MSACL 2025 Abstract

Quantitation of Perfluoroalkyl Compounds in Human Serum/Plasma by LC/MS/MS

Donna Smith, Melissa Hughs, Gwendolyn A. McMillin, Jessica M. Boyd,
ARUP Institute for Clinical and Experimental Pathology, Salt Lake City, Utah, USA

Donna Smith, BS.Chemistry, MS. Geoscience (Presenter) ARUP laboratories

Presenter Bio: My education is in Chemistry (B.S. Northern Arizona University) and Geochemistry (M.S. The University of Montana). I spent my early career in research and monitoring of fate and transport of environmental contaminants using field and mass spectrometry techniques. This led to a career as an HPLC/ Mass Spectrometry instrumentation specialist in service and applications. I am now part of a clinical mass spec research and development group, working on mass spectrometry techniques and clinical laboratory applications. It’s been exciting to combine my environmental background with the clinical aspects of PFAS contamination issues.

Relevant Financial Disclosures (within past 24 months, reported on Apr 24, 2025) No relevant financial relationship(s) to disclose.

INTRODUCTION:
Per- and Polyfluoroalkyl substances (PFAS) are a class of persistent organic pollutants which can cause adverse health effects including dyslipidemia and increased risk of developing certain types of cancer. The National Academies of Science (NAS) has provided clinical screening recommendations that sort patients into three risk categories according to a sum of concentrations PFAS analytes (MeFOSAA, PFHxS, PFOA, PFDA, PFUnDA, PFOS, and PFNA). We have developed a liquid chromatography tandem mass spectrometry (LC-MS/MS) method for quantitation of NAS PFAS panel analytes plus three additional PFAS analytes: Perflurooctanoic acid linear and branched isomers (n-PFOA and br-PFOA), Perfluorooctanesulfonic acid linear and branched isomers (n-PFOS and br-PFOS), Perflourohexanesulfonic acid (PFHxS), Perfluoronononic Acid (PFNA), Perfluorodecanoic acid (PFDA), Perfluoroundecanoic acid (PFUdA), Methyperfluorooctanesulfonamidoacetic acid (MeFOSAA). PFAS included in NHANEs panels: Perfluoroheptanoic acid (PFHpA), Perfluorododecanoic acid (PFDoA), Perfluorobutanesulfonic acid (PFBS)and novel PFAS compounds being clinically investigated: Perfluoropropionic acid (PFPrA) and Perfluorobutanoic acid (PFBA).

METHODS:
Protein is removed from serum or plasma samples by precipitation with 1% formic acid in acetonitrile. Supernatants are further cleaned up by filtration through a Captiva EMR-Lipid removal plate, eluent is evaporated and reconstituted prior to injection.
PFAS analytes are separated by reverse phase chromatography on a C18 column with a 16-minute gradient elution, mobile phases A was 5 mM ammonium acetate in water, mobile phase B was 5 mM ammonium acetate in 1:1 methanol:ACN. Analytes were detected using a Triple Quadrupole Mass Spectrometry in ESI negative mode. A C18 delay column was employed to reduce background PFAS interferences from the LC system and solvents. Each compound provided two fragment ions with the exception of PFBA, which only has a single transition. Isotopically labelled Internal Standards were employed to assess extraction efficacy, provide chromatographic confirmation and allow for quantitation based on compound response factor. Analytical measurement interval was 0.1 to 10 ng/mL for all analytes.

Preliminary validation experiments were performed. Standards were prepared from multiple PFAS sources at 0.15, 0.3 and 8 ng/mL and analyzed over 3 days (N=13) to assess accuracy and precision. External reference material NIST Standard Reference Material 1950 was analyzed over three days (N=5) to confirm accuracy. Seven patient pools were analyzed over two days (N=10) to assess precision.
A study of deidentified residual patient samples submitted for our Extended Lipid Panel is in progress. Patient ages range from eight to ninety-six years with a median age of fifty-five. Total cholesterol concentrations range from 78 to 416 mg/dL with a median concentration of 181 mg/dL.

RESULTS:
Accuracy in spiked samples was within 15% of target value for all compounds except br-PFOS. Precision in authentic patient pools and spiked samples was good with CV <20% for all compounds except br-PFOS. Results for NIST Standard Reference Material 1950 were within 30% of target value for all analytes,and were in good agreement with results from an external lab.

Thus far, we have tested 200 residual samples from ARUP’s lipid panel. Only 12% of the patients we tested were below the 2 ng/mL cutoff recommended by NAS as the threshold below which PFAS are unlikely to affect patient’s health. Most patients fell between 2 and 20 ng/mL indicating that they should receive some additional screening due to elevated PFAS exposure, and 6% of patients had an NAS-Sum value above 20 ng/mL indicating that PFAS exposure is likely to impact health.
The median cholesterol for the patients with NAS PFAS Sum concentration below 2 ng/mL was 153 mg/dL. For patients in the 2-20 ng/mL PFAS sum concentration range the median cholesterol was 187 mg/dL. Patients in the highest PFAS concentration category >20 ng/mL had a median cholesterol of 202 mg/dL, which is just above the Borderline High Cholesterol cutoff of 200 mg/dL.

DISCUSSION:
We have developed a robust LC-MS/MS assay for PFAS analytes including all NAS panel analytes plus three additional analytes. Background interferences were successfully mitigated, and we are confident that the method could be used to support clinical investigation of specimens collected with standard tube types.