MSACL 2016 EU Abstract

Detection, Identification and Confidence Rating of Unknown and Unexpected Compounds Found in Plasma Samples by LC-HRMS Analysis

Bertrand Rochat (Presenter)
CHUV

Bio: Bertrand Rochat obtained his PhD at the University of Lausanne in Switzerland, following which he carried out post-doctoral research in France, the United States and Scotland. From 2001 to 2003, he worked as lab head in the department of Pre-Clinical Safety at Novartis Pharma in Switzerland. Since 2003, he has been in charge of the Mass Spectrometry Facility at the University Hospital of Lausanne in Switzerland. In that role, Dr. Rochat helps medical doctors and biologists in their research and, in parallel, carries out his own research : • Intratumoral drug metabolism and disposition • Quantitative determination of native peptides by LC-MS in biological matrices • Implementation of a targeted metabolomics analysis in patient plasma. He has published numerous original articles, editorials, and reviews in these fields.

Authorship: Bertrand Rochat
University Hospital of Lausanne; CHUV, CH-1011 Lausanne

Short Abstract

With the capability of high-resolution (HR)-MS instruments to perform sensitive and reliable quantifications in HR-full scan mode, the detection (from the same acquisition) of unexpected and/or possible diagnostic compounds in human plasma samples, is becoming realistic. After 1) a rapid overview of data treatment for the discovery of unexpected compounds found in test/patient plasma samples, we will review how to 2) identify compounds and 3) rate the confidence of their identification. For this last part, we will base our analysis on 2 key documents: 1) Commission decision 2002/657/EC and 2) Sumner LW et al; Metabolomics, 2007, which are dedicated for screening and metabolomics labs, respectively.

Long Abstract

Many authors have shown that high-resolution (HRMS) instruments have an excellent quantitative performance, which underscores that HRMS are crucial instruments for research and routine analysis. HRMS instruments can perform sensitive and reliable quantification of a large variety of analytes in HR-full scan mode. Interestingly, HR-full scan acquisition records virtually all ionized molecules and allows post-acquisition data mining.

Whereas the great majority of clinical LC-MS analyses targets one or “a few” compounds, the HR-full scan records a global view that allows pushing forward the extraction of information from plasma samples. Pollutants, illegal substances, drugs at toxic levels or very low/high endogenous metabolites can be revealed even when unexpected. This global analysis could be of interest for MDs when patient status is unclear and unusual.

We will describe shortly how to detect all ions in a plasma samples using straight-forward software procedure and how to reveal unexpected compounds in test samples in comparison to controls (e.g. patient vs healthy).

Then, based on LC-HRMS data, we will review how to identify compounds of interest in test/patients samples. Parameters like retention time, mass accuracy of precursor ions, MSn product ions, relative isotopic abundance, fine isotopic distribution, additional information related, etc. will be presented. The vision in metabolomics as well as screening labs (food safety, forensic etc.) about compound identification will be discussed.

Eventually, we will evaluate how bioanalysts can rate the confidence of the identification. We will start our evaluation from two documents: 1) Commission decision 2002/657/EC and 2) Sumner LW et al; Metabolomics, 2007 that are dedicated for screening and metabolomics labs, respectively.

Concrete examples will be given from the discovery of compounds of interest in test plasma samples, their identification and the rating of the identification confidence at the light of HRMS data.


References & Acknowledgements:

-


Financial Disclosure

DescriptionY/NSource
Grantsno
Salaryno
Board MemberyesMSACL
Stockno
Expensesno

IP Royalty: no

Planning to mention or discuss specific products or technology of the company(ies) listed above:

no