What Biosimilar Insulins Must Prove

A biosimilar insulin is not approved because its amino-acid sequence looks close or because it lowers glucose in an informal comparison. The sponsor must show that the product is highly similar to a particular reference biologic, allowing only minor differences in clinically inactive components, and that there are no clinically meaningful differences in safety, purity, or potency.

FDA calls this a totality-of-evidence approach. Detailed analytical and functional comparison forms the foundation. Manufacturing quality, pharmacokinetics, pharmacodynamics, and immunogenicity address remaining uncertainty. A large stand-alone efficacy trial may add little if more sensitive comparisons already show similarity.

Why insulin follows a biologics pathway#

Small-molecule generic drugs are usually made by chemical synthesis and can often be characterized as the same active ingredient at the molecular level, but recombinant insulin is produced in living cells, then recovered, purified, formulated, and tested. The process can influence folding, aggregation, variants, impurities, and stability.

Normal batch-to-batch variation exists even for an established reference biologic. The regulatory question is not whether two batches contain atoms in perfectly identical arrangements. It is whether the proposed product's quality attributes fall within a scientifically justified similarity assessment and whether any observed differences could matter clinically.

In March 2020, insulin products that had historically been approved under the drug pathway transitioned to regulation as biologics under the Public Health Service Act, and that transition enabled reference products to support applications under the abbreviated 351(k) biosimilar pathway. It did not change the medicine inside existing packages on that date.

The reference product defines the comparison#

A 351(k) application names one US-licensed reference product. The sponsor builds a comparative program against lots of that product; a basal insulin cannot become a biosimilar to another basal insulin merely because both have a long action profile; molecular design and regulatory lineage matter.

The Purple Book records FDA-licensed biologics, their reference relationships, and whether FDA has licensed a product as biosimilar or interchangeable; similar-sounding brand names, a lower price, or a payer's preferred-product list do not tell you the regulatory relationship.

Some insulin products entered the market through other legal pathways before the transition and may be follow-on products without a 351(k) biosimilar designation, and everyday language can blur these histories. The application type and the Purple Book entry give you the precise answer.

Analytical similarity is the foundation#

The sponsor compares primary structure, higher-order structure, and charge variants. It compares aggregation, purity, impurities, and other product-specific quality attributes. Modern tools can detect very small differences, so the task is to understand whether a difference is expected process variation, clinically inactive formulation variation, or a feature that could alter function or immune response.

Functional assays test insulin-receptor binding and downstream biological activity. Potency assays assess whether the product produces the expected effect in a controlled system, and orthogonal methods, meaning methods based on different scientific principles, reduce the chance that one test misses an important difference.

The comparator program uses multiple lots to characterize the reference product's variation. A single vial-to-vial match would be weak evidence. Statistical methods and risk ranking help identify which quality attributes are most closely tied to clinical performance. None of that excuses poor manufacturing, because the proposed product must also meet biologics manufacturing requirements. Those cover identity, strength, and quality. They also cover purity and potency.

Manufacturing is part of the clinical claim#

Cell banks, fermentation or culture, and recovery must be defined and validated. So must purification, filling, container closure, and cold-chain controls. Process controls aim to keep critical quality attributes within limits. Release tests check each lot, while stability studies support storage conditions and shelf life.

Impurities may come from host cells, the production process, degradation, or formulation. Aggregates can be especially important because they may change potency or immune response. Container materials, silicone oil, particles, and device performance can also matter for an injectable product.

FDA inspects manufacturing facilities and reviews chemistry, manufacturing, and controls. Biosimilar licensure is therefore not a one-time comparison followed by unmonitored production. Manufacturers must maintain the approved process, report specified changes, investigate deviations, and meet continuing quality obligations.

PK and PD comparisons are central for insulin#

Pharmacokinetics, or PK, describes concentration over time; pharmacodynamics, or PD, describes biological effect over time. For insulin, comparative studies can measure both under tightly controlled conditions.

The euglycemic clamp is a sensitive PD method. Participants receive study insulin while glucose is infused and adjusted to keep blood glucose near a target. The glucose infusion rate over time reflects the insulin effect. Prespecified metrics can compare total effect, maximum effect, onset, and duration while PK assays compare insulin concentrations.

Clamp studies are technically demanding. Meal timing, activity, and background insulin can add variability. So can glucose targets, assay specificity, sampling, and the population studied. Randomized crossover designs may let the same participant receive both products, reducing between-person variation when scientifically suitable.

The goal is not to prove the reference product works again. Its effectiveness was established in the original approval. The comparative program asks whether the proposed biosimilar differs meaningfully from it.

Why a large efficacy trial may be unnecessary#

Traditional diabetes trials that compare average glycated hemoglobin can be less sensitive to product differences than a controlled PK and PD comparison. Glycated hemoglobin is influenced by adherence, diet, and activity. It is influenced by rescue changes, monitoring, and adjustment of other medicines. Similar results might therefore occur despite a modest product difference, or different results might arise from trial behavior rather than the insulin.

FDA uses a stepwise approach. If analytical, functional, and clinical pharmacology evidence resolves the relevant uncertainty, an additional comparative efficacy trial may not improve confidence. FDA has discretion to determine that an element is unnecessary.

That is an evidence-efficiency decision, not a relaxed standard. The statutory conclusion remains high similarity with no clinically meaningful differences. If uncertainty persists, FDA can require targeted studies.

Immunogenicity remains a specific concern#

Proteins can trigger antidrug antibodies. Insulin antibodies may have no clinical effect. Or they may affect PK, glycemic variability, or dose requirements. They may affect allergy or rare immune complications. Assays must be sensitive, specific, and able to distinguish relevant antibodies amid circulating insulin and other interferences.

FDA's insulin-specific immunogenicity guidance discusses when comparative clinical immunogenicity studies may be needed. The decision depends on analytical similarity, impurity and aggregate profiles, prior knowledge of the reference product, PK and PD results, and the ability of a study to detect a meaningful difference.

“No clinically meaningful differences” does not mean zero adverse events. It means the biosimilar is expected to have the same clinical safety and effectiveness profile as the reference product. Known insulin risks, including hypoglycemia, remain relevant to both.

Extrapolation is scientific, not automatic#

A biosimilar can be licensed for one or more reference-product indications without a separate efficacy trial in every indication, though the sponsor must scientifically justify extrapolation using mechanism of action, receptor biology, PK, immunogenicity, toxicity, and other factors across the conditions of use.

For insulin, the same core receptor mechanism supports broad inference, but the exact licensed indications, age groups, presentations, and label conditions still control, and a reference product's every historical presentation or device is not automatically copied.

Extrapolation should not be confused with off-label use. The former is an FDA-reviewed basis for putting an indication in the biosimilar label. The latter is clinical use outside approved labeling.

Biosimilar and interchangeable are different terms#

Every interchangeable product first meets the biosimilarity standard. Interchangeability adds statutory findings related to substitution: the product can be expected to produce the same clinical result in any given patient, and for products given more than once, switching risk must not exceed the risk of continued reference-product use under the applicable standard.

FDA's accumulated experience has led it to reconsider when dedicated switching studies are needed. Its June 2024 update was draft guidance, not a binding final rule, and for the status of a specific product, the current FDA determination in the Purple Book is what you should trust.

Interchangeable does not mean a non-interchangeable biosimilar is less safe or effective. A sponsor may not request the designation, or the evidence package may be on a different timeline. Clinicians can prescribe either; automatic pharmacy substitution depends on FDA status and state law.

Device and formulation details still matter#

Insulin is a high-alert medicine, and delivery errors can cause severe hypoglycemia or hyperglycemia. Two related products may differ in pen design, priming, or dose window. They may differ in needle compatibility, package color, storage after opening, or patient instructions. Concentrations and mixtures must never be assumed from a familiar brand family.

The biosimilar comparison supports the insulin product's clinical similarity. It does not make every delivery device operationally identical. FDA separately reviews device performance, human factors, container closure, and labeling as applicable.

Transitions can also reveal coverage, prior authorization, supply, and out-of-pocket differences. Those economic facts do not determine biosimilarity, but they can decide whether you stay on the same product. Clear product identification and education reduce avoidable errors.

Naming and safety monitoring#

Most newer biologics have a core nonproprietary name plus a four-letter suffix. Brand name, proper name, and manufacturer may all help trace a report. So may lot and device. That documentation earns its keep when several products share a core name.

FDA reviews adverse-event reports, manufacturing deviations, safety signals, and postmarket data for biosimilars as it does for other biologics. A report after switching does not by itself establish causation. Timing, glucose records, and technique may need assessment. So may intercurrent illness, diet, other medicines, and the specific lot.

Postmarket surveillance complements approval evidence. Rare events may be too uncommon for preapproval studies, and real-world use includes wider populations and more varied delivery conditions. A safety signal can lead to investigation, labeling change, manufacturing action, or other regulatory response.

How to verify a product claim#

Start with the exact proper or brand name printed on the carton. Search the FDA Purple Book to identify the reference product, application type, biosimilar status, and interchangeability status. Then read the current prescribing information for indication, strength, and presentation. Read it for device, storage, and instructions.

Marketing terms such as “alternative,” “follow-on,” or “similar insulin” are not substitutes for the FDA designation. “FDA registered facility” also does not mean a product is FDA licensed. The regulatory database and the label answer a different question from the one your coverage answers.

Sources#

1. FDA review and approval of biosimilar and interchangeable products
2. FDA biosimilars basics for patients, including insulin
3. FDA Purple Book database of licensed biological products
4. FDA clinical pharmacology data to support biosimilarity
5. FDA clinical immunogenicity considerations for biosimilar insulin products
6. FDA 2024 draft update on demonstrating interchangeability

Insulin changes require an exact product check and a safe transition plan.*