Happy New Year!

On January 1, 2022, the recently revised version of the PhRMA Code on Interactions with Health Care Professionals went into effect. We summarized the major revisions to the Code in a blog post when it was released in August 2021. Many of the updates relate to drug manufacturer practices with regard to speaker events, including meals, choice of venue, and attendance.

Although the PhRMA Code is a voluntary code of conduct, drug manufacturers should consider updating their marketing policies and practices to align with the new Code. The updated Code incorporates the latest guidance from the Office of Inspector General at the U.S. Department of Health and Human Services (OIG). This includes a November 2020 Special Fraud Alert wherein OIG explained its enforcement focus regarding speaker programs. (We blogged about the Alert here).

Another reason drug manufacturers may want to update their marketing policies are the numerous state requirements tied to the Code. Several states, including Connecticut, California, District of Columbia, Massachusetts, and Nevada, have adopted or incorporated the Code in their statutes or regulations. In some cases, a revised PhRMA Code adds obligations for manufacturers to update their practices. For example, California’s drug marketing law requires pharmaceutical companies to adopt an internal marketing compliance policy that aligns with the PhRMA Code. See Cali. Health & Safety Code § 119402(b). If the Code is revised, the California law gives companies six months to update internal policies to conform to the new version. Other jurisdictions like D.C. require sales representatives to comply with the Code “as it may be amended or republished from time to time.” See D.C. Municipal Regulation § 8305.11.

Given these state requirements, following the PhRMA Code is not only an approach to mitigate litigation risk – it is an explicit requirement for manufacturers that interact with health care practitioners in a state that has adopted or incorporated the Code.

As we approach our third year of COVID, one of the major questions from March 2020 eerily echoes today: where are all the COVID tests?  The situation now, of course, is very different and much more favorable than two years ago.  Unlike March 2020, numerous tests by multiple manufacturers have been reviewed by FDA and are being distributed.  There is also a wide variety of tests, including PCR assays, antigen tests, and antibody tests, and at least 16 over-the-counter assays (4 of which were authorized in the past month).  And yet, there is still a dramatic shortfall in the number of tests available.  There are innumerable accounts of desperate scavenger hunts for a COVID assay (link), and a photograph taken by one of us (Jeff) at the local library depicts an all too common scene.

There are multiple explanations for the current inadequate number of tests (link), including government policies on test procurement, manufacturers’ decisions to curtail production when demand fell this summer, the extraordinarily rapid spread of the omicron variant, and surges in demand brought on by the holidays.  But while the advent of omicron unquestionably precipitated the immediate critical shortage, limitations on the availability of COVID tests in the United States was a vulnerability long before the emergence of this variant (link).  As we chronicle in an article that appears in the current edition of the Food and Drug Law Journal, FDA’s policies for reviewing COVID assays, have been a contributing factor.

Over the past two years, FDA has played a central role as gatekeeper for COVID tests.  This role is one that has sporadically received attention (link) but which has not undergone a detailed analysis.  To help fill that void, our article takes an in-depth look at FDA’s policies since the emergence of the pandemic in the United States.  This article (link), which is made available with the permission of the Food and Drug Law Institute, examines multiple facets of FDA’s regulation of COVID assays.

The onset of COVID brought unprecedented challenges to FDA.  There is no question that the agency has devoted herculean efforts and resources to reviewing and authorizing COVID assays.  As of the latest count, FDA has granted more than 420 Emergency Use Authorizations (EUAs) for COVID In Vitro Diagnostics (IVDs) (25 of which were laboratory developed tests).  And that total understates the agency’s efforts to address the critical need for testing, which have included holding 75 Town Hall meetings and issuing eight EUA submission templates.

At the same time, as our article discusses, the agency has taken steps that have unnecessarily impeded the introduction of assays.  For example, the agency has abruptly changed its policies, leaving numerous tests in regulatory limbo and unavailable for use.  The agency’s prioritization scheme has been opaque and confusing, and its implementation has kept COVID tests off the market and hampered decision-making by IVD companies.  The strict data requirements for tests intended to be used with asymptomatic patients led to a lack of tests available for that use.  Ultimately, FDA took the unprecedented step of affirmatively encouraging the off-label use of tests authorized for symptomatic patients to be used off-label for asymptomatic patients.  As we learned first-hand through our counseling of scores of companies, and as has been reported in the media (link), the unpredictability of FDA policies had a chilling effect on product development and submissions.

FDA’s approach to laboratory developed tests (LDTs) has been perhaps one of the most remarkable examples of the impact of agency regulation on test development and availability).  As we’ve blogged about on many occasions (see, e.g., link, link, link, link), FDA’s LDT policy has been fraught for decades, and the consequences of unresolved jurisdictional issues were made manifest during the pandemic.

As we describe in the article, at the outset of the pandemic FDA blocked labs from offering COVID LDTs without an EUA, choosing instead to rely solely on the CDC’s assay.  When that assay turned out to be flawed (link), there were no alternate laboratory tests available because labs had been discouraged from developing them.  Only after weeks of the country flying blind in the face of the viral onslaught did FDA permit labs to offer testing while they prepared EUA requests for submission.  In August 2020, the Department of Health and Human Services intervened, directing FDA to stop requiring EUAs for LDTs in the absence of rulemaking.  This policy was reversed in November 2021, and laboratories were given 60 days to submit EUA requests (the deadline is January 14th) (link).  Given the inadequate testing capacity and the surging demand in the United States, it is inexplicable that FDA would take any steps that could increase burdens on test developers and potentially reduce test availability without an extraordinary reason – which FDA did not offer.

Our article closes with a strong recommendation that FDA take a close look at what has gone well with its review of COVID IVDs – and a lot has – and where improvements are needed.  Mistakes cannot be glossed over.  Under FDA’s quality system requirements, device manufacturers must review all available sources of information to identify quality problems, institute meaningful corrective and preventative actions when appropriate, and then verify effectiveness.  We should expect no less from FDA itself when it comes to self-evaluation of the impact of its testing policies on pandemic preparedness.

In the past few decades, multiple new and deadly etiologic agents have arisen.  SARS-CoV-19 will not be the last one.  The next time, the United States – and FDA – need to be much better prepared to facilitate widespread testing.

* In recognition of the wide and eclectic musical tastes of our HPM Blog followers, this blog post title alludes to the music of both Pete Seeger and Taylor Swift.

The authors acknowledge and thank Charlie Snow for his assistance in preparing this blog post.

On December 23, 2021, CDRH released as a draft guidance, Assessing the Credibility of Computational Modeling and Simulation in Medical Device Submissions (Draft Guidance).  Computational modeling and simulation (CM&S) can sometimes be useful to demonstrate the safety and effectiveness of medical devices or incorporated into devices.  FDA indicates that they receive regulatory submissions with such computational modeling, but the submissions “often lack a clear rationale for why models can be considered credible for the context of use.” Draft Guidance at 4.

The Draft Guidance describes a nine-step framework for evaluating the credibility of CM&S information submitted in pre-market applications.  A computational model is “the numerical implementation of the mathematical model performed by means of a computer.” Draft Guidance at 8.  The National Institute of Biomedical Imaging and Bioengineering describes computational modeling as “the use of computers to simulate and study complex systems using mathematics, physics and computer science.” NIH, Computational Modeling (May 2020).  Weather forecasting is an example of computational modeling and simulation for which most of us are familiar.  Similar techniques can be used to model complex biological systems. The Draft Guidance applies to physics-based or mechanistic CM&S and not statistical or data-driven CM&S, such as those incorporating artificial intelligence or machine learning.

The Draft Guidance describes four types of CM&S that can potentially be used to support a regulatory submission, by either being used to provide evidence to support a device’s safety and effectiveness or by being incorporated within the device, itself:

• In Silico Device Testing, which are computational models that simulate medical device performance. Draft Guidance at 5.
• CM&S used within medical device software, which is use of computational modeling within medical device software to perform device functions. at 6.
• In Silico Clinical Trials, where “device performance is evaluated using a ‘virtual cohort’ of simulated patients with realistic anatomical and physiological variability representing the indicated patient population.”
• CM&S-based qualified tools, which are tools for developing or evaluating a medical device that can be submitted to CDRH under the Medical Device Development Tools (MDDT) Program.

In the Draft Guidance, credibility is defined as “trust in the predictive capability of a computational model.” Id. at 4.  The guidance assesses credibility using key concepts from FDA-recognized standard ASME V&V 40 Assessing Credibility of Computational Modeling through Verification and Validation: Application to Medical Devices.  However, where the ASME standard assumes the ability to perform traditional validation activities, the Draft Guidance provides a more general framework that additionally incorporates non-traditional validation evidence.

A nine-step framework is presented for assessing credibility for purposes of a regulatory submission of the four types of computational modeling described above.  The first steps are to (1) describe the question of interest, (2) describe context of use and (3) model risk.  Next, (4) credibility evidence, either previously generated or planned, is identified and categorized, followed by (5) defining credibility factors and setting prospective credibility goals.  Prospective adequacy assessment is then performed (6) to answer the question, “will the credibility evidence be sufficient to support using the model for the context of use given the risk assessment?” Id. at 10.  Credibility evidence is then generated (7) by executing the proposed studies and/or analyzing previously generated data.  A post-study adequacy assessment (8) is conducted to determine if credibility goals were met, followed by preparation of a credibility report (9).

Key concepts from the framework are then presented in detail within the Draft Guidance, including points of consideration for each type of CM&S, where applicable.  The question of interest should describe the question that is being addressed using the model and along with other sources of information.  When considering the context of use, it should be the specific role and scope of the computational model used to address the question of interest.  The Draft Guidance recommends that a model’s risk, defined as “the possibility that the computational model and the simulation results may lead to an incorrect decision that would lead to an adverse outcome” is assessed according to the ISO 14971 and ASME V&V 40 standards.  Id. at 9.

Credibility evidence is evidence that could support the credibility of a computational model.  Id. at 8. There are three types of credibility evidence (code verification, calculation verification, validation) and ten distinct categories within these three types of credibility evidence that are discussed in the Draft Guidance.

Code verification provides evidence demonstrating that a computational model implemented in software is an accurate implementation of the underlying mathematical model.  Calculation verification determines the solution accuracy of a calculation.  Both calculation verification and validation of the model may be provided through a number of types of evidence, including: general non-context-of-use evidence, evidence generated using bench-top conductions to support the current context of use, evidence generated using in vivo conditions to support the current context of use, evidence generated using bench-top conductions to support a different context of use, and evidence generated using in vivo conditions to support a different context of use.  Validation can additionally be provided by population-based evidence, emergent model behavior, model plausibility and model calibration evidence.  Model calibration evidence is an assessment of the fit of simulation results against the data used to develop the model; while it can support the validation of the model, it alone cannot be used to validate the model.

Although a pre-submission is optional, the Draft Guidance suggests it may be useful to receive Agency feedback on the model risk assessment and prospective adequacy assessment.  A Credibility Assessment Plan is suggested for inclusion in pre-submissions.  For regulatory submissions, the Draft Guidance recommends inclusion of a Credibility Assessment Report.  The structure for both a Credibility Assessment Plan and Credibility Assessment Report are provided in Appendix 2 of the Draft Guidance. Id. at 34-36.

For regulatory and legal professionals, the Draft Guidance provides information that will help ensure regulatory submissions provide appropriate documentation to support the credibility of computational modeling and simulation information provided to support the safety and effectiveness of medical devices.  As FDA indicates, it is especially important to consider these issues within the clinical context (conditions of use).  A model that is credible in one context may not be credible in another.  We are interested to hear from engineers as to whether this guidance will also prove helpful in developing and validating CM&S.

We reported in August that CMS proposed to rescind the Most Favored Nation (MFN) drug pricing interim final rule issued in the latter days of the Trump regime.  Today, CMS finalized that proposal, effectively putting an end to the concept of international reference pricing as a means to limit drug prices.  The bipartisan idea of international reference pricing generated considerable controversy during its short lifetime.  The November 2020 rule was promptly challenged in four lawsuits, one of which resulted in a nationwide preliminary injunction against its implementation.  Though regulatory implementation was stymied, international reference pricing was carried over by Democrats into the drug pricing provisions of H.R. 5376, the Build Back Better Act, which passed the House on November 19.  However, the approach has been rejected in the version of the Build Back Better Act that is slowly taking shape in the Senate.  (See section 129001 of the most recent Finance Committee text.)  Rather than using foreign prices as benchmarks, that legislation would use a specified percentage of the non-federal average manufacturer price (NFAMP) reported by manufacturers to the Department of Veterans Affairs, in order to set a ceiling on Medicare Part B and D drug payment for selected high cost brand drugs.  With today’s action by CMS, the idea of international reference pricing for drugs has reached its demise in both the Congress and the Administration.

Since California passed AB 824: Preserving Access to Affordable Drugs in September 2019, the Association for Accessible Medicines (“AAM”) has been trying to invalidate the law, which imposes a presumption of anticompetitive effect on any Paragraph IV patent settlement in which the generic sponsor receives “anything of value,” including an exclusive marketing license or promise not to launch an authorized generic, from the patent holder.  Intended to target “reverse payment” settlement agreements, in which a brand company pays a first-filer ANDA holder to delay launch, the California law shifts the burden of proof to the drug sponsors to demonstrate that any Paragraph IV settlement agreement is not an antitrust violation.  In other words, the California law assumes that every potential Paragraph IV patent settlement is anticompetitive, and the pharmaceutical manufacturers must show that the settlement is not anticompetitive to avoid upwards of $20 million in fines.  Unsurprisingly—and understandably given the evidentiary hurdle imposed, as well as the amount of proprietary information that must be disclosed for all Paragraph IV patent settlement agreements—industry was not happy.

Thus, as soon as AB 824 went into effect in January 2020, AAM sued California alleging that AB 824 violates the dormant Commerce Clause because it extends to entities and agreements that are not located in California.  Ultimately though, the Eastern District of California denied AAM’s request for a preliminary injunction “primarily due to the nature of Plaintiff’s pre-enforcement attack on AB 824,” determining that AAM “failed to establish a likelihood of success on the merits or raise serious questions going to the merits.”   The Court also determined that AAM failed to establish an irreparable harm that was both likely and imminent.  AAM appealed, but the Ninth Circuit ultimately determined that AAM failed to demonstrate that its members had an Article III injury in fact and therefore lacked associational standing to bring claims on its members’ behalf.  The case was dismissed without prejudice.

Litigation Take 2: AAM filed suit again in the Eastern District of California on August 25, 2020, once again seeking injunctive relief based on allegations that AB 824 is unconstitutional.  AAM argued that AB 824 violates the dormant Commerce Clause by regulating out-of-state conduct; is preempted by federal patent law in view of FTC v. Actavis, 510 U.S. 138 (2013) and the BPCIA; violates the constitutional prohibition on excessive fines under the Eighth Amendment; and violates due process by burden-shifting with no meaningful opportunity to rebut the presumption applied.  This time, in spite of California’s contention otherwise, AAM argued that it has standing because several members have suffered “concrete economic” harm from AB 824.  Because an AAM member company submitting a Declaration stating that it “decided to pull out of a settlement negotiation for a pay-for-delay settlement agreement and chose instead to continue litigating a patent-infringement lawsuit at significant cost due to concerns about enforcement of AB 824,” the Court found the claim prudentially ripe and sufficient for purposes of standing.

With respect to the merits, the Court only addressed (here) the merits of the dormant Commerce Clause Claim because that argument alone was strong enough for the relief requested.  AAM argued that AB 824 directly regulates out-of-state commerce because it is not limited to settlements entered in California or between California entities; conversely, California argued that AB 824 does not regulate conduct occurring wholly outside of California.  The Court applied the “extraterritoriality theory” which precludes states from unduly burdening interstate commerce.  Under that theory, any statute that directly controls commerce outside the boundaries of the state exceed the limits of that state’s authority.  Recognizing that the Supreme Court rarely has held that statutes violate the extraterritoriality doctrine, the Court nevertheless explained that “AB 824 may reach the kind of settlement agreements . . . in which none of the parties, the agreement, or the pharmaceutical sales have any connection with California,” in violation of the doctrine.

California disagreed and told the Court that AB 824 applies only to agreements in California because manufacturers could omit California sales from any agreements subject to 824, but, said the Court, nothing in AB 824 limits the statute to only California settlements.  Because, as written, AB 824 could apply to settlements in which none of the parties, the agreement, or pharmaceutical sales have any connection with California, the statute violates the dormant Commerce Clause.  Further, AB 824’s civil penalties provision is violative, as it could levy substantially significant penalties on parties with no connection to California.  Thus, AAM “is likely to succeed in showing that AB 824 violates the dormant Commerce Clause.”

Because AAM members will be unable to recover monetary damages against the state even if AAM is successful, the Court determined that the monetary injury here constitutes irreparable harm absent a preliminary injunction.  And California could not demonstrate that the balance of equities tip in its favor due both the economic injury for pharmaceutical companies and the lost savings from slowed generic and biosimilar market entry.  As AAM explained, AB 824 will lead—and has already led—to delays in availability of generic medicines and driven manufacturers to withdraw Paragraph IV ANDAs.  This potential harm was not balanced by the need for California to have additional tools to address collusive agreements and its theory that the presumption ultimately will lower drug prices in California.

So AAM’s suit lives to see another day, and California is enjoined from enforcing AB 824.  California has pledged to continue the fight.  For now, California will have to continue to dream.  And we imagine we’ll see you back in 2022 for California Dreaming Part 5.

Hyman, Phelps & McNamara, P.C. is  pleased to announce that Adrienne Lenz will be speaking on Deciphering New and Proposed Regulatory Guidances For Medical Devices and Diagnostics at the Q1 Productions 4th Annual Life Science Regulatory Intelligence Virtual Event being held January 24-25, 2022.  The foundation of the regulatory intelligence and strategy is to remain abreast of current regulatory and legislative guidelines.  The magnitude of guidance documents affecting a product throughout its lifecycle makes this a continual challenge for regulatory teams. This session will highlight current guidance documents that are of key importance to device and diagnostic manufacturers and regulatory intelligence executives.  This virtual event brings together global intelligence, strategy, policy and legal experts to share best practices on challenges faced by regulatory teams.

FDA Law Blog readers are offered a discount of 15% off the registration price.  The case-sensitive discount code is Q1HPM15.  You can access conference information and register for the event here.

On December 7, 2021, the Natural Products Association (NPA) filed a complaint seeking an injunction based on FDA’s actions excluding N-acetyl-L-cysteine (NAC) from the definition of dietary supplement.  At issue is FDA’s interpretation of the “exclusionary clause,” section 201(ff)(3)(B)(ii) of the FDC Act, which among other things excludes from the dietary supplement definition articles that are approved as new drugs, or authorized by FDA for investigation for therapeutic uses for which “substantial clinical investigations have been instituted and for which the existence of such investigations has been made public” and which were not “marketed as a dietary supplement or as a food” before such approval or authorization.  Specifically, the issue is whether this provision applies to bar dietary supplements, such as NAC, that were on the market before the enactment of the Dietary Supplement Health Education Act (DSHEA), in 1994, which added this exclusionary clause into the FDC Act.

NAC is widely available as a dietary supplement and has been marketed as a dietary supplement since before the enactment of the DSHEA.  It also has been approved and is used to treat acetaminophen poisoning and some other conditions.   According to FDA, NAC is excluded from the dietary supplement definition because NAC was approved as a drug in 1963 and, as far as the Agency has been able to determine, there is no evidence that NAC was marketed as a food or supplement before the drug approval.

In July 2020, FDA sent warning letters (WLs) to seven companies about products containing NAC.  The main reason for the WLs seemed to have been the companies’ impermissible use of drug claims that NAC was effective against hangover.  However, FDA also asserted that NAC is excluded from the dietary supplement definition because NAC has  been approved as a drug in 1963 and FDA was not aware of evidence of the marketing of NAC before that date.  FDA did not consider that NAC was legally marketed as a dietary supplement before 1994.

Although FDA had previously questioned or suggested that NAC was excluded, FDA had not taken any enforcement action against NAC supplements during the more than 25 years that they have been marketed as a dietary supplement.

In December 2020,  the Council for Responsible Nutrition (CRN) sent a letter asserting, among other things, that NAC had been on the market before DSHEA was enacted and, therefore, was a “grandfathered” ingredient.  CRN argued that the exclusionary clause in DSHEA was not intended to apply to such grandfathered ingredients but instead was intended to bar new dietary ingredients which would be first introduced into the US market after the enactment of DSHEA.  Absent a specific provision in DSHEA, the exclusionary clause would not apply retroactively; the passing of DSHEA did not affect the status of ingredients on the marketed prior to DSHEA.  FDA did not respond to the letter.

Six months later, in an effort to get FDA to address the issue, CRN submitted a Citizen Petition (CP) again asking that FDA reconsider its position that NAC is excluded.   Several months later, NPA filed its own CP  making the same arguments.

In November 2021, FDA issued tentative responses to the two CPs, here and here.  In these responses and in the Constituent Update , FDA requests information on the earliest date that NAC was marketed as a dietary supplement or as a food, the safe use of NAC in products marketed as a dietary supplement, and any safety concerns.  Much to the frustration of the industry, the tentative response did not address the legal issue crucial to the future of NAC, i.e., that NAC is a legal (grandfathered) dietary ingredient because DSHEA does not apply retroactively to exclude dietary supplements that were on the market before the enactment of the law.  FDA merely indicated that it was still considering that issue.

As readers of this blog may know, FDA is not required to provide a substantive response to a CP within a specific time and it may take years before FDA issues a response (In fact, HPM submitted a CP in 2000 to which FDA has yet to respond).  Since FDA’s statements about NAC as dietary supplement had resulted in several retailers barring the sales of NAC supplement, getting to FDA to address this issue was an urgent matter.  Therefore, NPA filed a lawsuit.  NPA’s complaint includes essentially the same arguments and facts as the Citizen Petitions.  In addition, it points to statements by an FDA Office of Criminal Investigations agent in an application for search warrants that NAC is excluded from the definition of dietary supplement.  NPA points to these and subsequent statements by FDA in the prosecution as final agency action re NAC which can be challenged in court.

NPA asks the court to declare that the exclusionary clause does not retroactively apply to the dietary supplement NAC and seeks a preliminary and permanent injunction prohibiting FDA from taking any regulatory action against manufacturers, sellers or distributors of NAC based on the claim that the drug exclusion prohibition is retroactive.

There are interesting procedural and substantive issues in this case, both with respect to NAC specifically and with respect to FDA regulation generally.  First, what constitutes, reviewable agency action?  Here, FDA has not issued a rule or other formal administrative action, but it has—in what are purportedly non-final actions—pressed its interpretation.  Seemingly that should give a plaintiff standing.  Second, was DSHEA intended by Congress to apply retroactively?  As a matter of statutory construction, absent explicit language, statutes are generally intended to apply prospectively.  Third, even if Congress intended DSHEA to apply retroactively, does the Constitution permit it?  The Ex post facto clause of the Constitution generally makes it unconstitutional to have a law retroactively impose penalties. We will be monitoring developments in this litigation.

On December 6, 2021, FDA published a proposed rule to reorganize and amend the highly criticized agricultural water provisions of the Produce Safety Rule (PSR). This marks the Agency’s latest effort to balance public health protections with covered farms’ persistent concerns that the current provisions are overly complex, difficult to implement, and ill-suited to address the diversity of water uses and sources in the produce industry.

The proposed changes are extensive, but the three main takeaways to note are that the proposed rule:

• will replace microbial quality and testing requirements with systems-based agricultural water assessments;
• will enhance risk-based mitigation measures with expedited mitigation for known or reasonably foreseeable hazards from animal activity, biological soil amendments of animal origin, or improperly treated human waste associated with adjacent or nearby lands; and
• emphasizes flexibility for covered farms to choose how to comply with the agricultural water requirements.

These changes are described in greater detail below. We note that the proposal concerns pre-harvest activities and the existing standards for covered farms’ harvest and post-harvest activities are not affected by the proposed rule. Also, the proposal concerns all produce except for sprouts, which are subject to separate requirements.

Background

The PSR, finalized in 2015 under the Food Safety Modernization Act (FSMA) and codified at 21 C.F.R. Part 112, established (for the first time ever) mandatory federal science-based minimum standards for the safe production and harvesting of produce for human consumption. FDA has long recognized (in its own investigations and based on decades of scientific research) that the quality of agricultural water is an important factor in produce safety. Subpart E of the PSR was specifically developed to address agricultural water quality control.  Among other requirements, it establishes testing obligations and microbial quality criteria for agricultural water based on the water’s intended use, e.g., irrigation water, water used in preparing crop sprays, etc.). In addition, Subpart E includes measures that a covered farm must take if its agricultural water fails to meet these requirements.

FDA quickly found itself in troubled waters with these new requirements. The regulated industry criticized the complexity and the one-size-fits-all approach of Subpart E.  In 2017, in an effort to lighten the burden for affected parties, FDA proposed to extend compliance dates until at least Jan. 26, 2022.  Despite opposition, FDA finalized the proposal to extend the compliance dates in March 2019.

Now, six years after finalizing the PSR but before the compliance date, FDA published a proposal to address stakeholders’ concerns about the PSR’s agricultural water requirements.

Testing the waters with a comprehensive assessment framework

FDA proposes to replace Subpart E’s prescriptive and highly technical testing requirements and microbial water quality criteria with more flexible provisions. These provisions will instead require covered farms to perform comprehensive, systems-based, and management-reviewed agricultural water assessments to identify and address potential risks associated with pre-harvest agricultural water.

If finalized, covered farms will be required to conduct written pre-harvest agricultural water assessments annually. The assessments must include all possible sources which are reasonably likely to introduce known or reasonably foreseeable hazards regardless of whether the farm has control over those hazards, specifically:

• The location, nature, and current and previous use(s) of the water source;
• The type of water distribution system;
• The degree to which the system is protected from possible sources of contamination, such as adjacent and nearby land uses, and particularly animal impacts;
• The type of application method (e.g., overhead sprinkler or seepage irrigation);
• The time interval between the last direct application of agricultural water and harvest of the covered produce;
• Crop characteristics that make covered produce vulnerable to contamination (e.g., susceptibility of the produce to surface adhesion or internalization of contaminants);
• Environmental conditions (e.g., rainfall patterns that may impact the agricultural water system or damage produce); and
• Other relevant factors that could inform the assessment (e.g., optional testing).

Covered farms will also be required to conduct pre-harvest agricultural water assessments whenever a significant change occurs that may increase the likelihood that a known or reasonably foreseeable hazard will be introduced into or onto produce or food contact surfaces. In other words, any change in the factors considered in the agricultural water assessment will trigger a reassessment requirement if that change may impact hazard identification or a risk management determination. For example, a change in the manner of water application will require a farm to conduct another agricultural water assessment if that change introduces, or increases the chance of introducing, a known or foreseeable hazard into or onto produce or food contact surfaces.

This comprehensive assessment approach is adaptable to diverse and changing circumstances.  The proposal recognizes that many interconnected factors, such as variability in agricultural water sources, systems, and practices, nearby land uses, crop characteristics, and environmental conditions, pose possible sources of contamination which differ from farm to farm. In addition, the proposed rule is designed to accommodate future improvements in agricultural water quality science. For instance, although generic E. coli is currently the preferred indicator for monitoring water quality, future scientific developments may provide for the use of viral indicators, such as coliphages, to inform water quality assessments. As FDA explains, “as new science [becomes] available in the realm of water quality monitoring, farms should have the flexibility to take those findings into account when establishing or updating their sampling programs.”

The proposed rule further provides covered farms the flexibility to choose whether to test their pre-harvest agricultural water for generic E. coli and, as science evolves, other appropriate organisms or analytes. This option, which would be in addition to the required agricultural water assessments, includes detailed requirements for sample collection, testing frequency, and validation to ensure that the testing is adequate to supplement a farm’s agricultural water assessment. Optional testing helps alleviate the burden on produce farms that lack nearby laboratories capable of testing water samples, or that rely on water sources that are too numerous or inconsistent to provide representative water sampling.

Finally, proposed exemption provisions add flexibility to the regulations. FDA proposes to exempt covered farms from conducting a pre-harvest agricultural water assessment if they can demonstrate that their pre-harvest agricultural water for covered produce:

• meets the PSR requirements for harvest and post-harvest agricultural water (such as the microbial quality criterion and testing requirements for untreated ground water);
• is received from a public water system that meets Safe Drinking Water Act or state regulations (provided that the farm has public water system results or certificates of compliance demonstrating that the water meets relevant requirements); or
• is treated in accordance with the applicable PSR standards (i.e., the method, delivery, and monitoring of the treatment are done in a manner that is effective to make the water safe and of adequate sanitary quality for its intended use).

Corrective and mitigation measures to keep farmers’ heads above water

FDA proposes that covered farms use the assessments to determine whether corrective or mitigation measures will be necessary to reduce the risk of contamination of their pre-harvest agricultural water, or whether routine inspections and maintenance will suffice to ensure the water’s safety and sanitary quality. The assessments also allow covered farms to evaluate trends impacting their agricultural water systems and the effectiveness of any mitigation measures taken.

FDA is proposing to require that corrective or mitigation measures be taken based on certain findings, displayed in the chart below.

Even though the proposed rule prescribes measures that covered farms must take based on findings from their agricultural water assessments, it provides covered farms with choices of mitigation measures.  For example, they may adopt microbial die-off or removal post-harvest activities (e.g., commercial washing) as a mitigation measure.

By providing covered farms with options rather than being prescriptive, the proposed rule helps address the root cause of covered farms’ complaints that Subpart E’s requirements are too complex and rigid to implement effectively.

It’s Time to Sink or Swim: Proposed Effective and Compliance Dates

As mentioned above, the earliest compliance date for Subpart E is Jan. 26, 2022 (the compliance date for small farms is Jan. 26, 2023, and the compliance date for very small farms is Jan. 26, 2024). Since the current requirements are likely to change, FDA proposes to exercise enforcement discretion for the agricultural water requirements for covered produce while it works to extend the compliance dates until after the proposed rule has been finalized.

FDA has not identified a target date for publication of the final rule. Comments to the proposed rule may be submitted until Apr. 5, 2022 here.

It seems like every other discussion I have these days someone asks, “what is going on with the pre-submission program?”  “I heard CDRH isn’t reviewing pre-submission,” or “I heard FDA is only providing written feedback and not holding meetings.”  So, we thought it was time for a post dedicated to every device manufacturer’s favorite early interaction process – the pre-submission program.

This spring, CDRH provided an update on how the pandemic had affected the Center’s workload (announcement here).  The announcement indicated that OHT7 (formerly OIR) has been and would continue to be declining most pre-submissions.  The only pre-submission that this Office have been and would be continuing to review are those “related to COVID-19, companion diagnostics, a breakthrough designation request, or have a significant public health impact.”  In our experience, OHT7 has been reviewing pre-submissions for these product types.  Companies with non-COVID diagnostics should carefully consider whether their tests fit within one of these categories.

For pre-submissions in other Offices, the announcement stated that CDRH “anticipate[s] pre-submissions . . . be[ing] completed within 120 days, rather than the usual 70 days.”  In our experience, these estimates have held true.  Offices, other than OHT7, have continued to review pre-submissions and hold teleconferences, when requested.  These meetings are often taking longer than the pre-pandemic 70-day timeline, however.  In our experience, these delays, sometimes, come at the last minute just before a meeting has been scheduled.  Thus, sponsors will need to remain flexible even when a meeting has been scheduled.  We have also noticed, in some circumstances, Offices declining to answer all of a sponsor’s questions noting that, for example, a separate pre-submission was required for a question regarding study risk when the pre-submission is seeking input on the study design.  These efforts appear designed to limit the focus of pre-submissions.

Bottom line – the pre-submission program is still running just with some minor modifications.  Meetings are being held – virtually still.  In light of Omicron, it is unclear when in-person pre-submissions will return.  We confirmed that there have been no other changes to the Center’s policies.  Sponsors should be aware that CDRH is doing its best to remain flexible with its Divisions given their workload and many Divisions are currently understaffed due in part to departures and/or resources having been pulled to other Divisions.

Nothing says “happy holidays” quite like the issuance of Federal Trade Commission (FTC) orders – essentially, subpoenas – to large retailers, wholesalers and suppliers of food, cosmetics, personal care and OTC products, among other things, during the supply chain’s busiest time of the year. Antitrust has had an unusually crazy year, and things are getting crazier with the FTC section 6(b) study into the consumer goods supply chain approved by the FTC on November 29, 2021.

Back in July, the White House issued an Executive Order on Promoting Competition in the American Economy which set forth 72 initiatives for multiple federal agencies suggesting sweeping and decisive change in antitrust policy and priorities at the agency level. As one of those initiatives, the FTC is tackling concerns with the consumer goods supply chain. While the study will do nothing to alleviate the economy’s current bottlenecks, it could shape future regulatory actions intended to maintain or increase competition in key industries, consistent with the Executive Order.

Section 6(b) of the FTC Act allows the FTC to conduct “wide-ranging studies that do not have a specific law enforcement purpose.” The FTC is planning to study the effect of the supply chain disruptions of the past year on competition, and states it “will examine whether supply chain disruptions are leading to specific bottlenecks, shortages, anti-competitive practices, or contributing to rising consumer prices.”

The recent section 6(b) orders were sent to Walmart Inc., Amazon.com, Inc., Kroger Co., C&S Wholesale Grocers, Inc., Associated Wholesale Grocers, Inc., McLane Co, Inc., Procter & Gamble Co., Tyson Foods, Inc., and Kraft Heinz Co. It is possible the FTC will issue orders to other participants in the supply chain at a later date. The FTC seeks information about “the primary factors disrupting [these companies’] ability to obtain, transport and distribute their products; the impact these disruptions are having in terms of delayed and canceled orders, increased costs and prices; the products, suppliers and inputs most affected; the steps the companies are taking to alleviate disruptions; and how these companies allocate products among their stores when they are in short supply.” The orders also seek internal company documents such as “strategies related to supply chains; pricing; marketing and promotions; costs, profit margins and sales volumes; selection of suppliers and brands; and market shares.” The FTC is soliciting public comments on the impact of supply chain disruptions on competition in consumer goods and retail.

As mentioned above, section 6(b) studies do not have a specific law enforcement purpose, but they could lead to focused investigations and help reshape the FTC’s enforcement strategy. Conduct that may raise potential antitrust concerns such as exclusive agreements, allocation systems, or sudden price increases will certainly catch their eye. Additionally, a number of groups (see here, here, and here) have recently raised concerns to the FTC, House and Senate about industry practices that include price discrimination, trade promotion, category captain and online retail sales.

The information the FTC is collecting, the comments it is soliciting, and the hearings it will hold will eventually be distilled down to a written report, but that will take time – perhaps a year or longer. While the supply chain disruptions will hopefully be a distant memory by the time the report is issued, the report will no doubt provide insights into antitrust issues in the CPG supply chain and the direction of the FTC’s future enforcement efforts.

Hyman, Phelps & McNamara, P.C.’s Anne Walsh will be moderating “Updates in Litigation Risks: Product Liability, Private Litigation, and Consumer Class Actions,” at the upcoming Food and Drug Law Institute’s Enforcement, Litigation, and Compliance Conference on December 9-10.  As we move out of the COVID-19 pandemic, come hear from your peers about how they are staying compliance- and inspection-ready as FDA ramps up inspections, what companies need to do to plan for and manage enforcement risk, trends in criminal and civil litigation, and government priorities for the new year. Sign up with the discount code SAVE15 for 15% off registration.  Learn more here.

Earlier this fall, in accordance with its MDUFA IV commitments, CDRH launched an online platform that allows sponsors to track the status of its submissions.  The platform is called the Customer Collaboration Portal (CCP) and is “a secure, web-based tracker that displays the CDRH progress in reviewing traditional 510(k) submissions.”  See announcement here.  The CCP is currently only available for Traditional 510(k)s.

Now that sponsors have had the opportunity to use the system for a couple of months, we thought we would share our thoughts on it.  Unsurprisingly, we think it’s great.  It includes all the necessary information for each submission under review including the reviewer and all relevant dates in a concise format.  A sample is shown below.

This summary also appears to remain available even once a final decision is made on the 510(k).  To date, completed 510(k) submissions are still showing in the CCP.  This historical reference may also be useful to sponsors in the future.

While sponsors have always had all of these dates, they have had to keep track of them manually and then calculate when the next interaction will/should occur.  It is, certainly, convenient to have it all in one place.  Also, at the top of the page, there is a helpful tracking bar that shows how long until the next major date; for example, when the sponsor’s response to a hold letter is due or when FDA expects to render its decision on the 510(k).  Examples of the tracking bar are shown below.

Again, sponsors could always calculate these dates but having a location where they are easily accessible is expedient and ensures that there is a clear and consistent understanding between the sponsor and the Center.  We applaud the Center’s efforts to make this system available.  Sponsors who have not previously used the system will automatically be sent login credentials shortly after CDRH begins its review of the submission.

In terms of improvements – the system provides all the basic information for Traditional 510(k)s.  Given that this is only one type of submission that sponsors will file, it would be helpful to have all submission types included in the CCP going forward.  Also, in the future, it would also be nice if copies of the relevant correspondence to the sponsor were also linked to in the submission for ease of reference.  But, for now, in our view, the CCP is a welcomed tool for sponsors.

DEA is proposing to amend its regulations to expressly permit the transfer between pharmacies of an initial controlled substance prescription.  Currently, DEA regulations do not address transfer of controlled substance prescriptions — whether paper or electronic — between pharmacies for the initial filling of the prescription.  Historically, if a patient presented a paper prescription to a pharmacy and the pharmacy was unable to fill it, the pharmacist would simply return the prescription paper to the patient and the patient would carry the prescription to a second pharmacy.  There really was no need to “transfer” the paper prescription between pharmacies; paper prescriptions are “portable” by the nature of their paper format.

However, the growing use of electronic prescriptions (DEA notes that more than half of all states mandate electronic prescription opioids, all controls, or all prescriptions) requires DEA to rethink its position on initial prescription transfers.  This is especially important given that the SUPPORT Act of 2018 (section 2003) requires EPCS (with few exceptions) for those prescriptions covered under Medicare Part D beginning in January 2021.

The proposed rule also just makes sense.  More specifically, if a pharmacy receives an electronic prescription for a controlled substance that it cannot fill, the pharmacy simply cannot “return” the prescription (which now is likely an electronic data file) to the patient to take to another pharmacy.  DEA regulations currently do not include any provision for a pharmacy to transfer an EPCS to another pharmacy.  The regulations also do not describe how a pharmacy should handle an EPCS that it receives but cannot fill. At present, a pharmacy that receives an EPCS that it is unable to fill can only notify the patient that the prescription cannot be filled. In this scenario, the patient could then call the prescribing practitioner to request that a new EPCS be sent to a different pharmacy. DEA realizes that this scenario creates the potential for a duplication of prescriptions if, for example, the practitioner transmits a new EPCS to a different pharmacy and does not cancel or void the original EPCS that was sent to the first pharmacy. DEA also recognizes the inability to transfer prescriptions creates an additional burden for patients, who must get back in touch with the original prescribing doctor and request a new prescription.

DEA’s proposal states that, upon request, a registered retail pharmacy may transfer an EPCS in schedules II-V to another registered retail pharmacy for initial filling. This rule will also specify the procedures that retail pharmacies must follow, and the information they must document and maintain when transferring EPCS.

The following recordkeeping requirements will apply to the EPCS transfers:

• The transferring pharmacist must update the patient’s prescription record with the following information: name, address, and DEA registration of the pharmacy to which the prescription was transferred; the pharmacist receiving the transfer; the name of the transferring pharmacist; and the date of the transfer.
• The receiving pharmacist must record the transferring pharmacist’s name, address and DEA registration number, the name of the transferring pharmacist, the date of the transfer, and the name of the pharmacist receiving the transfer.

As with other DEA recordkeeping requirements, the electronic record must be maintained for two years.  DEA estimates that the annual cost savings from the transfer rule would be $22.0 million.  The anticipated savings are based on calculations related to the pharmacists, patients, and prescribers’ time communicating concerning the need to generate and send a new EPCS instead of having the ability to electronically transfer the prescription to another pharmacy.   Comments from industry are due January 18, 2022 (Docket No. DEA-637).

On November 15, HHS announced that it was withdrawing the prior administration’s policy that prevented FDA from requiring premarket review of laboratory developed tests (LDTs) without notice and comment rule making.  See the prior post on this policy here.  This statement comes at least six months after the policy was removed from the HHS website without any public notice (a copy of the prior policy is available here for reference).  Although we were surprised to learn that the prior policy had vanished from the HHS website at least six months ago, we were not surprised that the Biden Administration had reversed the prior policy.

The announcement is light on details, simply stating that the policy was withdrawn and that “HHS no longer has a policy on LDTs that is separate from FDA’s longstanding approach in this area.”  This suggests that the LDT status quo ante has been restored. As our readers know, our firm has long been critical of this status quo (see our prior posts here, here, here, here, and here, just to name a few).

The rationale for withdrawal appears to have been that the prior policy allowed LDTs with “poor performance” to enter the market without prior FDA review.  This rationale is weak, at best.  In the past, when pushed to provide examples, FDA has struggled to provide examples or evidence of poorly performing LDTs.  Nevertheless, FDA has said that in its review of 125 EUAs applications for COVID-19 LDTs, 82 contained design or validation issues that the agency believed needed to be resolved before an EUA could be authorized.[1]  Further, the prior HHS policy was supported by a legal analysis concluding that FDA could regulate LDTs only after notice and comment rulemaking. (legal memo available here).  It is unclear whether HHS revised this analysis prior to reversing the August 2020 policy.

The HHS announcement was accompanied by a corresponding statement from FDA as to how this change will affect its review of LDTs.  In relevant part, the FDA announcement states, “the FDA now generally expects newly offered COVID-19 tests, including LDTs, to have an EUA or traditional marketing authorization such as a granted De Novo or cleared 510(k), prior to clinical use.” (emphasis added).  The announcement goes on to say that the notification pathway available for certain tests intended for use in CLIA high-complexity laboratories had led to “some poorly performing testing being offered prior to FDA review.”  See our prior post on the notification pathway here.  Thus, the FDA states that it is “ending those notification policies going forward” and revised its COVID testing guidance, available here, and added new Q&As to its FAQs for SARS-CoV-2 Testing.

No doubt many questions will be raised regarding the rationale, legality, and impact of HHS and FDA’s actions, but two threshold questions for clinical laboratories are: what does this mean for COVID-19 LDTs they are currently offering or may seek to offer going forward?; and what impact will HHS’ removal of constraints on FDA’s oversight of LDTs mean for laboratories performing non-COVID LDTs? We discuss each of these key questions below.

What do these announcements mean for Labs performing COVID-19 LDTs?

If the lab has obtained an EUA for its COVID-19 LDT, which many did because of the PREP Act protections afforded by the authorization, there is nothing more for the lab to do because the only change introduced by the policy is the need for an EUA.  Certain modifications to the LDT could, however, require the submission of an amendment to the EUA.

If a lab is performing a COVID-19 LDT that is not subject to an authorized or pending EUA and that is not included in one of FDA’s notification lists (see our discussion of the notification list below), the Guidance states that the laboratory should either submit an EUA within 60 calendar days of November 15, 2021 (i.e., Friday, January 14, 2022) or cease marketing on or before this date.  FDA notes that it plans to review all LDT EUA submissions.  If FDA declines to issue an EUA after such review, a lab will need to cease offering the test within 15 calendar days of being notified by FDA.

FDA’s decision to again require EUAs for COVID-19 LDTs raises a serious question of capacity. OHT7 is already unable to perform its normal functions, such as reviewing pre-submissions for non-COVID diagnostic devices.  It is unclear why FDA would choose to devote resources to reviewing COVID-19 LDTs that are on the market instead of providing feedback to companies that have new types of diagnostic devices or reviewing IVD submissions in a timely manner.

It is also unclear why FDA would take steps that could reduce the availability of COVID-19 testing.  Even if the Agency believes that there is sufficient access nationwide, LDTs have played an important role in filling local needs.  The sudden loss of access to a laboratory is bound to be disruptive to institutions, physicians, patients, employers, and schools, among others, who have relied upon testing at that facility.

What does the withdrawal of the notification pathway mean for COVID-19 tests?

If a lab is performing a COVID-19 LDT that was added to the notification list and for which an EUA submission is pending, then the consequences depend on whether the EUA was submitted before or after February 1, 2021.  If the EUA was submitted after February 1, 2021, there is nothing for the lab to do as FDA does not intend to object to offering of such tests while the EUA is under review.  If the EUA was submitted prior to February 1, 2021, a lab must notify FDA on or before December 30, 2021, as described in the Guidance, letting FDA know that:

1. the developer wants FDA to continue reviewing its EUA request;
2. the EUA request is for the current version of the test; and
3. either the developer does not have additional data to add, or the developer submits updated information to FDA within that same timeframe including, if not previously provided, validation with clinical specimens using an appropriate comparator.

The Guidance indicates that if a lab needs to submit additional information under item 3, it should also explain how the test falls within the test priorities for the “current stage of the public health emergency and the tests for which FDA will be prioritizing review.”

As many on the notification list know, FDA review of certain EUAs has been slow, with many ultimately being “deprioritized.”  FDA’s revised guidance states that, for tests on the notification list, “FDA intends to notify test developers by email if FDA declines to issue or otherwise decides not to authorize a test for any reason,” (emphasis added) and the test manufacturer will be required to cease marketing within 15 calendar days of such declination.  The language “for any reason” raises significant questions, e.g., what if FDA declines to review an EUA because it is deprioritized?  Will these tests need to be taken off the market?  There is a difference, after all, between FDA determining that a test may not perform well and deciding that a test is low priority.  It is unclear how the deprioritization process affects the new call for EUAs. The lack of clarity here underscores, once again, the uncertainty surrounding various aspects of the LDT process.

What do these announcements mean for non-COVID LDTs?

FDA’s press release and guidance give no hint as to the FDA’s intentions with respect to the hundreds of thousands of LDTs being performed for conditions other than COVID-19.  As we noted above, HHS appears to have restored the status quo ante for LDTs – in other words, a general policy of enforcement discretion with exceptions (e.g., for companion diagnostics, direct-to-consumer tests).  But certain actions taken by HHS during the last administration arguably create “facts on the ground” that may have implications for FDA’s next steps.  For example, the prior HHS policy revoked all LDT guidance documents.  The latest notice makes no mention of the status of such guidance documents.  Furthermore, as noted above, the previous General Counsel prepared a memorandum for then FDA Commissioner Hahn that called into question the scope of FDA’s authority over LDTs and unequivocally concluded that FDA is legally required to engage in rulemaking before regulating LDTs. How, if at all, will the current administration respond to the legal arguments laid out in this memorandum?

HHS’ policy reversal and FDA’s guidance updates raise numerous practical questions that laboratories offering, or contemplating whether to offer, COVID-19 LDTs are now scrambling to address. These laboratories must decide whether to withdraw, submit, or update their LDTs, taking into account whether additional data will be needed to supplement their EUAs and whether their tests are likely to be authorized given FDA’s new test priorities.  The impact of FDA’s resumption of COVID-19 LDT regulation on laboratories and the public remains to be seen.

[1] COVID-19 Tests Highlight Need for Strengthened FDA Oversight and Diagnostics Legislation, PEW Charitable Trusts (May 19, 2021), https://www.pewtrusts.org/en/research-and-analysis/issue-briefs/2021/05/covid-19-tests-highlight-need-for-strengthened-fda-oversight-and-diagnostics-legislation [https://perma.cc/49Z9-3R2C].

Sixteen years after the publication of the Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices issued May 11, 2005, FDA issued a new draft guidance document on November 4, 2021 that describes the recommended documentation that a sponsor should include in their premarket submissions. To put the age of the existing guidance into perspective, it was published two years before the first iPhone was released. Over the past 16 years, there have been numerous advances in healthcare technology, particularly with respect to the use of software in and as a medical device. Please see our previous blog postings on various software regulatory developments (here, here, and here).

This new draft guidance document applies to “device software functions” that meet the definition of a device under section 201(h) of the Federal Food, Drug, and Cosmetic Act (FD&C 17 Act).

If you’re interested in software products that do not enjoy any of the 21st Century Cures exemptions and require premarket submissions, please continue to read this blog. Summaries of the impact of the 21st Century Cures Acts on software and the definition of a medical device are available here, here, here, and here.

The new draft guidance applies to both Software in a Medical Device (SiMD) which would be your traditional hardware with embedded software and Software as a Medical Device (SaMD). The scope of the new draft guidance is basically the same as the 2005 guidance which includes:

(1) firmware and other means for software-based control of medical devices, (2) stand-alone software applications, (3) software intended to be operated on general-purpose computing platforms, (4) dedicated hardware/software medical devices, and (5) accessories to medical devices when those accessories contain or are composed of software. It does not apply to software-related documentation that may be needed to evaluate post market software device issues (e.g., corrections and removals). Like the 2005 guidance, this new draft guidance applies to all types of premarket submissions, but now the Agency has specifically added De Novo Classification Requests and Biologics License Applications (BLA).

The most obvious update to the new draft guidance is the shift from three to two categories for determining what software documentation to include in a premarket submission. According to the 2005 software guidance document, FDA used Major, Moderate, or Minor Level of Concern to determine the recommended documentation for software. In the new draft guidance document, FDA introduced four risk-based factors to help determine the device’s documentation level—either Basic Documentation level or Enhanced Documentation level.

Basic Documentation is necessary for all device premarket submissions that include software functions. But if the device (i) is a constituent part of a combination product, (ii) is intended to test blood donations, or to determine donor and recipient compatibility, or a blood establishment computer software, (iii) is categorized as a Class III device, or (iv) presents a probable risk of death or serious injury upon failure, then such device software functions are categorized as Enhanced Documentation level. Note that, unlike the 2005 software guidance, the new guidance only references serious injury, which is an injury or illness that is life threatening, results in

permanent impairment of a body function or permanent damage to a body structure, or necessitates medical or surgical intervention to preclude permanent impairment of a body function or permanent damage to a body structure. Both the draft guidance and the 2005 guidance require the assessment of the risk to be evaluated prior to the implementation of risk control measures. Interesting to note, FDA specifically called out the use of the word “probable” versus purely hypothetical, when determining the risk of death or serious injury. However, “probable” risks also include the likelihood that the device could be compromised due to cybersecurity risks.

Below is an overview of the software documentation sponsors would need to provide for Basic and Enhanced levels:

The major difference between the two is that Basic Documentation Level does not need to submit the Software Design Specification. However, there are differences in the level of detail expected for some of the deliverables. For example, Software Development and Maintenance Practices allows a declaration of conformity to IEC 62304 to work for both categories.

However, if the sponsor does not provide a declaration of conformity, Basic level devices would only need to provide a summary of configuration management and maintenance activities while sponsors of devices in the Enhanced level would need to provide the summary as well as complete configuration management and maintenance plan documents.

For software testing as part of verification and validation, only a summary description of the testing activities at the unit, integration, and system levels is required for devices categorized as Basic level, but for sponsors of devices in the Enhanced level, they need to also provide unit and integration level test protocols and results. The Agency added details that clarify their expectations around test results, specifically that it would not only include the pass/fail determinations, but additionally include expected results and observed results.

Under the proposed framework, the amount and type of software documentation may increase or decrease for existing software devices when a sponsor files its next premarket submission.

“Minor” level of concern software devices, under the 2005 guidance, were defined as failures or latent design flaws that were unlikely to cause any injury to the patient or operator. If the draft guidance is finalized, the next time a sponsor submits for a formerly “Minor” level of concern software, it will need to provide documentation similar to what had been previously required of a “Moderate” level of concern device. “Moderate” level of concern was defined as failures, malfunctions or latent design flaws in the software that could directly or indirectly likely lead to minor injury to the patient or operator. (Refer to Table below).

Under the 2005 guidance, if a sponsor’s software was a “Moderate” level of concern device the required documentation would decrease under the draft guidance. The software device would now be categorized as Basic Documentation Level and the Software Design Specification would not need to be submitted.

Other significant changes include the FDA’s details around what is expected for System and Software Architecture and Risk Management. FDA has devoted an entire Appendix to providing sponsor’s examples of architecture charts. With respect to Risk Management, now in addition to submitting your Risk Analysis (previously referred to as a hazard analysis), sponsors also need to

submit their Risk Management Plan and Risk Management Report, which should address methods for collection of relevant production and post-production information. For those of you who thought Traceability Analysis went away, it did not. Traceability was simply included in the description of the Software Requirement Specification document.

FDA has also taken steps to update the guidance for Artificial Intelligence/Machine Learning software devices. This is a welcome change given that AI/ML software is becoming much more widely used in medical devices, as evidenced by the list of AI/ML-enabled medical devices recently published by FDA (here). For AI/ML software, specifically, the draft guidance indicates that the software description should include the populations or samples informed your model(s) and what steps were taken to identify and address potential biases and limitations of your model(s). Interestingly, the draft guidance does not reference other AI/ML-specific considerations, like, Predetermined Change Control Plan, which has been contemplated in FDA’s AI/ML action plan (see our prior post on the action plan here). We would encourage FDA to consider aligning the final guidance with the AI/ML action plan so the expectations for AI/ML software documentation is clear and comprehensive.

When final, this new document will replace the 2005 guidance document. Interested parties have until February 22, 2022 to comment on the new guidance at this link.