{"id":18819,"date":"2023-08-15T10:20:33","date_gmt":"2023-08-15T14:20:33","guid":{"rendered":"https:\/\/www.drugpatentwatch.com\/blog\/?p=18819"},"modified":"2026-04-13T20:53:35","modified_gmt":"2026-04-14T00:53:35","slug":"unsuccessful-regulatory-filings-for-biosimilar-approval-a-comprehensive-review","status":"publish","type":"post","link":"https:\/\/www.drugpatentwatch.com\/blog\/unsuccessful-regulatory-filings-for-biosimilar-approval-a-comprehensive-review\/","title":{"rendered":"Why Biosimilar Filings Fail: The Complete Technical Guide to 351(k) Rejections, CRL Patterns, and CMC Pitfalls"},"content":{"rendered":"\n

1. The Stakes: What a Failed 351(k) Filing Actually Costs<\/strong><\/h2>\n\n\n\n
\"\"<\/figure>\n\n\n\n

A Complete Response Letter (CRL) is not an administrative inconvenience. For a mid-sized biosimilar developer, it is a balance-sheet event. Development costs for a biosimilar program run between $100 million and $300 million, depending on therapeutic class and the extent of clinical data required. A single CRL, if it triggers a full re-inspection cycle at a manufacturing facility, adds 12 to 18 months of delay. On a product targeting a $2 billion reference biologic, that delay translates directly to foregone first-mover pricing premiums and, in concentrated markets, permanent market share loss to a competitor that filed first.<\/p>\n\n\n\n

Between 2018 and 2022, 37% of all biologics license applications (BLAs) and new drug applications (NDAs) received CRLs. For biosimilars specifically, the failure patterns are predictable enough that the deficiencies landing in CRLs have become, over the past decade, a de facto curriculum for what not to do in a 351(k) submission. The problem is that until July 2025, that curriculum was mostly locked away in confidential regulatory correspondence that companies had every incentive to minimize in their public disclosures.<\/p>\n\n\n\n

A 2015 BMJ cross-sectional study documented that sponsors omitted approximately 85% of safety and efficacy concerns cited in CRLs from their public statements. In roughly 40% of cases where the FDA recommended a new clinical trial, that requirement never appeared in the sponsor’s press release. The institutional incentive to suppress bad news has, for years, prevented systematic learning across the industry, and the same structural failure pattern has recurred across sponsors, programs, and therapeutic classes.<\/p>\n\n\n\n

That changed on July 10, 2025, when the FDA published more than 200 previously confidential CRLs issued between 2020 and 2024, making them available through the openFDA platform. The release is the most significant transparency event in biosimilar regulatory history. It gives IP teams, R&D leads, and investors the clearest view yet of exactly what kills a biosimilar filing.<\/p>\n\n\n\n

Key Takeaways: Section 1<\/strong><\/h3>\n\n\n\n

The financial exposure from a CRL is not limited to remediation costs. It includes foregone first-mover exclusivity, contracted supply chain commitments, and potential impairment of partnership valuations. R&D leads and portfolio managers should price CRL risk explicitly into program IRR calculations from Phase 1 onward, not retrospectively after a filing rejection.<\/p>\n\n\n\n

\n\n\n\n

2. Regulatory Architecture: FDA 351(k), EMA, and the Global Patchwork<\/strong><\/h2>\n\n\n\n

The FDA 351(k) Pathway Under the BPCIA<\/strong><\/h3>\n\n\n\n

The Biologics Price Competition and Innovation Act (BPCIA), enacted in 2010, established the 351(k) abbreviated BLA pathway for biosimilars. The standard requires a sponsor to demonstrate that its product is ‘highly similar’ to the reference biologic, with no clinically meaningful differences in safety, purity, or potency. The pathway uses a totality-of-evidence approach: analytical data, nonclinical studies, clinical pharmacology (PK\/PD), and, historically, a comparative clinical trial are assembled and evaluated together, with no single data type carrying absolute weight.<\/p>\n\n\n\n

The BPCIA also created a 12-year period of reference product exclusivity, during which the FDA cannot approve a 351(k) application. This exclusivity runs independently of patent protection, meaning a biosimilar developer can hold a valid, approved 351(k) application and still be blocked from marketing by unexpired patents or unresolved patent litigation under the BPCIA’s ‘patent dance’ framework.<\/p>\n\n\n\n

Under the patent dance, the biosimilar applicant must share its application with the reference product sponsor within 20 days of FDA acceptance. The parties then exchange lists of potentially infringed patents and negotiate which will be subject to immediate litigation versus a ‘later patent’ list. This structured pre-litigation process has been contested in courts and, in practice, many sponsors have opted out of portions of it, accepting the consequence of blocking the reference product sponsor from certain preliminary injunction remedies.<\/p>\n\n\n\n

EMA: The Senior Biosimilar Regulator<\/strong><\/h3>\n\n\n\n

The EMA established the world’s first biosimilar approval pathway in 2004. With more than 20 years of biosimilar regulation, Europe has the deepest institutional experience and the most developed guidance infrastructure. As of September 2024, 103 biosimilars had been approved in Europe versus 51 in the United States. The EMA evaluates biosimilars through its Committee for Medicinal Products for Human Use (CHMP), and a negative CHMP opinion or sponsor withdrawal before opinion is the European equivalent of a CRL.<\/p>\n\n\n\n

EMA guidance on comparability exercises (EMEA\/CHMP\/BMWP\/42832\/2005, updated through subsequent Q&As) requires the same totality-of-evidence approach as the FDA, but the EMA has historically been somewhat more prescriptive about the specific analytical techniques required in comparability exercises. The EMA also publishes European Public Assessment Reports (EPARs) for every approved or refused application, providing a richer public record of regulatory reasoning than the FDA has historically maintained.<\/p>\n\n\n\n

PMDA, Health Canada, and Emerging Regulators<\/strong><\/h3>\n\n\n\n

Japan’s Pharmaceuticals and Medical Devices Agency (PMDA) adopted biosimilar guidelines in 2009, broadly aligned with EMA and WHO frameworks. The PMDA requires a Japan-specific PK bridging study using Japanese reference product if the global reference data uses only the US or EU originator. This bridging requirement has caught several sponsors who planned global programs without accounting for the Japanese reference product as a separate analytical anchor.<\/p>\n\n\n\n

Health Canada issued its own biosimilar guidance in 2010, updated in 2016 and 2021, and follows a comparability framework substantially similar to the EMA’s. Emerging regulators in South Korea (MFDS), Brazil (ANVISA), and China (NMPA) have each developed independent frameworks that borrow from EMA and WHO guidance but add jurisdiction-specific requirements for local clinical data or local manufacturing oversight.<\/p>\n\n\n\n

The regulatory divergence across these authorities creates a genuine global filing strategy problem. A sponsor that achieves EMA approval for a biosimilar rituximab cannot simply cross-reference that package in a US 351(k) submission without bridging studies, because the reference products are sourced from different geographic markets and may carry different glycosylation profiles resulting from different commercial manufacturing processes over time. This reference product drift problem has generated CRLs and refusals that sponsors failed to anticipate at program inception.<\/p>\n\n\n\n

Investment Strategy Note: Regulatory Architecture<\/strong><\/h3>\n\n\n\n

Biosimilar developers with multi-market ambitions need to budget for the incremental cost of jurisdiction-specific bridging studies from Day 1. Programs that plan a ‘file-one-use-everywhere’ regulatory strategy routinely encounter expensive late-stage surprises. Portfolio managers evaluating pipeline assets should ask specifically whether the program includes Japan-specific PK bridging data and EU\/US reference product cross-comparisons, as absence of either is a CRL risk factor.<\/p>\n\n\n\n

\n\n\n\n

3. The 2025 Transparency Shift: FDA’s Public CRL Database<\/strong><\/h2>\n\n\n\n

What the Database Contains<\/strong><\/h3>\n\n\n\n

On July 10, 2025, the FDA made public 202 CRLs issued between 2020 and 2024, redacted to remove trade secrets and confidential commercial information, and available through open.fda.gov. Critically, this initial release covers only CRLs for products that were subsequently approved, meaning the database captures recoverable failures, not the graveyard of programs that never resubmitted. The FDA has indicated it may release additional CRLs from programs that did not advance, which would substantially expand the learning dataset.<\/p>\n\n\n\n

Among the 202 letters, approximately 46 apply to biological drug applications, of which roughly 40% are 351(k) biosimilar applications. This breakdown confirms that biosimilar approvals generate a disproportionate share of biologics CRLs relative to novel BLAs, consistent with the structural complexity of demonstrating similarity to an approved reference product while simultaneously satisfying CMC and facility inspection requirements.<\/p>\n\n\n\n

The Disclosure Gap Problem<\/strong><\/h3>\n\n\n\n

The release confirms what the 2015 BMJ study documented: sponsors systematically disclosed minimal information about CRL content in their public communications. In one documented pattern within the new database, CRLs cited safety signal concerns that the sponsoring company’s press release described only as ‘manufacturing-related.’ In several biosimilar CRLs, facility inspection failures were the stated basis, while the sponsor’s public statement referenced only ‘information requests’ from the FDA, framing a manufacturing deficiency as a data-gathering exercise rather than a remediation requirement.<\/p>\n\n\n\n

For investors, the historical disclosure gap has meant that biosimilar pipeline valuations have systematically underweighted CRL risk. Programs that received a first CRL were rarely repriced by institutional investors to reflect the full probability-adjusted delay cost, because the market had limited information about how deep the deficiency ran. The public CRL database changes that calculus.<\/p>\n\n\n\n

Patterns in the Biosimilar CRL Subset<\/strong><\/h3>\n\n\n\n

An analysis of 89 CRLs issued between January 2024 and January 2025 by The FDA Group found that facility inspection-related approvability issues appeared in 56% of all CRLs. Clinical or clinical\/statistical failures appeared in over 30%. Many CRLs contained multiple deficiency categories, indicating systemic preparation failures rather than isolated technical gaps. For biologics specifically, CBER CRLs (which cover therapeutic proteins and biosimilars regulated by the Center for Biologics Evaluation and Research) frequently cited facilities requiring ‘additional review and verification during a subsequent inspection,’ regulatory language that typically translates to a 12- to 18-month delay before a successful pre-approval inspection can be completed.<\/p>\n\n\n\n

Manufacturing issues appeared in nearly three-quarters of all CRLs across the full 202-letter dataset. For biologics and biosimilars specifically, the most common CMC deficiencies cited were comparability failures, immunogenicity data gaps, analytical method limitations, and facility inspection findings. These four categories recur across sponsors, programs, and therapeutic classes with a regularity that makes them the clearest target for preemptive quality investment.<\/p>\n\n\n\n

Key Takeaways: Section 3<\/strong><\/h3>\n\n\n\n

The FDA’s CRL database converts what was previously anecdotal institutional knowledge into a structured, publicly searchable record of regulatory failure modes. IP teams and regulatory affairs groups should map their own pipeline programs against the deficiency taxonomy visible in the database before any 351(k) submission. Portfolio managers should treat any biosimilar program with an active manufacturing site inspection cycle as carrying elevated CRL risk that is not fully priced into most pipeline valuation models.<\/p>\n\n\n\n

\n\n\n\n

4. Why Biosimilar Filings Fail: A Data-Driven Taxonomy<\/strong><\/h2>\n\n\n\n

4.1 Manufacturing and GMP Deficiencies<\/strong><\/h3>\n\n\n\n

Manufacturing deficiencies are the dominant cause of biosimilar CRLs and refusals, appearing in the majority of non-approval decisions across both FDA and EMA. The structural reason is that biologics are produced in living cell systems, and every aspect of that production, from cell line characteristics to upstream bioreactor parameters, downstream purification steps, formulation, fill, and finish, must be precisely characterized and reproducibly controlled. A change in cell culture media, bioreactor dissolved oxygen setpoint, or chromatography resin lot can alter glycosylation patterns, charge variant distributions, or aggregation profiles in ways that compromise analytical comparability to the reference product.<\/p>\n\n\n\n

The FDA’s pre-approval inspection (PAI) process is the final gate before a CRL or approval decision. Inspectors assess whether the manufacturing site’s quality systems are capable of consistently producing the proposed product within the specifications demonstrated in the application. ‘Systemic’ deficiencies, a word that appears repeatedly in the publicly released CRLs, are the most serious outcome. A systemic deficiency finding means the failure is not an isolated deviation but a fundamental flaw in how the quality management system is designed or executed. Remediation of a systemic deficiency typically requires the facility to implement corrective and preventive actions (CAPAs), submit evidence of their effectiveness, and pass a re-inspection before the FDA will revisit the application.<\/p>\n\n\n\n

The COVID-19 pandemic introduced a novel category of manufacturing-related CRL: facility inspection deferrals. Travel restrictions between 2020 and 2022 prevented the FDA from conducting pre-approval inspections at manufacturing sites in China, India, and parts of Europe. Multiple CRLs in the public database explicitly cite the FDA’s inability to complete a required facility inspection as the basis for non-approval, a category of failure that had nothing to do with product quality and everything to do with geopolitical disruption to inspection logistics. Several biosimilar programs lost 18 to 24 months from their commercial launch timelines for this reason alone.<\/p>\n\n\n\n

Process Validation and Control Strategy Gaps<\/strong><\/p>\n\n\n\n

Beyond inspection findings, CMC CRLs commonly cite process validation gaps. For biosimilars, process validation must demonstrate that the manufacturing process produces a product that is consistently within the analytical similarity range established during development. If the process validation campaign uses different operating conditions than those documented in the process description, or if the analytical methods used in validation are not the same as those in the proposed product specification, the FDA will identify the disconnect and issue a deficiency. Stability data gaps represent another recurring category: if the stability program does not cover the proposed shelf life, or if intermediate conditions have not been tested, the CMC package is incomplete regardless of the quality of the comparability data.<\/p>\n\n\n\n

Container closure integrity testing and extractables\/leachables studies are recurring product quality deficiencies that are heavily redacted in the public CRL dataset but visible as categories in regulatory feedback patterns. For biosimilars that use prefilled syringes or autoinjectors, device compatibility data must demonstrate that the drug product is not adversely affected by contact with the device components over the proposed shelf life, and that the delivery device functions consistently within the proposed use environment.<\/p>\n\n\n\n

4.2 Analytical Similarity Gaps and Method Validation Failures<\/strong><\/h3>\n\n\n\n

Analytical comparability is the scientific foundation of every biosimilar application. It is also where the complexity of biologics development most reliably exceeds sponsor expectations. A rigorous analytical comparability exercise requires characterizing the reference product and the proposed biosimilar across a comprehensive set of structural and functional attributes: primary sequence confirmation by peptide mapping, higher-order structure characterization by circular dichroism and hydrogen-deuterium exchange mass spectrometry, glycan profiling by HILIC-fluorescence or mass spectrometry, charge variant analysis by isoelectric focusing and cation-exchange chromatography, aggregation by size-exclusion chromatography and analytical ultracentrifugation, and biological activity by multiple orthogonal bioassays covering each relevant mechanism of action.<\/p>\n\n\n\n

Each of these analytical methods must itself be validated to demonstrate specificity, linearity, precision, accuracy, range, and robustness before the data generated by the method can support a regulatory claim. Analytical method validation failures are among the most consequential CMC deficiencies, because a single invalidated method can compromise 18 months or more of comparability data generated using that method, requiring complete revalidation and re-testing. The FDA Group documented a specific case in which a CRL citing ‘inadequate analytical method validation’ led to the discovery that 7 analytical methods required complete revalidation, invalidating 18 months of stability data and incurring several million dollars in additional costs.<\/p>\n\n\n\n

Fingerprint-like analytical similarity assessment, the practice of using a large panel of orthogonal characterization methods to build a comprehensive structural and functional profile, has become the regulatory standard for demonstrating biosimilarity. The FDA’s totality-of-evidence approach weighs the completeness of the analytical package heavily. A submission that relies on a narrow set of compendial tests without demonstrating higher-order structural similarity through advanced techniques like native mass spectrometry or cryo-electron microscopy is increasingly likely to receive an analytical deficiency. As of September 2024, the FDA noted that analytical assessments had identified issues in 6 biosimilar applications (out of 80 total) that led to non-approval, while clinical studies identified only one of those same six issues and contributed no unique deficiencies of their own. This asymmetry is the empirical basis for the FDA’s current direction toward analytical data as the primary similarity evidence, with clinical studies used to resolve residual analytical uncertainty rather than as a routine requirement.<\/p>\n\n\n\n

Reference Product Sourcing and Lot Selection<\/strong><\/p>\n\n\n\n

A frequently underestimated analytical pitfall is reference product sourcing and lot selection. The FDA requires that the proposed biosimilar be compared to the US-licensed reference product, not to the EMA-approved version, even if both originate from the same innovator. Lots of the US reference product must be obtained from US commercial channels and must represent the natural analytical variability of the reference product across its product lifecycle. If the sponsor characterizes only recently manufactured reference lots, without accessing older lots that may carry different glycan profiles or charge variant distributions, the comparability margins defined in the application may not encompass the full range of reference product variability. Reference lots manufactured using updated innovator processes may not be representative of what was in use when the innovator’s clinical database was built, creating a subtle but real comparability challenge.<\/p>\n\n\n\n

4.3 Clinical, PK\/PD, and Immunogenicity Shortfalls<\/strong><\/h3>\n\n\n\n

The role of clinical data in biosimilar approval has shifted substantially since 2024. The FDA’s June 2024 draft guidance eliminated the routine requirement for comparative clinical efficacy trials for biosimilar approval, reflecting the agency’s empirical conclusion that comparative clinical studies add little regulatory value once a rigorous analytical comparability package is assembled. For interchangeability designation, the same guidance eliminated the requirement for switching studies, permitting interchangeability determinations based on comparative analytical and clinical PK data. This regulatory shift has allowed developers like Formycon, Sandoz, and Bio-Thera to cancel or minimize Phase 3 trials for their pembrolizumab biosimilars, replacing them with Phase 1 PK studies plus expanded analytical packages.<\/p>\n\n\n\n

Despite this shift, PK comparability remains a mandatory element. A single-dose, three-arm crossover PK study comparing the proposed biosimilar to US-licensed and, where needed, EU-approved reference product is standard. CRLs citing PK deficiencies most commonly arise from underpowered studies, protocol deviations that compromise data integrity, or predefined equivalence margins that are too tight to be achievable given reference product natural variability. Sponsors that set 90% confidence interval criteria of 80-125% (bioequivalence-style) for a biosimilar PK study are often applying a standard appropriate for small-molecule generics to a biologics context where reference product variability routinely challenges that margin.<\/p>\n\n\n\n

Immunogenicity remains a clinical wildcard even in a post-clinical-trial regulatory environment. All therapeutic proteins carry some potential to elicit anti-drug antibodies (ADAs), and the immunogenicity profile of a biosimilar must be comparable to the reference product. CRLs citing immunogenicity issues typically fall into three categories: the ADA assay was not sufficiently sensitive to detect relevant antibodies, the clinical study was too short to capture late-onset immunogenicity, or the observed ADA rates in the biosimilar arm exceeded the reference product arm to a degree that raised a clinical concern. Post-marketing immunogenicity surveillance is also a regulatory commitment for approved biosimilars, and inadequate risk management plans (RMPs) addressing immunogenicity monitoring have been cited in deficiency letters.<\/p>\n\n\n\n

Extrapolation of indications is a related clinical issue. The FDA’s totality-of-evidence approach permits a biosimilar to receive approval for all reference product indications based on data from one indication, if the sponsor can justify that the mechanism of action, patient population, and clinical risk profile are sufficiently similar across indications. Failure to adequately justify extrapolation, particularly for biologics with multiple distinct mechanisms of action (such as anti-TNF antibodies used in both rheumatology and gastroenterology), can limit the approved label in ways that constrain the commercial opportunity.<\/p>\n\n\n\n

4.4 Regulatory Documentation and Dossier Errors<\/strong><\/h3>\n\n\n\n

Dossier quality failures are the most avoidable category of biosimilar CRL and, in some ways, the most embarrassing for large organizations with well-resourced regulatory affairs teams. Refuse-to-File (RTF) decisions, which precede a CRL and represent the FDA’s determination that a submission is not complete enough to accept for review, can result from missing modules in the eCTD format, labeling inconsistencies between the proposed package insert and the clinical section, or failure to include required bridging data for reference products from different geographic markets.<\/p>\n\n\n\n

Within the CRL dataset, several 351(k) applications received deficiency letters for missing or inadequate Risk Evaluation and Mitigation Strategies (REMS) data in cases where the reference product carried a REMS. Others received deficiencies for labeling that did not accurately reflect the indication scope supported by the comparability data. One case in the public CRL database received a CRL for a problem as straightforward as missing required copy on the product label, a failure that delayed a program by months.<\/p>\n\n\n\n

Incomplete or poorly structured bridging studies represent a structurally important documentation failure category. When a sponsor has generated PK or comparability data using the EU-approved reference product but is filing a US 351(k), the dossier must include bridging data that demonstrates the EU and US reference products are themselves sufficiently similar to make the EU-based data applicable to the US regulatory determination. Without a well-designed and documented bridging study, the FDA cannot accept the EU-based data as relevant to the US application, and a deficiency letter follows.<\/p>\n\n\n\n

Key Takeaways: Section 4<\/strong><\/h3>\n\n\n\n

Manufacturing readiness is the highest-priority pre-filing risk factor, with facility inspection deficiencies appearing in more than half of all CRLs across the 2020-2024 dataset. Analytical method validation must be treated as a pre-clinical investment, not a late-stage GMP exercise. PK studies for biosimilars require equivalence margins calibrated to reference product natural variability, not small-molecule bioequivalence conventions. Dossier completeness checklists should be reviewed against the most recent FDA eCTD technical specifications before any submission, and every bridging study requirement from the jurisdiction-specific guidance matrix should be verified before submission lock.<\/p>\n\n\n\n

\n\n\n\n

5. Case Studies: Anatomy of High-Profile Failures<\/strong><\/h2>\n\n\n\n

5.1 Alvotech: AVT02 (Adalimumab) and AVT05 (Golimumab)<\/strong><\/h3>\n\n\n\n

Alvotech’s experience with the FDA illustrates how a credible scientific program can be repeatedly delayed by manufacturing facility deficiencies that have nothing to do with the underlying analytical or clinical data package. AVT02, Alvotech’s high-concentration adalimumab biosimilar referencing AbbVie’s Humira, received multiple CRLs tied to observations at the company’s manufacturing facility in Reykjavik, Iceland. Each CRL cycle required the company to implement CAPAs, allow time for the FDA to assess the effectiveness of those actions, and schedule a re-inspection before the application could advance.<\/p>\n\n\n\n

The strategic context makes the delay particularly costly. AbbVie’s Humira entered its US biosimilar competition window in January 2023, when the last of its patent settlement agreements allowing delayed entry by biosimilar manufacturers expired. The market that opened in 2023 was among the most commercially valuable in the history of the biosimilar sector, given Humira’s status as the world’s highest-grossing drug by cumulative revenue. Sponsors that entered the market earliest, such as Amgen’s Amjevita (launched January 2023 at list price parity) and AbbVie’s own citrate-free formulation strategy, set pricing reference points that subsequent entrants had to undercut to gain formulary access. Each month of manufacturing-driven delay cost Alvotech not just revenue but the ability to lock in pharmacy benefit manager (PBM) and payer contracts at favorable terms.<\/p>\n\n\n\n

AVT05, Alvotech’s golimumab biosimilar referencing J&J’s Simponi, received a CRL from the FDA on November 2, 2025, following a pre-license inspection of the Reykjavik facility in July 2025 that identified ‘certain deficiencies.’ The CRL cited no other deficiencies with the BLA itself, confirming that the analytical and clinical package for AVT05 was satisfactory and that the obstacle was entirely facility-related. Separately, AVT05 had received its first global approval in Japan in September 2025 and a positive CHMP opinion in Europe in the same month, demonstrating that the regulatory data package met EMA and PMDA standards. The FDA’s facility concerns related specifically to manufacturing site GMP conditions, independent of the product quality demonstrated in the dossier.<\/p>\n\n\n\n

IP Valuation Note: Alvotech’s Biosimilar Portfolio<\/strong><\/p>\n\n\n\n

Alvotech’s commercial pipeline centers on biosimilars to high-value branded biologics with cumulative originator revenues that make each approved program strategically significant. AVT02’s Humira reference has generated over $200 billion in cumulative global sales for AbbVie. Golimumab generated approximately $3 billion in annual global revenues for J&J at the time AVT05’s applications were under review. For an investor assessing Alvotech’s equity, each month of manufacturing-driven CRL delay on AVT05 represents a direct reduction in the net present value of that program, with the magnitude depending on assumed market share capture rate and first-year pricing strategy relative to Simponi’s WAC price. Pre-approval inspection risk at a single manufacturing site creates correlated CRL risk across the company’s entire pipeline, since multiple programs are manufactured at the same facility, making manufacturing site GMP status a portfolio-level risk factor rather than a program-level one.<\/p>\n\n\n\n

5.2 Biocon\/Mylan: Trastuzumab and Pegfilgrastim (EMA)<\/strong><\/h3>\n\n\n\n

In 2017, Biocon and Mylan withdrew two biosimilar applications with the EMA: a trastuzumab biosimilar (referencing Roche\/Genentech’s Herceptin) and a pegfilgrastim biosimilar (referencing Amgen’s Neulasta). Both withdrawals followed GMP inspections that identified unresolved deficiencies at Biocon’s manufacturing facility in India. Rather than proceed to a negative CHMP opinion, the companies withdrew the applications strategically, addressing the manufacturing issues and resubmitting with updated packages.<\/p>\n\n\n\n

The trastuzumab program was ultimately approved by the EMA and FDA as Ogivri (later Herzuma in some markets), and the pegfilgrastim program advanced to approval as well. The withdrawal-and-resubmit strategy preserved the analytical integrity of the dossier while buying time to remediate the manufacturing concerns, and is now considered standard practice when a sponsor identifies a likely facility deficiency before a scheduled inspection. Proceeding to a negative opinion or refusal creates a more difficult resubmission path than a voluntary withdrawal in most regulatory systems.<\/p>\n\n\n\n

IP Valuation Note: Trastuzumab Biosimilars<\/strong><\/p>\n\n\n\n

Trastuzumab (Herceptin) generated peak annual revenues exceeding $7 billion globally before biosimilar competition. The market entry of multiple biosimilars including Ogivri, Kanjinti (Amgen\/Allergan), Herzuma, Trazimera, and others eroded Herceptin’s revenue substantially, with US Herceptin revenue declining more than 40% in the two years following biosimilar entry. For the biosimilar sponsors, trastuzumab presented an attractive but crowded opportunity: high revenue reference product, well-characterized target, established clinical data package from the originator’s clinical database, but intense competition among biosimilar developers that compressed achievable market share and pricing. The lesson for IP valuation: the gross size of the reference product market overstates the addressable opportunity when multiple 351(k) applications are advancing in parallel. Biosimilar market modeling must account for the number of competing applicants, their filing timelines, and the likelihood of each reaching approval in the same market access window.<\/p>\n\n\n\n

5.3 Xbrane: Ranibizumab (Multi-CRL Cycle)<\/strong><\/h3>\n\n\n\n

Xbrane’s biosimilar ranibizumab program, referencing Genentech’s Lucentis, is a textbook illustration of the re-inspection cycle trap. Xbrane received a first CRL from the FDA in April 2024 citing two specific deficiencies: problems with analytical methods for the reference standard, and pre-approval inspection findings at manufacturing partner sites. Xbrane resubmitted in December 2024 after implementing corrective actions and obtained evidence of those corrections from both production sites.<\/p>\n\n\n\n

The FDA conducted re-inspections at both sites during Q3 2025. Despite the sponsor’s submission of evidence of corrective actions in advance of the re-inspections, the FDA issued a second CRL on October 21, 2025, referencing unresolved observations at one of the production sites. The second CRL identified no other deficiencies with the BLA, isolating the problem to a single site’s inspection outcome.<\/p>\n\n\n\n

The Xbrane case demonstrates a structural risk inherent in multi-site manufacturing arrangements: if even one site fails its re-inspection, the entire application remains blocked regardless of the quality of the data package or the performance of the other manufacturing sites. For sponsors with contract manufacturing arrangements, the inspection risk is partially outside their direct control, creating a category of CRL exposure that cannot be fully mitigated through internal quality investments alone. The ranibizumab market at the time of Xbrane’s CRL difficulties already had two approved biosimilars: Samsung Bioepis’s Byooviz (approved September 2021) and Formycon\/Sandoz’s Cimerli (approved August 2022). The commercial cost of Xbrane’s inspection delays was not merely lost time but a progressively more crowded and price-compressed market at entry.<\/p>\n\n\n\n

5.4 BioPartners: Interferon-alpha (EMA Refusal)<\/strong><\/h3>\n\n\n\n

BioPartners’ interferon-alpha application, one of the earliest biosimilar submissions to the EMA, received a formal refusal rather than a withdrawal. The CHMP determined that BioPartners had not provided adequate analytical similarity data and that manufacturing inconsistencies were present in the submission. The formal refusal is significant because it represents a more difficult starting point for any future development program: a refused application creates a public regulatory record of scientific inadequacy, and the EMA’s Public Assessment Report documents the specific deficiencies in detail.<\/p>\n\n\n\n

The BioPartners case set an early precedent in the EMA biosimilar era: the agency would not accept the argument that structural similarity at a basic level was sufficient without comprehensive functional comparability data and robust manufacturing control demonstration. This early signal shaped how European biosimilar development programs subsequently approached their CMC and analytical comparability exercises, raising the baseline expectation for what constitutes an adequate submission.<\/p>\n\n\n\n

\n\n\n\n

6. The Patent Thicket Problem: IP Valuation as a Pre-Filing Variable<\/strong><\/h2>\n\n\n\n

Before a 351(k) application can achieve commercial relevance, the sponsor must navigate the patent estate surrounding the reference biologic. Originator companies have developed sophisticated layered patent strategies designed to extend market exclusivity well beyond the primary composition-of-matter patent expiration. AbbVie’s Humira is the most documented example: despite being launched in 2002, Humira was protected by more than 250 patents covering formulations (citrate-free, high-concentration), manufacturing processes, dosing regimens, delivery devices, and methods of use for specific patient subpopulations. This patent portfolio effectively delayed meaningful US biosimilar competition until January 2023, 21 years after the drug’s launch.<\/p>\n\n\n\n

Each patent in such a thicket costs approximately $25,000 to obtain but an average of $774,000 to challenge through inter partes review (IPR) at the USPTO, or substantially more through federal court litigation. For a biosimilar developer facing a reference product with 80 relevant patents, the cost of clearing the IP landscape through challenge alone can approach or exceed the cost of the clinical development program. Branded drug companies structure these thickets precisely because it is a numbers game: most biosimilar developers cannot sustain the legal costs of challenging every relevant patent, and even successful IPR proceedings (which historically cancel or modify claims in approximately 62% of patents undergoing reexamination) require years to reach a final non-appealable decision.<\/p>\n\n\n\n

Freedom-to-Operate Analysis as a Pre-Filing Requirement<\/strong><\/p>\n\n\n\n

A rigorous freedom-to-operate (FTO) analysis is not a legal formality; it is a capital allocation decision. A biosimilar program that reaches Phase 1 without clearing its FTO position is a program that may achieve FDA approval and then face a launch-at-risk decision, where the sponsor must choose between delaying commercial launch pending patent resolution or launching and accepting the risk of a preliminary injunction and treble damages. Both outcomes destroy value. The only rational risk mitigation is front-loading the FTO analysis, identifying the specific patents most likely to be asserted, and building the manufacturing process and formulation specifically to avoid infringing valid claims.<\/p>\n\n\n\n

IP Valuation Framework for Biosimilar Reference Products<\/strong><\/p>\n\n\n\n

For investors assessing the NPV of a biosimilar program, the relevant IP valuation question is not the peak revenue of the reference product but the effective date of open competition. A reference biologic with $5 billion in annual revenues but a patent thicket that delays open competition by six years generates a very different biosimilar program NPV than the same revenue product with a clean IP landscape. The effective exclusivity expiration, accounting for all credible patent barriers, is the variable that determines when meaningful biosimilar revenues begin.<\/p>\n\n\n\n

The BPCIA’s 12-year reference product exclusivity runs from the date of first licensure of the reference biologic and cannot be extended by secondary patents. This creates a period, typically between the 12-year exclusivity expiration and the last credible patent expiration, during which a biosimilar developer can hold an approved 351(k) and negotiate a patent settlement or litigate rather than launch at risk. Patent settlements with authorized biosimilar launch dates have become a standard commercial outcome for high-value reference products, and those settlement dates are often the single most important variable in biosimilar program financial modeling.<\/p>\n\n\n\n

Key Takeaways: Section 6<\/strong><\/p>\n\n\n\n

FTO analysis should be completed before IND filing, not before BLA submission. The cost of late-stage IP discovery is disproportionately high because it forces choices between abandoning an approved product, launching at risk, or paying royalty rates that compress already-thin biosimilar margins. Portfolio managers should request a structured IP risk map for any biosimilar program in due diligence, covering the reference product’s patent estate, the expiration dates of the ten most relevant patents, and any existing patent settlement agreements between the innovator and other biosimilar developers.<\/p>\n\n\n\n

\n\n\n\n

7. Interchangeability Designation: Regulatory Shifts and Strategic Implications<\/strong><\/h2>\n\n\n\n

Biosimilar interchangeability under the BPCIA is the designation that permits automatic pharmacy-level substitution for the reference product, without physician intervention, in states that have enacted substitution laws. Historically, achieving interchangeability required a biosimilar developer to conduct switching studies demonstrating that patients alternating between the reference product and the biosimilar experienced no greater risk than those maintained on either product alone. This requirement added cost and time to programs already carrying substantial development burdens.<\/p>\n\n\n\n

The FDA’s June 2024 draft guidance eliminated the switching study requirement for interchangeability designation, permitting the designation based on comparative analytical and clinical data already assembled for the base 351(k) approval. This aligns the US more closely with the EMA framework, where the concept of automatic substitution is handled at the national law level and does not require a separate regulatory designation. As of September 2024, the FDA had deemed five out of nine biosimilars interchangeable without additional switching studies, and the pace of interchangeability designations has since accelerated.<\/p>\n\n\n\n

The practical commercial implications of interchangeability have been contested. Branded manufacturers promoted interchangeability designation as a marker of superior quality relative to non-interchangeable biosimilars, a ‘whisper campaign,’ as HHS Secretary Kennedy described it at an October 2025 press conference, that implied non-interchangeable biosimilars carried inferior safety or efficacy profiles. The FDA’s empirical record shows no evidence that any approved biosimilar has failed a switching study or generated a unique safety signal relative to its reference product. The interchangeability designation is a regulatory category, not a quality hierarchy.<\/p>\n\n\n\n

For biosimilar developers, the elimination of switching studies removes a meaningful cost barrier and reduces the clinical development timeline. The removal of switching studies cuts biosimilar development costs from a range of $100-300 million to $75-250 million and shortens timelines from 7-8 years to 6.5-7.5 years. If comparative Phase 3 efficacy studies are also eliminated in a given program, costs can fall further to $50-75 million with timelines of 5-6 years. These reductions are already reshaping program design decisions across the biosimilar pipeline.<\/p>\n\n\n\n

Investment Strategy Note: Interchangeability<\/strong><\/p>\n\n\n\n

For investors evaluating biosimilar market penetration models, interchangeability designation accelerates formulary placement and pharmacy substitution rates in states with active substitution laws. Programs targeting automatic substitution markets should carry higher market share capture rate assumptions in years 1-2 post-launch than non-interchangeable biosimilars. However, the elimination of switching study requirements reduces the differential between interchangeable and non-interchangeable biosimilars as a market access variable, because the designation can now be obtained at lower incremental cost, making it a more widely available label claim.<\/p>\n\n\n\n

\n\n\n\n

8. Technology Roadmap: How Analytical Science Is Replacing Clinical Trials<\/strong><\/h2>\n\n\n\n

The progressive replacement of comparative clinical efficacy trials with analytical comparability data is not a regulatory concession; it reflects the maturation of analytical technologies to the point where they can detect structural differences between a biosimilar and its reference product with greater sensitivity than a clinical trial. This trajectory has been building since the EMA’s early biosimilar approvals and is now explicit in FDA policy.<\/p>\n\n\n\n

Stage 1: Foundational Comparability (Pre-2015)<\/strong><\/p>\n\n\n\n

Early biosimilar programs relied heavily on clinical efficacy trials to demonstrate that the biosimilar performed comparably to the reference product in patients. Analytical methods were less advanced, and regulators required clinical confirmation because the uncertainty in the structural characterization left meaningful residual questions about whether observed similarities at the molecular level would translate to equivalent clinical outcomes. The analytical toolkit was limited to compendial assays, basic size-exclusion chromatography, and relatively low-resolution mass spectrometry.<\/p>\n\n\n\n

Stage 2: Advanced Characterization Drives Label Expansion (2015-2020)<\/strong><\/p>\n\n\n\n

Higher-order structure characterization tools, including hydrogen-deuterium exchange mass spectrometry (HDX-MS), native mass spectrometry, and cryo-electron microscopy, entered biosimilar development programs at commercial scale. Glycan profiling by hydrophilic interaction liquid chromatography-fluorescence (HILIC-FLR) combined with mass spectrometry enabled quantitative comparison of N-linked glycan structures at a resolution that could detect subtle differences in sialylation, fucosylation, and galactosylation. Regulators began accepting these data sets as primary evidence of biological function comparability for certain product classes, reducing the weight placed on comparative efficacy trials.<\/p>\n\n\n\n

Stage 3: Totality-of-Evidence Matures, Clinical Trials Become Residual (2020-2024)<\/strong><\/p>\n\n\n\n

By 2020, the FDA had accumulated sufficient regulatory experience with approved biosimilars to assess the track record of analytical comparability exercises. The agency’s retrospective analysis found that analytical assessments had correctly identified all six biosimilar applications with significant similarity issues, while clinical trials identified only one of those same issues and generated no unique safety or efficacy signals not already captured by the analytical package. This empirical record, combined with the maturation of fingerprint-like similarity assessment methodology, provided the scientific basis for the FDA’s 2024 guidance eliminating routine comparative efficacy trial requirements.<\/p>\n\n\n\n

Stage 4: Machine Learning and Accelerated Analytical Development (2025-Ongoing)<\/strong><\/p>\n\n\n\n

The current frontier involves applying machine learning to reference product characterization data to define empirically derived similarity boundaries, rather than relying on pre-specified ranges based on a limited number of reference lots. Machine learning models trained on longitudinal reference product characterization data across commercial manufacturing sites and time periods can define the ‘natural product space’ with greater precision and with explicit statistical confidence intervals. Biosimilar developers that characterize their proposed product within this empirically defined space can present a stronger analytical similarity argument than those relying on descriptive comparability tables.<\/p>\n\n\n\n

Automated bioreactor control systems using multivariate statistical process control are reducing within-batch and batch-to-batch variability in biosimilar manufacturing, narrowing the analytical footprint of the biosimilar and making it easier to demonstrate that the product consistently falls within the reference product’s comparability acceptance criteria. Some developers are now running real-time release testing using process analytical technology (PAT) sensors that measure critical quality attributes inline during manufacturing, generating continuous product quality data rather than end-of-batch sample results.<\/p>\n\n\n\n

Key Takeaways: Section 8<\/strong><\/h3>\n\n\n\n

R&D leads planning biosimilar programs in 2025 and beyond should design their analytical comparability exercises around fingerprint-like assessment methodology from program inception, not as a late-stage addition to satisfy an analytical comparability guidance checklist. The clinical trial question should be addressed at program start with a written regulatory strategy that documents why comparative efficacy data is or is not required for the specific reference product and indication set, with explicit reference to the FDA’s 2024 guidance and any product-class-specific scientific advice obtained through Pre-BLA meetings.<\/p>\n\n\n\n

\n\n\n\n

9. Manufacturing Readiness: The Pre-Approval Inspection Gauntlet<\/strong><\/h2>\n\n\n\n

GMP Certification Timeline<\/strong><\/h3>\n\n\n\n

The most common mistake biosimilar developers make in manufacturing strategy is treating GMP certification as a late-stage regulatory requirement rather than an early-stage operational investment. The FDA’s pre-approval inspection process is not a one-time gate; it is a continuous quality system assessment conducted against the documented process in the BLA. If the manufacturing site has not received a satisfactory GMP inspection within a reasonable period before the BLA submission (typically 2-3 years), the FDA will schedule a pre-approval inspection as part of the review cycle. If that inspection finds deficiencies, the review clock effectively pauses until re-inspection clearance is obtained.<\/p>\n\n\n\n

For biologics manufacturers using contract manufacturing organizations (CMOs), GMP risk is distributed across multiple sites, each requiring independent FDA oversight. A drug substance site in Korea, a drug product fill-and-finish site in Germany, and a packaging and labeling site in the US each represent a separate inspection event. If any one site fails, the entire application is blocked. Biosimilar developers using multi-site CMO arrangements need inspection readiness programs at each site, mock inspections conducted 18-24 months before the planned BLA submission, and corrective action management systems that can demonstrate CAPA effectiveness to FDA inspectors on a re-inspection timeline.<\/p>\n\n\n\n

Quality by Design in Biosimilar Manufacturing<\/strong><\/h3>\n\n\n\n

Quality by Design (QbD) principles, formalized through ICH Q8, Q9, and Q10 guidelines, establish a structured framework for defining the design space of a biologic manufacturing process, identifying critical quality attributes (CQAs), linking them to critical process parameters (CPPs), and demonstrating that the process control strategy consistently maintains the product within the design space. A biosimilar program that implements QbD from process development through commercial manufacturing scale-up builds the documentation infrastructure that FDA and EMA inspectors use to assess whether the quality system is scientifically sound.<\/p>\n\n\n\n

The design space concept is particularly relevant to biosimilars because it provides a mechanism for demonstrating that manufacturing process changes, which are inevitable during the scale-up from development batches to commercial scale, do not alter the product’s comparability to the reference product. If the biosimilar’s design space is defined rigorously enough, process changes within the design space do not require a prior approval supplement, reducing the regulatory burden associated with post-approval manufacturing optimization.<\/p>\n\n\n\n

\n\n\n\n

10. Global Filing Strategy: FDA, EMA, PMDA, and Emerging Markets<\/strong><\/h2>\n\n\n\n

A sponsor’s global filing sequence is a strategic decision with direct revenue implications. Filing with the EMA first makes sense for programs with European clinical development infrastructure and for markets where European approval speeds reimbursement negotiations in individual member states. Filing with the FDA first makes sense when the US market accounts for the majority of the addressable commercial opportunity, which it does for most high-value biologics given US drug pricing dynamics.<\/p>\n\n\n\n

The PMDA’s requirement for Japan-specific PK bridging data means Japan cannot be a first-in-class market for most programs unless the sponsor specifically designs its Phase 1 PK study to recruit Japanese subjects using locally sourced reference product. Brazil’s ANVISA requires a full comparability dossier but also mandates local clinical study data for biologics above a specific molecular weight threshold, effectively requiring a dedicated Brazil-specific clinical investment for large proteins and monoclonal antibodies. China’s NMPA requires that clinical studies be conducted in Chinese patients using locally sourced reference product, and recent NMPA guidance has tightened the manufacturing site inspection requirements for foreign-manufactured biologics seeking Chinese registration.<\/p>\n\n\n\n

The practical consequence is that a biosimilar program designed for the US and EU markets without Japan, Brazil, or China-specific clinical data is either permanently excluded from those markets or requires a second development investment to generate the jurisdiction-specific data. Global filing strategies should be designed at program inception, with explicit resource allocation decisions about which markets are in scope for the first approval cycle and which, if any, will require dedicated supplemental development investments.<\/p>\n\n\n\n

\n\n\n\n

11. Investment Strategy for Analysts: Reading Biosimilar Pipeline Risk<\/strong><\/h2>\n\n\n\n

Due Diligence Checklist for Biosimilar Programs<\/strong><\/h3>\n\n\n\n

Investors evaluating biosimilar pipeline assets should structure their technical due diligence around the following primary risk categories, weighted by the frequency of each deficiency type in the historical CRL dataset.<\/p>\n\n\n\n

Manufacturing site GMP status is the highest-priority variable. An investor should confirm the date of the last satisfactory FDA or EMA inspection at each manufacturing site in the program’s supply chain. If no inspection has occurred within 3 years, or if the most recent inspection generated a Form 483 with multiple observations, the program carries elevated CRL risk that should be reflected in the probability-of-technical-success (POTS) estimate.<\/p>\n\n\n\n

Analytical comparability package completeness is the second priority. The sponsor should be able to articulate, in non-promotional terms, the full set of orthogonal characterization methods used to establish structural and functional similarity, the reference product lot selection strategy, and the statistical framework for defining comparability acceptance criteria. Programs that cannot clearly explain these elements to a technical investor team are unlikely to have assembled a package that will satisfy FDA or EMA reviewers.<\/p>\n\n\n\n

Patent estate clearance is the third priority. For any program targeting a reference biologic with more than $1 billion in annual revenues, the investor should request an FTO analysis summary that identifies the 10-15 most relevant patents by expiration date, validity risk, and likelihood of assertion by the innovator. Programs that have not conducted a structured FTO analysis are carrying undisclosed IP risk.<\/p>\n\n\n\n

Reference product PK bridging strategy is the fourth priority. If the program has generated clinical PK data using the EU-approved reference product but is filing a US 351(k), the bridging study design and data should be reviewed before submission, because inadequate bridging is a documented CRL category.<\/p>\n\n\n\n

Valuing a Biosimilar Program: Revenue Model Inputs<\/strong><\/h3>\n\n\n\n

A biosimilar program’s NPV is a function of several variables that are specific to the reference product and the competitive filing landscape. The relevant reference product revenue to model is not the current gross revenue but the projected revenue at the time of biosimilar market entry, adjusted for any pre-entry revenue decline driven by payer negotiation dynamics or off-label alternative prescribing. The biosimilar market share capture rate in year 1 depends heavily on whether the program has achieved interchangeability designation, whether the biosimilar developer has PBM contracts in place at launch, and how many competing biosimilar programs will enter the market in the same launch window.<\/p>\n\n\n\n

Pricing for biosimilars in competitive markets has compressed substantially from the 20-30% discounts to WAC seen in early market entries. In markets with five or more biosimilar competitors, like adalimumab post-2023, WAC discounts of 80% or more below the reference product are necessary to win formulary exclusivity. Investors should model revenue at realistic market prices, not at early-market reference product discount rates, to avoid overstating the commercial opportunity.<\/p>\n\n\n\n

The cost structure of a biosimilar program post-approval includes ongoing manufacturing costs, pharmacovigilance commitments, any REMS obligations, and sales force investments for programs that are not interchangeable and therefore require prescriber-level selling. These post-approval cost items are often underweighted in program NPV models, particularly for first-time biosimilar developers whose cost benchmarks come from branded drug commercial models rather than biosimilar-specific operational experience.<\/p>\n\n\n\n

Key Takeaways: Section 11<\/strong><\/h3>\n\n\n\n

Manufacturing site GMP status, analytical comparability package quality, FTO analysis completeness, and reference product PK bridging strategy are the four variables that drive the most CRL risk in the historical dataset. An investor who cannot obtain clear, documented answers to questions in each of these categories before committing capital to a biosimilar program is accepting undisclosed regulatory risk. Revenue models should use realistic WAC discount assumptions calibrated to the number of competing programs in the target market, not early-market pricing analogies.<\/p>\n\n\n\n

\n\n\n\n

12. Summary Key Takeaways<\/strong><\/h2>\n\n\n\n

Manufacturing deficiencies caused by GMP non-compliance or failed pre-approval inspections appear in more than half of all biosimilar CRLs and represent the single highest-priority risk mitigation target for any program. Inspection readiness programs should begin 24 months before the planned BLA submission, and mock inspections should cover all manufacturing sites in the supply chain, not just the primary drug substance site.<\/p>\n\n\n\n

Analytical comparability is the scientific core of every 351(k) application, and analytical method validation failures can compromise 18 months or more of development data at costs running into several million dollars. An independent CMC review conducted 18 months before submission is a cost-effective insurance policy against late-stage analytical deficiencies.<\/p>\n\n\n\n

The FDA’s June 2024 draft guidance eliminating switching study requirements for interchangeability designation, and the agency’s parallel move away from requiring comparative clinical efficacy trials for all programs, have cut biosimilar development cost estimates by 25-50% for programs that can rely on analytical and Phase 1 PK data as the primary regulatory package.<\/p>\n\n\n\n

Patent thicket analysis must be completed before IND filing for any reference product with annual revenues above $1 billion. The cost of conducting a full FTO analysis is orders of magnitude lower than the cost of a launch-at-risk situation or a post-approval injunction.<\/p>\n\n\n\n

The FDA’s July 2025 publication of 202 previously confidential CRLs creates a structured public learning dataset that biosimilar developers should use to benchmark their own regulatory preparation against the documented failure modes of prior programs. This dataset is the most useful single regulatory intelligence resource published in the history of the BPCIA era.<\/p>\n\n\n\n

Global filing strategy must be designed at program inception, with explicit resource allocations for jurisdiction-specific bridging studies, local clinical data requirements, and manufacturing site inspection events across all markets in scope for the first approval cycle.<\/p>\n\n\n\n

\n\n\n\n

13. Frequently Asked Questions<\/strong><\/h2>\n\n\n\n

What is the most common reason a biosimilar 351(k) application fails?<\/strong> Manufacturing deficiencies, typically identified during pre-approval inspections, appear in more than half of all CRLs affecting biologics and biosimilars. CMC deficiencies including process validation gaps, stability data shortfalls, and analytical method validation failures are the most common sub-categories within the manufacturing deficiency umbrella.<\/p>\n\n\n\n

How does a CRL differ from a Refuse to File (RTF) decision?<\/strong> An RTF is issued before the FDA accepts an application for substantive review, when the submission is so incomplete that a review cannot begin. A CRL is issued after a full review cycle, when the application has been substantively evaluated but cannot be approved in its current form. RTFs are typically addressable through a complete resubmission with the missing components. CRLs require resolution of specific deficiencies, which may include new clinical or manufacturing data, before the application can be resubmitted.<\/p>\n\n\n\n

Can a biosimilar receive approval in the EU and still face a CRL in the US?<\/strong> Yes. EMA approval and FDA approval require separate data packages, and the reference products in the two jurisdictions may differ in ways that require US-specific bridging studies. Alvotech’s AVT05 received a positive CHMP opinion and Japanese PMDA approval in September 2025 while simultaneously receiving an FDA CRL in November 2025 for manufacturing deficiencies at the Reykjavik facility.<\/p>\n\n\n\n

How has the FDA’s 2024 guidance on interchangeability changed biosimilar development strategy?<\/strong> The June 2024 FDA draft guidance eliminated the requirement for switching studies to obtain interchangeability designation, permitting the designation based on comparative analytical and clinical PK data already assembled for the base 351(k) approval. This has reduced biosimilar development costs by an estimated 25-50% for programs that no longer need to conduct full switching study clinical trials, and has compressed development timelines by 0.5-1.5 years.<\/p>\n\n\n\n

What is the ‘patent dance’ under the BPCIA?<\/strong> The BPCIA patent dance is a structured information-exchange and litigation-sequencing framework that governs patent disputes between biosimilar applicants and reference product sponsors. Within 20 days of FDA acceptance of a 351(k) application, the biosimilar applicant must provide the reference product sponsor with a copy of the application. The parties then exchange patent lists, identify which patents are subject to immediate litigation, and negotiate the litigation sequence. Participating in the patent dance can accelerate patent resolution but also involves sharing proprietary information. Several sponsors have elected to bypass portions of the process, accepting the consequence of limiting certain preliminary injunction remedies available to the reference product sponsor.<\/p>\n\n\n\n

What does the FDA’s July 2025 CRL database release mean for biosimilar developers?<\/strong> It converts previously private regulatory feedback into a public learning resource. Developers can now identify the specific deficiency categories that generated CRLs for programs similar to their own, benchmark their submission preparation against documented failure modes, and prioritize quality investments in the areas most likely to generate deficiency letters. The database covers only CRLs for products that were eventually approved, so it provides a dataset of recoverable failures. The FDA has indicated it may release additional CRLs for programs that did not advance to approval, which would expand the learning dataset further.<\/p>\n","protected":false},"excerpt":{"rendered":"

1. The Stakes: What a Failed 351(k) Filing Actually Costs A Complete Response Letter (CRL) is not an administrative inconvenience. […]<\/p>\n","protected":false},"author":1,"featured_media":38016,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_lmt_disableupdate":"","_lmt_disable":"","site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"default","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"categories":[10],"tags":[],"class_list":["post-18819","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-insights"],"modified_by":"DrugPatentWatch","_links":{"self":[{"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/posts\/18819","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/comments?post=18819"}],"version-history":[{"count":2,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/posts\/18819\/revisions"}],"predecessor-version":[{"id":38017,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/posts\/18819\/revisions\/38017"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/media\/38016"}],"wp:attachment":[{"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/media?parent=18819"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/categories?post=18819"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.drugpatentwatch.com\/blog\/wp-json\/wp\/v2\/tags?post=18819"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}