Impact of zervimesine on the neuroinflammatory biomarker GFAP and related proteomic molecular correlates in plasma of participants from a phase 2 clinical trial in Alzheimer’s disease

In the paper

“reports of findings from the subgroup analysis of phase 3 trial TRAILBLAZER-ALZ2 suggest that donanemab is more effective in participants with less cognitive impairment and lower tau burden [16, 18].”

What the sources say

• ?Not checked — no full text of the sourcecites [16] Vasudeva R, Doty E, Yang F, Evans C, Zimmer J, Lu M, et al. The Importance of Early Recognition and Treatment Initiation for Managing Alzheimer’s disease: Subpopulation Analysis of the TRAILBLAZER-ALZ 2 Randomized Trial (S23.005). Neurology. 2025;104. https://doi.org/10.1212/WNL.0000000000210512. source
• Partly supported

  The Discussion reports greater benefit in the low/medium-tau group than in the high-tau group and says this supports a hypothesis of greater benefit when treatment starts earlier. But the Subgroup Analyses section says outcomes were generally consistent across baseline-characteristic subgroups; our copy does not establish greater efficacy specifically among participants with less cognitive impairment. (LLM, conf=0.82)

  cites [18] Sims JR, Zimmer JA, Evans CD, Lu M, Ardayfio P, Sparks J, et al. Donanemab in Early Symptomatic Alzheimer Disease: The TRAILBLAZER-ALZ 2 Randomized Clinical Trial. JAMA. 2023;330:512–27. https://doi.org/10.1001/jama.2023.13239. sourceExperimental — accuracy not yet measured

In the paper

“p-tau217 was selected for this substudy as it is an extensively validated to be an excellent indicator of degree of amyloid as assessed via CSF analysis and amyloid PET, and tau levels as assessed via tau PET [9, 13, 14, 32]”

What the sources say

• Partly supported

  The Abstract and Discussion (page numbers unavailable) report that plasma p-tau217 predicts amyloid positivity defined by CSF or PET and identifies patients likely to have high tau-PET burden; the Discussion also cites prior supporting studies. The paper evaluates amyloid status and categories of tau burden, not how accurately p-tau217 indicates the quantitative degree of amyloid or tau levels across their ranges. (LLM, conf=0.85)

  cites [9] Mattsson-Carlgren N, Collij LE, Stomrud E, Pichet Binette A, Ossenkoppele R, Smith R, et al. Plasma biomarker strategy for selecting patients with Alzheimer disease for antiamyloid immunotherapies. JAMA Neurol. 2024;81:69–78. https://doi.org/10.1001/jamaneurol.2023.4596. sourceExperimental — accuracy not yet measured
• Partly supported

  The Results section reports that plasma p-tau217 was significantly associated with both amyloid PET and tau PET measures, supporting those parts of the claim. This paper does not establish its performance against amyloid measured by CSF analysis; the Introduction discusses CSF biomarkers but does not provide that comparison. Its PET findings also vary by disease stage, so “excellent indicator” is broader than the reported associations. (LLM, conf=0.86)

  cites [13] Mattsson-Carlgren N, Janelidze S, Bateman RJ, Smith R, Stomrud E, Serrano GE, et al. Soluble P-tau217 reflects amyloid and tau pathology and mediates the association of amyloid with tau. EMBO Mol Med. 2021;13:e14022. https://doi.org/10.15252/emmm.202114022. sourceExperimental — accuracy not yet measured
• Partly supported

  The Results section, “Accuracy in Discriminating Abnormal Aβ and Tau Pathologies” (page unavailable), reports high accuracy for identifying abnormal amyloid by CSF and amyloid PET and abnormal tau by tau PET across the studied cohorts. It also reports increasing plasma p-tau217 across tau-PET–defined Braak stages. The evidence is strongest for distinguishing abnormal from normal pathology, rather than for the claimed measurement of the degree of amyloid and tau pathology by each method. (LLM, conf=0.82)

  cites [14] Ashton NJ, Brum WS, Molfetta GD, Benedet AL, Arslan B, Jonaitis E, et al. Diagnostic accuracy of a plasma phosphorylated tau 217 immunoassay for Alzheimer disease pathology. JAMA Neurol. 2024;81:255–63. https://doi.org/10.1001/jamaneurol.2023.5319. sourceExperimental — accuracy not yet measured
• Partly supported

  The Background reports that plasma pTau217R reflects brain amyloid and tau pathology and cites prior findings that it predicts continuous amyloid levels. The Abstract Results report that a pTau217R-based model predicts tau levels measured by PET in amyloid-positive early AD. Our copy does not establish the claimed validation against CSF amyloid measurements, and its findings concern the pTau217 ratio rather than p-tau217 generally. (LLM, conf=0.86)

  cites [32] Devanarayan V, Charil A, Horie K, Doherty T, Llano DA, Andreozzi E, et al. Plasma pTau217 ratio predicts continuous regional brain tau accumulation in amyloid-positive early Alzheimer’s disease. Alzheimers Dement. 2025;21:e14411. https://doi.org/10.1002/alz.14411. sourceExperimental — accuracy not yet measured

In the paper

“plasma GFAP is a sensitive biomarker for detecting reactive astrogliosis, a hallmark of neuroinflammation, as well as Aβ pathology, even in early stages of AD [33].”

What the sources say

• Partly supported

  The Introduction calls blood GFAP a reactive-astrogliosis biomarker, and the Results and Discussion support plasma GFAP as an early indicator of Aβ pathology. However, the Results section on neuroinflammation found that plasma GFAP, unlike CSF GFAP, was not associated with the other glial biomarkers tested. The paper therefore does not establish plasma GFAP as a sensitive detector of neuroinflammation generally. (LLM, conf=0.86)

  cites [33] Benedet AL, Milà-Alomà M, Vrillon A, Ashton NJ, Pascoal TA, Lussier F, et al. Differences between plasma and cerebrospinal fluid glial fibrillary acidic protein levels across the Alzheimer disease continuum. JAMA Neurol. 2021;78:1471. https://doi.org/10.1001/jamaneurol.2021.3671. sourceExperimental — accuracy not yet measured

In the paper

“the oligomer receptor prion protein, PrPC, not only binds TMEM97 (S2R) [36], it is also known to bind amyloid precursor protein (APP) [37]”

What the sources say

• Partly supported

  The Results section reports a FRET signal between TMEM97 and PrPC in human postsynapses, supporting close proximity. The Discussion treats a direct interaction as possible, but the FRET finding does not establish that PrPC binds TMEM97. The APP-binding clause has a separate citation [37] and is not assessed against this source. (LLM, conf=0.86)

  cites [36] Colom-Cadena M, Toombs J, Simzer E, Holt K, McGeachan R, Tulloch J, et al. Transmembrane protein 97 is a potential synaptic amyloid beta receptor in human Alzheimer’s disease. Acta Neuropathol. 2024;147:32. https://doi.org/10.1007/s00401-023-02679-6. sourceExperimental — accuracy not yet measured
• Supported

  This source is cited for the APP-binding clause [37], not the TMEM97 clause [36]. Section 4.3, “Alzheimer’s Disease-Related Proteins,” says that the association of PrPC and APP was validated by co-immunoprecipitation. (LLM, conf=0.94)

  cites [37] Miranzadeh Mahabadi H, Taghibiglou C. Cellular prion protein (PrPc): putative interacting partners and consequences of the interaction. Int J Mol Sci. 2020;21:7058. https://doi.org/10.3390/ijms21197058. source

In the paper

“The level of plasma p-tau217 is correlated with amyloid and tau pathology on PET imaging and has proven useful in identifying study participants for trials assessing the effects of anti-amyloid therapies [11–13].”

What the sources say

• Partly supported

  The Results sections on abnormal Aβ and tau pathologies and on other plasma biomarkers report that plasma p-tau217 predicts PET-defined amyloid and tau pathology and correlates with both PET measures (page numbers unavailable in our copy). The Discussion describes promising utility for early Alzheimer disease trials, but says further studies of trial-inclusion designs are warranted. It does not establish that p-tau217 has already proven useful for selecting participants in anti-amyloid therapy trials. (LLM, conf=0.88)

  cites [11] Ashton NJ, Brum WS, Di Molfetta G, Benedet AL, Arslan B, Jonaitis E, et al. Diagnostic accuracy of a plasma phosphorylated tau 217 immunoassay for Alzheimer disease pathology. JAMA Neurol. 2024;81:255. https://doi.org/10.1001/jamaneurol.2023.5319. sourceExperimental — accuracy not yet measured
• ?Not checked — no full text of the sourcecites [12] Hu Y, Cho M, Sachdev P, Dage J, Hendrix S, Hansson O, et al. Fluid biomarkers in the context of amyloid-targeting disease-modifying treatments in Alzheimer’s disease. Med. 2024;5:1206–26. https://doi.org/10.1016/j.medj.2024.08.004. source
• Partly supported

  The Results section reports associations between plasma P-tau217 and both amyloid and tau PET measures, supporting the first part of the claim. The paper does not report that plasma P-tau217 has proven useful for selecting participants for anti-amyloid therapy trials; its Participants section says amyloid positivity in the studied cohort was determined by PET. (LLM, conf=0.87)

  cites [13] Mattsson-Carlgren N, Janelidze S, Bateman RJ, Smith R, Stomrud E, Serrano GE, et al. Soluble P-tau217 reflects amyloid and tau pathology and mediates the association of amyloid with tau. EMBO Mol Med. 2021;13:e14022. https://doi.org/10.15252/emmm.202114022. sourceExperimental — accuracy not yet measured

In the paper

“The Aβ42/40 ratio is decreased in AD patients reflecting increased brain amyloid plaque load, and has been helpful in the diagnosis of AD [34, 35].”

What the sources say

• Supported

  The BBMs for AD and related disorders section says the plasma Aβ42/Aβ40 ratio is lower in amyloid-positive groups and is associated with brain plaques. The CSF-based biomarkers section says the ratio correlates strongly with amyloid PET status and AD neuropathology and that CSF assays are used clinically. The paper also cautions that plasma assays vary substantially in diagnostic performance, so the claim should not be read as applying equally to every assay. (LLM, conf=0.89)

  cites [34] Hansson O, Blennow K, Zetterberg H, Dage J. Blood biomarkers for Alzheimer’s disease in clinical practice and trials. Nat Aging. 2023;3:506–19. https://doi.org/10.1038/s43587-023-00403-3. source
• Partly supported

  The Results (Demographic data and CSF biomarkers; page unavailable) report a lower CSF Aβ42/40 ratio in participants with greater PET-measured cortical amyloid deposition. The Introduction says the ratio improves the accuracy of clinical AD diagnosis. However, the study’s cohort was 88% cognitively normal, so its finding does not directly establish that the ratio is decreased specifically in AD patients. (LLM, conf=0.86)

  cites [35] Lewczuk P, Matzen A, Blennow K, Parnetti L, Molinuevo JL, Eusebi P, et al. Cerebrospinal fluid Aβ42/40 corresponds better than Aβ42 to amyloid PET in Alzheimer’s disease. J Alzheimers Dis. 2016;55:813–22. https://doi.org/10.3233/JAD-160722. sourceExperimental — accuracy not yet measured

In the paper

“Treatment of neurons with zervimesine restores the functional and synaptic deficits incurred by Aβ oligomers, including the restoration of membrane trafficking, and the prevention of synapse loss [1].”

What the sources say

• Supported

  The Results report that CT1812 restored Aβ-oligomer-induced vesicular trafficking deficits in neuronal cultures (Figure 1) and prevented or reversed synapse loss in cultured neurons (Figure 2). The paper uses the name CT1812 rather than zervimesine, but the reported effects match those attributed to the treatment. (LLM, conf=0.88)

  cites [1] Izzo NJ, Yuede CM, LaBarbera KM, Limegrover CS, Rehak C, Yurko R, et al. Preclinical and clinical biomarker studies of CT1812: a novel approach to Alzheimer’s disease modification. Alzheimers Dement. 2021;17:1365–82. https://doi.org/10.1002/alz.12302. source

In the paper

“EEA1 is a key protein is a key marker of early endosomes that plays a crucial role in the endocytic trafficking and amyloidogenic processing of amyloid precursor protein (APP) [42], and altered expression of endosomal proteins including EEA1, can contribute to the AD progression, particularly in relation to Aβ accumulation [43].”

What the sources say

• ?Not checked — no full text of the sourcecites [42] Burrinha T, Martinsson I, Gomes R, Terrasso AP, Gouras GK, Almeida CG. Upregulation of APP endocytosis by neuronal aging drives amyloid-dependent synapse loss. J Cell Sci. 2021. https://doi.org/10.1242/jcs.255752. source
• Partly supported

  Split into 2 atoms — 0 confirmed, 2 partially supported, 0 disputed, 0 misattributed, 0 unverifiable. Expand the claim to see per-atom verdicts.

  cites [43] Xicota L, Lagarde J, Eysert F, Grenier-Boley B, Rivals I, Botté A, et al. Modifications of the endosomal compartment in fibroblasts from sporadic Alzheimer’s disease patients are associated with cognitive impairment. Transl Psychiatry. 2023. https://doi.org/10.1038/s41398-023-02355-z. sourceExperimental — accuracy not yet measured

In the paper

“IGF2R is important for memory, and its dysregulation may contribute to AD pathology. IGF2R levels decrease with the presence of the APOE4 allele, which is associated with increased risk for AD and overexpression of IGF2R is also associated to increase Aβ production [51, 52].”

What the sources say

• Partly supported

  The section on IGF2 expression and memory reports that blocking or deleting hippocampal IGF2R impairs long-term memory. The CNS diseases section links IGF2R to amyloid reduction in an AD mouse model, but does not report lower IGF2R levels with APOE4 or increased Aβ production from IGF2R overexpression; it instead reports that IGF2 overexpression reduced Aβ, apparently via IGF2R. Page numbers are unavailable in our copy. (LLM, conf=0.87)

  cites [51] Alberini CM. IGF2 in memory, neurodevelopmental disorders, and neurodegenerative diseases. Trends Neurosci. 2023. https://doi.org/10.1016/j.tins.2023.03.007. sourceExperimental — accuracy not yet measured
• Supported

  The Discussion reports that hippocampal neuronal CIM6P/IGF2R is crucial for memory consolidation and implicates the receptor in Alzheimer’s disease. It also states that receptor levels decrease with APOE4 alleles and that receptor overexpression increases Aβ production. The APOE4 and overexpression findings are reported from earlier studies, not presented as this paper’s own experiments. (LLM, conf=0.94)

  cites [52] Yu XW, Pandey K, Katzman AC, Alberini CM. A role for CIM6P/IGF2 receptor in memory consolidation and enhancement. Elife. 2020;9:1–25. https://doi.org/10.7554/eLife.54781. source

In the paper

“Nominally significant trends (p ≤ 0.10) in reducing plasma levels of Aβ40, Aβ42, and NfL were also observed in zervimesine vs. placebo treated participants”

In the paper

“Scatter plots of two of these proteins – glutathione peroxidase 3 (GPX3) and cathepsin G (CTSG) – are shown, demonstrating examples of a positive and negative correlation of change in abundance with improvement in plasma GFAP levels”

In the paper

“of which 58 proteins were significantly differentially abundant with zervimesine treatment (Fig. 2A: zervimesine vs. placebo p ≤ 0.05, 53 increased (red), 5 decreased (green)).”

In the paper

“Fifteen proteins (Fig. 4C, brown overlap) met both correlation and differential abundance criteria.”

In the paper

“only the 283 “Drug Selective” protein correlates (r≥|0.50|) (orange semicircle; top correlates shown in Supplemental Table 3), were selected for pathway analyses”

In the paper

“These candidate biomarkers include early endosome antigen 1 (EEA1), identified in mITT and low p-tau217 groups, and progranulin (GRN) identified in the mITT population.”

In the paper

“Data provide insights into the mechanism by which zervimesine may ameliorate AD through S2R, perhaps via dampening neuroinflammation in addition to slowing neurodegeneration”

In the paper

“Plasma biomarkers are less invasive, less costly, and more time-efficient measurements than CSF or neuroimaging biomarkers [9].”

What the sources say

• Not checked yetcites [9] Mattsson-Carlgren N, Collij LE, Stomrud E, Pichet Binette A, Ossenkoppele R, Smith R, et al. Plasma biomarker strategy for selecting patients with Alzheimer disease for antiamyloid immunotherapies. JAMA Neurol. 2024;81:69–78. https://doi.org/10.1001/jamaneurol.2023.4596. source

In the paper

“Participants were randomized 1:1:1 to receive placebo, or 100 or 300 mg zervimesine, orally, once daily for 6 months.”

In the paper

“Increased GFAP, associated with astrogliosis, is a hallmark of and linked to pathological processes in Alzheimer’s disease, particularly amyloid build-up, and neuroinflammation [27].”

What the sources say

• Not checked yetcites [27] Pereira JB, Janelidze S, Smith R, Mattsson-Carlgren N, Palmqvist S, Teunissen CE, et al. Plasma GFAP is an early marker of amyloid-β but not tau pathology in Alzheimer’s disease. Brain. 2021;144:3505–16. https://doi.org/10.1093/brain/awab223. source

In the paper

“This hypothesis was pre-specified in the statistical analysis plan for the SHINE trial using an indirect measure of brain pathology by assessing baseline plasma p-tau217 levels and using the median value to generate “low” and “high” p-tau217 subgroups.”

In the paper

“Given that PrPC can bind APP and can alter levels of Aβ40 and Aβ42 [38]”

What the sources say

• Not checked yet

  Checked before — its earlier verdict is held back until a new check runs.

  cites [38] Foley AR, Roseman GP, Chan K, Smart A, Finn TS, Yang K, et al. Evidence for aggregation-independent, PrP C -mediated Aβ cellular internalization. Proc Natl Acad Sci. 2020;117:28625–31. https://doi.org/10.1073/pnas.2009238117. source

In the paper

“Findings from the SHINE p-tau217 subgroup analysis identified a patient population most likely to respond to zervimesine treatment, with favorable findings of a biological impact – decrease in GFAP and trends in decreases in NfL and Aβ42 – consistent with improved clinical outcomes [7, 8].”

What the sources say

• Not checked yetcites [7] Cognition Therapeutics. A Study to Evaluate the Safety and Efficacy of CT1812 in Subjects With Mild to Moderate Alzheimer’s Disease. ClinicalTrials.gov identifier: NCT035077. 2025 [cited 2026 Feb 4]. Accessed 4 Feb 2026. https://clinicaltrials.gov/study/NCT03507790.
• ?Not checked — no full text of the sourcecites [8] Vijverberg EGB, Scharre DW, Woodward M, Catalano S, Hamby ME, Grundman M, et al. Randomized, Double-Blind Study of Zervimesine in Mild to Moderate Alzheimer’s Disease. Journal of Alzheimer's Disease. 2026. https://doi.org/10.1177/13872877261437145. source

In the paper

“Elevated levels of GFAP in plasma and serum are associated with the degree and rate of neurodegeneration, Aβ positivity and cognitive decline in individuals with AD [27, 28].”

What the sources say

• Not checked yetcites [27] Pereira JB, Janelidze S, Smith R, Mattsson-Carlgren N, Palmqvist S, Teunissen CE, et al. Plasma GFAP is an early marker of amyloid-β but not tau pathology in Alzheimer’s disease. Brain. 2021;144:3505–16. https://doi.org/10.1093/brain/awab223. source
• Not checked yet

  Checked before — its earlier verdict is held back until a new check runs.

  cites [28] Phillips JM, Winfree RL, Seto M, Schneider JA, Bennett DA, Dumitrescu LC, et al. Pathologic and clinical correlates of region-specific brain GFAP in Alzheimer’s disease. Acta Neuropathol. Springer Science and Business Media Deutschland GmbH; 2024;148. https://doi.org/10.1007/s00401-024-02828-5. source

In the paper

“which provides support for a role of zervimesine in impacting amyloid biology, neuroinflammation and neurodegeneration in patients with mild to moderate AD that have lower than 1 pg/ml p-tau17 levels at baseline using the AlzPath Dx assay.”

In the paper

“Several biomarkers identified in either or both the mITT population and p-tau217 subgroup analyses, supports that zervimesine modulates i) Aβ toxicity (CTSS, B2M, LPL, APOA5, PSMA3, PSMB2, PSMAB5, LTF), (ii) neuronal trafficking (GRN, IGF2R, CLSTN1, RAB5B, SAA4), and (iii) synaptic plasticity (CANX, AP2M1, ACTB).”

In the paper

“Progranulin (GRN) regulates lysosomal biogenesis, inflammation and stress response [44]. GRN has been genetically linked to the neurodegeneration, with loss of function mutations or decreased levels have increased risk for neurodegenerative diseases including AD [44].”

What the sources say

• ?Not checked — no full text of the sourcecites [44] Rhinn H, Tatton N, McCaughey S, Kurnellas M, Rosenthal A, Elsevier Ltd. Progranulin as a therapeutic target in neurodegenerative diseases. Trends Pharmacol Sci. 2022. https://doi.org/10.1016/j.tips.2021.11.015. source

In the paper

“Limitations of this study include a limited sample size, and sample sizes were even smaller in subgroups.”

In the paper

“GPX3 is an enzyme involved in the antioxidant system and its expression is altered in AD [49].”

What the sources say

• Not checked yet

  Checked before — its earlier verdict is held back until a new check runs.

  cites [49] Panyard DJ, Reus LM, Ali M, Liu J, Deming YK, Lu Q, et al. Post-GWAS multiomic functional investigation of the TNIP1 locus in Alzheimer’s disease highlights a potential role for GPX3. Alzheimers Dement. 2024;20:5044–53. https://doi.org/10.1002/alz.13848. source

In the paper

“CLSTN plays a role in Aβ processing, axonal transport and formation of synapses [53].”

What the sources say

• Not checked yet

  Checked before — its earlier verdict is held back until a new check runs.

  cites [53] Vagnoni A, Perkinton MS, Gray EH, Francis PT, Noble W, Miller CCJ. Calsyntenin-1 mediates axonal transport of the amyloid precursor protein and regulates aβ production. Hum Mol Genet. 2012;21:2845–54. https://doi.org/10.1093/hmg/dds109. source

In the paper

“In sum, findings of the SHINE study support the continued clinical development of zervimesine for the treatment of neurodegenerative diseases.”

In the paper

“it is possible that while some biomarkers identified may reflect biological changes in the brain, it is also possible that some biomarkers may reflect changes occurring in the periphery, that may not be disease-relevant.”

In the paper

“Given this analysis was performed using an unadjusted p-value of p ≤ 0.05, it is likely that 5% of the proteins may be false-positives.”

In the paper

“Plasma proteomic analysis of SHINE participant further supports a role of zervimesine in impacting amyloid biology, synaptic biology, vesicle and lipid trafficking, neuroinflammation and neurodegeneration and highlights a newly discovered set of biomarkers to track in future trials, biomarkers which could reflect S2R pathway engagement or an impact on the disease progression.”

In the paper

“In a subgroup analysis of the SHINE trial, in alignment with the clinical effects and CSF biomarker changes, treatment with zervimesine for 6 months leads to positive changes (decreases) in canonical biomarkers of neuroinflammation, trends towards dampening of neurodegeneration, and an impact on amyloid biology; biological effects consistent with the greater clinical effect size seen in the low p-”

In the paper

“and is involved in Aβ deposition and like CTSS can modulate immune responses [50].”

What the sources say

• ?Not checked — no full text of the sourcecites [50] Savage MJ, Iqbal M, Loh T, Trusko SP, Scott R, Siman R. Cathepsin G: localization in human cerebral cortex and generation of amyloidogenic fragments from the β-amyloid precursor protein. Neuroscience. 1994;60:607–19. https://doi.org/10.1016/0306-4522(94)90490-1. source

In the paper

“All proteomics differential abundance and correlations analysis performed considered unadjusted p-values and did not control for multiple comparisons, thus permitting a greater likelihood of type 1 errors (false positives).”

In the paper

“Of note, LPL is a key enzyme involved in lipid metabolism. By interacting with lipoprotein receptors, such as the direct interactor with S2R (TMEM97), LDLR [46], LPL promotes the internalization of lipoproteins, contributing to lipid homeostasis and receptor-mediated signaling in neurons. In addition, LPL maintains the balance of membrane composition and fluidity, which is crucial for proper”

What the sources say

• Not checked yetcites [46] Riad A, Zeng C, Weng C-C, Winters H, Xu K, Makvandi M, et al. Sigma-2 receptor/TMEM97 and PGRMC-1 increase the rate of internalization of LDL by LDL receptor through the formation of a ternary complex. Sci Rep. 2018;8:16845. https://doi.org/10.1038/s41598-018-35430-3. source
• Not checked yet

  Checked before — its earlier verdict is held back until a new check runs.

  cites [47] Althaher AR, S. AL Shehadeh R. Lipoprotein Lipase and Neurological Health: Investigating its Impact on Brain Function and Alzheimer’s Disease. Open Biomark J. 2024;14. https://doi.org/10.2174/0118753183367402241216115139. source

In the paper

“While the p-tau217 subgroup analysis was prespecified, it was not the primary analysis.”