Biosimilar Insulins

Abstract

Until now most of the insulin used in developed countries has been manufactured and distributed by a small number of multinational companies. Beyond the established insulin manufacturers, a number of new players have developed insulin manufacturing capacities based on modern biotechnological methods. Because the patents for many of the approved insulin formulations have expired or are going to expire soon, these not yet established companies are increasingly interested in seeking market approval for their insulin products as biosimilar insulins (BI) in highly regulated markets like the EU and the United States. Differences in the manufacturing process (none of the insulin manufacturing procedures are 100% identical) can lead to insulins that to some extent may differ from the originator insulin. The key questions are if subtle differences in the structure of the insulins, purity, and so on are clinically relevant and may result in different biological effects. The aim of this article is to introduce and discuss basic aspects that may be of relevance with regard to BI.

Biopharmaceuticals (http://en.wikipedia.org/wiki/Biopharmaceutics) were invented in the 1980s of the past century. They are proteins derived from cell culture/fermentation processes; the first example was recombinant human insulin. Other examples are cytokines, monoclonal antibodies, and other insulins. Since a number of years biosimilar products are on the market for clinical usage, that is, “copies” of therapeutically used proteins (eg, erythropoietin, somatropin).^1-3 One definition of biosimilar is “a new biological product similar to an already authorized medicine . . . similar but not identical to the biological reference medicine.” As outlined below, such proteins can and will never be truly 100% identical to the original protein.

In a recent article written by members of the Biosimilar Medicinal Products Working Party (BMWP; http://www.ema.europa.eu/ema/index.jsp?curl=pages/contacts/CHMP/people_listing_000024.jsp) they highlight the need for a precise terminology and propose a narrow definition of biosimilars: “A biosimilar is a copy version of an already authorized biological medicinal product with demonstrated similarity in physicochemical characteristics, efficacy and safety, based on a comprehensive comparability exercise.”⁴ Their pledge for a consistent and correct usage of the term biosimilar is mainly driven by the wish to avoid misinterpretation and confusion.

The human insulin (HI) molecule, as a nonglycosylated, disulphide-bonded heterodimer of 51 amino acids, is the result of numerous years of evolution. To achieve its many biological activities (blood glucose lowering is only 1 of these), this protein has not only a defined primary structure, but also a well-defined secondary and tertiary structure. Each and every modification of this structure may have a relevant impact on the effects of HI.

In the past decades a number of insulin analogs were developed. These are insulin molecules with a primary structure that differs from that of HI to some extent. Aim of these developments was to achieve insulin formulations that have improved pharmacokinetic (PK) and pharmacodynamic (PD) properties in comparison to the native HI formulations. These rapid-acting and long-acting insulin analogs allow a better coverage of prandial or basal insulin requirements of patients with diabetes.

When insulin analogs were first developed some 20 years ago, there was a fierce discussion about the risks that the introduced molecular changes in the primary structure of the insulin analogs potentially bring to patients treated with these. There are certain differences in the 3-dimensional structure of insulin analogs in comparison to HI that have a more or less small impact on the interaction between the insulin molecules per se and on the binding to the insulin receptor. Such alterations in the binding properties of the insulin analogs to the insulin receptor (and to the receptor of the insulin-like growth factor-1 [IGF-1]) are known to impact the intracellular signaling.⁵ Thus, it is well known that changes in the structure of HI are of relevance for the safety and efficacy of this protein.

HI and insulin analogs that are on the markets in the EU and the United States are manufactured by a relatively small number of manufacturers. These markets have an annual volume of several billion dollars. However, a number of other pharmaceutical companies also manufacture insulin. Their insulins are on the market in many countries and it is clear that these manufacturers will apply for market approval of their biosimilar insulin (BI) in the EU and/or the United States. According to the narrow BI definition presented above, in principle none of the insulins mentioned below should be named BI, as none has been approved yet; however, this imprecise usage of the term is hopefully possible without causing confusion.

BIs Are Not Generics

Because of the fact that there are and will always be certain differences between the originator products and biosimilars, it is clear that the regulatory handling of therapeutic proteins like insulins is more complex than that of, for example, small molecules. BIs differ from small molecule drugs (= generic) in many respects (Table 1).

Table 1.

Differences Between Generics and Biosimilar Insulins.

	Generics	Biosimilar insulins
Product characteristics	Small molecules	Large complex molecules
	Often very stable	Stability requires maintain cooling chain
	Typically taken orally	Devices are often the differentiating factor
Production	Produced by chemical synthesis	Produced in living organisms
		Highly sensitive to manufacturing changes
		Often high production costs
Development	Very limited clinical trials (only bioequivalence studies)	Significant R&D (ie, cell lines)
		Clinical trials to a limited extend
Regulation	Shorter registration procedures in Europe and the United States	Regulatory pathway defined by EMA
		“Comparability” status
	Usually enjoy “substitutability” status	In United States; not part of the BPCI Act
Marketing	No or limited detailing to physicians	Detailing to (specialist) physicians required
		Pharmacists may not substitute (depends on national law)
	Market substitution in pharmacies	Price sensitivity is product specific

Insulin Manufacturing

For many decades the major sources for (animal) insulin were pancreata of pigs and cows. After the discovery of the primary structure of the insulin molecule, total chemical synthesis appeared to be the next logical step for insulin production; however, this process was too complex and expensive at this point in time. Nevertheless, chemical synthesis (beside other novel approaches) might be an attractive option for the future. However, with the invention of the biotechnological production of HI some 25 years ago using genetically modified bacteria or yeast as “production machines,” this high-tech approach is now the predominant way of insulin production. In this sense, also the HI formulations manufactured by different companies that are already on the market are not 100% identical to the insulin made by the human body, and presumably can carry differences between each other that could be owed to different manufacturing processes, even if minimal.

The progress made in the past decades with the production technology of recombinant proteins makes it relatively easy nowadays to manufacture insulins. Because of the fact, that the patents for HI and many insulin analogs have expired by now or will expire relatively soon (insulin glargine [Lantus], 2014-2015; insulin lispro [Humalog], 2013; insulin aspart [NovoRapid], 2017) other companies than the company which has developed a certain insulin analog, for example, insulin glargine (= the originator insulin), can attempt to bring these off-patent insulins to the market. Also, the methods used to manufacture HI/insulin analogs, which are covered by different patents, will expire. Manufacturers constantly try to improve their individual manufacturing process; however, details of this process are usually not communicated. It is not clear if these methods actually improve the manufacturing of insulin to a better quality or higher yield process (= higher economy of scales). Nevertheless, such a reduction in manufacturing costs is of high importance when it comes to the price by which insulins once they are approved as BIs could be sold in comparison to the originator product.

It is important to understand that the nature of the manufacturing process defines the final product (“process is product”). Even if the primary amino acid sequence is clearly identical, this does not hold true for the secondary and tertiary structure of the insulin molecule. Each and every manufacturing process differs from each other to a greater or larger degree; in none are the identical strains of bacteria/yeast, incubation technologies/conditions, and so on used. One reason for this is that the knowledge about the manufacturing details—as mentioned above—is proprietary knowledge of the innovator. The consequence is that insulins produced can and will never be truly identical to the originator protein: at best they are (very) similar. This can have consequences for the antigenicity (eg, by different degradation products), bioavailability, and storage stability. For example, it was reported that insulin glargine expressed in Pichia pastoris (a yeast)—that normally is not glycosylated—had 3 sugar molecules (mannose) attached to it as the result of a different manufacturing process employed by 1 manufacturer of recombinant insulins.⁶ The publication refers to an experimental glycosylated product; however, also a published analysis of the glargine marketed by Biocon showed that this is glycosylated.⁷

The issue is that even “minor” differences/changes in the manufacturing process may have considerable (potentially clinically relevant) effects on the biological effects induced by such insulins. In addition to changes of the insulin molecule per se, attention has also to be given to product related substances/impurities and process related impurities; in particular desamido forms and other forms that may derive from the expression vector or arise from the conversion steps removing the C-peptide and regenerating the 3-dimensional structure.

It should be mentioned that the concerns as discussed above in principle also apply to potential differences of originator insulins as they might result from manufacturing process improvements.

Current Situation

Until now no insulin has been approved as BI in Europe or United States. However, this situation is likely to change soon, that is, if not in 2014 then in 2015. As it holds true for the biosimilar area in general, the world of BI is a complex one with rapid changes and unexpected turns. Interestingly, also established large insulin manufacturers develop insulin analogs that were once invented by their competitors: Lilly is developing an insulin glargine formulation and Sanofi rapid-acting insulin analogs originally developed by Lilly/Novo. The submission of Lilly/Boehringer Ingelheim to the EMA for approval of their new insulin glargine formulation in July 2013 can mean that this might be one of the first “BIs” coming to the market in the EU assuming usual handling times of such applications by this regulatory agency. However, it can be expected that generic insulin glargine formulations will become available in the EU/United States within the next 1 to 2 years. As such large insulin manufacturers have well-established marketing channels, manufacturing expertise, and brand name recognition; they might have an advantage in marketing such insulins compared to new companies entering the insulin arena. So in a sense the race is open on who will be the first to bring a BI to the market; however, it remains to be seen what the market success of these BIs will be.

Review on Reviews

One fundamental issue with BIs is the profound lack of scientific publications about the topic; that is, practically no results of clinical studies have been published so far. Therefore, the information summarized in this review about BI from a scientific and regulatory point of view represents more an overview about the current knowledge mixed with more general information about BIs.^5,7,8 Most existing publications were supported by the manufacturers of the original insulin formulations.

Glucose Clamp Studies

The recent draft of the EMA guidelines for BI describes in detail why glucose clamps are a central part of the approval process and provides a detailed description of how the glucose clamp technology should be employed with respect to study design, study population, insulin dose, and so on. This emphasis on PK and especially PD comparability with the originator biologic is a reflection of what the regulatory authorities believe is needed. For demonstration of similarity of PD properties all manufactures of BIs must have initiated glucose clamp studies with their insulin formulation in comparison to a reference insulin formulation. Unfortunately, no data from a glucose clamp study performed in the EU or the United States for the purpose of being part of an application for BI approval in these regulated markets have yet been published. It is worth to mention that glucose clamps with long-acting insulin analogs and long-acting HI preparations represent a challenging task, there was a fierce discussion about all aspects relevant in this respect some years ago when the PD properties of 2 long-acting insulin analogs were compared.⁹ The flat time-action profile of these insulins and the relatively high variability in metabolic effect in the tail part of the clamp study represent a challenge for any long-acting insulin when assessed with this technique. However, the results of one glucose clamp study used for an application and approval as BI in China was published.¹⁰ In addition, the EMA has published withdrawal reports from 2 applications by an Indian company.

In the Chinese study the efficacy and bioequivalence of Basalin® (manufactured by Gan & Lee Pharmaceutical Co Ltd) were studied, referred to as a “domestic” insulin glargine biosimilar, in comparison to Lantus®, referred to as “imported” glargine. The metabolic effect induced by this long-acting insulin analog was also compared with that induced by NPH insulin (Novolin® N). The 2 insulin glargine formulations were found to be bioequivalent. These insulins were administered as single subcutaneous injections of 0.4 U/kg body weight to 16 healthy subjects who participated in 2 glucose clamp experiments of 24 hours duration each. Six subjects participated in an additional glucose clamp during which NPH insulin. It was not stated in the article that a sample size calculation was performed. Thereby the risk could not be ruled out, that the “negative” result (= no difference between 2 formulations of the same substance) is based on a too small number of subjects studied. In case of insulin glargine which is well-known to show substantial intra- and also intersubject variability, most probably a formal sample size calculation would have asked for a higher number of subjects. It is of note that for the statistical analysis 2 different confidence intervals (CIs) were used, for AUC 80% ~ 125% and for Cmax (70% ~ 143%). This is different at least from the EU guidelines that require that the ratio of a given parameter for the 2 insulin formulation is within the 80%-125% CI. Most probably the EMA or the FDA would express concerns about the data as presented, that is, there is need to work toward standardized glucose clamp methodologies. The application of robust and comparable research standards to BI is of paramount importance to ensure that these products can be safely prescribed to patients.

Some years ago the Indian company Marvel (http://www.mjbiopharm.com/anti_diabetics.htm) was the very first one to apply for getting approvals as BI for their soluble insulin, NPH insulin, and insulin mixture in the EU.⁸ This (first) application was withdrawn and incomplete data from 3 different glucose clamp studies are accessible at the home page of the EMA as part of the withdrawal reports. Not only was the quality of this application mediocre according to the withdrawal reports, also the glucose clamp data presented indicated that the 2 different insulins in their 3 different formulations showed significant (not only statistically but visually appreciable) differences. More recently Marvel submitted a second application in which data about Solumarv, Insumarv, and Combimarv are presented, and biosimilarity is being claimed with Humulin S (soluble = short-acting; Eli Lilly), Humulin I (isophane = medium-acting) and Humulin biphasic (30/70%). The PD properties of all 6 insulin formulations were evaluated in a randomized crossover comparison in manual glucose clamps in 22 (soluble insulin)/48 (long-acting and insulin mixture) Indian patients with type 1 diabetes. The subcutaneously applied insulin dose was 0.2 IU/kg body weight. The clamps were performed in the year 2008 by a CRO with proclaimed experience in the euglycaemic clamp technique in India (Bombay Bio-Research Centre). However, as Figure 1 shows, the results of all clamps performed are quite different from what is known from numerous other clamp studies, that is, the time-action profiles do also not reflect the metabolic effects known from clinical practice. This is not the place to speculate why such results were obtained; however, it is clear from the withdrawal documents that the reviewers at the EMA were not pleased with such data/glucose clamp performance. Several issues have been identified in the study reports of these studies including statistical errors, unclear calculations (statistical analysis plan has not been provided) and inconsistent or missing information were noted. It is important to note that the PK data do not fit to the PD data.

Figure 1.

Glucose infusion rates of Solumarv (soluble insulin), Isomarv (NPH insulin), Combimarv (combination of 30% Solumarv and 70% of Isomarv). These are original figures from the EMA Withdrawal Assessment Report (Solumarv: http://www.ema.europa.eu/docs/en_GB/document_library/Application_withdrawal_assessment_report/2013/02/WC500138886.pdf, Isomarv: http://www.ema.europa.eu/docs/en_GB/document_library/Application_withdrawal_assessment_report/2013/02/WC500138885.pdf, Combimarv: http://www.ema.europa.eu/docs/en_GB/document_library/Application_withdrawal_assessment_report/2013/02/WC500138884.pdf).

Batch-to-batch Variability and Quality Assurance

Manufacturing of proteins is not a continuous process, but one that is performed in batches. It is quite cumbersome (and expensive) to maintain the same level of production quality for each and every batch of insulin manufactured. Even if all laboratory methods used to check the quality of the final product according to good manufacturing practice (GMP) are documented and in place, this is not sufficient to guarantee identical effects of the protein product manufactured. Therefore the “robustness” of the manufacturing process is extremely important to achieve a consistently reliable product. Measures to assess the “precision”/“reproducibility” of the manufacturing process have yet to be defined. It also needs to be considered that complex manufacturing processes are not static; the procedures continuously progress to optimize the yield of the process. Also materials necessary for the manufacturing process (like certain peptidases, eg, Trypsin) may differ from batch to batch in their properties. Therefore the manufacturing processes are not stereotypic, but have an astonishing, dynamic complexity. It is noteworthy that questions around the batch to batch variability of approved insulin formulations have to our knowledge not been systematically investigated from a clinical point of view.

Insulin Antibodies

Animal insulins like porcine insulin or bovine insulin are known to induce formation of insulin antibodies. Impurities are known to boost such reactions. However, invention of HI and high purification of insulin formulations has decreased the relevance of insulin antibodies to a large extent. Nevertheless, also subcutaneous injection of HI induces some antibody formation, but without major consequences for efficacy or safety. Interestingly, HI does not induce less response of the immune system than insulin analogs (with a different primary structure in at least one amino acid); the changes introduced in the analogs are in sections of the insulin molecule that do not induce a more pronounced immune response.¹¹ However, the immune system is able to detect small differences in the molecule structure of a given protein, if this is visible for it.

One might say that even if the insulin manufacturing processes between originator and BI companies differ to a certain extent, that any of those differences will translate into structural differences in the final molecule that will be so small (“microheterogeneity”), that they wouldn’t have any clinical relevance. However, our current knowledge about the immunological potency of such differences is limited, and as far as we know no systematic reviews about immunological questions of insulins have been published in recent years. The clinically relevant question at the end is if a given insulin (= BI) will trigger the production of insulin neutralizing antibodies.¹² An increase in the titers of nonneutralizing antibodies would be of lesser concern.

Regulatory View of the EMA/CHMP

The manufacturer of a given insulin formulation must provide evidence to the regulatory authorities to convince them about the therapeutic equivalence of their product with the originator product to receive approval as a BI. The question is if the differences between the originator and its copy are of clinical relevance or not. Unfortunately such differences cannot be depicted for sure by even the most state-of-the-art in vitro laboratory methods. Identifying potential or real differences does require clinical studies with human beings to demonstrate that the “new” insulin has an equivalent safety and efficacy profile when compared to the original product.

In Europe, the EMA’s Committee for Medicinal Products for Human Use (CHMP) is responsible for the scientific assessment of recombinant proteins like insulin following a “centralized procedure of marketing authorization.” To address the fact that biosimilars are a different class with different compound and manufacturing characteristics when compared to generics, the EMA developed a set of guidance documents that are specific to the development of biosimilar medicines (a list of all relevant documents can be found at the home page of the EMA’s CHMP). One annex to these biosimilar guideline documents is specifically addressing nonclinical and clinical requirements for soluble HI containing products as active pharmaceutical ingredient (API) claiming to be similar to another one already marketed. In a recent draft guideline updating this annex, insulin analogs also are handled.

The EMA requires that biosimilar manufacturers submit data that will fully describe the chemical manufacturing characteristics (CMC, or chemical, manufacturing, control) of their products. Like innovators, manufacturers of biosimilars must completely describe their processes, including detailed and rigorous validation and monitoring of batch-to-batch variability (see above)—and especially the effect of any changes they may have introduced to the manufacturing process. The reference product for comparison must be one that is approved for clinical use within the EU. As the safety concerns about insulin produced by a different manufacturer relate mainly to the potential of different immunogenicity, clinical trials with a sufficient duration (at least 6/12 months) shall be performed (see CHMP guidelines for HI). The primary outcome measure is the incidence of antibodies to the 2 insulins.

The standards specified by EMA to prove that a given insulin is safe and efficient and can be approved as a BI might be regarded as stringent. They in fact may be looked at as measures to protect existing markets for already established insulin manufacturers as they require a considerable amount of logistics and organization that a company that is new to the market has issues to provide. However, the question really is if relatively small studies with several hundreds of patients are sufficient to demonstrate safety on a large scale. Clearly the introduction of a risk management program as required by the European guidelines could be aimed to cover this risk. A systematic and prospective postapproval evaluation of a marketed insulin would provide information about safety and efficacy after prolonged periods of usage by larger groups of patients (see CHMP guidelines for HI). However, at least in Europe/Germany it appears as if until now pharmacovigilance has not been taken very seriously in daily life; serious underreporting of adverse drug reactions has been reported.¹³ Therefore the question is with which certainty clinically meaningful safety issues caused by a BI (or also a marketed insulin) could be identified by the current pharmacovigilance procedures.

Regulatory Approval by the FDA

Many other countries worldwide essentially follow the European guidelines, except for the United States. After a number of years with an intensive discussion the FDA has recently issued a guideline document for biosimilars. As a matter of fact these are 3 documents (1 clinical, 1 for quality, and 1 for Q&A). This guideline is part of the Biologics Price Competition and Innovation Act (BPCI Act). Under the BPCI Act, a biological product may be regarded to be “biosimilar” if data show that, among other things, the product is “highly similar” to an already-approved biological product. Interestingly enough insulin is not regarded as a biosimilar; insulins are regarded as generic and therefore are not covered by the biosimilar guideline.

One expects that “generic” biologic product require substantial evaluations of safety and possibly efficacy before entering the market because by nature they are more complex. However, the requirements as to the overall evaluation program of such products are less demanding when compared to those of an innovator’s program. In fact, the best evaluation for a generic biologic would be one specific to the differences between the innovator’s product and the copy product with specific regard to the level of certainty of molecular structure, the impurity profile and the manufacturing process. Section 505(b)(2) of the Food and Drug Cosmetic Act was the first attempt by the US legislative to address the differences between innovator and copy products. Section 505(b)(2) can be understood as a mechanism that allows for a robust evaluation of the safety and efficacy of a pharmaceutical product copy but also allows for a dramatically reduced overall evaluation program compared to that of a new chemical entity (NCE). Manufacturer of new insulins can approach the FDA for pre–investigational new drug (IND) meetings and scientific advice. Their insulin will be handled by the Center for Drug Evaluation and Research (CDER); however, this approval process does not deal with biosimilars.

In 2009 the US Congress passed the Patient Protection and Affordable Care Act to provide a regulatory mechanism to approve biologics that are similar to the innovators product and the similarity between 505(b)(2) pathway and the biosimilars approval path is apparent. In this act (section 7002 of HR.3590), Congress provided FDA with a path to approve biosimilar biological products. A biological product can be licensed as a biosimilar once the manufacturer can demonstrate that the biological product is biosimilar via data derived from chemico-physical studies, animal studies (including toxicology) and clinical studies including assessments of immunogenicity. In addition, this section allows for a determination of interchangeability. Interchangeability is allowed when the copy product is found to be biosimilar to the marketed product and when patients can be treated with either drug in a single treatment regime, without producing any deleterious effects.

The FDA therefore can advise on the detailed design of an evaluation program that fits the specific differences between 2 biological products. Surprisingly, as mentioned above, insulin is currently not regarded as a biological drug, but as a chemical drug. However, after April 2020 the status of insulin will be changed to that of a biological drug and the approval process will be handled by Center for Biologics Evaluation and Research (CBER). Other biotechnologically produced proteins (that are larger than insulin) are already regarded as biosimilars and handled by CBER. The CDER has published in July 2011 guidance to develop better methodologies to ensure Quality of Innovator and Generic Drug Products (including biologics).

Other Considerations

The manufacturing costs for all insulins are relatively high, at least compared to the costs of many generics. The production technology for insulins is complex and typically requires a huge investment. In addition to setting up and managing GMP compliant manufacturing facilities the costs for the clinical development/market approval are considerable too. Even if this can be managed and accomplished, transport of this heat-sensitive product in an appropriate cooling container, storage, distribution, and marketing are cost-intensive elements too. Therefore, the economic advantage that can be achieved with BIs may not be as high as for many generics but will very likely still be significant. It is clear from the payer perspective that it would be attractive to have more competition between insulin manufacturer as this is assumed to provide access to insulin at more affordable prices; this is especially relevant in view of the rapid increase in the number of patients with diabetes who do need insulin treatment.

In looking at today’s prescription habits of physicians, it appears as if these are not mainly driven by economic considerations. The guarantee of long-term availability of a given insulin formulation is of relevance too, as physicians typically try to avoid switching patients from one insulin formulation to another without pressing needs.

Automatic substitution allows for dispensing of generic drugs in place of prescribed innovator products by pharmacists without the knowledge or consent of the treating physician. Interchangeability in contrast means that only the treating physician can change freely from on insulin to the other (which might be a BI). Whereas this approach can be appropriate for generics, this may not to be appropriate for insulins or any biosimilar for that matter. The respective rules in different countries for substitution and interchangeability differ; however, unfortunately also the definitions are not used unanimously. For example in Germany the contracts between the joint organization of the health insurance companies and pharmacies state that the exchange between an originator product and a drug containing the API (= biosimilar) is possible if the API in both cases stems from the same production facility (= identical raw substances and production processes). However, this is not the case for any insulin formulation. It is of note that automatic substitution is not permitted in most European countries and other parts of the world.

Another issue is that a given BI might have different names in different countries, like with many of the marketed insulins right now. In summary, evaluation of a specific side effect profile of a BI in practice might be difficult. In a comment by the EMA about biosimilars, it was stated, “Since biosimilar and biological reference medicines are similar but not identical, the decision to treat a patient with a reference or a biosimilar medicine should be taken following the opinion of a qualified healthcare professional.” European Agency for the Evaluation of Medicinal Products. Questions and answers on biosimilar medicines (similar biological medicinal products) (EMEA/74562/2006). http://www.emea.europa.eu/pdfs/human/pcwp/7456206en.pdf. Accessed 29 October 2007.

Last but not least, another relevant aspect is devices used for insulin application. Depending on their geography, many patients use insulin pens for insulin application. The question is if cartridges containing a given BI can be used in nondedicated reusable pens. Switching between disposable pens without adequate teaching of the patients might induce issues. Another question is if pens developed by manufacturers of BI are equivalent to those developed by manufacturers of originator insulin. Also, it is not clear who is responsible for educating patients when switching from one device to another.

Summary

One tends to assume that the insulins that are moving toward being approved as BIs will be as effective and safe as the originator insulins; however, based on the complexities of the manufacturing process of insulin, and looking at the drug safety and efficacy considerations that need to be applied, a word of caution is appropriate, and it should not at all be surprising that the requirements developed by regulatory bodies for such an approval process are stringent and demanding. Considering certain potential risks (eg, immunogenicity), more extensive clinical trials (eg, postapproval) might be of help and could be part of a risk management program as suggested, for example, by EMA. It is of interest to note that the majority of marketing applications for biosimilars (others than BI) were in fact successful (they received market approval). These approvals were for more complex proteins than insulin.

It is reasonable to expect that a number of insulin formulations will be approved as BIs in regulated markets like the United States and the EU within the next years; probably insulin glargine will be the first BI in these markets. While the economics may play out to the advantage of the health care sector, the confusion of patients, physicians, pharmacists, and so on with a plethora of additional insulin formulations and choices may actually increase. Questions around interchangeability, substitution, and devices will cover critical aspects in this context. Until recently academic diabetology has not paid much attention to the impact that the market introduction of BI might have on the treatment of patients with diabetes.