21 Nov 2024

Yes, blood tests for Alzheimer’s disease are ready for prime time. So pronounced Suzanne Schindler, Washington University, St. Louis, in a keynote presentation at CTAD 2024 last month in Madrid. In answering a question that has been raised often in the last few years, Schindler said that the field has reached a tipping point, and that highly accurate plasma tests deserve much of the credit.

“Things are moving very quickly. I expect in 2025 we may have FDA-approved blood tests that are reimbursed by payers,” Schindler said. “This would really change the landscape.”

Indeed, Jonathan Schott, Queen Square Institute of Neurology, London, noted that in the U.K., only about 2 percent of people who qualify are tested for CSF markers or a PET scan. He thinks blood tests could bring that number close to 100 percent. “We have the potential now to democratize molecular diagnosis,” Schott said. “I think we have an unbelievable opportunity here.”

A number of laboratory-developed tests (LDT) for blood p-tau217 are already commercially available in the U.S. and Europe, and used in research and some specialty clinical care centers. Current LDTs include those sold by ALZpath Inc., C2N, Eli Lilly Clinical Diagnostics Laboratory, LabCorp, Lucent Diagnostics, Mayo Clinic, and others. However, the U.S. Food and Drug Administration stipulated last May that, by 2028, any new plasma AD diagnostic will need approval. Existing tests are exempt, but many vendors will comply to stay competitive (May 2024 community news).

Which vendors are filing? Fujirebio last September requested approval from the FDA for its plasma p-tau217/Aβ1-42 test. At CTAD, Fujirebio’s Manu Vandijck said that following the FDA submission, the company also plans to seek IVD approval from the Competent Authorities for Medical Devices in the EU, and the Pharmaceutical and Medical Devices Agency in Japan.

In Madrid, Margherita Carboni, Roche Diagnostics International, told Alzforum that her company, too, had filed for IVD approval from the FDA for its p-tau181 and p-tau217 assays. As has Fujirebio for its test, Roche received FDA breakthrough device designation in July 2002 for its Elecsys Amyloid Plasma Panel, which measures plasma p-tau181 and ApoE4, and for its plasma p-tau17 last April.

Jeff Tarmy, Beckman Coulter Diagnostics, Brea, California, told Alzforum that his company expects to file for IVD certification in the near future. Startup company ALZpath Inc., Carlsbad, California, licensed its p-tau217 antibody to Beckman last July for use in blood-based in vitro diagnostics. Scientists who spoke to Alzforum said they expect other vendors to file soon, as well. Those in the running include C2N Diagnostic’s mass spectrometry-based test for the p-tau217/tau217 ratio, Eli Lilly’s p217 immunoassay, and others.

Leaders of AD drug development programs told Alzforum they are urgently awaiting regulatory approval for those tests, especially in Europe, where IVD status is often required for a plasma test to be used in investigational drug trials. In the U.S., laboratory-developed tests can be used in clinical trials.

How have things evolved so quickly? One driving force was diagnostic criteria issued by a Global CEO Initiative panel. Led by Schindler and Oskar Hansson from Lund University, Sweden, the panel earlier this year recommended that for AD triage in primary care, blood biomarker tests have a sensitivity and specificity of at least 90 and 85 percent, respectively, and that two separate cut points be used to identify people who are positive, negative, or indeterminate. Spurred by this, vendors are now optimizing their tests to meet these criteria.

As with every test, the positive and negative predictive value of a p-tau217 test, PPV and NPV for short, will depend in part on the prevalence of the tested condition in the population being screened (discussed in Aug 2024 conference news). Hence, vendors are validating their tests in various clinical settings.

In Spain, Francesca De Simone, Fujirebio Diagnostics Inc., Malvern, Pennsylvania, reported that, in their hands, the p-tau217/Aβ1-42 ratio outperforms p-tau217 alone. In samples from 208 volunteers—44 from the Swedish BioFinder2 study, 66 from the BIOCARD study at Johns Hopkins University, and 98 from the MissionAD trial of a discontinued BACE inhibitor—p-tau217 identified amyloid-positive people with PPVs and NPVs of 92 and 93 percent, respectively, which is within the CEOi criteria. However, for 34.2 percent of them the test was indeterminate. The CEOi panel recommends that no more than 20 percent of people fall into this gray zone.

At CTAD, De Simone showed that the p-tau217/Aβ1-42 ratio did better in this regard. The PPV and NPV reached 95 and 96 percent, respectively, and only 20 percent of samples were indeterminate. These results highlight a high performance of the p-tau217/Aβ1-42 ratio in predicting amyloid positivity, De Simone said. Vandijck reported finding the same numbers in a separate cohort of 500 volunteers. Twenty percent fell into the gray zone using the ratio, while 34 percent were indeterminate based on p-tau217 alone.

This held up in analysis of plasma samples in the Phase 3 AD trial of AriBio Inc.’s phosphodiesterase inhibitor AR1001. Sharon Sha, Stanford University, California, reported that for 256 samples the PPV and NPV for p-tau217/Aβ1-42 were 91 and 97 percent, respectively, with 21 percent indeterminate. For p-tau217 alone, PPV and NPV were comparable, 37 percent indeterminate.

For its part, Roche seems to be sticking with p-tau181+ApoE4 and p-tau217 alone. In Madrid, Carboni reported that the latter performs well across the AD spectrum. One burning question in the field is how well plasma markers will work in different populations, especially those where prevalence of AD may be low. Roche has tested its p-tau217 assay across five cohorts comprising people who are 60 and older. They span the AD spectrum from cognitively normal and an amyloid prevalence of 11 percent, to those clinically diagnosed with dementia, where the prevalence runs close to 60 percent. Carboni did not report PPVs and NPVs but said the AUCs, a measure of sensitivity and specificity, was 0.88 or higher across the spectrum.

“These evaluations have demonstrated robust clinical and technical performance across all cohort,” Carboni told the audience. Roche plans to test this assay across broader populations to prove it is useful and reliable. Carboni said they will use the two-cutoff method and are working to get the number of indeterminate reads to below 20 percent. 

P-tau217 Alone? Eli Lilly’s CertuitAD relegates only 18 percent of samples into the indeterminate zone. [Courtesy of Eli Lilly Clinical Diagnostics Laboratory.]

Lilly scientists have also settled on analysis of p-tau217 alone for their CertuitAD test, which is available commercially in the U.S. as an LDT for clinical diagnosis. It uses Lilly antibodies and runs on the Quanterix imaging system. Lilly was first to establish a plasma assay for p-tau217, based on antibodies developed by Jeff Dage, who is now at Indiana University, Indianapolis.

In Madrid, Lilly’s Rose Beck reported that among 2,071 volunteers who had been screened for the Trailblazer-Alz2 trial of donanemab, CertuitAD identified amyloid positives with a PPV and an NPV of 95 and 84 percent, respectively. Only 18 percent of this cohort, who were 60 and older with cognitive complaints, were indeterminate. On this basis, this assay would meet the CEOi criteria for an IVD. Beck told Alzforum that it is not clear if Lilly will file for approval with the FDA.

Post Approval
What happens once IVD blood diagnostics are approved? Scientists at CTAD agreed that work needs to be done to integrate them into clinical practice. “We need many more real-life studies to see how these tests perform in diverse settings,” said Schott during a roundtable discussion led by Marc Suárez-Calvet, Barcelonaβeta Research Center, Spain.

Researchers around the world are stepping up. In South Korea, scientists led by San Won Seo at Samsung Medical Center, Seoul, asked if having different high and low cutoffs for plasma p-tau217 at different stages of the Alzheimer’s helps improve test performance in their country. They tested samples from 2,699 participants in the Korea-Registries to Overcome Dementia and Accelerate Dementia Research. Part of the worldwide ADNI program, K-ROAD has recruited more than 5,800 people, between 63 and 80 years old, across 28 centers (Jang et al., 2024).

In Madrid, Jehyun Ahn, also at Samsung Medical Center, showed that among cognitively unimpaired people aged 65 and younger, lowering both cutoffs improved both PPV and NPV. For those over 65, only the NPV improved by using a group-specific cutoff. Among people with mild cognitive impairment or with a clinical diagnosis of AD, group-specific cutoffs did not improve diagnosis. Ahn did not report data on how other factors, including sex, ApoE genotype, comorbidities, or ethnicity might affect these cutoffs, but thinks this should be investigated.

Others are investigating how plasma tests will change clinical practice, much like IDEAS in the U.S. and AMYPAD in Europe assessed how amyloid PET scans influence disease management (April 2019 news; Sep 2022 conference news). Lilly’s clinical utility study for CertuitAD is testing how plasma p-tau217 results affect diagnosis, treatment, and management; it will read out in early 2025, said Beck. In the U.K., Schott is planning a randomized trial to test how disclosing blood test data three and 12 months later affects diagnosis and patient outcomes. “We have learned a lot about how to improve utility from IDEAS and AMYPAD. Touch wood, this should be easier with blood tests than with PET,” said Schott. Michael Schöll, University of Gothenburg, is running a similar study in southwestern Sweden (Jan 2024 news).

Scientists agree that these tests should only be used as part of a broader clinical evaluation. “There is a danger that a positive test will be seen as a clinical diagnosis,” warned Schott. An important aspect will be education for clinicians who may be naïve about molecular diagnoses. Suárez-Calvet, whose center uses blood-based biomarkers for both research and to support diagnosis, said that rolling the tests out with abundant physician education is important to prevent false use and a possible backlash. Marwan Sabbagh, Barrow Neurological Institute, Phoenix, agreed that physicians will need guidance. “I would start using these tests for their negative predictive value, not positive, and I would only use them in patients who have memory complaint,” Sabbagh said.

Appropriate-use recommendations for these tests will be issued. The Alzheimer’s Association is leading an effort to do that. “We start with clinical practice guidelines for specialty care,” said the association’s Rebecca Edelmayer. They will be based on current evidence, which at this point comes from secondary care settings. “We need to drive toward primary care, but that will require an enormous amount of clinical utility research,” Edelmayer said. The guidelines will evolve into a collection to support the right tests for the right patients at the right time, Edelmayer said

Who will pay for these tests? Dage said that the Centers for Medicare and Medicaid Services was proposing to reimburse at a rate of $17.27 per test. At that rate, labs will lose money and won’t implement the tests, Dage wrote to Alzforum. How did the CMS come up with that rate? “It is called cross-walking to an existing test,” according to Dage. Generic immunoassays are currently reimbursed at that rate, and the CMS extrapolates to AD.

CSF tests for some of those same analytes are reimbursed at $130. “Given the complexity of the blood tests, including the need for ultrasensitive detection and specialty high-affinity reagents, they should be reimbursed at a minimum of the $130 rate,” wrote Dage. The deadline to comment on the proposed fee passed on October 25. At time of writing, the final CMS payment schedule had not been posted.—Tom Fagan

Comments

No Available Comments

Make a Comment

To make a comment you must login or register.

References

News Citations

1. FDA Will Regulate Diagnostic Tests. Yes, Those for Alzheimer’s, Too. 10 May 2024
2. By Setting Standards, Experts Aim to Tame a Wild West of AD Blood Tests 21 Aug 2024
3. Results from IDEAS Study Published 3 Apr 2019
4. AMYPAD Confirms: Amyloid PET Improves Diagnosis 8 Sep 2022
5. AD Blood Tests Are Here. Now, Let's Grapple With How to Use Them 25 Jan 2024

Therapeutics Citations

1. Donanemab

Paper Citations

1. Jang H, Shin D, Kim Y, Kim KW, Lee J, Kim JP, Kim HJ, Cho SH, Kim SE, Na DL, Seo SW, K-ROAD Study Groups. Korea-Registries to Overcome Dementia and Accelerate Dementia Research (K-ROAD): A Cohort for Dementia Research and Ethnic-Specific Insights. Dement Neurocogn Disord. 2024 Oct;23(4):212-223. Epub 2024 Oct 22 PubMed.

External Citations

1. plasma p-tau217/Aβ1-42 test

Further Reading

No Available Further Reading