flaws in the methodology used, the inclusion of inappropriate alternatives based on currently available evidence, and the inappropriate exclusion of relevant costs. What is really needed to inform decision-making in the area of chemotherapy-induced (febrile) neutropenia is a complete cost-effectiveness analysis that includes clinically appropriate alternatives and accounts for all of the costs and consequences involved for these alternatives, from relevant perspectives, with a relevant time horizon, and use of the best available evidence.

Transparency

Declaration of funding

This letter has no funding to declare.

Declaration of financial/other interests

MB, CE, and CB are employees of Amgen, Inc. MB and CE are stockholders in Amgen, Inc. JC is a consultant for Amgen, Inc. and is the recipient of research funding from the Institute for Clinical and Economic Review (ICER). JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

Medical writing support was provided by Matthew Gitlin (BluePath Solutions).

Note

Notes

1 2012 cost information from Tai et al.¹⁵ was inflated to 2017 $US using information from the Bureau of Labor Statistics [BLS] Consumer Price Index [CPI] for Medical Care using quarterly values for Q1 2012 to Q1 2017, resulting in an inflation factor of 1.15511).