A Final LDT Rule in April!? Will FDA be Prepared?

With comments due on the proposed LDT rule last week, FDA wasted no time updating the Unified Agenda to indicate that a final rule may be published in April (here).  We take this date with a grain of salt given the frequency with which these dates are missed and given the volume of comments filed. (It is not entirely clear how many comments were actually logged in; on December 7, FDA showed 19,655 comments, and the next day 6,732.)  While some comments were boilerplate, others were comprehensive, detailed critiques of the Proposed Rule.  Under the Administrative Procedure Act, FDA is obligated to address major substantive issues when – and not so much if – it publishes a final rule.  Complying with that requirement between now and April will be a daunting task.  But, FDA has expressed urgency in moving forward with this rule so it is not out of the question that it could happen.

With a possible final rule right around the corner, it made us think – will FDA be ready to actually implement the rule?  FDA must believe it will be, but sound public policy is predicated on accurate and reliable data.  FDA recognizes that “data is at the heart” of the agency’s work as a “science-based Agency.”[1]  Yet FDA’s conclusions about the Agency’s ability to regulate the entire laboratory industry are based on fundamentally flawed assumptions about the number of entities and tests that will be subject to FDA regulation.

FDA estimates that approximately 12,000 CLIA-certified laboratories are currently certified to perform high-complexity testing in the United States.  This is based on FDA’s reported 2018 review of the CLIA database.^[2]  It is unclear how FDA derived this number or why it chose to rely on a review conducted five years ago when the database is—according to FDA—updated monthly. In fact, as of November 2023, the CMS’s Standards and Certification’s (S&C’s) Quality, Certification and Oversight Reports (QCOR) CLIA Laboratory Lookup Database shows 17,206 CLIA-certified laboratories.  This is 50% greater than the 2018 numbers relied upon by FDA in its estimates.[3]

FDA further assumes that these laboratories collectively perform roughly 80,000 LDTs.  But, FDA understates the number of laboratories by roughly one-third.  Using FDA’s per-lab number of tests, the number of LDTs would be well over 100,000.

Even using a lower, out-of-date number of laboratories, FDA anticipates receiving an astounding number of premarket review submissions for LDTs, specifically:

• 32,160 510(k) premarket notifications;
• 4,210 PMAs, PDPs, Panel-Track PMA Supplements; and
• 4,020 de novo[4]

FDA derives these estimates by applying the same metrics applicable to currently‑regulated IVDs: 50% undergo premarket review, and of those 40% are submitted through the 510(k) process, 5% are submitted for de novo classification, and 5% are submitted for premarket approval.[5]  This stratification, though, assumes that LDTs will follow the same pattern as IVDs currently regulated by FDA.  However, this prediction is unfounded, because it assumes that there is an existing classification regulation and/or product code for 90% of currently available LDTs.  Further, it assumes that 80% of LDTs requiring premarket review will be able to identify a “predicate” device and be reviewed under the 510(k) process.  Given that many LDTs introduced are for new indications for which IVDs are not currently available, this assumption is unwarranted. In other words, FDA’s extrapolation from existing IVD submission data ignores some key differences between LDTs and distributed IVDs.

FDA further assumes that half of LDTs will not need to undergo premarket review because they are exempt.  That seems very unlikely to be the case, because exemptions apply only to well-established tests.  In seeking to justify regulation of LDTs, FDA emphasizes that many of these tests are novel.[6]  It is inconsistent to say that LDTs must be regulated as medical devices because of the newness of the tests and then reduce the cost of the proposed regulation by declaring that half of all LDTs will be exempt from premarket review.  FDA’s assumption that 50% of the tests will be exempt is particularly baffling because the laboratories are ones licensed to perform high complexity tests.  By their very nature, these tests are less likely to be exempt from FDA review.

Thus, FDA understates the number of laboratories and almost certainly overstates the number of LDTs that will be exempt.Yet, even assuming the accuracy of these estimates, the predicted volume of submissions would be unprecedented.  FDA’s 5-year average across all of CDRH is 73 PMAs, PDPs, Panel-Track PMA Supplements; 3,877 510(k) premarket notifications; and 66 de novo submissions per year.[7]  This means that OHT7, the Division responsible for reviewing in vitro diagnostic tests, will need to be staffed to review substantially more submissions per year than the entire Center, specifically: 57 times the number of PMAs, PDPs, Panel-Track PMA Supplements; 8 times the number of 510(k) premarket notifications; and 60 times the number of de novo submissions.  This would require a massive increase in staffing and training.[8]  FDA would need to commence a hiring spree at the same time that laboratories would be looking to hire personnel with the same sorts of expertise to be able to navigate the FDA process.  The foreseeable outcome will be a shortage of qualified personnel and significant salary increases.  Regulatory affairs professionals with significant expertise in diagnostics are already in short supply.

Furthermore, given FDA’s proposed 5-Stage, 4-year timeline, these submissions will all arrive within a very narrow window, with PMAs being due 3.5 years after publication of the final rule and all 510(k)s and de novo submissions due a mere six months later (i.e., 4 years after publication of the final rule).  This sudden bolus of submissions will strain FDA’s capacities beyond the breaking point, as happened during the COVID-19 pandemic.

FDA also does not take into account the volume of pre-submissions that laboratories will inevitably need to begin filling almost immediately after issuance of the final rule to ensure that the data they will be presenting to FDA in their premarket submissions is what the Agency will expect.  Unless FDA provides timely responses to pre-submissions, laboratories (and current IVD manufacturers) will not be able to submit applications that address FDA’s expectations.  During COVID-19, however, FDA essentially stopped reviewing pre‑submissions severely limiting feedback to non-COVID test developers.  Under FDA’s own calculations, the number of LDT premarket submissions will far exceed the number of COVID Emergency Use Authorizations (EUAs).  Thus, it is unclear how FDA will have resources to field pre-submissions and the anticipated premarket submissions.

During the October 31, 2023 webinar discussing the proposed rule, stakeholders sought clarity from FDA officials on the Agency’s plan to address the resource gap.  FDA cited two solutions to the resource problem: (1) enhancing the existing third-party 510(k) review program; and (2) resolving resource issues during the next user fee renegotiation process (MDUFA VI).[9]  Neither of these proposed approaches can be expected to adequately address the anticipated volume of submissions.

First, FDA notes that the Agency is “currently working to enhance our Third-Party Review Program, which was reauthorized under MDUFA V.”[10]  While this goal is admirable, the Third-Party Review Program has been available for decades and has long been regarded as a flop.[11]  In fact, in Fiscal Years 2018 – 2022, fewer  than 100 510(k)s went through the Third-Party Review Program annually.[12]  In FY2020, 2021, and 2022, Third-Party Review 510(k)s accounted for 2.6%, 2.6%, and 2.5% of 510(k) decisions, respectively.[13]  Notably, a search of FDA’s 510(k) database shows that zero IVDs were cleared through the Third-Party Review Program between November 1, 2022 and November 1, 2023.[14]  Thus, while the Third-Party Review Program is not well‑utilized generally, it is not utilized at all by the in vitro diagnostic industry.

Furthermore, even if the Third-Party Review Program could gain traction within the IVD industry, it is, per statute, available only to select Class I and II devices.[15]  Even of those IVDs fitting within the statutory criteria, FDA must specify if an IVD type is eligible for third-party review.  Given the innovative nature of LDTs, it is unreasonable to anticipate that many LDTs would be eligible for the program.  Absent legislative change, therefore, the Third-Party Review Program would not be able to review the overwhelming majority of the PMAs, PDPs, Panel-Track PMA Supplements or de novo submissions.  Thus, it is wholly unrealistic for the Third-Party Review Program to provide any meaningful reduction in Agency workload.

Second, FDA assumes that there will be a sufficient number of accredited third-party review organizations available to conduct the reviews.  During the October 31, 2023 webinar, FDA cryptically noted that there are “certain CLIA‑accredited organizations that may be interested in serving in this third-party review role.”[16]  However, the experience of the New York State Department of Health (NYSDOH) should serve as a cautionary tale.  In 2017, FDA announced the addition of NYSDOH as an accredited third-party review organization in the context of the Agency’s clearance of an NGS-based tumor profiling test, explaining that the Agency’s goal in allowing third party review of such tests was “to reduce the burden on test developers and streamline the regulatory assessment of these types of innovative products.”[17]

This initial fanfare culminated in the unheralded disappearance of NYSDOH from FDA’s list of accredited third parties earlier this year.[18] Despite NYSDOH’s expertise in reviewing diagnostic tests, the program was used sparingly.  Notably, NYSDOH was reported as having fewer than five 510(k) submissions in each fiscal year of MDUFA IV (FY18 – FY22).[19]  It is unclear why FDA expects other CLIA-accredited organizations would succeed in this program where the most prominent and “only state that requires test approval for IVDs offered as LDTs” failed.[20]

In short, without both statutory changes and a specific plan to revamp the Third-Party Review Program so that it can have a meaningful effect on the workload related to LDT pre-market review, the reliance on this program to provide critical resources for this massive regulatory undertaking is, at best, wishful thinking.  Without dramatic changes to the Third-Party Review Program, FDA will need to shoulder on its own the unprecedented burden of receiving thousands of applications for novel, complex diagnostics in a very short period of time.

In reality, FDA’s staffing plan relies almost entirely on hiring a significant number of additional reviewers with user fees authorized by Congress under MDUFA VI.  However, while Congress can authorize the funding mechanism, it cannot ensure the availability of qualified personnel.  (Congress could also decide not to authorize the funding mechanism or take other steps that would hamper hiring.)  Even assuming a 3% reduction in the 510(k) workload from the Third-Party Review Program, under FDA’s projections OHT7 will need to be staffed to review nine times the current capacity of the entire Center.[21]  Calling this task impossible may be an understatement.  Indeed, FDA has publicly noted its challenges with hiring under MDUFA IV, explaining that “competition with well-known tech innovation locations, the creation of new scientific and technical professional fields, and fewer candidates with a hybrid of specialties have resulted in hiring delays for the MDUFA program.”[22]  The “competition” for these candidates will inevitably worsen due to the hiring needs of laboratories that must submit tens of thousands of applications. Regulating LDTs as medical devices will mean that neither FDA nor laboratories will be able to recruit, hire, and train enough qualified personnel to meet their respective obligations.

Given the acknowledged and ongoing recruitment and retention challenges, FDA’s reference to MDUFA VI amounts to nothing more than “kicking the can down the road,” rather than a meaningful plan to ensure adequate resources. Meanwhile, given the compressed enforcement discretion phaseout timeline proposed by FDA, upon issuance of a final rule (which could occur while MDUFA V is still in place) laboratories would need to immediately begin compliance efforts, including by filing pre-submissions.  FDA has proposed no plan for how resources, including those needed to review pre-submissions for LDTs or the numerous other questions laboratories will have, will be handled under MDUFA V.  Even if MDUFA VI were to allow the hiring of a significant number of new personnel, implementation of the Proposed Rule would begin before those employees arrive.

OHT-7’s well documented meltdown during the COVID-19 public health emergency should serve as an object lesson.  The massive onslaught of EUA submissions, in part caused by FDA’s decision to not exercise enforcement discretion for LDT-based COVID-19 tests, led FDA to decline most pre-submissions and to delay review of most submissions for non-COVID-19 IVDs for many months.[23]  This significant disruption to the broader IVD industry was triggered by an increase of approximately 1,200 submissions annually.[24]  These submissions all related to a single disease.  FDA’s own estimates would result in an influx of premarket submissions, in a six-month period, that would be 40 times greater than the number of COVID submissions and related to an extraordinarily diverse set of diseases and conditions.

Nor can FDA simply reallocate staff and resources from other, non-IVD product areas, as it did during COVID-19.  The requirements and data specific to in vitro diagnostics are unlike those for other device types.  Further, even if all CDRH review staff were reallocated to OHT7 to support this effort, it would still be insufficient given the vast volume of submissions that are expected.

The bottom line is this: without far better resource planning, this massive regulatory undertaking is going to be a disaster for all stakeholders, including FDA, industry, health care systems, providers and, most importantly, patients.  Certain points are indisputable.  First, under FDA’s own, extremely conservative projections there will be an exponential increase in submissions for diagnostic tests.  Second, based on years of experience, it would be hopelessly optimistic to expect the Third-Party Review Program to provide any meaningful relief.  Third, even with user fee support, it will be extremely challenging for FDA to recruit, hire, train, and manage the staff needed to cope with an unprecedented onslaught of submissions.  Fourth, the inability to manage this influx will harm the diagnostic industry, including manufacturers of distributed IVD products.

Without proper preparation one of two things will happen: (1) the regulatory review of non-LDT premarket submissions will come to a halt, or (2) there will be significant delays in the implementation of the final rule.  We do not expect that the broader IVD  industry will tolerate the first, particularly in light of the MDUFA VI negotiations that would be on-going leading up to the time at which labs will be required to file premarket submissions.  Thus, it is likely that FDA will have to push out the timelines for review of LDT premarket submissions to align with its available resources.  It would be far better for everyone if FDA acknowledged this reality now, and significantly extended the timelines for implementation.

As we explained in our comments (here), FDA should not finalize the proposed regulation.  If it does, instead of requiring submissions and then hoping to obtain resources and the necessary congressional authorizations, FDA should have a clear plan for obtaining the required resources.  The failure to do so is likely to result in chaos for all diagnostic products.

[1] https://www.fda.gov/news-events/fda-meetings-conferences-and-workshops/modernizing-fdas-data-strategy-06302020-06302020#:~:text=Data%20is%20at%20the%20heart,making%20and%20public%20health%20mission.

[2] U.S. Food & Drug Administration (FDA), II. Preliminary Regulatory Impact Analysis, Reference 34, Laboratory Developed Tests Proposed Rule, Docket No. FDA-2023-N-2177, at 21 (Oct. 4, 2023) [hereinafter PRIA], https://www.regulations.gov/document/FDA-2023-N-2177-0077. (“Laboratories that meet these requirements are the only laboratories that can perform LDTs under CLIA regulations, because LDTs are considered high complexity tests”)

[3] S&C’s Quality, Certification and Oversight Reports (QCOR), Active CLIA Laboratory Search Database, https://qcor.cms.gov/advanced_find_provider.jsp?which=4&backReport=active_CLIA.jsp.

[4] PRIA at 28.

[5] Id. at 75.

[6] 88 Fed. Reg. 68006, 68012 (Oct. 3. 2023) (hereinafter the “Notice of Proposed Rule Making”).  Notice of Proposed Rule Making Reference 1 “Grand View Research, “Laboratory Developed Tests Market Size, Share & Trends Analysis Report By Technology (Immunoassay, Molecular Diagnostics), By Application (Oncology, Nutritional & Metabolic Disease), By Region, and Segment Forecasts, 2023–2030: Report Summary,” available at https://www.grandviewresearch.com/industryanalysis/

laboratory-developed-tests-market-report (last accessed on Nov. 20, 2023) (“The lack of an equivalent IVD on the market is the primary reason labs develop LDTs.”)

[7] PRIA. at 28.

[8] FDA assumes that user fees will cover much of the increased costs. These fees, however, do not cover all costs. FDA does not discuss the consequences of a shortfall in congressional appropriations, which is a foreseeable risk.

[9] FDA anticipates the beginning of the MDUFA VI user fee cycle aligning with Stage 4 – FDA’s end of its enforcement policy for high-risk LDTs.  FDA, Transcript of CDRH Webinar regarding Proposed Rule: Medical Devices; Laboratory Developed Tests, at 9-10 (Oct. 31, 2023), https://www.fda.gov/media/173623/download?attachment.

[10] Id. at 9.

[11] Jeffrey N. Gibbs, Allyson B. Mullen & Melissa Walker, 510(k) Statistical Patterns, MD+DI (Dec. 2, 2014), https://www.mddionline.com/news/510k-statistical-patterns.

[12] FDA, Third Party Review Organization Performance Report (version 1 of FY2022, Q4), at 38, https://www.fda.gov/media/162658/download?attachment

[13] See id.; FDA, Performance Report to Congress: Medical Device User Fee Amendments FY 2022, at 13-18, https://www.fda.gov/media/167825/download?attachment.

[14] FDA, 510(k) Premarket Notification Search Database (last updated Nov. 13, 2023), https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPMN/pmn.cfm. Searches were done for all IVD panels: immunology, microbiology, clinical chemistry, pathology, toxicology, hematology, and molecular genetics.

[15] 21 U.S.C. § 360(m)(a)(3).

[16] FDA, Transcript of CDRH Webinar regarding Proposed Rule: Medical Devices; Laboratory Developed Tests, at 9-10 (Oct. 31, 2023), https://www.fda.gov/media/173623/download?attachment.

[17] FDA, News Release, FDA unveils a streamlined path for the authorization of tumor profiling tests alongside its latest product action (Nov. 15, 2017), https://www.fda.gov/news-events/press-announcements/fda-unveils-streamlined-path-authorization-tumor-profiling-tests-alongside-its-latest-product-action.

[18] Compare FDA, Current List of FDA-Recognized 510(k) Third Party Review Organizations (updated July 31, 2023), https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfthirdparty/accredit.cfm  [https://web.archive.org/web/20230803134353/https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfthirdparty/accredit.cfm], with FDA, Current List of FDA-Recognized 510(k) Third Party Review Organizations (updated Nov. 13, 2023), https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfthirdparty/accredit.cfm?party_key=7.

[19] FDA, Third Party Review Organization Performance Report (version 1 of FY2022, Q4), at 38, https://www.fda.gov/media/162658/download?attachment.

[20] PRIA at 22.

[21] Total number of submissions from PRIA estimates less 3% of 510(k)s divided by the annual number of submissions, given that all of these submissions are slated to be filed within six months of each other under the proposed rule.

[22] FDA, Financial Report to Congress: Medical Device User Fee Amendments of 2017 FY 2022, at 27, https://www.fda.gov/media/165116/download?attachment.

[23] https://www.fda.gov/news-events/fda-voices/year-pandemic-how-fdas-center-devices-and-radiological-health-prioritizing-its-workload-and-looking

[24] “[W]e continue to receive more than 100 IVD pre-EUA and EUA submissions per month.” See Jeff Shuren & William Maisel, Looking Ahead to 2022 as FDA’s Center for Devices and Radiological Health Manages a Sustained Increase in Workload, FDA Voices (Dec. 21, 2021), https://www.fda.gov/news-events/fda-voices/looking-ahead-2022-fdas-center-devices-and-radiological-health-manages-sustained-increase-workload.