Biomarker Development, Validation, and Proteomics Approach

Top Notes

Marker selection: alpha tubulin or beta tubulin
Assay modality selection: LC/MS or ELISA (Simoa)
Timeline of assay development

Stages of Biomarker Development

The upper table is headed Stages of biomarker development (below?). It includes biomarker discovery, replication, assay validation/measurement performance, qualification/correlation, and utilization/context-of-use concepts. The dense table should be treated as partially transcribed.

stage / area	conservative readable content
BM discovery	human disease and healthy samples from a well-defined target population; confirm cross-sectional changes of BM; characterize longitudinal BM changes; identify a subset of patients with altered BM level; “To characterize BM species that is changed.”
BM replication	reproducibility in an external, independent cohort should be tested; cohort should be tested
Assay / technical validation	method technical validation; measurement performance; preclinical sample; clinical sample
Qualification / correlation	handwritten/Korean note: Qualification 이라고 봐야겠음; Correlation & causality; correlation between intervention, biomarker, endpoint; documented process demonstrating evidentiary link of a biomarker measurement with a clinical endpoint
Utilization	context of use; clinical utility measures the impact of incorporating the biomarker assay in clinical practice in terms of improved disease management and/or patient outcomes

Readable Korean/annotation snippets in this section mention acetyl-tubulin, brain, CSF assay development, intervention-biomarker correlation, CSF GBA mRNA/protein, Tau mRNA/tau protein, GlcSph/GlcCer, and clinical sample/clinical validation. Exact wording and intended meaning require manual review against the image.

Figure 17.1 Biomarker Development Pipeline

The page includes a figure labeled:

Figure 17.1 Biomarker development pipeline. Source: Adapted from Sin et al. 2015 [7].

The visible pipeline labels are:

Content development -> Validation and qualification -> Launch
Biomarker discovery -> Biomarker replication -> Analytical validation -> Clinical validation -> Clinical utility -> Clinical use

Fluid Biomarker Working Group Table

The next dense table is labeled:

[Fluid Biomarker Working Group 20190211v7.pptx]

It appears to be a molecular BMx workflow table with columns:

column	selected readable content
BMx PJ Start	In silico assessment of assay feasibility; reagents (mAbs); standards (recombinant or endogenous); assay format (ELISA, Simoa, Singulex, LC-MS/MS). Deliverable: SoW/Research plan and contract with selected vendor.
BMx FA	Procure reagents (mAbs, standards, etc.); verify compatibility of reagents and assay format; verify adequate assay sensitivity in human patient samples; verify adequate assay sensitivity in non-clinical species. Deliverable: Prototype assay.
BMx TA	Achieve assay acceptance criteria for dilutional linearity and MRD analysis, standard linearity and assay range, accuracy, repeatability, intermediate precision, limit of quantitation, controls and parallelism. Deliverable: Spec sheet with assay wet lab quantified parameters.
BMx Execution	In vivo studies with validated assay in pre-clinical species (WT mouse/rat/NHP); timecourse data reflecting dynamic biomarker changes after single dose and after approximately 10-14 repeated doses; collaborate with BTUs for animal models if necessary; test patient sample cohorts, ideally longitudinal. Deliverable: In vivo BMx data package including PK/PD analysis in collaboration with DMPK.
BMxS	If in vivo biomarker data package accepted by sDIB, transfer information to Clinical Biomarker Plan and incorporate into overall Clinical Development Plan. Deliverable references Clinical Biomarker Plan and transfer of assay methods to clinical CRO for use in TAU programs.

BM Discovery to Validation / Proteomics Approach

The lower table is headed approximately BM Discovery -> Validation (verification) and Proteomics approach. It compares sample/source types including CSF, Tissue (Brain), Culture medium, BLOOD, and Urine.

row / source	conservative readable content
Sample characteristics / CSF	Cons: 10-100 times lower amount of total protein content than serum/plasma. Pros: organ proximity.
Tissue (Brain)	Cons include small pieces of tissue with limited material for proteomics studies; clinical material from CNS available only post mortem; heterogeneous disease stages. Pros/annotation mention after death proteolysis and that proteins that withstand it may be robust.
Culture medium	Easily liberated proteins, as opposed to cytoskeletal proteins; Korean annotation suggests these are what would mainly be observed and may have CSF translatability.
BLOOD	Protein concentration in brain and blood may differ due to factors including generation from neurons/microglial cells, influx/efflux, albumin-related movement, kidney clearance of free Ab, no organ proximity, and often not reflected in blood. Citation visible: Ahn and Simpson 2007, PMID 21136753.
Urine	Does not have homogeneous composition; protein concentration can vary from milligrams to grams per liter depending on disease state; contains high concentrations of excreted compounds such as salts, urea, or solubilizing agents; proteome partly originates from plasma through filtration/leakage/secretion/reabsorption pathways.

Additional readable rows:

row	conservative readable content
Clinical cohort	Well defined and representative of the target population; otherwise background noise.
Sample preparation	Immunodepletion of the 12 most abundant proteins is necessary; visible examples include serum albumin, IgG, fibrinogen, transferrin, IgA, IgM, haptoglobin, apo AI, apo A-II, alpha-1-antitrypsin, alpha-1-acid glycoprotein, alpha-2-macroglobulin; these proteins comprise over 96% of total protein content in plasma/serum [5]. Protein fractionation. CSF, like serum/plasma, requires depletion of highly abundant proteins.
Quantitative analysis	LC-MS
Reference for data normalization	none; actin appears in the tissue/brain column
Bioinformatics analysis	Statistical analyses; good statistics can be performed with 3-7 repeats of each sample.
Interpretation	Change in expression vs activity? Magnitude of difference. Anything below 20% of the difference is heavily contaminated with experimental error.
Validation / replication row	Independent assay & cohort; can be with selected potential BMs.

Elements of Validation of a Biomarker

The visible lower section is headed approximately Elements of validation of a biomarker: and begins a table of validation types.

validation type	conservative readable content
Validation	Korean note after heading appears to say only two types exist below; exact text requires manual review. Definition begins: “a process to establish that the performance of a test, tool, or instrument is acceptable for its intended purpose”.
Analytical validation / assay validation / method validation	Establishing performance of the biomarker test / biomarkers’ assay performance, including biomarker measurement methods, calibration standards, accuracy, precision, reproducibility, linearity, and quality specifications.
Clinical validation	Includes (2018 Yee) wording about establishing in clinical samples that the biomarker test identifies its concept of interest. Includes (2016 fda-NHI BEST) wording about establishing that the test/tool/instrument acceptably identifies, measures, or predicts the concept of interest.
Concept	In a regulatory context, concept is the aspect of an individual’s clinical, biological, physical, or functional state, or experience, that the assessment is intended to capture or reflect.
Predictive validation	Ability of the assays to detect and predict diseases.

Uncertain Spans

• Exact Korean handwritten/typed annotations in the upper biomarker-development table, especially around CSF GBA, tau, GlcSph/GlcCer, calibration, and clinical validation.
• Dense table cell boundaries and row/column continuation across the Fluid Biomarker Working Group workflow.
• Exact wording of the BMx TA acceptance criteria and BMx Execution/BMxS deliverables.
• Exact citation formatting for Sin et al. 2015 [7], Ahn and Simpson 2007 / PMID 21136753, and references [5].
• Exact validation-definition wording and Korean note in Elements of validation of a biomarker.
• Page/section mismatch between navigation pane (Steps of PRS STUDY) and body content (biomarker validation).