Pre‐analytical guidelines for blood and CSF Biomarkers 2025: Recommendations from the NACC ADRC Biofluid Biomarker Best Practices Workgroup

Abstract

The Biofluid Biomarkers Best Practices Workgroup of the National Alzheimer's Coordinating Center–Alzheimer's Disease Research Center (ADRC) Biomarker Core Steering Committee was convened to update pre‐analytical handling guidelines for biofluid biomarkers, focusing on cerebrospinal fluid (CSF) and blood. We reviewed current literature pertinent to best practices for biomarker studies and surveyed the ADRCs for biomarker analytes, platforms, and protocols used at each center. Across 37 ADRCs, 16 CSF and 28 plasma/serum analytes were reported to be studied at multiple centers. The pre‐analytical handling steps and concerns related to each, as supported by empirical studies and expert opinion, were integrated to generate a revised guideline document. The guideline aimed to standardize steps in biospecimen and biomarker analyte collection, storage, and pre‐analytical handling across the ADRCs. The 2025 ADRC guidelines represent the current working knowledge on biomarker best practices, providing guidance and harmonized protocols, and promoting robust analysis and reporting of composite data.

Highlights

• The largest source of laboratory variance in analyte measurement is from delays in processing; thus, there is a need for standardized processing and centrifugation protocols.

• Tube types matter, as does dead volume in stored biospecimen aliquots.

• Standardized ATN analyte assays, including pT217‐tau, are stable at 4°C for short periods, cell free.

• Most proteins withstand two or more freeze‐thaws before activity‐integrity loss.

• Operational bench time should be standardized to avoid temporal variance between runs.

1. INTRODUCTION

Research into Alzheimer's disease (AD) and age‐related neurodegenerative disorders is progressing rapidly, aided by large databases of clinical information and associated biorepositories. The use of biomarkers detectable in blood for diagnostic testing and disease monitoring has grown considerably in recent years, both in the number of biomarkers in broad use in AD research and in the level of validation data available. A major goal across research consortia is to standardize methods of collection, storage, and other pre‐analytical factors that can affect data sharing and multicenter initiatives where data are combined in large datasets. The robust and effective use of this information requires standardized procedures. The National Alzheimer's Coordinating Center (NACC) and Alzheimer's Disease Research Center (ADRC) Biofluid Biomarkers Best Practices Workgroup was established by the ADRC Biomarker Core Steering Committee to revise and update the National Institute on Aging–ADRC guidelines last released in 2014. ¹ The collective NACC resource and the 37 US‐based ADRCs facilitate standardized data and specimen storage for global research use by the scientific community. This is one of the world's largest curated National Institutes of Health (NIH)‐sponsored collections, with processes and procedures that need periodic expert review to maintain best practices. The current holdings include specimens from more than 32,000 subjects with primary etiologic diagnosis of mild cognitive impairment (MCI) or dementia, including AD (21,395), Lewy body diseases (2085), vascular brain injury or stroke (1425), frontotemporal lobe dementias (3175), traumatic brain injury (148), and other brain disorders (4092; Uniform Data Set summary, NACC website).

The workgroup was convened in 2023 and worked to assess and then present the needed updates for revised guidelines, including obtaining and incorporating feedback from ADRC members and other international experts. The two main goals in the Workgroup Charter were as follows:

1. Develop updated best practice guidelines for collecting, preparing, and storing blood products (plasma or serum) and cerebrospinal fluid (CSF) samples specific to AD biomarkers.

2. Create a standardized guideline for assay procedures (e.g., sample handling), alongside methods for quality control.

2. METHODOLOGY

The activities of the workgroup included:

• Surveying 37 ADRCs to determine biomarkers, assays, and platforms in use.

• Reviewing current technical literature regarding pre‐analytical factors (collection, stability, storage).

• Soliciting feedback from ADRCs and additional key opinion leaders (KOLs).

• Updating technical concerns where data were available.

The work accomplished includes:

• Revisions of biomarkers and methods currently in use at ADRCs, and documenting specific considerations for each.

• Providing standardized ADRC working protocols and recommendations for CSF and blood.

• Collaborating with the ADRC Data Variables Workgroup to update the data elements collected from submitting ADRCs, revising the data variable structure for study standardization and curated use.

• Draft guidelines were sent for independent expert review by key opinion leaders.

Final guideline criteria:

Biomarker analyte is routinely studied by at least two ADRCs: This approach allowed the inclusion of biomarkers gaining use in the AD research community and therefore meeting a minimum need for standardization. This was also a forward‐looking approach.

Biomarker analyte was validated in large cohort studies: The validation in one or more large cohort studies (i.e., >200 well‐adjudicated subjects, appropriate statistical design to assess differences). This supported the need for standardization. Repeated, statistically supported findings were found in the peer‐reviewed literature, demonstrating that the biomarker analyte was altered in AD, related to premorbid disease states or related diseases studied by the ADRC network.

3. GUIDELINE OVERVIEW

The following guidelines were derived for use across the ADRC network to ensure comparability across biofluid sampling procedures, such that studies are standardized for large‐scale collaborative studies and consistency of findings. These best practices should be followed in the creation and conduct of biorepositories in the global AD research community.

Part IV. Protocol 1: Sampling and use of cerebrospinal fluid.

Part V. Protocol 2: Sampling and use of blood‐derived fluids.

Part VI. Collection and pre‐analytical processing summary

Part VII. Additional forward‐looking considerations.

Part VIII. Summary and conclusions.

Protocols

4. PROTOCOL 1: SAMPLING AND USE OF CSF

4.1. Results from the 2024 survey of ADRC members for bioassays currently performed in CSF

For CSF, more than 16 biomarkers in widespread use were considered (e.g., amyloid beta [Aβ]1‐42/40, neurogranin [NRGN], soluble triggering receptor expressed on myeloid cells 2 [sTREM2]). Table 1 lists the biomarkers reported to be studied at the surveyed ADRCs. Highly multiplexed technologies generally used for discovery were excluded. The individual biomarkers from some smaller multiplex technologies captured in the survey were, however, included in Table 1 (e.g., Quanterix Simoa Neuro 4‐Plex).

TABLE 1.

Current biomarkers assayed in cerebrospinal fluid (CSF) and blood biofluids at Alzheimer's Disease Research Centers and considered for updated guidelines.

Biomarkers currently studied at ADRCs	Biofluids used
Amyloid beta (Aβ38, Aβ40, Aβ42)	CSF, plasma, serum
Brain‐derived neurotrophic factor (BDNF)	Plasma, serum
Cytokine and chemokine panels	CSF, plasma, serum
Glial fibrillary acidic protein (GFAP)	CSF, plasma, serum
Intercellular adhesion molecule 1 (ICAM‐1, CD54)	Plasma, serum
Neurofilament light chain (NEFL, NfL)	CSF, plasma, serum
Neurogranin (NRGN)	CSF, plasma, serum
Neuron‐specific enolase (NSE, ENO2)	Plasma, serum
Synaptosome associated protein‐25 (SNAP‐25)	Plasma, serum
Synuclein α (SNCA)	Plasma, serum
Synuclein α, phosphorylated (pSNCA, p‐Ser129‐SNCA)	Plasma, serum
Synuclein β (SNCB)	CSF, plasma, serum
TAR DNA‐binding protein 43 (TDP‐43)	CSF, plasma, serum
Tau, brain‐derived isoform‐specific (BD‐tau) or total tau (tTau)	CSF, plasma, serum
Tau, microtubule binding region fragment tau (MTBR‐tau)	CSF, plasma, serum
Tau, phosphorylated (pTau) (e.g., p181‐tau, p205‐tau, p231‐tau, pTau‐217)	CSF, plasma, serum
Soluble triggering receptor expressed on myeloid cells 1 (sTREM1)	Plasma, serum
Soluble triggering receptor expressed on myeloid cells 2 (sTREM 2)	CSF
Vascular cell adhesion molecule (VCAM‐1)	Plasma, serum
Vascular endothelial growth factor (VEGF)	Plasma, serum
Visinin‐like protein 1 (VILIP‐1)	Plasma, serum
Chitinase‐like 3 (YKL40)	Plasma, serum

4.2. Recommendations for acquisition of CSF biospecimens

1. Fasting. Fasting is not necessary for CSF antibody collection ² ; centers may determine whether fasting is recommended for other novel biomarker assessments.

2. CSF should preferably be collected at a consistent time in the morning (e.g., 8:00 to 11:00 a.m.). ³ , ⁴ This will reduce variability due to diurnal variation, affecting some biomarker levels.

3. Consider taking matching plasma and/or serum samples for simultaneous measurement of CSF and blood biomarkers and understanding temporal relationships.

4. Use of an atraumatic spinal needle (e.g., Sprotte 25‐, 24‐, or 22‐gauge needle) is recommended for the lumbar puncture (LP) to minimize risk of post‐LP headache (<5%). ⁵ Having spinal needles of various lengths (e.g., 90, 103, 120 mm) available at the time of the procedure will increase the success of CSF acquisition in individuals with varying body habitus.

5. Gravity drip method is preferred for fresh CSF analyses and measurement of Aβ and tau. ² , ⁵ CSF may be withdrawn under negative pressure with sterile polypropylene syringes. Up to 30 mL CSF may be withdrawn without increased risk of adverse events (e.g., post‐LP headache is common in elderly subjects). ⁴ , ⁵ The plastic extension tubing/siphon provided in LP kits should not be used.

4.3. Preparation and storage of CSF biospecimens

A number of studies have addressed the effects of collection techniques and potential sources of variation in the study of CSF for the detection of essential neurodegenerative disease biomarkers. Tables 2 and 3 list evidence reported for individual biomarkers for factors that could affect detected analyte levels and represent avoidable sources of variance, diminishing intra‐assay precision and inter‐assay comparability. ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²²

TABLE 2.

Collection and processing issues reported for standardized CSF studies.

Biomarker	Issue of concern	Specific consideration	References
Aβ42	Tube material and fill volume	Adsorption of Aβ to recipient material. Re: fill volume, some studies find no impact (summary table in Hansson 8 ).	5 , 6 , 7 , 8 , 9 , 10
Aβ42	Tube surface absorption during transportation of fresh samples	Aβ42 levels were lower with horizontal versus upright transportation.	8 , 9 , 10
Aβ42	Tube surface absorption during transportation/storage conditions	Aβ42 levels were higher with maximal tube filling.	8 , 9 , 10
Aβ42	Tube surface absorption	Aβ42 levels varied by tube type; using a low‐bind, polypropylene Sarstedt 2.5 mL false‐bottom tube (FBT) reduced variation.	8 , 9 , 10
Aβ42	Tube surface absorption during mixing	For fresh CSF, Aβ42 levels were higher with no mixing versus roller/inversion mixing. Roller mixing after thawing is recommended for frozen samples.	8 , 9 , 10
Aβ42	Hemolysis and blood‐borne proteases	In frozen CSF samples, 0.25% blood contamination decreased Aβ1‐42 concentrations.	6 , 8
Analytes, various	Time of day sampling	Diurnal variation shown with 6.2% to 12.2% variation for several (e.g., synaptic biomarkers NRGN, SNCB, SNCG, but not GFAP, NfL, or pTau/Tau)	11 , 12 , 13
Aβ, pTau, tTau,	Time and temperature collection – permanent storage	Optimal storage is at −80°C	14 , 15 , 16

Abbreviations: Aβ, amyloid beta; CSF, cerebrospinal fluid; GFAP, glial fibrillary acidic protein; NfL, neurofilament light chain; NRGN, neurogranin; pTau, phosphorylated tau; RT, room temperature; SNCB, synuclein β; sTREM2, soluble triggering receptor expressed on myeloid cells 2; tTau, total tau.

Based on available information in published guidelines and unpublished working knowledge from experienced research centers.

TABLE 3.

Storage and stability issues affecting cerebrospinal fluid biomarkers.

Biomarker	Issue of concern	Specific considerations	References
Aβ40	Freeze‐thaw cycles	Aβ40 significantly decreased after four freeze‐thaw cycles.	8 , 14 , 15
Aβ42	Freeze‐thaw cycles	Aβ42 significantly decreased after two freeze‐thaw cycles.	16 , 17
Aβ42/Aβ40 ratio	Freeze‐thaw cycles	Ratios stable over four freeze‐thaw cycles.	17 , 18
GFAP	Freeze‐thaw cycles	Average decline of 188.21 pg/mL beyond first freeze‐thaw cycle.	17 , 18
NfL	Stability	Not stable in neat CSF for 1 day at RT. Store at 4°C or freeze in aliquots immediately.	19
pTau, tTau	Stability, freeze‐thaw	Tau was stable long term at −80°C, and for six freeze‐thaw cycles.	15 , 20 , 21
sTREM2	Freeze‐thaw	sTREM2 takes five freeze‐thaw cycles to be impacted.	22

Abbreviations: Aβ, amyloid beta; CSF, cerebrospinal fluid; GFAP, glial fibrillary acidic protein; NfL, neurofilament light chain; pTau, phosphorylated tau; RT, room temperature; sTREM2, soluble triggering receptor expressed on myeloid cells 2; tTau, total tau.

Considering the multitude of research initiatives and biomarkers the ADRC will support, the most stringent storage criteria are recommended for all sites.

4.3.1. For fresh and frozen CSF collection for Aβ and tau measures

1. All tubes handled by a clinician during the LP must be sterile.

2. Using a gravity drip, discard the first 1 to 2 mL CSF or until clear to limit potential blood contamination (Table 2). ⁵ , ⁶

3. Continue to use the gravity drip method to collect 2.5 mL directly into a 3‐mL polypropylene low protein binding tube (e.g., Sarstedt false‐bottom tube). The number of tubes drawn in this manner will be determined by each center. While no specific vendor is endorsed, tubes must be validated (by a published report or center experiments) to ensure they are low‐binding for Aβ and other analytes. ⁶ , ⁷ , ⁸ , ⁹

4. For fresh CSF samples, there should be no further handling (no centrifugation, freezing, mixing/inverting, or tube transfers).

5. Transport and store fresh CSF samples at 2°C to 8°C up to 14 days (if not feasible, transport and store at room temperature [RT; 20°C to 25°C] for up to 2 days). Samples should be kept upright to avoid interaction with tube caps, which are not polypropylene. ⁷ , ⁸

6. Once fresh CSF arrives at the testing site, there should be no further handling. Measure Aβ40, Aβ42, tTau, and pTau immediately, when possible.

7. If some CSF samples collected for Aβ and tau analyses (using methods above) are going to be stored in a biorepository, these samples may be frozen at −80°C for long‐term storage. CSF aliquot storage tubes must be validated (by a published report or center experiments) for freezing and storing at −80°C. ⁵ , ⁶ , ¹⁰

4.3.2. For additional CSF analytes and biobanking

1. After discarding the first 1 to 2 mL and collecting Aβ/tau samples via the gravity drip method, proceed with either the continued drip technique or the syringe suction technique as determined by center investigators. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵

2. Collect a total of up to 30 mL CSF in 5 or 10 mL polypropylene tubes and/or syringes.

3. A volume of 1 mL of CSF (or the minimum amount needed) should be sent to the local clinical laboratory for analysis of cell count to assess sample quality for blood contamination. Other local laboratory measurements, such as CSF protein or glucose, should be determined by center investigators based on study needs. The local laboratory sample may be placed in the plastic (polystyrene) tubes that are included in some commercial LP kits. The polystyrene tubes can ONLY be used for the sample sent to the clinical laboratory.

4. If CSF was collected in syringes, transfer the remaining CSF into one 30‐mL polypropylene tube, and do not mix.

5. Centrifuge for 2000 × g for 10 min in a temperature‐controlled room (20°C to 25°C).

6. Remove CSF with a pipette and transfer to a second 30‐mL polypropylene tube, being careful not to disturb the pellet.

7. Invert briefly to mix the 30‐mL tube.

8. Divide the CSF into 0.5‐mL aliquots in 1‐mL polypropylene tubes (or fill tubes to at least 75% of tube capacity to reduce variability caused by CSF volume to tube surface area differences).

9. Use of screw‐cap tubes with rubber O‐rings are recommended to reduce evaporation.

10. Freeze at −80°C (no need to flash freeze on dry ice).

4.4. Sample management

4.4.1. Sample annotation and recordation

• Uniform, non‐redundant, Health Insurance Portability and Accountability Act of 1996 (HIPAA)‐compliant annotation of samples is recommended. Use of sample management software is recommended.

• Document the exact volume of fluid obtained at each CSF collection, as this can vary.

4.4.2. Documentation

Appropriate and complete documentation surrounding biospecimen collection, processing, and storage is essential and relevant to the quality of research data obtained. Note date and details of any changes to or deviations from the pre‐analytic procedure. ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸

4.4.3. Freeze‐thaw recommendations

The existing literature addressing the effect of storage and freeze‐thaw cycles on specific biomarker proteins in stored CSF is reviewed in Table 3. Thawing and refreezing of samples is strongly discouraged, as the number of freeze‐thaw cycles affects analyte recovery for Aβ. Freeze‐thaw stability should be determined for each biomarker studied, if unknown. In general, CSF proteins can be more vulnerable to degradation due to the absence of stabilizing carrier proteins found in other biological matrices, such as blood. ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²²

4.4.4. Alarm systems

A monitored back‐up alarm system and plan for freezer failure is recommended. Centers are encouraged to have an open freezer available in the event of a freezer failure. Temporary freezers may be rented from specialized appliance repair companies if an extra freezer is not available. Annual or semi‐annual maintenance is recommended, depending on the frequency of entry, tendency for ice build‐up, and the manufacturer's recommendations.

4.5. Sharing and dissemination of cerebrospinal fluid samples

4.5.1. ADRC as a resource

The ADRC repositories are a shared national resource for the purpose of answering valid scientific questions related to cognitive aging and dementia. ADRC investigators should discuss in advance the amount of CSF to be collected per visit to reserve specimen material for internal and/or future studies, balanced with a transparent resource sharing plan.

4.5.2. Resource access

Specific evaluation criteria for specimen requests should be documented and consistently applied by an ADRC‐designated committee.

4.5.3. Resource sharing

CSF biospecimen sharing is recommended to be limited to the smallest number of samples and sample volume required to adequately answer the research question under investigation, as defined by power calculations.

4.5.4. Management of biorepositories

Use of sample management software can assist with sample tracking and dissemination. There are many commercial software options available.

4.5.5. Standardized policies and procedures

Each center should develop operating procedures to facilitate timely resource sharing, including Institutional Review Board IRB/HIPAA approval of resource sharing, institutional material transfer agreements, institutional data use agreements, ADRC‐specific resource use agreements (ADRC acknowledgement and what can be done with sample), specific federal human sample shipping training, and shipping manifests related to shipping CSF samples domestically and internationally to optimize successful transport of these valuable biospecimens.

4.6. Additional considerations for CSF

4.6.1. Processing cells

Processing cells from CSF is generally beyond the scope of this guideline. Briefly, processing cells from CSF can be accomplished with standard centrifugation of 300 × g ( = 300 relative centrifugal force [RCF]) for 10 min, and the cell pellet can be saved in an appropriate freezing medium after carefully decanting the CSF. CSF‐derived cell pellets can be stored in appropriate freezing medium in cryovials at −80°C. ² , ⁴

4.6.2. Processing CSF for extracellular vesicles

Processing parameters for preparation, isolation, gradient centrifugation, and volume considerations for extracellular vesicle (EV) research are beyond the scope of these guidelines and should be developed according to the research needs of individual labs.

Generally, 1‐ to 2‐mL volumes of CSF are required for isolation of EVs and therefore should be isolated from EV study‐specific research protocols (protocols at individual centers) to preserve curated ADRC specimen volumes for the larger community of users. Published guidelines for EV isolation include the Minimal Information for Study of Extracellular Vesicles (MISEV) and the International Society for Extracellular Vesicles (ISEV) website has many additional resources. ²³ Some guidelines for EV isolation are also available from vendors, for example:

https://www.beckman.com/resources/sample‐type/extracellular‐vesicles/getting‐started/isolation/isolation‐methods

4.6.3. Biological factors affecting CSF biofluid biomarker detection

A number of biological factors related to the human research participant were demonstrated to relate to differences in CSF biomarker results (e.g., age, sex, race/ethnicity, body mass index [BMI], exercise, medical conditions and comorbidities, food and beverages consumed prior to collection, medications including over‐the‐counter medication (OTC) and supplements, circadian and diurnal cycle relative to time of lumbar puncture). ³ , ⁴ , ⁵ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ To support the full spectrum of use cases from ADRC specimens, attempts should be made to record as much information related to these variables as possible in order for appropriate adjustments to be made during analysis of results.

Online link top CSF protocol: ADRC CSF collection guidelines

https://files.alz.washington.edu/best‐practices/csf‐collection‐processiong‐storage‐guielines‐9.22.25.docx

5. PROTOCOL 2: SAMPLING AND USE OF BLOOD‐DERIVED FLUIDS

Rigorous control of blood collection, processing, and storage is important for providing systematic results that are comparable between centers. Table 1 includes a listing of biomarkers currently studied by multiple ADRCs in either serum or plasma. For these protein biomarkers, a review of the existing technical literature was performed to provide recommendations to reduce technical variance. Tables 4 and 5 provide a practical summary of the reported factors that affect blood‐based biomarkers, reviewed by the Workgroup from 2023 to 2025 and gleaned from expert reviewers, to support recommendations for best practices. Factors to be considered include blood tube selection, additives, and inter‐center variance, ¹⁴ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ as well as processing storage and freeze‐thaw restrictions, important factors to rigorously characterize for each analyte. ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴²

TABLE 4.

Literature reports addressing blood collection and processing issues.

Biomarker	Issue of concern for blood	Specific consideration	References
Aβ40, Aβ42	Tube material and fill volume Differences in measurable quantity	Re: Fill volume, some studies find no impact (summary table in Hansson 2018). Loss in clot formation in serum, lower levels in plasma EDTA versus citrate tubes.	29
BDNF	Differences in measurable quantity	EDTA plasma tubes decrease measurable BDNF compared to other tube types (e.g., citrate).	30 , 31
Cytokines	Tube type/additive	Certain cytokines have been shown to be less stable in serum. Recommend EDTA plasma tubes.	32
NRGN	Storage stability	Stable at –20°C, no published data on short‐term storage.	36
pTau217	Centrifugation prior to measurement	Better correlations of plasma pTau217 with CSF pTau217 and CSF Aβ42/Aβ40 for samples centrifuged at 2000 × g at 4°C for 10 min compared to non‐centrifuged samples.	37

Abbreviations: Aβ, amyloid beta; BDNF, brain‐derived neurotrophic factor; CSF, cerebrospinal fluid; NRGN, neurogranin; pTau, phosphorylated tau.

TABLE 5.

Reported storage and stability issues affecting the study of proteins in blood.

Biomarker	Issue of concern	Specific considerations	References
Aβ1‐40, Aβ1‐42	Quantity – see above	Decreased detection level with delayed freezing.	29
BDNF	Quantity – see above	Decreased detection level with delayed freezing.	30 , 31
Cytokines/chemokines	Stability	Stable frozen storage at 4°C affects IL‑8, VEGF, TNFα, and EGF	32
GFAP	Freeze/thaw cycles	After two freeze/thaw cycles, the concentration increases	19 , 33
NfL	Freeze/thaw cycles	Freeze‐thaw cycles may increase detected levels. Levels increase upon −20°C storage compared to −80°C due to breakdown	33 , 34 , 35
NRGN	Stability	Stable at −20°C, though plasma/serum fragment patterns exist.	36
pTau (potential), tTau, BD‐tau, MTBR‐tau	Freeze‐thaw cycles	After two freeze/thaw cycles, the concentration starts to significantly decrease.	37
pTau181 pTau217	Thawing temperatures	Thawing conditions (RT or on ice) do not affect the performance of plasma pTau181/pTau217.	37

Abbreviations: Aβ, amyloid beta; BDNF, brain‐derived neurotrophic factor; BD‐tau, brain‐derived isoform‐specific tau; MTBR‐tau, microtubule binding region fragment tau; NfL, neurofilament light chain; NRGN, neurogranin; pTau, phosphorylated tau; RT, room temperature; TNFα, tumor necrosis factor alpha; tTau, total tau; VEGF, vascular endothelial growth factor.

5.1. Collection of blood samples for biomarker studies

5.1.1. Biological factors affecting biofluid biomarker detection

A number of biological factors related to the human research participant were demonstrated to relate to differences in blood biomarker results (e.g., age, sex, race/ethnicity, BMI or body composition, exercise, medical conditions and comorbidities, food and beverages consumed prior to collection, medications including OTC and supplements, circadian and diurnal cycle relative to time of blood draw). ¹² , ¹³ , ¹⁴ To support the full spectrum of use for cases from ADRC specimens, attempts should be made to record as much information related to these variables as possible for appropriate adjustments to be made during analysis of results.

5.1.2. Fasting versus non‐fasting prior to blood draw

Many AD studies utilize fasting blood collection, and this is recommended for consistency. However, given patient and clinic needs, this is not always possible. Therefore, whether fasting or non‐fasting, time of day and time since last meal should be recorded.

5.1.3. Avoiding laboratory protocol deviations

Estimates suggest that up to 46% of laboratory deviations come from pre‐analytic processing. ²⁴ , ²⁵ , ²⁶ , ²⁷ Table 4 reviews pre‐analytical factors affecting the collection and processing of blood samples for a biomarker study. Table 5 reviews reported effects of storage conditions on the stability of analytes, reported to affect consistency in biomarker proteins measured in blood. Factors related to blood collection devices (needle gauge, tube lubricants, tube walls) can impact blood biomarker levels. ²⁶ , ²⁷ Some methods of transporting specimens after they are collected can impact measurements. ²⁸ The largest source of pre‐analytical variation for specific blood markers (primarily Aβ species) was reported to derive from the initial time delay from collection to centrifugation and time from centrifugation to storage, during which the tubes are kept at RT or at 4°C. ²⁹ Standardized and uniform techniques of sample processing are recommended, since biorepositories have broad use and effects on biomarker stability differ and remain mostly unknown. ²⁹ Deviation from standard recommendations should be noted upon data entry.

5.1.4. General procedure and order of use of blood draw tube type

Detailed step‐by‐step procedures for the collection of blood samples are available in the CLSI H3‐A6. ²⁷ Broad recommendations for standardization of sample collection are as follows:

Blood should be collected, with participant is seated, from the median cubital vein, as opposed to other, more fragile, veins.

Alcohol used to clean the skin should be allowed to evaporate before venipuncture.

A tourniquet applied around 7.5 to 10 cm (3 to 4 inches) above the site of venipuncture should be loosened once blood starts to flow.

Blood is generally drawn with a vacutainer system.

Tubes for plasma should be adequately filled with blood to ensure the optimal blood/additive ratio.

For most studies, a needle gauge of 19 to 23 is preferable, with 21 gauge being the most common. ²⁴ , ²⁷

Order of blood draw for a common standardized ADRC workflow should be as follows (skip tubes not being utilized) (CLSI H3‐A6) ²⁷ :

• Blood culture tube

• Coagulation tube

• Serum tube with or without clot activator or gel

• Heparin tube with or without gel separator

• EDTA plasma tube with or without separator

• Glycolytic inhibitor tube

• PAXgene blood RNA tube

This order moves from the least to the most additives and avoids additive contamination.

5.2. Processing of serum tubes

• Serum tubes include a clot separator gel or clot activator.

  After filling serum tubes, incubate the tubes to allow clot formation.

• Serum should be clotted in a tube rack in a vertical position for at least 30 min and no more than 60 min, followed by centrifugation if the patient is not on anticoagulant therapy.

• RCF (number × g; g‐force) should be utilized rather than using revolutions per minute (RPM) in standard operating protocols (SOPs) and publications, since RPMs vary by rotor size.

  Note: Horizontal rotors are preferable.

• Centrifugation at 2000 × g is performed for 10 min to separate serum and clotted material. Generally, 1 mL of whole blood will produce 500 µL of serum volume.

• Note: Centrifugation at RT versus refrigerated (4°C) can cause variation in downstream assay detection levels for certain markers in serum. The recommendation in general is to centrifuge at 4°C. ²⁶ , ²⁹

5.3. Processing of plasma tubes

• Plasma tubes are vacutainers with different additives, including EDTA and heparin.

• Plasma tubes should be filled and then gently inverted eight to 10 times to mix the additives.

• Plasma tubes are incubated at RT for 30 min prior to centrifugation.

• Relative centrifugal force (RCF; number × g; g‐force) should be utilized rather than using revolutions per minute (RPM) in SOPs and publications, since RPMs vary by rotor size.

  Note: Horizontal rotors are preferable.

• Centrifugation at 2000 × g is performed for 10 min to separate plasma. Generally, 10 mL of whole blood will produce 4 to 5 mL of plasma volume.

• Note: Centrifugation at RT versus a refrigerated (4°C) can cause variation in downstream assay detection levels for certain markers in plasma.

• Note: Refrigerated centrifuges are also recommended for platelet preparation.

• Note: When processing DNA from plasma, see (5.4), below.

5.4. Processing of DNA from plasma tubes

• During preparation of plasma, an interphase between red cells and clear plasma fluid forms, containing a buffy coat, a layer of white blood cells.

• After the plasma is carefully pipetted into destination tubes, aliquots of buffy coat cells should be frozen at −80°C until DNA extraction. Generally, 5 mL plasma will produce 0.1 to 0.2 mL of buffy coat volume. ²⁶ , ²⁷

5.5. Processing of tubes with glycolytic inhibitors

• Mixing: Immediately after collection, gently invert the tube eight to 10 times to ensure proper mixing of the blood with the glycolytic inhibitor.

• Centrifugation: Process the samples promptly. Centrifuge the tubes at the recommended speed and duration to separate the plasma or serum from the cells.

• Storage: Store the processed samples at the appropriate temperature to maintain sample stability. Glycolytic inhibitor tubes should be stored at 4°C until analysis.

5.6. Processing of PAXgene RNA tubes

• Immediately after blood collection, gently invert the PAXgene RNA tubes 20 to 30 times.

• Place tubes in a rack to stand in the vertical position at RT for a minimum of 2 h and a maximum of 72 h.

• Place the PAXgene RNA tubes in a 13‐cm (∼5‐inch) cryobox and store cryobox in a −20°C freezer.

• After 24 h, transfer to –80°C storage.

• Record the specimen location and the time the specimens are placed in the freezer.

• Further processing of PAXgene RNA tubes should follow the manufacturer's protocol for RNA purification, quantification, and integrity evaluations.

5.7. Additional sample processing considerations

Tables 4 and 5 list empirical evidence from published literature regarding specimen acquisition, processing, stability, and storage. Included are studies that represent the current consensus and have been reported in peer‐reviewed studies. These support the following concerns and recommendations.

Rapid processing of samples is optimal (total processing time no more than 2 h from collection to freezing). Detailed procedures for processing blood specimens are provided by CLSI H18‐A4. ²⁶ General recommendations follow, although individual steps may need modification for specific markers.

Serum/plasma should be physically separated from contact with cells as soon as possible (no more than 2 h). Specific volumes for aliquots should consider multicenter and multi‐instrument use to avoid freeze‐thaw cycles. As a reasonable example for current instrument use:

These factors should be documented:

• Type of collection tube (manufacturer's name, type of anticoagulant)

• Time from collection to centrifugation

• Centrifugation time (minutes)

• Temperature between collection and centrifugation

• Presence and type of separator, if present

• Temperature of centrifugation

• Number of centrifugations (single or double)

5.8. Other considerations for additives

1. Whether it is necessary to add protease inhibitors to samples after aliquoting is not certain and depends upon the nature of the study. This may be worth consideration if plasma or serum samples are to be used for proteomic analyses, and this must be noted.

2. General guidelines recommend not adding inhibitors (other than EDTA), where the broader research community accesses samples for studies, to prevent unforeseen effects on downstream analyses over time.

5.9. Post‐centrifugation considerations

1. The following should be documented:

   • Type of secondary container (tube, straw)
   • Time between centrifugation and freezing
   • Storage temperature
   • Number of freeze‐thaw cycles
   • Duration of storage
   • Storage location of aliquot vials
   • Degree of hemolysis and lipemia

5.10. Aliquoting recommendations

• Aliquots should be made in low‐protein‐binding polypropylene tubes (or straws) using polypropylene tips for pipettes. Rubber o‐ring screw caps prevent dehydration over significant storage periods.

• Small aliquots (generally not larger than 0.5 mL) are recommended for storage, to avoid unnecessary freeze‐thaw cycles of samples. Consider aliquoting plasma and serum in even smaller volumes (e.g., a number of 55‐, 200‐, or 500‐µL bulk tubes), as noted above.

5.11. Storage and shipping recommendations

• Long‐term storage should be at −80°C or in liquid nitrogen.

• If storage on dry ice is utilized for shipment, the headspace should be vented, or the sample should be allowed to sit in −80°C freezer for 9 h prior to thaw. ²⁹

• Consider using disposable thermometers to keep track of temperature during transportation.

5.12. Document the volume of plasma or serum that was obtained

• Record total volume of plasma or serum collected.

• Record individual aliquot volumes for plasma or serum.

5.13. Factors influencing the quality of serum or plasma

5.13.1. Hemolysis

Red or pink tingeing of plasma or serum is an indicator that significant hemolysis has occurred (scored on a depth of color scale [denoted H1 to H6, with grade H3 and higher representing significant hemolysis and can be further investigated by measuring hemoglobin level (Hb) by spectrophotometry (several methods have been published).

5.13.2. Lipemia

A milky white substance floating in the plasma or serum, which may render the samples less useful for many biomarker studies, should be determined on a case‐by‐case basis.

5.14. Important points for PAXgene collection and processing

1. Store PAXgene Blood RNA Tubes at RT (18°C to 25°C) before use.

2. If the PAXgene Blood RNA Tube is the only tube to be drawn, a small amount of blood should be drawn into a discard tube prior to drawing blood into the PAXgene Blood RNA Tube. Otherwise, the PAXgene Blood RNA Tube should be the last tube drawn in the phlebotomy procedure.

3. Allow at least 10 s for a complete blood draw to take place in each tube. Ensure that the blood has stopped flowing into the tube before removing the tube from the holder. The PAXgene Blood RNA Tube with its vacuum is designed to draw 2.5 mL of blood into the tube.

4. Immediately after blood collection, gently invert/mix (180° turns) the PAXgene Blood RNA Tube 15 to 20 times.

5. Incubate the PAXgene Blood RNA Tube UPRIGHT at RT (18°C to 25°C) for 24 h. Record the time/date of the draw.

6. Repeat steps a–e for each PAXgene Blood RNA Tube to be collected per subject.

7. After 24 h at RT, transfer the PAXgene tubes to −80°C (or −20°C) freezer. Record the time and date of freezing.

5.15. Long‐term storage recommendations

Table 5 lists supportive literature reporting the effects of storage and freeze‐thaw cycles on biomarker proteins in serum or plasma matrices. Centers should also refer to the Office of Biorepositories and Biospecimen Research (OBBR) guidelines for maintenance and long‐term storage recommendations:

(See http://www.ncbi.nlm.nih.gov/pubmed/24749882)

5.16. Sharing and dissemination of plasma and serum samples

1. ADRCs and associated biorepositories are a national resource to be shared for the purpose of answering valid scientific questions related to cognitive aging and dementia. Center investigators should discuss in advance the amount of sample to be collected per visit to reserve specimen material for internal and/or future studies, balanced with a transparent resource‐sharing plan.

2. Specific evaluation criteria for specimen requests should be documented and consistently applied by a center‐designated committee.

3. Plasma or serum biospecimen sharing is recommended to be limited to the smallest number of samples and sample volume required to adequately answer the research question under investigation, as defined by power calculations.

4. Use of sample management software can assist with sample tracking and dissemination.

5. Each center should develop operating procedures to facilitate timely resource sharing, including IRB/HIPAA approval of resource sharing, institutional material transfer agreements, institutional data use agreements, center‐specific resource‐use agreements (center acknowledgement and what can be done with sample), specific federal human sample shipping training, and shipping manifests related to shipping plasma or serum samples domestically and internationally to optimize successful transport of these valuable biospecimens.

Online link for the protocol: ADRC blood collection guidelines

https://files.alz.washington.edu/best‐practices/blood‐collection‐processing‐storage‐guidelines‐9.22.25.docx

6. COLLECTION AND PRE‐ANALYTICAL PROCESSING SUMMARY

The processes outlined above for CSF, serum, and plasma are summarized in a visual workflow in Figure 1. The following highlights reflect the general recommendations compiled in this update for biofluid biomarker best practices.

FIGURE 1.

Visual workflow for recommended standardization of CSF, serum, and blood processes to minimize variance caused by pre‐analytical factors in research.

6.1. Specimen processing

• Standardized centrifugation

• Tube types matter, as does dead volume in aliquots

• Consistent and QC‐tested materials for biofluids

• Overnight fasting – for assay‐dependent SOPs only (impractical for enrollment in clinics – should reflect real‐world sampling/context of use)

• Standardized collection SOPs provided with additional resources

6.2. Storage and stability

• Standardized ATN assays including p217‐Tau are stable at 4°C for short periods, cell free.

• Avoid dead volume in aliquots, affecting the ratio of surface area to sample.

• Redundancy is recommended (split samples into multiple locations to protect resources).

• Freezer alarms/temperature records: routinely monitored by remote notification and data download.

6.3. Thawing, assaying, and analysis of samples

• Most proteins withstand two or more freeze‐thaws before activity‐integrity loss. It should be noted that this requires study in each specific assay, since aspects such as structural changes and antibody epitope location on the protein structure may be differentially affected by protein degradation. ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹

• Monitor/control temperature and established performance limits.

• Standardize operational bench time to avoid temporal variance between runs.

• For example, temperature‐dependent changes in tau signal. ²⁰ , ²¹

7. ADDITIONAL FORWARD‐LOOKING CONSIDERATIONS

It should be noted that the information presented in the current guidelines, with input from the Alzheimer's Association working group members and other experts, is intended to support the standardization of research conduct and analytical results for the network of ADRCs. However, these guidelines may be used as a reference by other centers. The cited references represent empirical support for the information in the guidelines. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ , ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸

7.1. Additional biomarkers

This update is limited in its coverage of biomarkers specific to those currently in use at the ADRCs, as reported in the survey. However, with the rapid expansion of information in the literature, certain biomarkers will require additional research that indicates their utility in assessing disease stage, disease specificity, treatment monitoring, or other disease stratification or subclassification (subtyping) utility in order to be adopted for further evaluation. The current presentation of revised guidelines was limited by the available scientific literature and the goals of the ADRC Workgroup Charter. As such, the authors recognize that other promising biomarkers may exist that have not yet had adequate pre‐analytical studies performed or reported on. The ongoing work at the ADRCs utilizes a mixture of instrumentation that may have optimized processes for specific biomarkers. For example, assays use different amounts of biofluids per run, which may affect aliquoting volume decisions or other considerations.

Another emerging area of study examines biomarkers localized in EVs. Processing parameters for preparation, isolation, gradient centrifugation and volume considerations for EV research are beyond the scope of the current guideline revision and should be developed according to the research needs of individual labs (see MISEV Guidelines). ²³ As mentioned previously, generally larger volumes of CSF or plasma are required for isolation of EVs than are needed for standard bioassay runs, and therefore EVs should be isolated from set‐aside vials for specific research protocols at individual centers in order to preserve curated ADRC specimen volumes for the larger community of users. Some guidelines from international working groups, such as ISEV, were recently published and can provide a guide to researchers in this area. ²³

7.2. Standardization of protocols for specific instrumentation

According to the ADRC survey we conducted, the main service instruments in use across the ADRCs include MesoScale Discovery electrochemiluminescence‐linked assay plate readers (ECLIAs, MSD Sector or Quickplex120), Amprion SAA, Quanterix Simoa (e.g., HD‐X and HD‐X1), Fujirebio Lumipulse G 1200, and multiplex discovery platforms including Alamar Biosciences ARGO HT, Somalogic Somascan, and mass spectroscopy. Note that mass spectroscopy is often used for discovery, as well as having some clinically validated low‐plex tests (e.g., C₂N Diagnostics, Quest). Clinical main lab instruments were also noted. Figure 2 shows the list of analytical instrument platforms reported to be currently and commonly in use at the responding centers, taken from the survey of all ADRCs. The workgroup defers to manufacturer guidelines for instrument and kit use, but including this provided a useful historical snapshot of activity.

FIGURE 2.

Instruments currently utilized by multiple Alzheimer's Disease Research Centers (ADRCs) for cerebrospinal fluid and blood biomarkers. The survey of 37 ADRCs returned data supporting the use of 10 main instrument platforms for biomarker analysis, both internal (blue) or external (red) to the ADRC.

Figure 3 shows the percentage of ADRCs that indicated they were performing each of the listed assays, indicating whether they were performed internally at the ADRC or external service labs were utilized. In general, the majority of biomarkers being routinely studied at multiple ADRCs were neuronal and glial biomarkers, with some additional inflammatory proteins and fewer vascular biomarkers represented. As both the inflammatory and vascular components of neurodegenerative disease are essential aspects of onset and progression, both areas could use additional study. This information also highlighted the need for a more robust study of pre‐analytical factors for the majority of the biomarkers in use.

FIGURE 3.

On‐site and external service providers. The percentage of ADRCs that perform each of the studied assays on instruments internal to their facility or utilized through external service labs is shown.

In addition to the aforementioned protein biomarker platforms, genetic, epigenetic, and transcriptomic profiling technologies are in widespread use among ADRCs, but they are covered in other guidelines. Similarly, metabolomic and lipidomic molecules are being studied in structured cohorts, but they were beyond the scope of the CSF and blood guidelines. Nevertheless, many of the recommendations included within the present guidelines enable such studies (DNA preparation from buffy coat, PaxGene RNA tube specimens).

7.3. Additional biofluid types

The research community has pursued biofluids other than CSF and blood derivatives, including some that could provide advantages in elderly subjects, those with medical conditions or treatments that affect clotting proficiency (blood thinners), or subjects with vascular frailty. It is worth mentioning that saliva, urine, and oral mucosal fluid have been explored, as has stool, for detecting AD pathology biomarkers. Some of these were discussed with the workgroup but were considered to be beyond the scope of this guideline update due to a lack of consistent or widespread use.

8. SUMMARY AND CONCLUSION

The advent of higher‐throughput technologies for biomarker discovery has led to an increasing number of newly studied biomarkers, known and novel proteins that may still provide advantages in the future as clinical indicators. ³⁸ The spectrum of potential biomarkers, as noted, may also, in the near future, include approaches to diagnostic subclassification, which could use biofluid biomarkers of different types, including specific modified proteoforms. Thus, combinations of the most specific and informative biofluid indicators may include proteins and modifications, as well as metabolites, lipids, and/or RNA species, potentially on the same detection platform. Areas of important emerging research that will impact these approaches include increased understanding of the impact of co‐pathologies on individual disease, since it is known that multiple neurodegenerative mechanisms commonly coincide, and the extent and nature of vascular disease are a major factor affecting chronic neurodegeneration. In addition, efforts to create an accessible set of global reference standards for all clinically informative bioassays are under way, which will aid in the standardization of assay results and interpretation. Notable recent efforts in this area are the Standardization of Blood Biomarkers (SABB), the Alzheimer's Association‐supported Global Biomarker Standardization Consortium (GBSC), and similar reports on specific biomarkers and pre‐analytical factors. ³⁹ , ⁴⁰

The publication of the current 2025 ADRC guidelines represents an attempt by the authors to represent the current state of knowledge in the field but was limited to CSF and blood‐based biomarkers in order to standardize practices for the majority of research conducted in the community. With this purpose in mind, and using the stated weight of evidence criterion, the information was compiled, reviewed by the workgroup, and then sent for expert review by key opinion leaders recognized for advancing global biomarker standardization efforts. ³⁹ , ⁴⁰ , ⁴¹ , ⁴² This updated set of guidelines was compared during its development with efforts of the SABB and GBSC for harmonization, but it aims specifically to support standardized practices across the US‐based ADRCs. Therefore, the list of protein biomarkers included is limited to current ADRC activities, and this report is by no means a comprehensive review of the reported biomarkers studied more broadly in the context of AD pathology. Collectively, these efforts by the international community will contribute to the robust advancement of biofluid biomarkers as tools for clinical decision‐making in AD and other neurodegenerative pathologies.

CONFLICT OF INTEREST STATEMENT

Authors claim no conflicts of interest in relation to the public guidelines presented in this manuscript, assembled from a review of the cited literature and from external expert opinion. Author disclosures are available in the Supporting Information

CONSENT STATEMENT

Because this is a review manuscript, no human subjects were used in preparation of this manuscript, only information from published articles, operational research within the ADRC network, and interviews with experts in the field of neurology and clinical biochemistry. All authors reviewed this manuscript.

Supporting information

Supporting Information