Source-linked AI summary
How shall we use the proteomics toolbox for biomarker discovery?
Pierre Lescuyer, Denis Hochstrasser, Thierry Rabilloud
TL;DR
Proteomics-based biomarker discovery has produced relatively few clinically useful successes, particularly from plasma. This perspective analyzes causes of limited success and proposes shifting discovery to cells or tissues, or selectively enriching body-fluid proteins.
Problem
Plasma profile comparisons have proven inefficient for biomarker discovery, partly because plasma proteins span a twelve-order-of-magnitude concentration range and 22 proteins comprise 99% of protein mass.
Method
The paper analyzes separate discovery and validation phases and evaluates cell, tissue, secretome, neighboring-fluid, and selective protein-enrichment strategies.
Results
Body-fluid discovery has yielded few successes, while selective protein strategies and cell-, tissue-, or secretome-based discovery are presented as alternate approaches.
Takeaways & Limitations
Biomarker studies should separate discovery from validation and design each phase around the clinical question and the most suitable sample type.
Takeaways & Limitations
Selective combinatorial protein-enrichment approaches can lose quantitative information after binding-site saturation, limiting their biomarker-discovery utility.
Abstract
from arXiv · showhide
Biomarker discovery for clinical purposes is one of the major areas in which proteomics is used. However, despite considerable effort, the successes have been relatively scarce. In this perspective paper, we try to highlight and analyze the main causes for this limited success, and to suggest alternate strategies, which will avoid them, without eluding the foreseeable weak points of these strategies. Two major strategies are analyzed, namely, the switch from body fluids to cell and tissues for the initial biomarker discovery step or, if body fluids must be analyzed, the implementation of highly selective protein selection strategies.
1. Frame and Facts.
The paper argues that conventional plasma proteomics has often failed to yield clinically useful biomarkers because of biological, technical, and clinical limitations. It proposes redesigned workflows emphasizing clinically defined questions, alternative discovery samples, selective protein enrichment, and rigorous validation.
- 2. Success & Failure.: Proteomic biomarker success is more likely in organ-proximal fluids, whereas plasma-derived cancer biomarker discovery has produced results of limited interest.Cerebrospinal-fluid 14-3-3 proteins and stroke biomarkers illustrate successful or promising organ-proximal discovery contexts.
- 4. Proteomic strategy.: Conventional comparisons of plasma protein profiles between healthy and diseased subjects have proven inefficient or unsuccessful for biomarker discovery.
- 4. Proteomic strategy.: Plasma’s enormous protein dynamic range limits detection of minor proteins, while depletion and combinatorial methods introduce efficiency, saturation, and quantitative-loss problems.Depletion must approach 100% efficiency to be useful, and saturated binding sites in combinatorial methods eliminate quantitative information.
- 6. What to do?: Biomarker studies should begin with a clearly defined clinical question and separate discovery from plasma or serum validation aimed at selecting clinically useful candidates.Clinical chemists and physicians should guide sample choice, while validation assesses candidates against the clinical problem and real-world testing needs.
- 6.3. Exception or clever tricks.: Alternative workflows should shift initial discovery toward cells, tissues, or released proteins, or selectively fractionate body fluids such as plasma or urine.These strategies trade weaker direct protein analysis for more focused cellular or animal biology and targeted protein subclasses.