Anaphylaxis is a hypersensitivity reaction that is rapid in onset and can be fatal. In recent years there has been a concerning increase in anaphylactic cases worldwide. Due to the limitations of currently used biomarkers of anaphylaxis such as narrow time collection window for blood and/or poor applicability to food-induced anaphylaxis, there is a need to assess new biomarkers of anaphylaxis. Urinary 11β-prostaglandinF2α (11β-PGF2α) and leukotrieneE4 (LTE4) were recently proposed as possible new biomarkers of anaphylaxis. However, these studies relied on immunoassay-based methods, which may not be able to distinguish all isomers, and small cohorts, thus requiring more rigorous follow-up studies. The objective of this project was to develop and validate a highly selective ultra-high-performance liquid chromatography-quadrupole time of flight (UHPLC-QTOF) method for the accurate analysis of urinary 11β-PGF2α and LTE4. 
C18 reversed-phase chromatography was used for the separation of isomers.  Mobile phase A was 0.02% acetic acid in water and mobile phase B was 0.02% acetic acid in acetonitrile/iso-propanol (90:10). The isocratic flow and shallow gradient of mobile phase B gave a baseline separation of PGF isomers and LTE4 isomers, respectively. Both positive and negative electrospray ionization (ESI) were tested for the ionization of LTE4. Negative ESI was selected based on the signal-to-noise ratio, which was 7x higher in negative ESI than positive ESI. 
Solid phase extraction (SPE) was selected for sample preparation to achieve sufficient enrichment to enable the measurement of low endogenous levels of these two putative biomarkers. Three different SPE sorbents were tested for recovery of 11β-PGF2α and LTE4: mixed-mode strong anion-exchange, mixed-mode weak anion-exchange and reversed-phase C18. C18 SPE was selected for the sample preparation based on the analyte recovery and minimum extraction of interferences. Separate SPE elution of 11β-PGF2α and LTE4 was chosen as the best strategy to minimize background signal and improve the limits of detection of LTE4. A forced degradation study of LTE4 showed that LTE4 is not stable at a pH below 4 and hence strict control of pH was required throughout the procedure. The use of 6 μL of 30% glycerol in method (v/v) improved the method precision by reducing analyte losses in non-specific adsorption. 
11β-PGF2α is also excreted as glucuronic acid conjugate in urine. Hence, to quantitate total endogenous 11β-PGF2α concentrations in urine, β-glucuronidase enzyme hydrolysis was used.
The method was validated for accuracy, precision, linearity, recovery, matrix effects and stability. The validation was performed using deuterated internal standards as surrogate analytes since both 11β-PGF2α and LTE4 are present endogenously. Intra-day accuracy for ranged from 90-111% and 86-108% for 11β-PGF2α-d4 and LTE4-d5, respectively.  The method showed good intra-day precision (% RSD = 4.0-7.4% and % RSD = 6.8-14.9% for 11β-PGF2α-d4 and LTE4-d5 respectively). The method was linear from 9.8-5000 pg/mL concentration of both analytes. Inter-day accuracies were 91-100% (% RSD = 4.2-8.4%) for 11β-PGF2α-d4 and 93-113% (% RSD = 4.9-8.8%) for LTE4-d5. The lower limit of quantitation of both analytes was 15 pg/mL. The recoveries were 92-95% and 77-83% and the matrix effects were 83-88% and 99-115% for 11β-PGF2α-d4 and LTE4-d5, respectively in pooled urine from six different individuals. Recovery and matrix effects were also evaluated in nine individual urine lots. The 11β-PGF2α-d4 recovery ranged from 70-116% and LTE4-d5 recovery ranged from 61-99% across nine urine lots. Matrix effects for 11β-PGF2α-d4 ranged from 69-106% and matrix effects for LTE4-d5 ranged from 87-126% across nine urine lots, suggesting that slight matrix effects were present in individual samples as compared to matrix effects in pooled urine. 11β-PGF2α-d4 was stable for up to three freeze-thaw cycles whereas LTE4 was stable only up to one freeze-thaw cycle indicating that urine samples should be aliquoted and free-thaw cycles should be avoided for LTE4 quantitation. Both analytes were stable in the freezer for up to three-week period.
Overall, this novel validated and highly selective LC-MS method for urinary 11β-PGF2α and LTE4 can be used for anaphylaxis biomarker analysis in clinical studies and addresses some of the key shortcomings in other LC-MS assays reported in literature such as low LTE4 recovery and lack of data on isomer separation.