By Andrew L. Concoff, MD, FACR, CAQSM
The study by Jeffrey R. Curtis, MD, MS, MPH, of the University of Alabama at Birmingham and colleagues featured in Healio seeks to address a question that, at first blush, may seem rather straightforward. Namely, does the cost savings achieved through the use of one of the FDA-approved biosimilars for infliximab in RA outweigh the increase in cost incurred as a consequence of the common practice of escalating the dose of that biosimilar, when compared to the alternative use of golimumab with the same method of delivery? Upon further analysis, the answer is a resounding “It depends”.
The answer is dependent upon context and detail. First, local and regional policies have given rise to dramatic differences in the availability of biosimilars, the rate of biosimilar adoption and the cost savings achieved by transitioning to biosimilars from bio-originators. Countries with single-payer, government-led health care systems are able to directly pit bio-originator medications against biosimilars in ‘winner-take-all’ negotiations that have resulted in dramatic price reductions, including a remarkable 80% reduction in the price for Humira by AbbVie in Sweden.
In the U.S., however, the Byzantine system of payers, pharmacy benefit managers, and the cadre of various middlepersons, coupled with aggressive, patent-related litigation has limited both the availability of biosimilars and the cost-savings associated with using them. Even with the approval and marketing of two infliximab biosimilars in the U.S., adoption has been disappointingly slow as has been the pace of reduction in price for infliximab, with net price reduction averaging just 13.6% per year since the introduction of its biosimilars in 2017.
The cost equation with infliximab and its biosimilars also depends upon the frequency and amount of dose escalation. It may be more than just a curiosity that the sole TNF-inhibitor for which biosimilars are currently marketed in the U.S. is one that allows for on-label dose escalation. While infliximab may initially be dosed at 3 mg/kg, the starting dose may be higher for certain patients, and escalation is formally approved up to 10 mg/kg. Furthermore, although the dosing interval begins at every 8 weeks, it is not uncommon for that interval to be reduced to six, or even four weeks.
In contrast, golimumab is an alternative, branded TNF-inhibitor for which no biosimilar is available, which is much less commonly escalated by either dose or interval. No dose escalation of golimumab is formally approved.
In this study, unsurprisingly, dose escalation over 18 months was far greater among those receiving infliximab (49%) vs. golimumab (5%). Further, 72% of those that remained on infliximab throughout the study period were found to have dose escalated. However, despite the difference in dose escalation rate, the adjusted least square mean costs were significantly lower for infliximab ($21,216) vs. golimumab ($28,146). This relationship was maintained when only those who persisted in their treatment with infliximab superior by $9,269.
However, an additional finding from the study regarding a specific predictor of dose escalation warrants significant reappraisal from a cost perspective, as it may be misleading as stated. After controlling for confounding variables through multivariable-adjusted logistic regression and mixed model assessment, physician ownership of the infusion center was associated with a greater likelihood of dose escalation (OR=1.25, 95%Ci 1.09-1.44). Physician “ownership” was assigned based upon whether the billing for the infusion was listed under their own NPI or was “part a group”.
This definition is fraught with error, as practices may bill as a group for infusion revenues or under only the individual NPIs of certain managing partners from a suite they own, while certain practices may bill infusion services by individual NPIs, for instance, to track utilization, without the physician owning those revenues from the infusion center. Thus, considerable allocation bias is likely in this assessment.
From a broader economic perspective, the study fails to account for other critical inputs to the value equation of infusions for RA. The researchers have not determined whether the dose escalations were appropriate according to a treat-to-target paradigm. It may be that closer monitoring of infusion centers allowed by physician ownership yields a tighter relationship between disease activity and dose escalation response, thus indicating that the increased frequency of dose escalation represents tighter treat-to-target responsiveness. In such a scenario, the increased costs of dose escalation must be balanced against the well-recognized reduction in cost associated with better disease control.
Further context is also necessary in assessing the costs of infusions in clinic settings, which are those typically owned by physicians, as this arrangement has been recognized to be dramatically less costly than the alternative, those performed in hospital outpatient settings. In fact, a recent study from the UnitedHealth Group suggested that by migrating privately-insured patients with RA from hospital- to clinic-based infusion centers would save nearly $6,000 per patient per month, and transitioning the 180,000 patients across multispecialty indications could save $4 billion per year.
The focus by the researchers on physician ownership as a driver of dose escalation is of uncertain veracity given the likelihood of allocation bias, the uncertainty of the appropriateness of the dose escalation and resultant change in disease activity and related cost impacts; its implications are economically misleading given the lack of an assessment of the major, infusion-related cost inputs.
The final issue in addressing the price of TNF inhibitors that remains to be determined is what happens next with TNF inhibitor biosimilar adoption and pricing. The future market share of these biosimilars is uncertain following the introduction of new, oral targeted immunomodulators (TIMs) for RA, notably selective JAK inhibitors. It may be that pharmaceutical manufacturers have managed to run out the clock on TNF inhibitor biosimilar utility.
The delay brought about by legal wrangling regarding patent protection of bio-originators and the uncertainty created by unrealized concerns as to biosimilar effectiveness and/or safety, may have gone on long enough to usher in an entirely new era marked by widespread, first-line adoption of an entirely different class of medications. Thus, the long-promised, TNF-biosimilar-inspired reduction in cost for adalimumab and etanercept, may occur after such time as it has any ability to engender a significant reduction in the cost of RA care.
This will only occur, of course, if the early results suggesting that selective JAKinibs confer greater benefit without significantly greater toxicity than TNF inhibitors bear out in real-world practice. In such a scenario, selective JAKinibs may achieve a dominant market share, thereby rendering TNF inhibitors obsolete and their prices irrelevant to the cost of RA care. How likely is that to happen? It depends.