Clinical reporting following the quantification of cerebrospinal fluid biomarkers in Alzheimer's disease: An international overview

Abstract

Introduction

The current practice of quantifying cerebrospinal fluid (CSF) biomarkers as an aid in the diagnosis of Alzheimer's disease (AD) varies from center to center. For a same biochemical profile, interpretation and reporting of results may differ, which can lead to misunderstandings and raises questions about the commutability of tests.

Methods

We obtained a description of (pre‐)analytical protocols and sample reports from 40 centers worldwide. A consensus approach allowed us to propose harmonized comments corresponding to the different CSF biomarker profiles observed in patients.

Results

The (pre‐)analytical procedures were similar between centers. There was considerable heterogeneity in cutoff definitions and report comments. We therefore identified and selected by consensus the most accurate and informative comments regarding the interpretation of CSF biomarkers in the context of AD diagnosis.

Discussion

This is the first time that harmonized reports are proposed across worldwide specialized laboratories involved in the biochemical diagnosis of AD.

1. INTRODUCTION

Alzheimer's disease (AD) has gradually become one of the major global public health issues due to its prevalence, which increases with age and life expectancy, and the economic cost of caring for patients whose cognitive decline progressively leads to loss of functional autonomy. ¹

The diagnosis of AD is based on a multidisciplinary approach, involving, among other things, evaluation of the medical history together with clinical symptoms and signs, neuropsychological tests, and neuroimaging. The quantification of cerebrospinal fluid (CSF) core biomarkers (amyloid beta peptides [Aβ_1‐40 and Aβ_1‐42], total tau [t‐tau] and its phosphorylated form on threonine 181 [p‐tau(181)]) has progressively proven useful for the diagnosis of AD and its prodromal forms. ¹ CSF biomarkers are now included in international guidelines for the diagnosis of AD in research settings and clinical practice ² , ³ and the Alzheimer's Association appropriate use criteria for the use of lumbar puncture and CSF testing in the diagnosis of AD have been published. ⁴ Such biochemical diagnostics are currently implemented in many specialized centers around the world. Different methods of analysis have been developed over the last decade and each laboratory has implemented the one best suited to its own practice. Related to this diversity there are also variations in pre‐analytical and analytical conditions (such as sample tubes, storage, dilution of the biological sample, definition of cut‐off values) between centers. The subsequent interpretation of the analytical results may depend on the calculation of ratios (such as t‐tau/Aβ_1‐42 or Aβ_1‐42/Aβ_1‐40 ⁵ , ⁶ , ⁷ ), the use of scales (PLM, ⁸ Erlangen ⁹ scores), or on additional experiments (eg, dilution if t‐tau is above the limit for detection ¹⁰ ). Some laboratories mentioned the use of the A/T/N ¹¹ classification, which is, however, based on data additional to CSF biomarkers, and is used more in the research setting than in the clinic. Depending on the laboratory, the type of report sent back to physicians (prescribing or referring physicians, and general practitioners) varies greatly, which may raise questions about the commutability of the tests and cause misunderstanding. It is therefore very important to harmonize comments on the reporting of results, so that the conclusions are similar regardless of where the analysis is performed.

Our work provides an overview of the procedures used in 40 centers worldwide performing CSF analysis to support AD diagnosis. For each clinical laboratory participating in this work, we report the pre‐analytical (eg, type of sample tubes, storage conditions, potential non‐compliance with pre‐specified local protocols) and analytical (quantified biomarkers, methods) conditions. We also detail each partner's post‐analytical procedures, such as cutoff values, and use of ratios or scales for the interpretation of the results. Then, we list the clinical reports (for each biochemical profile) sent to the physicians responsible for the prescriptions. On the basis of the most frequently used reports and in‐depth exchange and discussion between the participants, we propose harmonized reports adapted to each biochemical CSF profile.

This work is an essential step towards a consensual harmonization of clinical reporting after CSF analysis in the context of AD diagnosis, as advocated by the Biofluid Based Biomarkers Professional Interest Area (BBB‐PIA) working group of the Alzheimer's Association. This harmonization is of great importance given the prevalence of AD and the increasing number of laboratories performing these diagnostic assays worldwide.

2. METHODS

2.1. Partners involved

Centers and laboratories specialized in AD diagnosis were contacted through the French Society of Clinical Biology (SFBC, https://www.sfbc‐asso.fr/), the International Society to Advance Alzheimer's Research and Treatment (ISTAART) BBB‐PIA, or the Society for Neurochemistry and Clinical CSF analysis (http://www.neurochem.info/). A total of 40 centers (17 French and 23 from 15 different countries, see authors' affiliations and Supplementary Figure S1 in the Supporting Information) provided different levels of information regarding their practice. For the interpretation of the surveys, each laboratory was anonymized. No personal or clinical patient data were used for this project, which therefore did not require ethical clearance. Data were collected between June and December 2020.

2.2. Inquiries

Clinical laboratories performing CSF testing were asked to provide information on the pre‐analytical and analytical protocols in their clinical practice (eg, type of tubes used, centrifugation or storage protocol, type of kit) and their criteria for non‐conformity with local protocols. Post‐analytical information was also requested, such as the cutoff values of the analytes, and the use of ratios or scales. All potential combinations of amyloid β, t‐tau and p‐tau(181) were then regrouped in eight different profiles labeled as follows: (1) “all normal,” in which amyloid (Aβ_1‐42 or ratio Aβ_1‐42/Aβ_1‐40), t‐tau, and p‐tau(181) are within reference range values; (2) “all pathological,” in which all biomarkers show pathological values; (3) “amyloid,” in which only amyloid values are pathological; (4) “t‐tau,” in which only t‐tau values are pathological; (5) “amyloid & t‐tau,” in which amyloid and t‐tau values are pathological; (6) “amyloid & p‐tau(181),” in which amyloid and p‐tau(181) values are pathological; (7) “t‐tau & p‐tau(181),” in which values for both tau biomarkers are pathological; and (8) “p‐tau(181),” in which only p‐tau(181) values are pathological. These profiles were also associated with their corresponding values of the scales and ratios (see Supplementary Table S3). We asked the participants to provide the different reporting texts they used according to the most common biochemical profiles they encountered.

HIGHLIGHTS

• Diversity of global practices in cerebrospinal fluid biomarker quantification

• Definition of a common terminology for biochemical biomarkers’ reporting

• Harmonization of clinical reporting for Alzheimer's disease diagnosis

RESEARCH IN CONTEXT

1. Systematic review: The quantification of cerebrospinal fluid (CSF) biomarkers (beta‐amyloid peptides, total tau, and hosphor‐tau proteins) is gradually being implemented in specialized clinical laboratories as an aid in the diagnosis of Alzheimer's disease (AD). However, for the same biochemical profile, the interpretation and reporting of results may differ from one center to another, which can lead to misunderstandings and raises questions about the commutability of tests.

2. Interpretation: Consensus approach to provide the most accurate and informative comments regarding the interpretation of CSF biomarkers will benefit AD care, prevention, and treatment strategies.

3. Future direction: This work is an essential first step towards harmonization of the clinical reporting of CSF biomarkers panel for the diagnosis of AD. We also consider this work useful to new CSF biomarkers and from the perspective of defining reporting comments for emerging blood biomarkers of AD.

2.3. Data processing and decision making

Clinical comments were compiled into a single table and returned to participants for selection. The percentage of similar reports and an initial vote to identify the two most relevant comments resulted in a short list of comments for each biochemical profile. This list was then commented on electronically and a series of video conferences was held to reach a consensus on the proposed comments for different profiles. The various comments were then translated into the different national languages of the participants. The laboratory of Prof. Sylvain Lehmann (France) in Montpellier was in charge of piloting the study and preparing, collecting, and analyzing all the survey responses.

2.4. Role of the funding/sponsoring source

This research did not receive any specific grant from funding agencies in the public, commercial, or not‐for‐profit sectors. The study supporters had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

3. RESULTS

3.1. Pre‐analytical and analytical conditions overview

Table 1 summarizes the current practice across participating centers and Supplementary Table S1 provides detailed data on the analytical and pre‐analytical procedures of each participating laboratory. The majority of centers (94%) used very similar polypropylene (PP) tubes for sample collection, and storage conditions at –20°C for short term and –80°C for long term. This is in line with standard operating procedures (SOPs) and the various experimental works in the field. ¹² , ¹³ , ¹⁴ , ¹⁵ The use of collection tubes different from those recommended by each laboratory was identified as a non‐conformity by 97% of centers. For secondary tubes that have a lower impact on pre‐analytical variability, ¹³ we observed a greater diversity of origin, but these were PP microtubes from 0.5  to 2 mL in most cases.

TABLE 1.

Pre‐analytical and analytical features used by the different centers (in %) for the measurement of AD biomarkers in CSF

	Participant centers (in %)
Use of similar collection tube	94.0
Measurement on frozen sample	97.0
Automated immunoassay analyzer	88.2
Cutoff based solely on manufacturer's information	16.6
Dilution of t‐tau if above limits	59.4
Systematic measurement of Ab1‐40	58.1
Use of the Ab1‐42/Ab1‐40 ratio	82.3
Use of other ratios than Ab1‐42/Ab1‐40	57.6
Use of scales	38.2
Turnover < = 1 week	34.5

For the analytical part, Fujirebio immunoassays (Lumipulse) were used in 76.5% of the centers for the quantification of the four analytes (Aβ_1‐42, Aβ_1‐40, t‐tau, and p‐tau[181]). Other laboratories measured the four analytes using Roche (Elecsys), Euroimmun (ELISA), Fujirebio (ELISA), IBL (ELISA), or MSD (V‐Plex), either alone or in combination. One center used a liquid chromatography mass spectrometry (LC‐MS) approach for Aβ_1‐42. Overall, 88.2% used automated immunoassay analyzers. We observed an important heterogeneity of the cutoffs selected for the different analytes, not only for Aβ_1‐42, but also for t‐tau and p‐tau(181).

The selection of cutoffs by the centers was based on information given solely by manufacturers in 16.6% of cases, on literature (4.2%), on other laboratories/colleagues (12.5%), on internal data (45.8%), or on a combination of these approaches (25%). Cutoffs for Aβ_1‐40 have generally not been defined, as this analyte is mainly used to calculate the Aβ_1‐42/Aβ_1‐40 ratio, which has its own cutoff. Aβ_1‐40 was systematically quantified in 58.1% of laboratories or added in case of discordance (ie, when Aβ_1‐42 was normal but tau biomarkers were abnormal, or vice versa) in 22.6% of laboratories (Table 1 and Supplementary Table S1). Seventeen percent of the laboratories mentioned the use of a “grey zone,” which corresponds to profiles where the values, often close to the cutoffs, correspond to a situation that remains undetermined.

Dilution of samples, when the upper limit of detection was exceeded, was only performed for t‐tau and only in 59.4% of the laboratories. The Aβ_1‐42/p‐tau(181) and t‐tau/p‐tau(181) ratios were computed in 39.4% and 30.3% of the laboratories, respectively. The validated PLM ⁸ and Erlangen ⁹ scales combining biomarkers are used by only 17.6% of the participating laboratories, a low percentage that is certainly an underestimate depending on the country, the Erlangen score being widely used in Germany for example. CSF biomarkers were also used by seven centers to establish the ATN ¹¹ research classification that also includes imaging information.

3.2. Clinical reports according to CSF biomarker profiles

Supplementary Table S3 (row 6) shows the mean and minimum/maximum frequency of the eight biochemical profiles, observed in consecutive series of samples (55 to 3000 samples) in 15 of the participating laboratories. Supplementary Table S2 lists the comments initially provided by participants for the biochemical profiles: the most frequent ones (“all normal,” “all pathological,” “amyloid,” “t‐tau”) and the less frequent (“amyloid & t‐tau or amyloid & p‐tau[181],” “t‐tau & p‐tau[181]” and “p‐tau[181]”). We observed very similar comments provided for “all normal,” “all pathological,” and “amyloid” profiles in 30%, 26%, and 36.6% of the centers, respectively. The comments for the other profiles showed more heterogeneity. In a second step, each laboratory was asked to select, from among all the reports previously listed, the two that best reflected their own current clinical practice. They were also given the opportunity to add additional comments based on available clinical information.

On the basis of these proposals, and after several rounds of exchanges (electronically and by videoconference) with all the partners, we generated harmonized comments for each profile, associated in some cases with additional information (Table 2 and Supplementary Table S3). We also translated these comments into 11 languages corresponding to the different countries of the participating laboratories (Supplementary Table S4). It should be noted that our survey found that less than 5% of centers use plots/graphs in addition to numerical values. In our group discussion, the consensus was not to use additional plots/graphs that might interfere or make the commentary be returned with the numeric values less clear.

TABLE 2.

Summary of consensus comments for interpretation of biochemical profiles of AD biomarkers in CSF

amyloid	t‐tau	p‐tau(181)	Consensus comments
N	N	N	Biochemical profile not consistent with Alzheimer's disease.
P	P	P	Biochemical profile consistent with Alzheimer's disease.
P	N	N	Biochemical profile consistent with an amyloidopathy.
N	P	N	Biochemical profile not consistent with Alzheimer's disease; may be consistent with other neurodegenerative disease and/or neuronal damage. (If t‐tau is close to/above upper limit of detection with a high t‐tau/p‐tau[181] ratio, the profile may indicate Creutzfeldt‐Jakob disease)
P	P	N	Atypical biochemical profile; may be consistent with Alzheimer's disease.
P	N	P	Atypical biochemical profile; consistent with Alzheimer's disease.
N	P	P	Atypical biochemical profile; not consistent with Alzheimer's disease.
N	N	P	Atypical biochemical profile; not consistent with Alzheimer's disease.

Note to be added to all comments: This biochemical profile must be interpreted in its clinical context and in conjunction with a physician.

Abbreviations: N, normal; P, pathological; p‐tau(181), tau phosphorylated at threonine 181.; t‐tau, total tau.

4. DISCUSSION

In this work, we collected information from 40 centers located in 15 different countries (Supplementary Figure S1) that measure CSF amyloid and tau biomarkers in the clinical setting. Moving from the measurement of biomarkers for clinical research, which is mainly performed on retrospective cohorts, to routine clinical measurement is a real challenge. ⁴ Even with established SOPs ¹² and international guidelines for the handling of CSF, ¹⁵ pre‐analytical and analytical deviations may be present in real world settings, affecting the result, and sometimes going unreported. To provide high‐quality results, it is important to ensure that tests achieve sufficient levels of performance to make a meaningful contribution to diagnosis and ultimately to patient care. Finally, a critical step in the medical use of CSF testing is how the results are communicated to clinicians and, when appropriate, to patients themselves. The harmonization of reporting is in this regard essential to avoid misunderstandings while comparing results between centers and to provide accurate, informative, and harmonized information that will have an impact on prevention, care and treatment strategies. ¹ , ¹⁶

We have focused our work only on CSF biomarkers currently used in clinical practice, which are part of official guidelines ² , ³ and measured using IVD (In Vitro Diagnostics) assays. The main context of use (COU) of these tests is the diagnosis of AD. It is particularly important to keep this COU in mind because it influences the choice of pathological cutoffs (which may vary depending on the clinical question). This also explains why comments in our proposed consensus clinical reports focus on the diagnosis of AD rather than dementia or neurodegenerative diseases in general.

The methodology to be used in the development of clinical practice guidelines is well established. ¹⁷ The first phase generally involves conducting a systematic review and synthesis of the literature. Different works focus on the interpretation of biochemical profiles, ⁸ , ⁹ , ¹⁵ but we could not find previous publications dealing with the clinical reporting of CSF results for AD diagnosis. This observation is not surprising since these tests have only recently been widely used in clinical routine. Therefore, we employed a “consensus” methodological approach directly based on the agreement among experts through iterative ratings with feedback.

Our review of the pre‐analytical protocols of the different centers firstly shows that study over the last ten years of the confounding factors related to this phase ¹² and the definition and harmonization of SOPs ¹⁴ , ¹⁵ for CSF AD biomarkers have been successfully implemented. Indeed, most of the centers use very similar PP tubes (only differing in their size and shape), which have previously shown low adsorption with amyloid β peptides. ¹⁸ However, the volume of stored samples differs among centers, and this may still affect amyloid β quantification_. Pre‐analytical procedures, including centrifugation, secondary aliquots in microtubes and freezing at –20°C for a few days or at –80°C were also very similar, with the exception of the secondary tubes. Rather than using fresh samples, use of frozen secondary samples may represent a more easy‐to‐use protocol, ¹⁹ which probably adapts to current numbers of tests requested.

Regarding the analytical part, in addition to the three core biomarkers Aβ_1‐42, t‐tau and p‐tau(181), we observed that more than 83% of the laboratories also measure Aβ_1‐40 and thereafter compute the Aβ_1‐42/Aβ_1‐40 ratio. The rationale is likely related to the fact that the ratio improves diagnostic performance ⁶ , ²⁰ and reduces biases linked to collection tube, volume of sample, and storage. In terms of detection method, it is also unsurprising that more than 88% of the laboratories are using automated chemiluminescence immunoassays that have a reduced analytical variability ²¹ , ²² and offer a more flexible test throughput. Despite this, one striking finding was the wide dispersion of the cutoff values used by the different laboratories, even for Aβ_1‐42, which will benefit from a metrological harmonization as a result of the development of both mass spectrometry reference methods and certified reference materials. ²³ This diversity is linked on the one hand to the different assays/protocols used, but also to the fact that most centers adapt the values proposed by the test providers using results from their own cohort (16.6% vs 45.8%, Supplementary Table S1). In addition, there is a highly variable distribution of the percentages of CSF profiles between centers (Supplementary Table S3). This is consistent with the prevalence of AD, which varies widely between centers, from a low of 25% to a high of 53%. This indirectly leads to different optimal threshold definitions; therefore, the positive and negative predictive values of the tests also vary considerably.

Regarding the interpretation and reporting of the CSF biomarker results, the inquiries from the different centers (Supplementary Table S2) showed first of all that the different comments involve “Alzheimer's disease,” which is in line with the COU of the CSF biomarkers. We also note the term “biochemical profile.” It should be kept in mind that if CSF biomarkers do mainly reflect the amyloid and tau pathologies of AD, they may show pathologic change without clinical symptoms and this may occur sometimes more than a decade before clinical manifestation of AD. ¹ Nevertheless, in accordance with the National Institute on Aging and Alzheimer's Association Research Framework, AD can be defined as a biological construct, ²⁴ which led us in the comments to refer to AD per se, rather than with its pathological changes.

There are several points to consider in the derivation of the consensus comments from the initial reporting of the biomarker profiles. First, when all biomarkers are pathological, most initial comments from the centers indicated that the profile is “suggestive” or “consistent” with AD, showing some caution in asserting the diagnosis. The performance of CSF biomarkers for AD is very good, ¹ but there may still be room for improvement using other p‐tau biomarkers such as p‐tau(217) ²⁵ or p‐tau(231). ²⁶ In addition, the physician's diagnosis of AD remains multidisciplinary, combining clinical history, symptomatology, neuropsychological testing, imaging, and biology, ¹ and accordingly CSF biomarkers alone are not diagnostic of the disease. The consensus comment in this pathological situation was therefore “Biochemical profile consistent with Alzheimer's disease,” and we have chosen to mention in all cases that “This biochemical profile must be interpreted in its clinical context and in conjunction with a physician” (Table 2).

Second, when all biomarkers are normal, one may clearly indicate that the profile is “not consistent” with AD. This makes sense because AD is intrinsically associated with amyloid and tau pathology and retrospective studies show that normal CSF profiles virtually rule out AD. ²⁷

Third, when only amyloid biomarkers are pathological there is an obvious consensus to indicate that the biochemical profile is consistent with an “amyloidopathy” (Table 2 and Supplementary Table S3). However, this pathological situation may also be considered indicative of the presence of AD in the disease continuum. ²⁴

Fourth, when Aβ_1‐40 is assessed, there is usually no defined cutoff value for this analyte alone. Variations of Aβ_1‐40 in frontotemporal dementia ²⁸ or cerebral amyloid angiopathy ²⁹ have been described but they are minimal and therefore only useful when associated with clinical information in favor of these diagnoses. However, assessment of Aβ_1‐40 in combination with Aβ_1‐42 (Aβ_1‐42/Aβ_1‐40 ratio) has proven to be highly informative for AD diagnosis in clinical routine ⁶ , ⁷ , ²⁰ and this ratio is currently used in 88.2% of the centers.

Fifth, the profile showing an isolated increase in t‐tau is present in 2.7% to 15.7% of cases, depending on the center. Initial comments indicated that this profile was not consistent with AD but rather with other neurodegeneration and/or neuronal damage (such as cerebrovascular disease or—if strongly increased—Creutzfeldt‐Jakob disease [CJD]). For the latter possibility, it should be noted that a very high level of t‐tau (close to or above the high detection limit of the assays) associated with a high t‐tau/p‐tau(181) ratio is strongly in favor of this diagnosis, if other causes of major neuronal injury, for example, stroke and encephalitis, are excluded. ¹⁰ The consensus comment for this profile could therefore refer to this particular situation if clinical information or the diagnostic hypothesis suggests the presence of CJD (see Table 2).

Sixth, to cover all situations, we also defined consensus comments for the less frequent profiles (< 5% of cases). Thus, when amyloid biomarkers are pathological and associated with an increase in p‐tau(181) but not t‐tau, one may consider that this atypical profile “is consistent” with AD. When both amyloid and t‐tau biomarkers are pathological (with p‐tau[181] normal), the consensus is to consider that this atypical profile “may be consistent” with AD. On the other hand, profiles with normal amyloid but abnormal t‐tau and/or p‐tau(181) are considered as “not consistent” with AD (Table 2). This is an important consensus decision emphasizing that AD can exist only in the presence of amyloid pathology. This situation is also reminiscent of the “suspected non‐Alzheimer's disease pathophysiology (SNAP).” ³⁰ It should be noted that in these cases, special attention should be paid to the search for amyloidopathy using in particular the Aβ_1‐42/Aβ_1‐40 ratio.

A limitation of this work is that we are not exhaustive in consulting laboratories using CSF biomarkers. Therefore, some results such as the percentage of use of different ratios or scales may be biased. In addition, CSF biomarkers are not always used for the diagnosis of AD and, therefore, many centers and countries are missing from this international study.

In conclusion, this work is an essential first step towards harmonization of the clinical reporting of the CSF biomarkers panel for the diagnosis of AD. The proposed framework is adaptable and applicable to new CSF biomarkers passing regulatory criteria and prospective validations for clinical application. We also consider this work useful from the perspective of defining reporting comments for emerging blood biomarkers of AD.

CONFLICTS OF INTERESTS

Charlotte Teunissen's institution received consulting fees from Roche and received Kits for blood‐based biomarkers from Quanterix and ADx Neurosciences. She leads the Global Biomarker Standardisation Consortium, Blood based Biomarker PIA, CSF Society.

Kaj Blennow has served as a consultant (and received fees), at advisory boards, or at data monitoring committees for Abcam, Axon, Biogen, JOMDD/Shimadzu, Julius Clinical, Lilly, MagQu, Novartis, Roche Diagnostics, and Siemens Healthineers, and is a co‐founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, all outside the present work.

Daniel Alcoléa participated in advisory boards from Fujirebio‐Europe and Roche Diagnostics and received speaker honoraria from Fujirebio‐Europe, Roche Diagnostics, Nutricia, Krka Farmacéutica S.L., and Esteve Pharmaceuticals S.A. Dr Alcolea declares a filed patent application (WO2019175379 A1 Markers of synaptopathy in neurodegenerative disease).

Elodie Bouaziz‐Amar received fees from Roche Pharma/Roche Diagnostics France (Board Alzheimer 23 octobre 2019).

Giovanni Bellomo received fees form Innuvatech srl.

Mercè Boada received consulting fees form Roche, Grifols, Araclon, Biogen, Merck, Lilly, Nutricia, Cotexyme, Renew Research.

Olivier Bousigues received consulting fees from KeyQuest Health Ltd, Azure Knowledge Corporation, and benefited from manuscript writing fees from Spectra diagnostic and Revue francophone des laboratoires.

Miles D. Chapman is a member of UK NEQAS SAG Committee.

Sebastiaan Engelborghs received consulting fees from Biogen, Danone, Eisai, icometrix, Pfizer, Novartis, Nutricia, Roche, and received research funding from ADx Neurosciences and Janssen Pharmaceutica. He received fees as an advisory Boards from Biogen, Danone, Eisai, icometrix, Pfizer, Novartis, Nutricia, Roche. He serves as Secretary‐General of Belgian Neurological Society, Vice‐President of Belgian Dementia Council, ExCo member of European Alzheimer Disease Consortium.

Melanie Hart is member of the Royal College of Pathologists Special Advisory Board for Immunology—clinical scientist representative.

Takeshi Ikeuchi received a honoraria for lectures from the following pharmaceutical companies: Eisai, Takeda, Novartis, Roche, Daichi‐Sankyo, Ono, Chugai.

Andreas Jeromin, has served as an advisor to Quanterix Corp and holds stock options.

Michael Khalil has received speaker honoraria from Bayer, Novartis, Merck, Biogen Idec, and Teva Pharmaceutical Industries Ltd. He serves on scientific advisory boards for Biogen Idec, Merck Serono, Roche, Novartis, Bristol‐Myers Squibb, and Gilead.

Agnieszka Kulczynska‐Przybik received a consulting and/or lecture honoraria from Roche company.

Piotr Lewczuk received consulting and/or lecture honoraria from IBL International, Fujirebio Europe, AJ Roboscreen, and Roche.

Alberto Lleó has served as a consultant or at advisory boards and received fees from Roche, Biogen, Nutricia, Zambon, MassConsultoria, SEMNIM. In addition, Dr. Lleó has a patent WO2019175379 A1 Markers of synaptopathy in neurodegenerative disease issued. He received royalties from a European Patent licensed.

Barbara Mroczko has received consulting and/or lecture honoraria from Abbott, Wiener, Roche, Cormay and Biameditek.

Léonor Nogueira is a co‐author of a patent concerning the detection of antibodies toward citrullinated protein for the diagnosis of Rheumatoid Arthritis.

Adelina Orellana has submitted MMP‐10 CSF biomarker in the European Patent Office (Patent number 21382305 7, priority date April 9th 2021)

Markus Otto received fees as consultant or advisory board from Axon, Roche, Axon, Biogen. He is a member of the CSF Society.

Jean‐Baptiste Oudart received fees as an expert advisory board for Osimertinib (Astrazeneca).

Claire Paquet received fees as a consultant from Pharmaceutical laboratories Biogen and Roche concerning anti‐AD treatments, as a lecturer for Roche, as advisory board member from Biogen and Roche.

Armand Perret‐Liaudet received consulting fees from QUADRANT consulting (Paris), support from Fujirebio company for attending meetings.

Katell Peoc'h received consulting fees from Vifor France.

Albert Puig‐Pijoan received support from Nutricia (payment of inscription fees to attend meetings).

Jonathan M. Schott received royalties from Oxford University Press, Henry Stewart Talks, Alzheimer's Research and Therapy, consulting fees from Eli Lilly, Biogen, GE, Roche, Alzheimer's Research UK, UK Dementia Research Institute Ltd and as advisory board from Axon Neuroscience SE. He serves as Chair ISTAART Advisory Council.

Leslie M. Shaw received fees from Biogen teaching program.

Marc Suárez‐Calvet received fees as a consultant and at advisory boards for Roche Diagnostics International Ltd, as a lecturer in symposia sponsored by Roche Diagnostics, S.L.U and Roche Farma, S.A. and as member of advisory boards for Roche Diagnostics International Ltd.

Magda Tsolaki is member of Alzheimer Hellas Greek Alzheimer's Disease Federation.

Hayrettin Tumani received fees or honoraria for lectures and advisory boards from Alexion, Bayer, Biogen, Celgene, Genzyme‐Sanofi, Merck, Novartis, Roche, Teva. He is member of DGLN (German Society of CSF Diagnostics and Clinical Neurochemistry).

Chinedu T Udeh‐Momoh received support (travel grant) from Alzheimer's Association and British society for neuroendocrinology. He is trustee for British society for neuroendocrinology.

Lisa Vermunt received support (travel grant) from the Alzheimer Association.

Jens Wiltfang received fees as a consultant from Roche Pharma GmbH, Boehringer Ingelheim, Immungenetics AG, MSD Sharp & Dohm, Abbott AG, Biogen, as a lecturer from Pfizer, Janssen‐Cilag GmbH, med Update GmbH, AGNP e. V., CSF Society, Roche Pharma, Vitos Klinikum Kurhessen, AWO Psychiatrie Akademie, Actelion Pharmaceuticals, Amgen, Fachverband Rheumatologische Fachass., Neuroakademie e. V., Guangzhou Glorylen Medical Technology Co. (China), Beijing Yibai Science and Technology Ltd. He has Patents (no.: EP1270592B1; patent no.: US 6,849,416 B2; patent no.: EP2095128B1; EP3105589 A1). He is member of AGNP.

Henrik Zetterberg is a co‐founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, which holds licenses (unrelated to the current submission). He served at scientific advisory boards for Eisai, Denali, Roche Diagnostics, Wave, Samumed, Siemens Healthineers, Pinteon Therapeutics, Nervgen, AZTherapies, and CogRx. He has given lectures in symposia sponsored by Cellectricon, Fujirebio, Alzecure, and Biogen. He is chair of the Alzheimer's Association Global Biomarker Standardization Consortium and the Alzheimer's Association Biofluid‐Based Biomarker Professional Interest Area.

Sylvain Lehmann received consulting fees from Fujirebio, Euroimmun, and Roche diagnostics.

All other authors declare that they have no competing interests and nothing to disclose.

AUTHOR CONTRIBUTIONS

S. Lehmann and C. Delaby had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: S. Lehmann, C. Delaby, C. E. Teunissen, H. Zetterberg.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: S. Lehmann and C. Delaby.

Critical revision of the manuscript for important intellectual content: All authors.

Supporting information

SUPPORTING INFORMATION

SUPPORTING INFORMATION

Delaby C, Teunissen CE, Blennow K, et al. Clinical reporting following the quantification of cerebrospinal fluid biomarkers in Alzheimer's disease: An international overview. Alzheimer's Dement. 2022;18:1868–1879. 10.1002/alz.12545