In the Biosimilar Marketplace Will There Be 50 Ways to Leave Your Insulin?

Abstract

The future biosimilar insulin marketplace could be a bane, benefit, or something in between, to patients under our care in the United States. Formulary preferred product status, with or without FDA interchangeability designation, coupled with current and proposed state pharmacy substitution laws may lead to an environment with as many as 50 different substitution guidelines depending on in which state a patient presents his or her prescription to be filled. If online global prescription supply options, often referred to as “Canada Drug,” are utilized, other country substitution guidelines come into play, which may yield a recipe for confusion, if not disaster.

The future biosimilar insulin marketplace could be a bane, benefit, or something in between, to patients under our care in the United States. Formulary preferred product status, with or without FDA interchangeability designation, coupled with current and proposed state pharmacy substitution laws, may lead to an environment with as many as 50 different substitution guidelines depending on in which state a patient presents his or her prescription to be filled. If online global prescription supply options, often referred to in the United States as “Canada Drug,” are utilized, other country substitution guidelines come into play, which may yield a recipe for confusion, if not disaster.

Background

The FDA has created the Purple Book to address biosimilar product evaluation.¹ What is somewhat concerning is that there is no specific definition of biologic therapeutic protein drug equivalence standards other than a guidance statement suggesting a best case of a comparison analysis showing a “highly similar with fingerprint-like” match to the original product to help determine interchangeability.² Given the variable nature of biologic drug products a fingerprint-like match assay is not likely between the innovator brand and an alternate manufacturer product. Since there is limited information regarding the impact of bio-product fingerprint variance to patient response, clinical outcome comparison studies become the only remaining measure of expected biological activity. Current standards for nonbiologic generic drug products outlined in FDA’s Orange Book for bio-equivalence specify that the pharmacokinetic (PK) parameters compared to innovator compounds has an allowed “absorbed to the same extent” range of 80% to 125%.³ This means the generic version can show 20% less or 25% more absorption than the original branded product and still be presumed therapeutically equivalent. Although this is meant to serve as a guideline for nonbiologic generic drug equivalency there’s nothing that suggests a different PK absorption range standard for a similar “generic” biologic drug (biosimilar) under Purple Book guidance.

Regulatory and Legislative Efforts

There are ongoing efforts to develop cross reference standards between biosimilar and innovator biologic drug products in the EU but in the United States there is uncertainty as the FDA works to get the science right. During this period of FDA guidance development 31 individual US states have proactively filed biologic and/or biosimilar related bills or resolutions to initiate biosimilar substitution regulation at the state level. Of the 31 state legislative introductions 17 states have passed legislation that was signed into law. Of particular note, legislation in 1 state (Florida) does not require prescriber notification of substitution of “interchangeable” biological products while other states do require prescriber notification. To complicate matters even further 13 states and Puerto Rico have existing laws that mandate the pharmacist to dispense a generic version of a drug if all prescription requirements are met that are not invalidated by the new biosimilar substitution regulations enacted.⁴ In those 13 states and Puerto Rico the pharmacist will still be required by law to provide the least expensive interchangeable product available unless medical necessity for a specific manufacturer is documented. Due to these conflicting legal directives the pharmacist may not be certain if he or she must notify the prescriber that a change in biologic product manufacturer has occurred.

While the principle behind these preemptive state legislative efforts is well-meaning, without rules and regulations at the federal level to establish biosimilar interchangeability guidelines and standardized biologic product review the states may take the lead, potentially creating a massive regulatory puzzle. Mail order pharmacy operators will be the first to find ways to circumvent more stringent state laws by “dispensing” from states that have the least restrictive statutes regarding biosimilar substitutions by the pharmacist.

On a global scale, regulations have become more convoluted with Australia’s recent recommendation for “a” flagging of Basalgar® (insulin glargine) to be therapeutically interchangeable with Sanofi’s Lantus® without notification of the prescriber if brand necessary is not indicated.⁵ By determining that Basalgar is interchangeable with Lantus Australia will be able to negotiate for the lowest cost insulin glargine product for preferred formulary status. This will likely leave the patient to cover a potentially substantial difference in cost if the prescribed brand product is not preferred on the Australian formulary but deemed to be medically necessary. If Sanofi wishes to maintain its insulin glargine market share in Australia it now must be price competitive, which creates an opportunity of diversion of lower priced Lantus intended for the Australian market to the United States and Europe for illegal, but potentially substantial, financial gain for an unscrupulous intermediate broker.

Monitoring the Global Biosimilar Supply Chain

Adding to the mix, the manufacturer Biocon, located in New Delhi and working in partnership with Mylan, recently announced plans for biosimilar submissions for the United States and Europe including recombinant human insulin (NPH and Regular) and insulin glargine (Lantus and Basalgar).⁶ This creates potential for lower cost consumer options but adds foreign manufacturing sites producing biosimilar products for the FDA to schedule for inspection and comparison protein “fingerprint” data to evaluate. While manufacturers located in the United States are subject to unannounced FDA inspection and must produce documentation for review upon demand, the FDA must give advance notice of inspections to manufacturing sites outside the United States. Not to imply that Biocon is not an impeccable company with outstanding quality control, such advance notice that FDA officials are coming to inspect facilities provides the opportunity to “adjust” paperwork and samples for examination and thus the impact of a surprise inspection is lost.

Due to this international inspection loophole it took years for the FDA to take action against India based generic giants Ranbaxy and Sun Pharmaceuticals for submitting falsified data, repackaged innovator products and/or carefully crafted samples to FDA for testing to acquire approval for distribution. Despite ongoing (but announced) FDA site inspections production of contaminated, substandard, and unstable products for global distribution went undetected. These acts of maleficence were initially uncovered primarily due to acts of conscience by company insiders, not by routine FDA inspection of facilities and review of testing data.⁷ Given the potential for large volume sales and profits, one should be very suspicious of a nearly identical fingerprint match of any biosimilar product when compared to the innovator.

This should be a case of once burned, twice shy, but the FDA must rely on the honor system for some portions of their production facility inspections. Though manufacturers can overcome past illegal and immoral actions to regain FDA accreditation, the opportunity to fudge records in a foreign country for difficult to compare therapeutic proteins should be at the top of everyone’s list of concerns.

Retrospective Versus Prospective Quality Review

Pharmacovigilance and risk management plans are essential to an effective, comprehensive drug/device safety plan. Current retrospective quality assessment methods (where quality issues are reported to the FDA after a drug or device has been distributed and used) make it difficult to identify manufacturer and lot specific problems until a critical mass of independent reports of poor quality and/or adverse patient outcome is reached. Recent FDA Guidance outlines a plan to add a manufacturer specific suffix to the proper name of a biosimilar to help identify the source manufacturer as a method for improving pharmacovigilance.⁸ This biosimilar product naming effort may have little impact in tracking adverse events given potentials for multiple insulin manufacturer substitutions to a patient over any given year. Even with manufacturer specific naming, retrospective record review will still require time frame matching between pharmacy substitution and actual patient use of suspect product. Oldest verses newest dispensed product may not match patient home use sequence due to separate travel supply, cold storage issues, hoarding, and so on.

Detection of substandard, adulterated, counterfeit, or diverted biosimilar products using current retrospective methods will be extremely difficult. Given their volume of use substandard insulin products could affect thousands of individuals on an hourly basis. Subpotent insulin would be patient adaptable and could go unreported for months but superpotent insulin would require urgent action and be difficult to self-manage. If distributed on large enough scale the negative impact of poor quality to a manufacturer’s market share and erosion of public trust in regulatory agencies would be considerable. Prospective review of product quality by randomly pulling samples from the distribution pipeline for testing, though not a perfect answer, would help in early detection of deviation from approval baseline fingerprint of biosimilars by narrowing the search area. In today’s global supply chain environment, creation of universally accepted standards and methods for ongoing prospective product pipeline surveillance will require unprecedented cooperation between nations, regulatory agencies and manufacturers.

Getting the Regulatory Science Right

Given there may be subtle fingerprint differences between manufacturer lots of biologic products perhaps the FDA should consider fingerprint analysis of 3 (or more) separate lots from each manufacturer to be kept in a secure database for postmarket comparison. Biosimilar products with a near fingerprint match to the innovator could qualify as interchangeable but given the difficulty and cost of performing highly detailed manufacturer site inspections the FDA may be reluctant to provide an interchangeable designation. Utilizing a surveillance error grid metric format, variance of pipeline samples could be compared to the original FDA approved baseline range for lot differences in expected component content and potency. Variance greater than ±5% from the established baseline fingerprint average would indicate possible substandard content and initiate follow-up by FDA and the manufacturer to confirm nonconforming biologic product. Periodic random pipeline sampling using methods similar to the DTS Surveillance Program for Cleared Blood Glucose Monitors for quality measurement could be of great benefit in this area of concern as part of a prospective biosimilar quality assurance program.

To reduce the physical analytical burden on the FDA, impartial organizations could be identified to independently confirm randomly drawn pipeline samples of insulin matching, within acceptable tolerances, the original approved fingerprint baseline on file with the FDA. Financial support for maintaining such a quality program for all insulin products could be obtained by imposing a manufacturer fee of perhaps 5 cents per vial and cartridge produced for the US market. Individually a small cost but given the volume of insulin production several million dollars would be generated annually to pay for a reasonably robust random sample program. Methods developed to create a fingerprint reference database and randomized pipeline sampling for biosimilar insulin should also be highly applicable in the evaluation of more complex biosimilar therapeutic proteins. The program could also provide both the FDA and EMA a mechanism to measure the impact of cold chain variance on quality, facilitate earlier detection and tracing of counterfeit or diverted products, narrow recall quantities and regions to be alerted along with other quality related issues. This information would not just assist in regulatory enforcement; it would also reduce the financial impact to manufacturers when investigating possible counterfeit, adulterated or diverted products as well as potential for reducing size and scope of a recall.

Conclusion

Managing quality in a global open market environment is a tremendous challenge and both FDA and EMA regulators need all the help they can get. Without a prospective pipeline quality analysis program providers and patients are playing a dangerous game by leaving outcome up to a roll of the global market dice. As providers we can only control what we can, which in today’s cost driven environment means we control less and less each day, but will still be held liable for the patient’s outcome. A new equitable and affordable prospective product safety program for biosimilar products must be created. Otherwise, there may be 50 ways to leave your insulin.