Brief Report: Biosimilar Uptake in the Academic and Veterans Health Administration Settings: Influence of Institutional Incentives

Abstract

Objective:

To compare uptake in the ordering of biosimilars at a Veterans’ Affairs medical center (VAMC) to that at an academic medical center, where institutional incentives for infused medications differ.

Methods:

We performed a cross-sectional study of medical record data and estimated institutional financial incentives at two medical centers in Philadelphia: 1) the University of Pennsylvania Health System (UPHS) and 2) the local VAMC. All ordering events for filgrastim or infliximab products were quantified over time, stratified by product (biosimilar versus reference product) and center. Financial incentives to the institutions over time were determined based on actual drug costs for the VAMC and average sales prices (ASPs) and Medicare Part B reimbursement rates for UPHS.

Results:

There were 15,761 infusions of infliximab, of which 99% were for the reference product. There was sharper decline in use of reference products at the VAMC; 62% of the 446 infliximab infusions ordered at the VAMC were for the reference product. ASPs were consistently lower for biosimilar infliximab products, but the estimated institutional financial incentives remained similar over time for biosimilar and reference infliximab at UPHS. At the VAMC, the costs for 100 mg vials of reference infliximab and infliximab-abda were $623.48 and $115.58, respectively: a $507.90 (81%) savings per vial.

Conclusions:

The uptake of infliximab biosimilars has been slow at an academic medical center, compared to a nearby VAMC, where financial savings are realized by the institution from its use. Slow adoption of biosimilar medications may impact the rates of decline in costs.

Over 20 biosimilars have been approved for marketing in the United States (US), and over 50 have been approved in Europe (1). Non-medical mandatory switching and required use of a biosimilar for patients newly initiating treatment have resulted in the rapid uptake of biosimilars in some Scandinavian countries (2). However, in the US, changes in biosimilar prescribing patterns have occurred gradually in the absence similar large-scale mandates from healthcare institutions or payers (3).

Few studies have investigated the uptake of biosimilars in the US. One study of US Medicare Part B prescription claims for filgrastim products found biosimilar filgrastim-sndz to account for 32% of all filgrastim prescriptions by 21 months after its approval by the US Food & Drug Administration (4). The uptake of infliximab biosimilars is likely to differ since incentives and financial drivers at the institutional level are different for infused therapies.

Currently in the U.S., Medicare (one of the largest payers for infused therapies) is not permitted to negotiate drug prices. Infused therapies are reimbursed based on the average selling price (ASP) over the prior quarter. Biosimilar therapies that are priced lower than their reference products (lower ASP) will have lower reimbursement rates and may have less room to negotiate with institutions for a lower purchase price compared to the reference product. Since the financial incentive to use a product depends on the difference between the reimbursement rate and purchase price, there will not necessarily be an incentive for an institution to use the lower priced product. This may result in slow adoption of biosimilars and negatively impact the ability of these products to effectively stem increasing health care costs.

The uptake of biosimilar products has not been evaluated in the Veterans Affairs (VA) medical system, which uses centralized negotiation and contract management to establish a national formulary (5). To save costs and promote the use of less expensive therapies, the VA has designated biosimilar options for preferred use and has mandated switching patients to these therapies, including biosimilar infliximab and filgrastim products, unless there are special clinical circumstances. In such instances, physicians are required to submit non-formulary requests if they wish to use the non-preferred therapy. Thus, a comparison of biosimilar ordering at a VA Medical Center (VAMC) to that at a nearby academic medical center provides a natural experiment to assess how tendering and national policy mandating use of the biosimilar can affect uptake of these therapies and impact healthcare costs (6).

We hypothesized that the rate of uptake of biosimilar infliximab therapies at an academic medical center would be less than that at a nearby VAMC, where lower acquisition costs had been negotiated for infliximab biosimilars. We further hypothesized that institutional incentives at the academic medical center, based on Medicare reimbursement policies for infused therapies, would not favor a switch to the biosimilar therapies despite their lower ASP.

Methods

We compared the frequency of prescribing of biosimilars at a VAMC to that at a nearby academic medical center between January 1, 2015 and May 31, 2019. We leveraged data from a prior analysis of pharmacy data from the University of Pennsylvania Health System (UPHS) electronic medical record and the Corporal Michael J. Crescenz VAMC pharmacy databases (7). These institutions are geographically adjacent (<0.5 miles) and share healthcare providers who work at both facilities.

We focused on three medications for which biosimilars have been marketed over the past 3–5 years: infliximab, filgrastim, and pegfilgrastim. Filgrastim and pegfilgrastim were examined in order to assess patterns of use of non-infusion therapies, at UPHS and the VAMC. We included the reference products and all approved biosimilars. We summarized the frequency of ordering of infliximab, filgrastim, and pegfilgrastim reference and biosimilar or alternative products after the approval date for the first biosimilar or alternative product of each.

Medicare was the predominant payer, with approximately 35% of infusions covered by Medicare or Medicare Advantage; 21% of infusions were covered by Medicaid and 44% of infusions were covered by commercial insurance. We therefore estimated institutional reimbursement for infliximab reference product and biosimilars at UPHS based on the Medicare Part B reimbursement policy – infusions are reimbursed at the quarterly Medicare Part B ASP for the reference product or biosimilar plus 4.3% of the ASP for the reference product (8). The ASP reflects the actual average price for medications, considering all negotiated discounts and rebates. Competition and price negotiations between payers, pharmacy benefits managers (PBMs), and institutions are expected to result in reductions in ASP over time. We used the published ASP from two quarters in the future in order to estimate the acquisition cost to an institution for each medication during the current quarter (9). We defined the difference in reimbursement and estimated acquisition cost as the financial incentive to the institution to utilize that therapy. We then compared these estimated incentives for UPHS to the VA institutional incentives, that were based on actual purchasing costs to the VA for each individual therapy (10).

Results

Between July 1, 2015 and May 31, 2019, there were 15,761 infusions of infliximab at UPHS and 446 at the VAMC. At UPHS, 99% of infliximab infusions were for the reference product, Remicade® (Janssen Pharmaceuticals, Inc.), compared to 62% at the VAMC over the same time-period. The initial shift to use of biosimilar infliximab-dyyb (Inflectra®, Pfizer, Inc.) at the VAMC occurred in December 2017, with an abrupt change in ordering around this date (Figure 1) (11). Use of biosimilar infliximab was similar across the divisions of gastroenterology (39% of all infliximab orders) and rheumatology (31% of all infliximab orders).

Figure 1:

Frequency of orders for reference and biosimilar infliximab (top panels) and filgrastim/pegfilgrastim (lower panels) at the VA Medical Center (VAMC, right panels) compared to the nearby academic medical centerin Philadelphia (UPHS, left panels). Frequency of reference and biosimilar products are shown for Inflximab (dark gray), Infliximab-dyyb (dark blue), infliximab-abda (red), filgrastim (dark gray), pegfilgrastim (light gray), filgrastim-sndz (green), tbo-filgrastim (yellow), and pegfilgrastim-jmdb (orange-red).

Since July 2017, after initially having a higher ASP, ASPs for biosimilar infliximab products have been consistently lower than those for the reference product (Figure 2). However, estimated institutional incentives based on Medicare Part B reimbursement and estimated acquisition costs for both reference and biosimilar infliximab have been similar since 2018. In fact, in 2019 the estimated institutional incentive favored the reference product (Remicade®) by $49–64 dollars per 100 mg vial.

Figure 2:

(A) Average sales prices (ASPs) for reference infliximab and the biosimilar products infliximab-dyyb (Inflectra®) and Infliximab-abda (Renflexis®) by quarter. (B) Estimated financial incentive to the institution per 100 mg vial for use of each product, based on Medicare Part B reimbursement policy and subsequent ASP.

At the VAMC the situation is very different: the costs for 100 mg vials of reference infliximab (Remicade®) and infliximab-abda (Renflexis®, Merck & Co., Inc.) are $623.48 and $115.58 each, respectively, yielding a $507.90 (81%) savings per 100 mg vial for the biosimilar. The current cost to the VA for infliximab-abda is 79% lower than the Medicare ASP for the same therapy ($116 v. $546).

Between January 1, 2015 and May 31, 2019, there were 46,683 orders for filgrastim or pegfilgrastim at UPHS, of which 90% were for either reference product (Neupogen® or Neulasta®, Amgen, Inc.) (Figure 1). There was modest use of filgrastim-sndz at UPHS, beginning in 2016. At the VAMC 88% of prescriptions were for one of the reference products but with a sharper decline in use of the reference product and a sharper increase in use of biosimilar products after September 2016, when biosimilars were designated as the preferred product.

Discussion

Compared to that at a nearby academic medical center hospital, the uptake of biosimilars, particularly infliximab products, has been dramatically faster at the VAMC, where tendering and a nationwide policy change mandating preferred use of the biosimilar have resulted in clear financial savings for the institution. These data suggest that, with current Medicare Part B reimbursement policy, the absence of financial incentives to encourage use of infliximab biosimilars has resulted in slower uptake of biosimilar use at institutions outside of the VA system. The implications of this are a slower reduction in costs to the health care system since decreases in ASP over time are predicated on negotiations at the institutional level, which have been gradual and stepwise.

The use of biosimilars has been widely accepted in Europe, especially where tender processes and changes in policy at a national level have mandated their use (12). The lesser discounting of biosimilars in the US, and the observed slow adoption have delayed realization of the potential economic benefits of biosimilars on healthcare spending. This analysis demonstrates that, even when the ASP of an infusible biosimilar is lower than that of its reference product, current Medicare Part B reimbursement policy does not necessarily provide an institution with a financial incentive to transition to use of the biosimilar.

Because an institution is now reimbursed for a Medicare Part B medication at 104.3% of its ASP, use of a medication with a lower ASP will yield lower reimbursement. In this context, the incentive for an institution to use an individual therapy depends primarily on the negotiated purchasing price relative to the reimbursement rate. Our analysis reveals that, because more expensive therapies have more room to negotiate, a similar relative reduction in price for a more expensive therapy can result in a greater incentive for the institution to use that therapy. These observations are in spite of reimbursement for an infused biosimilar being its ASP plus an additional 4.3% of the ASP of the more expensive reference product.

In addition, while a gradual reduction of the ASPs of reference infliximab and both infliximab biosimilars has occurred with ongoing negotiations between institutions and manufacturers, the rate of decline in ASP has been slow, and substantially less than that which has been observed at the VA, where costs have already been reduced by 80% compared to the ASP of reference inflximab. This may be, in part, because major reductions in pricing on the part of the manufacturer are expected to result in only short-term incentives for use of the product.

As observed, despite lower ASPs, providers at the academic medical center have not incorporated biosimilar infliximab substantially into their practice. In addition to the lack of clear financial incentives, other economic or market factors and uncertainty and misperceptions about biosimilars among healthcare providers, pharmacists, and patients may also slow the transition to their acceptance and use (13, 14). This study suggests that institutional mandates for use of specific products, such as that observed in the VA medical system for biosimilar infliximabs, can surmount these barriers that are observed at the level of individual providers. In the Netherlands, structured educational programs have increased patient acceptance of transitioning from a reference product to its biosimilar (15). Thus, any institutional mandate to use biosimilars should be accompanied by education of healthcare providers, pharmacists, and patients about the efficacy, safety, and utility of biosimilars.

Although ASPs reflect prices, after discounts and rebates, averaged throughout the US for drugs reimbursed by Medicare Part B, we were not able to ascertain actual acquisition costs to the UPHS. The actual financial incentives to an institution depend upon price negotiations with individual manufacturers. However, our estimate of acquisition costs using ASPs can be generalized nationally. Although some of our results may not be applicable to other geographical regions of the US, the comparison of two affiliated institutions in geographical proximity and with shared healthcare providers is a strength. Our findings should be replicated using national VAMC data or data from other healthcare systems.

Our observations may not be applicable to medications that are not administered by infusion and thus are not covered by Medicare Part B. However, because PBMs often receive rebates that are calculated as a percentage of the manufacturer’s list price, they may receive a larger rebate for a more expensive drug. Thus, while not studied directly here, PBMs may receive incentives that continue to promote the use of reference products that have higher manufacturer’s list prices, which likely will limit the uptake of both infused and injectable biosimilar therapies over time. This hypothesis is supported by our data which demonstrate that use of biosimilar filgrastim-sndz remains low relative to its reference product at UPHS, in comparison to the relative use of the biosimilar at the VAMC, despite the ASP of biosimilar filgrastim being approximately 40% lower than that of its reference product (9). This finding has important implications for when non-infused biosimilars (e.g. etanercept and adalimumab) are eventually introduced to the US market.

In conclusion, we observed slower uptake of an infused biosimilar at an academic medical center compared to that at a neighboring VAMC, where preferred use of biosimilar infliximab is mandated and clear financial savings are realized by its use. These observations demonstrate how complexities of financial incentives for use of biosimilar medications may delay their adoption. A slower incorporation of use of biosimilar is expected to result in delays in the reductions in pricing and costs to Medicare, particularly compared to the reduction in costs observed for the VA. The incentivization of policy changes at the institutional level, as observed in the VA, may be the most effective way to increase use of biosimilars in the US and to realize savings to the health care system.