Despite extensive use of biomarkers in research studies, most Alzheimer’s disease diagnoses in the United States are determined by clinical examination. That may be about to change. The Food and Drug Administration issued its first approval for a fluid biomarker test May 4, greenlighting Fujirebio’s Lumipulse G cerebrospinal fluid Aβ42/40 assay. This approval will enable broad clinical usage and ensure more consistent insurance coverage.

  • The FDA approved the first CSF AD test, Fujirebio’s Aβ42/40 assay.
  • Quest Diagnostics will offer a plasma Aβ assay nationwide.
  • Primary care physicians want training on how to use these tests in clinical practice.

Adding another option for fluid marker testing, Quest Diagnostics, the largest diagnostic company in the world, announced May 2 that it plans to market a plasma Aβ assay in the United States. Quest AD-Detect will join C2N Diagnostics’ Precivity AD assay as the only blood tests certified for clinical use in the United States, though these tests are not yet FDA-approved or covered by insurance.

Researchers hailed these advances. “The approval of a CSF test is an important first step for fluid biomarkers. The rigorous FDA evaluation gives it a quality stamp,” Henrik Zetterberg at the University of Gothenburg, Sweden, told Alzforum. Meanwhile, Michelle Mielke at Wake Forest University in Winston-Salem, North Carolina, noted that blood-based biomarkers will facilitate more widespread AD screening, particularly in rural areas where PET imaging or CSF testing may not be available.

Nonetheless, researchers cautioned that primary-care physicians will need training to properly interpret Aβ biomarker test results, which can aid diagnosis but by themselves do not indicate the presence of disease. Importantly, all these fluid biomarker tests are intended for use in people with cognitive impairment, not for detecting preclinical disease in cognitively healthy people. Researchers also noted the need for more data on how the tests perform in diverse populations and in people with comorbidities, which are more common in routine clinical care than in research studies. “It’s very important that we understand how best to interpret these markers for clinical practice,” Mielke said.

FDA Approval Could Open Floodgates for Clinical CSF Testing
CSF AD tests have been part of clinical practice in Europe for more than 20 years, and are now used routinely. In the United States, their use has been more scattershot and mostly in research settings or specialty clinics. Many CSF tests are certified by the federal Clinical Laboratory Improvement Amendments (CLIA) for diagnostic testing, and thus can be used clinically. These CLIA-certified tests are typically reimbursed. So why is FDA approval even necessary? Private insurers have occasionally balked at paying, calling the CLIA-approved assays experimental. “CSF tests for AD have been in that gray zone,” Suzanne Schindler at Washington University in St. Louis told Alzforum.

Researchers expect FDA approval to boost confidence in the tests, smoothing the path to widespread clinical usage and reimbursement. In addition, the agency’s approval carries weight in Europe and elsewhere around the world, Zetterberg said. This is because the FDA evaluates the effectiveness of an assay in the clinical population that will use it, while certification, which only attests to the quality of the test, does not. “In the U.S., the FDA process is the only way to be confident you are using an assay that had all of its clinical claims vetted by an independent group of unbiased scientists,” Rianne Esquivel at Fujirebio told Alzforum.

Fujirebio’s CSF Aβ test was approved through the FDA’s De Novo pre-market review pathway, which is used for devices or assays with little risk to the patient. The test, an immunoassay run on the fully automated Lumipulse platform, is intended for use in people 55 or older with cognitive impairment. It discriminates amyloid-positive from -negative people with an accuracy of around 90 percent, and closely agrees with amyloid PET (Dec 2017 conference news; Nov 2018 conference news).

Andrew Budson at Boston University said this first FDA approval represents an enormous step forward. “Any physician, primary care provider, or specialist can use this test to improve the accuracy of their diagnosis,” he wrote to Alzforum (full comment below).

Another plus to FDA approval: It is expected to streamline the process for future fluid biomarkers, because they can be compared to an existing standard. Other CSF assays may soon be on the market. “The Elecsys AD CSF assays are available now in CE-marked countries, and we hope to bring them to the U.S. soon,” Sasha Bozeat at Roche told Alzforum.

Mass Spec Plasma Tests Gather Steam

Nonetheless, CSF testing has drawbacks, requiring an invasive lumbar puncture that can cause side effects, such as headache. While physicians in Europe routinely sample CSF, doctors in the U.S. are more averse to the procedure. Also, the test is not an option for some people, for example anyone on blood thinners. Blood tests would be simpler, more widely available, and more appealing to patients.

Enter Quest AD-Detect. In 2017, Quest Diagnostics, based in Secaucus, New Jersey, developed a liquid chromatography tandem mass spec test for evaluating the Aβ42/40 ratio in CSF (Weber et al., 2019). The plasma test was derived from this, Michael Racke at Quest told Alzforum. Similar to other mass-spec-based tests, it can measure plasma Aβ in the range of 10 to 2,500 pg/mill. Aβ42/40 ratios below a cutoff of 0.160 indicate the likelihood of AD. Quest analyzes all blood samples at its central lab in San Juan Capistrano, California.

The company collaborated with Todd Golde at the University of Florida, Gainesville, and James Brewer at the University of California, San Diego, to investigate the test’s clinical performance. When combined with APOE genotype and age, the assay detected the presence of amyloid in the brain with about 90 percent accuracy, Racke said. Those data are being prepared for publication. Meanwhile, the Alzheimer’s Clinical Trials Consortium is evaluating Quest’s test along with several others. Racke noted that ACTC was particularly interested in the potential for broad AD screening through Quest’s network of 2,250 patient service centers. These are located in Walmart stores and Albertsons grocery stores nationwide.

Quest AD-Detect joins C2N’s Precivity as the only CLIA-approved blood tests for AD in the United States (Nov 2020 news). Precivity is available in every state except New York, which has its own state certification process. As with Quest’s test, all blood samples for Precivity are sent to a central lab, in St. Louis, for analysis. The Precivity test recently received accreditation from the College of American Pathologists (CAP), and the company is working toward FDA approval, said company head Joel Braunstein.

C2N announced on May 18 that it will offer mass-spec-based assays of plasma p-tau217 and p-tau181 as well, but at the moment these tests are for research use only.

Other companies have also developed mass spec plasma Aβ tests. Araclon Biotech in Zaragoza, Spain, reports testing its assay in the Swedish BioFINDER cohort, with the results to be published soon.

Immunoassays Offer Speed and Automation
For truly widespread clinical use, fully automated immunoassays such as those offered by Fujirebio and Roche would be more practical than mass-spec-based tests, avoiding the need to send blood samples to a central lab. However, the tiny drop of about 10 percent in the plasma Aβ ratio in amyloid-positive people is problematic for these assays, falling short of the needed sensitivity (Jun 2019 news). Roche is getting around this problem by using a combination of plasma p-tau181 and ApoE4 instead. “This was the combination that gave the best clinical performance while still being robust,” Bozeat told Alzforum.

Roche is assessing this blood test in a clinical population and preparing an FDA submission. The company’s plan is to have the biomarker ready to accompany a potential commercial launch of its anti-amyloid antibody gantenerumab next year, should Phase 3 trials give a positive readout this fall. Bozeat noted that the blood test’s high negative predictive value will help rule out people who have a low likelihood of amyloid positivity, reducing unnecessary CSF draws and PET scans. “We see it as streamlining the diagnostic pathway,” Bozeat said.

Fujirebio has taken a different approach. In Europe, the company is pursuing CE certification for the Lumipulse plasma p-tau181 test; meanwhile, its blood tests are available for research use in Europe and the United States. However, the company is not yet convinced of their utility for routine clinical use. “We don’t see quite the sensitivity and specificity we would like to have in a true diagnostic test in plasma,” Esquivel said.

Other companies specializing in neurological diagnostics, such as Euroimmun and Quanterix, are working on AD plasma immunoassays as well. However, larger clinical chemistry companies, which maintain platforms analyzing hundreds of analytes, are hesitant to move into the AD field because of the challenges of this biomarker, Zetterberg said. “It’s a tricky biomarker that requires a lot of work,” he noted.

Will Clinicians Use Plasma Tests?
Right now, clinical use of plasma AD tests is limited in the United States. This is due to cost and the lack of insurance coverage—a Precivity test can run $1,250—as well as more limited validation data compared to CSF tests. “We’re happy to see these blood tests become commercially available, but so far they haven’t had much of an impact on either our research activities or clinical practice,” noted Russell Swerdlow at Kansas University Medical Center in Kansas City (full comment below). “We’re in a wait-and-see mode.”

Schindler noted that WashU, where the Precivity test was developed, has only used the test on about 80 patients so far. These are mostly people who cannot tolerate lumbar punctures. “We really need insurance to reimburse before we can use blood tests routinely,” Schindler said.

Still, researchers expect usage to rise rapidly, especially as FDA approvals and insurance coverage come through. Quest Diagnostics claims 87 percent of primary care providers surveyed think blood tests will become the standard of care for AD, and three-quarters want training on how to use them. Patients also expressed interest, with 83 percent of the 2,052 American adults surveyed saying they would be willing to take such a blood test. The demand could soar into the millions if an AD treatment is fully approved by the FDA. “I don’t think there’s any way we could manage that scale of testing with CSF or amyloid PET,” Schindler said.

Potential Pitfalls Await
Scientists foresee several hurdles to widespread clinical adoption of AD blood tests. For one thing, they need more information on how the tests perform in diverse clinical populations with comorbidities that affect results. For example, chronic kidney disease boosts AD biomarkers in the blood and could lead to false-positive test results, Mielke noted (Syrjanen et al., 2021). Other factors such as body mass index affect these biomarkers as well. Some studies suggest race and ethnicity influence biomarker profiles (Jan 2019 news; Apr 2022 news). “We need to understand this before these assays are taken to the population level,” Mielke said.

Researchers also stressed that CSF and plasma tests should not be used in cognitively healthy people worried about their risk of AD. Because the disease has such a long preclinical phase, some older people who test positive will not develop symptoms in their lifetime. A positive test could cause needless distress, and also have financial ramifications, such as making it more difficult to get long-term care insurance, Mielke noted. In addition, biomarker tests perform differently in presymptomatic and symptomatic populations, noted Colin Masters at the University of Melbourne, Australia (Nakamura et al., 2018; Leuzy et al., 2022). Braunstein agrees selecting the correct population is key. “The field is still figuring out how best to use these markers in people who don’t yet have symptoms,” he said.

Researchers cautioned against diagnosing AD based on biomarker tests alone. They emphasized that clinicians should do a full workup to identify other factors that might be causing clinical decline, such as medications or sleep apnea. “I hope as these tests become used more widely, people continue to look for other factors besides AD,” Schindler said.

Right now, there is no large-scale effort to educate primary care physicians on how to interpret AD biomarker tests. “The system needs to evolve,” Mielke said. “We need more dementia specialists, and we need to think about this from a comprehensive perspective.”—Madolyn Bowman Rogers

Comments

  1. I believe that this approved test is an enormous step forward in providing an easier, safer, and (hopefully) less expensive way to diagnose the presence or absence of amyloid plaques in the brain. With 97 percent of individuals with positive results showing the presence of amyloid plaques by PET scan, and 84 percent of individuals with negative results having a negative amyloid PET scan, I think the test error is in the right direction, with the false-positive rate being quite low. I think that any physician, primary-care provider, or specialist can use this test to improve the accuracy of their diagnosis. However, they should be cautioned to start with the basic work-up, including the history from the individual and an informant, physical and neurological exam, MRI or CT scan of the brain, and blood test to rule out medical problems, vitamin deficiencies, or thyroid disorders. All providers need to keep the false-positive and false-negative rates in mind when discussing the results of the test with individuals and their families.

  2. Here at KU we’re happy to see these commercially available blood tests become available, but so far they haven’t had much of an impact on either our research activities or clinical practice. For our research study participants, we do our ATN biomarkers in-house. In our clinic, we are kind of in a wait-and-see mode for a few obvious reasons. First, we’re wondering how well the diagnostic value of these tests compares to the main ATN test we currently use, namely CSF Aβ, tau, and p-tau measurements. The CSF gives us information on tau, which seems to be an advantage. Also, cost is a factor. So far, the cost to the patient seems to be higher with the blood-based tests, because CSF studies are, to an extent, covered by insurance. For these reasons I suspect we haven’t yet embraced the plasma tests. It’s certainly not that we’re against them, it’s just that at this point the reasons to switch haven’t overridden the reasons not to switch. I suspect we might start ordering plasma tests for the clinic if the cost to the patients came down.

    If there is a potential pitfall, it would relate to the issue of putting the test data into context. For instance, we use biomarker tests to help refine our differential diagnosis in tough cases, or to inform prognoses in those who have cognitive complaints in the absence of a clinically detectable Alzheimer’s syndrome. However, the field recognizes these tests can turn positive decades before a patient manifests a clear AD syndrome. This is something I worry about when I’m contextualizing biomarker data. For example, let’s say a patient presents with a clinical picture of semantic variant primary progressive aphasia (svPPA), and they have a positive AD biomarker. Does that mean they must have AD rather than svPPA, or that they have svPPA now, and at some point in the next 20 years they may also have AD dementia? If a 70-year-old with subjective cognitive complaints has a positive AD biomarker, does that mean they are going to manifest an AD syndrome in one year, five years, 20 years, or ever? Plus, this is all playing out against a background of “What is AD?” Is it a biomarker change? A clinical syndrome plus a biomarker change? To me these are important questions. The field needs to decide if the Aβ and tau measurements constitute AD, or if they are biomarkers of AD. I’m not sure we can have it both ways.

  3. Studies have demonstrated that diagnosis of symptomatic Alzheimer's disease based on clinical criteria alone leads to a significant rate of misdiagnosis. In one large study, 25 percent of patients diagnosed with symptomatic AD were amyloid-PET-negative (Rabinovici et al., 2019). If it is important to have an accurate diagnosis, it is essential to use AD biomarkers. Unfortunately, the threshold for AD biomarker testing has been quite high for both CSF biomarkers and amyloid PET: Patients and clinicians must believe that increased diagnostic accuracy will be worth the perceived risks (and small but real risks) of a lumbar puncture (LP), or the cost of an amyloid PET scan. The consequence of this high threshold is relatively infrequent biomarker testing. In our memory clinic, we only perform about six to 12 LPs for AD biomarkers per month. I suspect that less than 5 percent of patients with cognitive concerns in the community undergo AD biomarker testing.

    Although dementia specialists have been using CSF biomarkers of AD for many years, the FDA approval of the Fujirebio test may make some providers more comfortable using CSF biomarkers and it should remove any remaining obstacles to insurance reimbursement. Blood tests, on the other hand, have minimal risks and therefore may reduce the high threshold for biomarker testing, such that they can be used more broadly. However, clinical use of AD blood tests will be limited until they are reimbursed by insurance. The proliferation of AD blood tests and the ever-increasing data on their performance will hopefully lead to reimbursement, so that more patients can receive an accurate diagnosis.

    References:

    . Association of Amyloid Positron Emission Tomography With Subsequent Change in Clinical Management Among Medicare Beneficiaries With Mild Cognitive Impairment or Dementia. JAMA. 2019 Apr 2;321(13):1286-1294. PubMed.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Automated CSF Tests: Check. Blood Tests: In the Works
  2. Amyloid PET Aids Diagnosis. But Could CSF Do Just as Well?
  3. Plasma Aβ Test Wins Approval—Are p-Tau Tests Far Behind?
  4. Drawing Closer: Alzheimer’s Blood Test for Primary Care
  5. Do Alzheimer’s Biomarkers Vary by Race?
  6. Diversity: FDA Guidance and New Data on Incidence/Biomarkers by Race

Therapeutics Citations

  1. Gantenerumab

Paper Citations

  1. . High-Throughput Mass Spectrometry Assay for Quantifying β-Amyloid 40 and 42 in Cerebrospinal Fluid. Clin Chem. 2019 Dec;65(12):1572-1580. Epub 2019 Oct 18 PubMed.
  2. . Associations of amyloid and neurodegeneration plasma biomarkers with comorbidities. Alzheimers Dement. 2021 Sep 27; PubMed.
  3. . High performance plasma amyloid-β biomarkers for Alzheimer's disease. Nature. 2018 Feb 8;554(7691):249-254. Epub 2018 Jan 31 PubMed.
  4. . Blood-based biomarkers for Alzheimer's disease. EMBO Mol Med. 2022 Jan 11;14(1):e14408. Epub 2021 Dec 3 PubMed.

External Citations

  1. issued
  2. announced 
  3. patient service centers
  4. announced 
  5. reports

Further Reading