Dr Ivo Abraham Column: Biologics and Biosimilars—Harnessing Regulatory Data for Value, Access

Ivo Abraham, PhD, chief scientist of Matrix45 and a professor at the University of Arizona, demonstrates why regulatory bodies in the United States shouldn't look at economic evaluations when reviewing regulatory data for biosimilars in his latest column.

Each year, I teach a course on medication use and policy in our PharmD program, in addition to my courses on health technology assessment and value appraisal in our graduate programs. For each of these courses, I have a “read-up” list that I go through: what’s new and should I include it in my courses? High on that list is pricing of drugs, which in the United States translates into who is and who is not, and who perhaps should be, in the business of costing and pricing of drugs—especially those that are covered under public, taxpayer funded health care financing programs. Also high on the list is the transition from regulatory approval to commercialization and to patient care—and aspects of value, access, equity, and parity of increasingly more expensive drugs and the role of biosimilars herein.

The 2022 Inflation Reduction Act (IRA) made for some new and interesting reading material: the provisions that, under certain conditions, Medicare will be able to negotiate drug prices (as do some other parts of the federal government) and the provisions that aim to protect and foster the biosimilars market.

A solid step forward.

If it Ain’t Broke, …

Upfront, I should state my position (bias, if you will) that regulatory agencies should not consider costs in their evaluation of new drugs. However, invariably, every year there is more (of the same) on whether regulatory agencies, like the FDA and the European Medicines Agency (EMA) or even country-specific regulators, should add cost to their purview. Some people in the United States claim, incorrectly so, that “the Europeans do it and it works over there”: within the European Union, regulatory drug approval is transnational by the EMA, but formulary-setting and pricing are at the national level. The US publications on integrating cost into the regulatory process are long on historical review, short on why and how regulators should consider cost in addition to efficacy and safety, and bereft of actionable recommendations. The naïve rhetoric that if the public and Congress want it, it should and will happen, is of blinding political ignorance (to reference Leonardo Da Vinci).

Just consider how many permutations of regulatory decisions are possible if a regulator were to consider efficacy (good, marginal, poor), safety (safe, marginal, poor), and cost (high, middle, low). Leave out poor efficacy and poor safety (those drugs would not get approved), but that still leaves 2 x 2 x 3 = 12 scenarios.

Regulators should not consider cost.

Not for small molecules.

Not for complex biologicals.

We need the tightest controls when approving drugs built on the strongest possible evidence, generated from the strongest possible data, and analyzed, verified, interpreted with the strongest possible methods, and…free of economic considerations of any kind.

There is another reason why regulatory agencies should not be concerned with the cost of biosimilars, let alone with pricing them. Marketwise, it is just not necessary. In the United States and Europe, there are enough biosimilars referencing the same originator. In good part, the biosimilar manufacturers are already competing with each other on price—with the reference manufacturer still in the market without the pressure of recouping development costs, willing to drop the price, all the while generating bigger margins than the biosimilars.

Free markets are effective.

Free markets are efficient.

Product, not cost, needs to be regulated.

The continued efforts to accelerate (rigorous and uncompromising) approvals of biosimilars in general and through the 2022 IRA are encouraging for several reasons that I have laid out previously. Using biosimilars as opposed to originators generate (cost-)efficiencies; important in itself, but even more important in that it can reallocated to provide more care to more people budget-neutrally. This occurs despite the counterwinds of price erosion and downward pricing pressures from payers. This, and the fact that biosimilars are commodities, make that volume is of the essence. With the many biosimilars approved in the United States, Europe, and other affluent markets, it will become evident that these markets are too small to accommodate all approved biosimilars—how many pegfilgrastims? How many adalimumabs? How many of the other “mabs” do we need in these markets? For several manufacturers, commercial survival will come from middle-income countries first and low-income countries after that—differentially priced.

Regulate in the West—use at home and around the world.

Regulatory Data Integration for Value Appraisal of Biologics and Biosimilars

There is a function, though, where regulators can facilitate (fair) pricing: sharing data to inform value appraisals that enable access to drugs and assure equity and parity.

Economic Benefit vs Economic Value

In this, we need to make a distinction between economic benefit from pharmaceutical products vs value of pharmaceutical car—considering the 7 “P” stakeholders: producers, payers, and providers, patients, prescribers, policers, and policy-makers.

Economic benefit is about the business side, starting with the producer of a drug: a fair profit after the investments that led to the innovation—say, a new biological, originator or biosimilar. Economic benefit is also about the payer: a fair profit after the investments to create a system to pay for drugs. And economic benefit is about the provider: a fair profit after the investments to use the drug in patient care.

Conversely, value is about the balance between quality, cost, and access—the “iron square” of health care, to which I add choice. It is about the patient: illness, access, choice, affordability, health and well-being. It is about the prescriber: care, options, affordability, and outcomes. It also about policers, whether regulators of drugs or regulators of care: access, quality, safety, and outcomes. And, lastly, it is about policy-makers: demand, planning, accessibility, quality, safety, affordability, and outcomes.

Constraints

If we look at economic evaluations of new drugs and treatments, 3 things show the constraints of doing economic evaluations based on aggregate, sample-level data.

By necessity, these economic evaluations are based on clinical trial data. Because of the randomization, these evaluations consider the respective treatment arms similar (“balanced”, is the term) on a number of recorded demographic and clinical variables, but also on unknown variables; within certain margins of variation. By extension, they attribute any differences between the treatment arms on the endpoints of interest to the fact that participants were randomized to either the novel or the comparator drug. This is what good trials are about.

Further, these economic studies assume that all patients within each treatment arm are quite the same; again within certain margins of variation—some older, some younger, some sicker, some not so sick, some with and some without predisposing factors, and so on.

Lastly, and despite this variation between and within treatment arms, the results of the trial are generalized to the entire treatment arm: participants who got the new drug did better than those who did not—as if that is true for all subjects in each arm, which is unlikely to be the case.

Unless we have patient-level data, our understanding of how and in whom a novel drug works better than the comparator, and vice versa, will be general. Trial data tell us something about a novel drug and its comparator at the aggregate level. They do not help us much in understanding in which patients the novel drug worked exceedingly well, in which rather well, in which just a bit, and in which not at all. These are significant constraints.

Consequently, our economic evaluations are constrained as well. Add to this that economic evaluations are often sponsored by a stakeholder (say, a payer or a manufacturer), and therefore, not independent and at risk for bias.

The solution is evident: access to patient-level data.

Will a manufacturer willingly release its trial data? We know that answer. Will a manufacturer support an independent economic evaluation that may run counter to theirs? No, but probably accept it and hope that memory fades fast.

Constrained public data.

Constrained economic evaluations.

Sample-level trial results.

Sample-level economic results.

Patient-level data will have to come from elsewhere.

Harnessing Regulatory Data—for Independent Analysis

This is where the regulatory agencies come in, with their independence and their duty to the public interest, as well as their trove of data:

  • Patient-level data included in the submission; data submitted post-approval; volunteered safety data; to name a few.
  • Of high-quality data because of regulatory standards, regulatory scrutiny, and the manufacturer’s objective of getting the drug approved.
  • Repurposed to inform value appraisals in the public interest; made available “on loan”; by all ethical standards and with all legal protections; to bridge the gap from regulator to (public) payer.
  • Enabling targeted stratified value appraisals that inform access, equity, and parity—for all patients and for subgroups of patients.

The significance of regulatory patient-level data to value appraisals is immense. Such data…

  • Enable identifying good responders, moderate responders, and poor responders—and the factors that account for their response, directly or through mediation.
  • Enable patient-stratified efficacy analyses: who benefited most from the soon-to-be-or-just approved treatment (where value is high); who somewhat (where value is moderate); and who minimally (where money is lost)?
  • Enable patient-stratified safety analyses: who tolerated the treatment well (and will cost much less in adverse event management); who had some difficulty (and will cost more), in what respects, and how can these adverse events be averted (and save money); and who had severe tolerance problems that should be averted (and save money)?
  • Distinguish between who was adherent to treatment (and had the best chance of responding), who less so, and who minimally so; and what are the predictors?

This, of course, applies to all biologicals—and small molecules too. However, when applied to biosimilars, safety prevails over efficacy. Biosimilars are approved on the basis of studies powered to assess equivalence in efficacy, with safety data collected in parallel but without being powered for this. Knowing in whom safety fails and in whom not, would lead to better targeting of patients.

Efficacy is a more limited differentiator even though patient-level biosimilar trial data would tell what factors are associated with gradients of efficacy. Failure of efficacy (not efficacy in general), is a major differentiator. It too requires patient-level data.

Safety and failure of efficacy, at the patient level, are the unknowns in the value equation. Pre-market and post-market regulatory safety and failure of efficacy data, and their respective determinants, are major determinants of value—and of equity, access, and parity.

Patient-level regulatory data—distinguished by the rigor with which they were defined, δand needs of all the “P” stakeholders.

Especially when analyzed independently.

Beyond “One-issue” Value Appraisals

My colleague Nimer Alkhatib, now back in his (beautiful) country of origin, Jordan, and on faculty at Al-Zaytoonah University, and I developed Six Delta, a 6-dimensional model for independent outcomes-based value assessment of pharmaceuticals. As presented in the lead paper of a set of 8 different articles, Six Delta is a comprehensive and integrated platform for outcomes-based costing and pricing that assesses and reconciles cost and price variations along 6 differentatiators δ (from δ1 to δ6) which are subsequently integrated into an overall price estimate. Two δ’s are long-term focused and the remaining 4 δ’s concern assessments against a shorter time horizon. Each δ applies different scenarios in creating price dispersions, which are generated from Monte Carlo simulations to generate probabilistic dimension-specific price estimates, as well as estimates that average all dimensional estimates.

The pricing dimensions are the following—and can be applied specifically to biosimilars as described here:

δ1. Cost-effectiveness and cost-utility. Difficult with sample-level data, patient-level data, especially when stratified into responder subgroups or other determinant-driven subgroups, enable “classical” comparative cost-effectiveness and cost-utility analyses that probabilistically estimate incremental cost-effectiveness or cost-utility ratios. These analyses may be relative to a different comparator (for instance, reference biologic), to other levels of responders, or to yet other patient strata.

δ2. Willingness-to-pay. Conventional pharmaco-economic analyses use willingness-to-pay (WTP) more as a “we will not go higher than a set dollar figure to get the intended outcome”. In Six Delta, WTP is seen as a matter of choice of “how much do I want this treatment and how much do I want to pay to get the clinical outcome that I hope for” – and this across all 7 “P” stakeholders, each in their own way.

δ3. Referencing. It is important to have contextual data about prices, specifically how prices in one jurisdiction (say, the United States) compare to prices in other jurisdictions. Comparability is key here, but also difficult considering that health and pharmaceutical care costs in the United States, in general and per capita, are among the highest in the world. Pricing in other economically and health-system comparable jurisdictions, adjusted for economic differences by such methods as medical purchasing-power-parities, provide a reference against which to compare local data.

δ4. Safety. This is a critical dimension with regards to drugs and biologics in general, but certainly when it comes to biosimilars. Trial samples are too small to detect but the most common adverse events and may only give an initial idea of the range and intensity of a biosimilar’s adverse event profile. Complemented with voluntarily contributed safety data from patients and prescribers, a more complete (and statistically more powerful) safety data set may emerge.

Importantly, if adverse event rates of a biosimilar exceed that of a comparator or another upper threshold for a biosimilar, the biosimilar could be “penalized” in dollar terms—a payback to the payer by the producer. If, on the contrary, the biosimilar’s adverse event rates fall below a lower threshold, the biosimilar could be “incentivized” in dollar terms because of the additional costs averted—payback to the producer by the payer.

δ5. Risk of efficacy failure. Biosimilars are expected to achieve a minimum level of efficacy, yet with individual patient-level data it is also possible to determine who the “below-responders” are and the determinants thereof. In time it will be possible to estimate the risk of efficacy failure for a given patient. If a patient’s risk of efficacy failure, set by statistical modeling of determinants of treatment failure, exceeds a threshold and it is evident that this patient should not have been treated, a manufacturer could “penalize” a provider but also a payer and request a payback.

δ6. Adherence. This dimension may be less evident with injectable biosimilars, if adherence is defined as patient behavior. However, there is a link to injectables if a biosimilar is associated with adverse events that cause treatment to be delayed, discontinued, or otherwise impaired at greater than known rates and triggers a “penalization”.

The estimates of each δ are averaged into an overall δ, with standard deviations providing room for discussion.

Conclusion

In this perhaps rather long exposé, I tried to make a case that regulatory agencies should not (be obliged to) incorporate cost or price in their decision-making process, yet that their pre-market and post-market data can be leveraged to optimize treatment choices for individual or clusters of patients, at pricing derived multidimensionally, that is fair to all 7 stakeholders in health and pharmaceutical care. Among the most important features of regulatory data is the data quality and the granularity at the patient level. This makes it possible to model those most and those least likely to benefit from a drug, and to select those in-between both more likely to benefit. While this holds for drugs in general, it certainly applies to high-priced biologics in general as well as those with biosimilar competition.

Bio

Ivo Abraham is Chief Scientist of Matrix45 and Professor of Pharmacy, Medicine, and Clinical Translational Sciences at the R. Ken Coit College of Pharmacy at the University of Arizona, where he is associated with the Center for Health Outcomes and PharmacoEconomic Research. He has worked in biologicals since the late 1990s and in biosimilars since their introduction in the European marketplace—collaborating closely with Karen MacDonald (also his wife) on large international and national observational studies. On both the private and academic sides, their group published the first economic evaluations of biosimilars, a line of studies that continues to date and have been instrumental in the breakthrough and market adoption of biosimilars in Europe and the United States. More recently, Matrix45 has ventured into biosimilars in emerging markets, including low- and middle-income countries. Ivo Abraham may be reached at cntr4biosim@matrix45.com.

Perspective

I am a strong proponent of biosimilars. That does not mean I am against innovation—on the contrary. There would be no biosimilars without the innovators. I have worked on several of these innovators. I am working now on innovators that someday may have biosimilar analogs. I am of the generation that has had the joy of seeing treatments emerge (and some fail) for diseases that over 40 years ago had the poorest of poor prognoses—but are now treatable.

Innovation in therapeutics (that is, the originator products) is about moving the boundaries of hope. Biosimilars are a channel for spreading more hope to more patients.

Statement of Disclosures of Relevance to This Monthly Column

In early November,I was invited to co-lead a grant proposal to the FDA for regulatory science and innovation to the FDA. The ideas in this column preceded this invitation by several months.

In September, I was invited by a USDHHS contractor to consult on a project on the cost of biosimilar drug development. As of this writing, this has not been formalized.

Matrix45, LLC and predecessor companies in which Ivo Abraham and Karen MacDonald hold or have held equity, have been contracted for research, analytics, dissemination, and consulting services by Janssen/Johnson & Johnson, Amgen, Novartis, and Roche on the originator side and by Sandoz/Novartis, Coherus Biosciences, Mylan/Viatris/Biocon, Hospira/Pfizer, and Teva on the biosimilars side; with past and current conversations with Merck KGaA, Therapeutic Proteins International, Celltrion, Apobiologix, Apogenix, Fresenius Kabi, and Spectrum. By company policy, associates of Matrix45 cannot hold equity in sponsor organizations, nor provide services or receive compensation independently from sponsor organizations. Matrix45 provides its services on a non-exclusivity basis.

All contributions to this column are prepared independently and without funding from sponsors.

Links to Prior Columns

Dr Ivo Abraham Column: 1 Billion People Can Access Biosimilars; What About the Other 7 Billion? (November 2022) https://www.centerforbiosimilars.com/view/dr-ivo-abraham-column-1-billion-people-can-access-biosimilars-what-about-the-other-7-billion-

Biosimilars and the Commoditization of Treatments (October 2022). https://www.centerforbiosimilars.com/view/dr-ivo-abraham-column-biosimilars-and-the-commoditization-of-treatments

When more may yield less: price erosion of biosimilars following US market entry. (September 2022).https://www.centerforbiosimilars.com/view/dr-ivo-abraham-column-when-more-may-yield-less-price-erosion-of-biosimilars-following-us-market-entry

It’s what we do with the savings: economics and equity. (August 2022) https://www.centerforbiosimilars.com/view/dr-ivo-abraham-column-it-s-what-we-do-with-the-savings-economics-and-equity

Good bait and fair switch: biosimilar interchangeability, substitution, and choice. (July 2022) https://www.centerforbiosimilars.com/view/contributor-good-bait-and-fair-switch-biosimilar-interchangeability-substitution-and-choice

To try or not to try, that’s not the question: phase 3 trials of biosimilars and beyond. (June 2022) https://www.centerforbiosimilars.com/view/contributor-to-try-or-not-to-try-that-s-not-the-question-phase-3-trials-of-biosimilars-and-beyond

The enemy of your enemy should be your friend: why biosimilar companies should collaborate. (May 2022) https://www.centerforbiosimilars.com/view/contributor-the-enemy-of-your-enemy-should-be-your-friend-why-biosimilars-companies-should-collaborate