For decades, the U.S. Food and Drug Administration has taken a light touch to regulating diagnostic tests that use body fluids or tissues. Such assays, very much including cerebrospinal fluid and blood tests for Alzheimer’s disease, were able to be marketed without providing proof of accuracy in the intended population. Now, the FDA is bringing down the hammer. In a May 6 ruling, the agency said future laboratory-developed tests (LDTs) will need to meet the same standards as medical devices, including passing quality benchmarks and reporting adverse events. The ruling will take effect July 5, but the new standard will be phased in over four years to give laboratories time to comply. Existing assays will be largely exempt, but it is expected they will comply to maintain a competitive edge.

  • FDA ruling requires in vitro diagnostics to meet the same standards as medical devices.
  • New tests will have to prove accuracy for detecting underlying disease.
  • The policy will phase in over four years; existing tests are exempt.

The ruling has sparked political controversy, and analysts expect industry to file lawsuits challenging the policy. However, scientists and clinicians contacted by Alzforum were uniformly enthusiastic about the change. “Increased oversight of LDTs is long overdue but very welcome,” Thomas Beach at Banner Sun Health Research Institute in Sun City, Arizona, wrote to Alzforum. Suzanne Schindler at the University of Washington, St. Louis, was of a similar mind. “This seems like a much-needed change that will help patients, providers, and the field have greater confidence in the results of Alzheimer’s disease blood tests,” she wrote. Researchers expect the FDA ruling to have positive effects beyond the U.S. as well. “It will surely be a guideline to other countries,” Colin Masters at the University of Melbourne, Australia, wrote to Alzforum (comments below).

Bringing Order to a Chaotic Field
Until now, in vitro diagnostics in the U.S. have existed in a nebulous regulatory space. Unlike imaging tests, which have always been regulated as medical devices, assays using body fluids or tissues received “enforcement discretion” from the FDA, meaning the agency mostly let them be. Instead, the Clinical Laboratory Improvement Amendments (CLIA) program administered by the Centers for Medicare and Medicaid Services certified tests for use. Importantly, however, CLIA only assesses a test’s reproducibility, not its accuracy for detecting disease. In theory, this means an inaccurate test can be sold, so long as it is reproducibly and consistently inaccurate. Over time, this has resulted in shoddy tests flooding the market, with no way for physicians to know which ones were best.

For example, Beach noted that the accuracies of α-synuclein skin assays for Parkinson’s disease vary from 24 to 98 percent. For Alzheimer's blood tests, only some have published accuracies as high as 90 percent. “Other tests are little better than flipping a coin, and no validation data has been published … [this] has led to confusion and skepticism about use of Alzheimer’s disease blood tests in clinical care,” Schindler wrote.

In a press release, the FDA noted that low-quality cancer and COVID-19 tests have harmed patients. Because LDTs are becoming more widely used to make important healthcare decisions, the agency believes a more rigorous standard is now required.

How will it work? Tests that come on the market after July 5 will be subject to the new ruling. By 2028, laboratories will have to ensure their tests meet minimum accuracy standards, determine how well they work in different populations, and report adverse events. As with medical devices, quality testing would be done by third parties, such as contract research organizations.

So as not to disrupt patient care, existing assays will be grandfathered in. However, any protocol changes or updates to those assays will be subject to the new standard, hence the FDA expects that over time, most tests will conform. A few LDTs will remain exempt, including tests developed to deal with public health emergencies such as pandemics, and those that meet a special need for which no FDA-approved assays are available.

The American Clinical Laboratory Association expressed disappointment in the ruling. The trade association, which represents laboratories that make tests, predicted it would hike the costs of testing, and suppress innovation. Senator Bill Cassidy (R-LA) said the FDA is overstepping its authority by issuing this rule, and called on Congress to act. Analysts believe legal challenges are likely (Endpoints article).

How Do Alzheimer’s Tests Stack Up?
If the ruling stands, how will it affect tests for Alzheimer’s disease? The answer will likely vary by company. For example, Fujirebio received FDA approval for its automated Aβ42/40 CSF assay, and Roche for its automated Aβ42/p-tau181 and Aβ42/t-tau CSF tests (May 2022 news; Dec 2022 news; Jun 2023 news). Because these tests have already been validated against amyloid PET, the companies may need to do little to bring them into full compliance with the new standard. Roche and Fujirebio representatives did not respond to a request for comment.

Joel Braunstein at C2N Diagnostics believes the effect on his company will be limited. “We do not believe the FDA’s recent final rule will change much of our strategy and approach to commercializing in vitro diagnostic tests,” he wrote to Alzforum. He pointed out that C2N has been working under a similar FDA quality standard since 2021, and routinely validates both the analytical and clinical performance of its tests (Nov 2020 news).

How about other companies? Quest Diagnostics at one point marketed an AD blood test directly to consumers, sparking backlash from the field (Aug 2023 news). The company has paused consumer marketing and currently focuses on delivering that test to primary care physicians, according to Quest communications director Wendy Bost. “Our menu isn’t changing at this time. We are reviewing the [FDA] rule to understand its requirements, and it will take some time to assess what effect, if any, it may have on our menu,” Bost wrote to Alzforum.

More AD tests are entering the market. For example, the laboratory services provider Labcorp, based in Burlington, North Carolina, recently announced that it's now offering a blood test for inflammatory marker GFAP based on Roche’s electrochemiluminescence system. The company did not respond to a request for comment.

Alzheimer’s researchers believe the ruling has global implications. “This regulation aligns well with the in vitro diagnostic regulation of the European Union, which also requires evidence on analytical and clinical validity and a quality and monitoring system for tests used for clinical decision making,” noted Charlotte Teunissen at Amsterdam University Medical Center.

Philip Scheltens at EQT Life Sciences Dementia Fund said reliable tests should report sensitivity, specificity, cutoff thresholds, and positive and negative predictive values in diverse populations. “One can only say that for the AD blood-based biomarkers, that is not yet the case, so this FDA ruling should be an incentive for those who develop these tests to get their act together,” he wrote to Alzforum (comments below).—Madolyn Bowman Rogers

Comments

  1. The announcement by the FDA of their Final Rule for increased oversight of laboratory developed tests (LDTs) is long overdue but very welcome. Once fully implemented, this should greatly increase the scientific rigor required for certifying LDTs for clinical use. The lack of rigor to date has resulted in the proliferation, over the past decades, of highly discrepant diagnostic accuracy reports for LDTs across medicine, including neurodegenerative diseases. While some LDTs have taken an extra step by obtaining CLIA certification, it has long been known, and the FDA explicitly has stated (Federal Register, 2023), “CLIA is not a substitute for FDA oversight” and “… under CLIA, CMS does not regulate critical aspects of laboratory test development, does not evaluate the performance of a test before it is offered to patients and healthcare providers, and does not assess clinical validity, i.e., the accuracy with which a test identifies, measures, or predicts the presence or absence of a clinical condition or predisposition in a patient.”

    A major reason for the FDA’s concern about LDTs has been their “high variability in performance.” This has also been well documented for biopsy-based diagnostic tests for synucleinopathies. Colonic biopsies were the first that were widely pursued for PD diagnosis, quickly resulting in widely varying reports of diagnostic accuracy. Multicenter studies sponsored by the Michael J. Fox Foundation eventually established, however, that mucosal colonic biopsies were poorly sensitive and poorly specific (Corbillé et al., 2016; Beach et al., 2016; Lee et al., 2017). This same problem now confuses the assessment of the utility of skin biopsies for PD, as the reported accuracy of these has varied between labs and methods from 24 percent to 98 percent (Chahine et al., 2020; Doppler 2021; Brumberg et al., 2021; Donadio et al., 2019; Gibbons et al., 2024). 

    A continuing, and still largely unrecognized, shortcoming of diagnostic test evaluations for PD and other synucleinopathies has been the persistent use of an inadequate gold standard. The clinical diagnoses of Parkinson’s disease, dementia with Lewy bodies, and multiple system atrophy are all much lower when compared to the autopsy diagnoses, especially for recognition of early disease stages, when the clinical diagnostic accuracy of PD is documented as being only 71 percent, while for even advanced DLB and MSA the respective figures are 45 percent and 61 percent (Adler et al., 2021; Koga et al., 2015; Sekiya et al., 2023; Fujishiro et al., 2008). 

    Unfortunately, unlike the situation for Alzheimer’s disease, where autopsy-validated FDA evaluation of PET amyloid (Clark et al., 2012) has provided a critical cornerstone on which new biomarkers may be more quickly compared, there is no autopsy-validated, FDA-approved diagnostic test for any of the synucleinopathies. This glaring inadequacy prevents confidence upon which the field could move forward. Even for AD, additional autopsy-validated, FDA-approved tests that are less expensive and more easily administered than amyloid PET would ease the validation of new tests. Recently, CSF and plasma tests for AD have been FDA-approved but only against amyloid PET. To use these newly approved biofluids tests to validate yet more new tests would suffer from a progressive degradation of the gold standard as amyloid PET, if even 90 percent accurate for diagnosing AD, could only enable a theoretical accuracy, in a biomarker tested against it, equivalent to the product of the two tests, so a new test that was found to be 90 percent accurate against amyloid PET would only have a theoretical accuracy of 0.9 x 0.9 = 0.81, or 81 percent. To then validate a third level of biomarkers against, for example, one of these CSF or plasma biomarkers, would see this theoretical accuracy reduced to 0.9 x 0.81 = 0.73 or 73 percent.

    For these reasons, I see the FDA Final Rule on LDTs to be essential for establishing much-needed rigor into the biomarker field.

    References:

    Medical Devices; Laboratory Developed Tests: A Proposed Rule by the Food and Drug Administration on 10/03/2023. Federal Register, October 3, 2023 Federal Register

    . Evaluation of alpha-synuclein immunohistochemical methods for the detection of Lewy-type synucleinopathy in gastrointestinal biopsies. Acta Neuropathol Commun. 2016 Apr 4;4:35. PubMed.

    . Multicenter Assessment of Immunohistochemical Methods for Pathological Alpha-Synuclein in Sigmoid Colon of Autopsied Parkinson's Disease and Control Subjects. J Parkinsons Dis. 2016 Oct 19;6(4):761-770. PubMed.

    . The Search for a Peripheral Biopsy Indicator of α-Synuclein Pathology for Parkinson Disease. J Neuropathol Exp Neurol. 2017 Jan 1;76(1):2-15. PubMed.

    . In vivo distribution of α-synuclein in multiple tissues and biofluids in Parkinson disease. Neurology. 2020 Sep 1;95(9):e1267-e1284. Epub 2020 Aug 3 PubMed.

    . Detection of Dermal Alpha-Synuclein Deposits as a Biomarker for Parkinson's Disease. J Parkinsons Dis. 2021;11(3):937-947. PubMed.

    . Dermal and cardiac autonomic fiber involvement in Parkinson's disease and multiple system atrophy. Neurobiol Dis. 2021 Jun;153:105332. Epub 2021 Mar 17 PubMed.

    . Abnormal α-synuclein deposits in skin nerves: intra- and inter-laboratory reproducibility. Eur J Neurol. 2019 Feb 15; PubMed.

    . Skin Biopsy Detection of Phosphorylated α-Synuclein in Patients With Synucleinopathies. JAMA. 2024 Apr 16;331(15):1298-1306. PubMed.

    . Clinical Diagnostic Accuracy of Early/Advanced Parkinson Disease: An Updated Clinicopathologic Study. Neurol Clin Pract. 2021 Aug;11(4):e414-e421. PubMed.

    . When DLB, PD, and PSP masquerade as MSA: an autopsy study of 134 patients. Neurology. 2015 Aug 4;85(5):404-12. Epub 2015 Jul 2 PubMed.

    . Validation Study of the MDS Criteria for the Diagnosis of Multiple System Atrophy in the Mayo Clinic Brain Bank. Neurology. 2023 Dec 12;101(24):e2460-e2471. Epub 2023 Oct 10 PubMed.

    . Validation of the neuropathologic criteria of the third consortium for dementia with Lewy bodies for prospectively diagnosed cases. J Neuropathol Exp Neurol. 2008 Jul;67(7):649-56. PubMed.

    . Cerebral PET with florbetapir compared with neuropathology at autopsy for detection of neuritic amyloid-β plaques: a prospective cohort study. Lancet Neurol. 2012 Aug;11(8):669-78. PubMed.

  2. Although no blood tests for Alzheimer’s disease pathology have yet been approved by the FDA, an increasing number of minimally regulated laboratory-developed tests (LDTs) are clinically available. There is very wide variation in the accuracy and validation of these LDT blood tests for AD: Some tests have accuracy similar to FDA-approved CSF tests (approximately 90 percent sensitivity and specificity) and have been described in peer-reviewed publications (e.g., Ashton et al., 2024; Barthelemy et al., 2024; Meyer et al., 2024). Other tests are little better than flipping a coin and no validation data has been published. Unfortunately, most clinicians and patients don’t know which tests are accurate and trustworthy. Additionally, some health systems only provide access to blood tests from specific diagnostics companies based on broader institutional agreements that may not take into account the accuracy and validation of the specific Alzheimer’s disease blood test. Overall, the major variation in the accuracy of AD blood tests has led to confusion and skepticism about their use in clinical care. For example, groups using lower accuracy tests have declared that blood tests aren’t clinically useful (Bouteloup et al., 2024)—these lower-accuracy tests will likely never be appropriate for clinical use (Buckley and Schindler, 2024).

    Some lower-accuracy tests use a cut-off with high sensitivity and low specificity, which they promote as useful in ruling out AD. This approach yields many false positive results, sometimes exceeding the number of true positive results. Even if the positive result is described as being “higher risk,” the patient may think they have Alzheimer disease and seek a referral to a dementia specialist. Given long wait times for dementia clinics and capacity constraints for CSF testing and amyloid PET, there is no guarantee that patients with a positive result will be able to access a dementia specialist and undergo follow-up testing within a reasonable period. Therefore, the use of tests with low specificity will inevitably lead to some patients incorrectly thinking they have Alzheimer’s disease for an extended period, which could lead to unnecessary distress and misinformed decision-making.

    Diagnosing a patient with dementia due to Alzheimer’s disease has a major impact on their life and must be based on accurate information. Therefore, I am grateful that the FDA will be more closely regulating LDTs in the coming years. This regulation will include common-sense requirements, such as disclosing the performance characteristics of the test. These changes will be phased in over the next four years and will not have an immediate impact on the tests that are currently available. However, over the next four years I expect that blood tests for AD will become much more broadly used in the diagnosis of dementia, especially as more treatments become available. Greater regulation will help to assure patients that these tests provide accurate information. While there is some concern that greater regulation may increase the burden on diagnostics companies and stifle innovation, the current environment of minimal regulation is leading to a proliferation of lower-accuracy tests that may be disincentivizing development of higher-accuracy tests and undermining confidence in AD blood tests generally. More regulation may lead to higher standards, better tests, more accurate diagnosis of patients receiving these tests, and more confidence in using these blood tests in clinical care. On the whole, this seems like a much-needed change that will help patients, providers, and the field have greater confidence in the results of AD blood tests.

    References:

    . Diagnostic Accuracy of a Plasma Phosphorylated Tau 217 Immunoassay for Alzheimer Disease Pathology. JAMA Neurol. 2024 Mar 1;81(3):255-263. PubMed.

    . Highly accurate blood test for Alzheimer's disease is similar or superior to clinical cerebrospinal fluid tests. Nat Med. 2024 Apr;30(4):1085-1095. Epub 2024 Feb 21 PubMed.

    . Clinical validation of the PrecivityAD2 blood test: A mass spectrometry-based test with algorithm combining %p-tau217 and Aβ42/40 ratio to identify presence of brain amyloid. Alzheimers Dement. 2024 May;20(5):3179-3192. Epub 2024 Mar 16 PubMed.

    . Explaining the Variability of Alzheimer Disease Fluid Biomarker Concentrations in Memory Clinic Patients Without Dementia. Neurology. 2024 Apr 23;102(8):e209219. Epub 2024 Mar 25 PubMed.

    . Lower Accuracy Alzheimer Disease Blood Tests Will Never Be "Ready for Prime Time". Neurology. 2024 Apr 23;102(8):e209295. Epub 2024 Mar 25 PubMed.

  3. Generally, increased oversight and improved Quality Assurance/Quality Control (QA/QC) is a good thing for the field, particularly as we move into routine clinical practice with significant public health implications. Blood screens for AD need to be shown to be effective in diverse conditions, especially when expensive therapies are involved. So, overall, I welcome this FDA action. It will surely be a guideline to other countries. I don’t think it will hamper the introduction of new assays, which usually need strong support from the diagnostic companies in industry.

  4. From our perspective at C2N, we do not believe the FDA’s recent final rule will change much of our strategy and approach to commercializing in vitro diagnostic tests. C2N has a Breakthrough Device designation with FDA for its Precivity™ test to identify amyloid pathology among individuals with signs or symptoms of cognitive impairment. Our laboratory has been working under the ISO13485:2016 quality management system (QMS) standard since 2021. This standard aligns with CFR 820, which is an important component of FDA’s new requirements. Further, we approach validation of our laboratory-developed tests in a way that includes both rigorous analytical and clinical validation to support an appropriate intended use. Our team is still reviewing FDA’s full policy to understand how it might impact future commercial tests. We have invested heavily in our QMS and expect to continue innovating high-performance diagnostic tests that address areas of significant unmet medical need.

  5. It is a good development that the FDA takes oversight of laboratory-developed tests that are used for patient decisions. I do not worry that it affects the development of new tests, but rather expect that it will stimulate it. This is because the investments of doing good research on novel diagnostic tests will be more valuable and better rewarded with the new oversight in place.

    This regulation also aligns well with regulation of in vitro diagnostics by the European Union, which also requires evidence of analytical and clinical validity and a quality and monitoring system for tests used for clinical decision making. To reiterate: Research use of a test is outside the scope of the rule.

    The oversight will prevent laboratories advertising, sometimes even direct to consumer, Alzheimer’s disease blood tests that have no published evidence. Evidence of good reliability in this intended population is lacking and interpretation of their test results can be very difficult. Harm can come to patients, especially those who do not have symptoms, if they rely on results that have not been validated in this setting, making this population vulnerable to poor advertising practices.

  6. An immunoassay, or any biomarker assay, needs to show acceptable, pre-defined performance criteria before it can be used for patient management or treatment selection, whether it is an IVD, LDT, or a research assay. The assay documentation must show results on the effect of comorbidities and a detailed protocol for sample handling (collection, aliquoting, freezing, thawing, testing). It must provide information on the protein form being measured in a specific biological sample and how that protein relates to the pathology or disease progression. The latter is heavily dependent on the critical raw materials integrated into the assay design, and on the robustness for sample testing. Better analytical precision will always result in less risk for misdiagnosis. The risk for patients needs to be minimized as much as possible.

    Difficulties with LDT testing include the lack of confirmation in other labs, the risk that the assay will not be available “long-term,” and variability between lot numbers.

    Clearly, the new rule will extend the development time before implementation in the lab. More experimental data will be required before the LDT can be considered and approved. However, I do not think it a good idea to distinguish between old (already available on the market) and new tests. All tests used in a clinical environment need to be of good quality, otherwise we will end up working with different categories of LDTs and different speeds of implementation. If sufficient performance criteria are unavailable, then the lab should have to go back to the drawing board to improve the production process.

  7. Without knowing the specifics of this, I would say, in general, that this move by the FDA is to be commended. For any laboratory test, one should be guaranteed that it has been developed with scientific rigor and that any patient should be able to rely on the outcome. Hence sensitivity/specificity, positive predictive value and negative predictive value should have been tested in the relevant populations, and cut-off values determined that would be applicable to diverse populations. One can only say that for the AD blood-based biomarkers, that is not yet the case, so this FDA ruling should be an incentive for those who develop these tests to get their act together.

    We have seen earlier that commercial parties have already started direct-to-consumer selling of these tests, and in an earlier comment I and others have strongly argued against that.

    I would suspect that there will be ripple effects and I think there should be. This [approval] should be a standard requirement worldwide for all lab tests, especially for AD. Diagnostic tests, such as amyloid and tau PET, had to go through the process at FDA, so why not also lab tests? Especially since the idea behind blood-based biomarkers for AD is that, ultimately, they will be used by primary care physicians, the more important it becomes that the tests are regulated for quality and reliability.

  8. Laboratory-developed test (LDTS) in the current framework are an important tool to accelerate innovation, in particular in rapidly evolving diagnostic markets such as AD. Understanding risks/benefits, the target, performance in racial populations, and multiple chronic comorbidities is important. LDTS are essential to promote clinical implementation and understand the impact of testing on care pathways.

  9. The FDA's decision to impose stricter regulations on in vitro "diagnostic" tests was anticipated, and this regulatory change will be beneficial for the Alzheimer's in vitro test ecosystem, as well as for clinicians and their patients. Europe has already implemented similar regulatory changes for in vitro tests, requiring prior validation by a notified body for the marketing phase, instead of simple self-certification. These FDA changes will benefit and clarify things for clinicians:

    ▶ They will help define what in vitro "diagnostic" tests for Alzheimer's disease really mean. Currently, diagnostic Alzheimer's tests can have different purposes. For example, a test predicting an "abnormal" level of amyloid plaques in the brain provides different information than a test predicting which patients with mild cognitive impairment (MCI), or no cognitive impairment, will develop clinical AD dementia symptoms in the next few years (Ritchie et al., 2014). These tests answer different medical questions, so it is crucial for clinicians to choose the appropriate test based on their specific medical question.

    ▶ Blood tests will now need to declare their "intended use," helping doctors understand how to interpret the results. Blood tests predicting cerebrospinal fluid (CSF) assay results will need to apply for a 510(k) pathway (incremental innovation) by comparing themselves to existing tests, such as the De Novo predicate device granted to Fujirebio Diagnostics in 2022 (DEN200072). Blood tests identifying MCI patients or cognitively unimpaired individuals that will convert to clinical symptoms of Alzheimer's dementia will follow the De Novo pathway since they currently have no predicate device (disruptive innovation). This will help clinicians understand test performance (sensitivity, specificity, false-positive rate, positive predictive value, negative predictive value) in relation to the intended use, allowing them to compare the benefits and limitations of each test. This way, clinicians can combine tests to get the most accurate information, especially when starting treatment that has significant side effects.

    ▶ Companies developing tests in the AD field have already implemented protocols and quality controls in quality management system standard laboratories aligned with European and FDA standards. This should therefore have no major impact on their development plans or regulatory approach.

    In conclusion, these FDA regulations changes for in vitro diagnostic tests will improve the U.S. ecosystem by keeping only those tests that have provided sufficient clinical evidence, making it easier for clinicians to interpret the results. This will likely stimulate disruptive innovation by highlighting unmet diagnostic needs in patient care.

    See: DEN200072, FDA’s evaluation of automatic class II designation for Lumipulse G 13-Amyloid Ratio (1-42/1-40), decision summary, https://www.accessdata.fda.gov/cdrh_docs/reviews/DEN200072.pdf

    References:

    . Plasma and cerebrospinal fluid amyloid beta for the diagnosis of Alzheimer's disease dementia and other dementias in people with mild cognitive impairment (MCI). Cochrane Database Syst Rev. 2014 Jun 10;6:CD008782. PubMed.

  10. The new regulations of diagnostic Alzheimer´s disease tests by the U.S. FDA are an important step to increase the accuracy of the tests. The stratification of patients with neurodegeneration is crucial for a positive therapy response, as already seen for cancer. It will be important to exclude false-positive results, especially in very early disease stages at which therapy response is most successful. Different tests with different readouts are needed to determine the best and most reliable test to classify Alzheimer’s disease in a symptom-free or only mildly impaired stage.

    However, the new regulations could restrict the development of innovative new tests, especially by start-up companies with limited resources compared to major pharmaceutical companies. It is therefore important that the new FDA ruling provide a broad and not too narrow framework that allows small start-up companies to also bring tests to the market. Care must be taken that the regulations not overshoot the target. That would be a drawback for small, innovative companies and would give preference to large pharmaceutical companies only, like the new IVD rules in Europe.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. FDA Approves Fujirebio’s CSF Test for AD—Quest Diagnostic Offers Plasma Test
  2. Blood Amyloid Test May Help Diagnose Alzheimer’s, but Questions Remain
  3. FDA Clears Roche’s CSF Aβ42/tTau Assay
  4. Plasma Aβ Test Wins Approval—Are p-Tau Tests Far Behind?
  5. Direct-to-Consumer Alzheimer’s Blood Test Opens Pandora’s Box

External Citations

  1. May 6 ruling
  2. press release
  3. Endpoints article
  4. announced

Further Reading