CLB-09236; No. of pages: 4; 4C: Clinical Biochemistry xxx (2016) xxx–xxx Contents lists available at ScienceDirect Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem Short Communication Analytical and clinical validation of an LC–MS/MS method for urine leukotriene E4: A marker of systemic mastocytosis Alan J. Lueke a, Jeffrey W. Meeusen a,⁎, Leslie J. Donato a, Amber V. Gray a, J.H. Butterfield b,c, Amy K. Saenger a a b c Department of Laboratory Medicine & Pathology, Mayo Clinic, Rochester, MN, United States Department of Medicine, Division of Allergic Diseases, Mayo Clinic, Rochester, MN, United States Mayo Clinic Program for Mast Cell and Eosinophil Disorders, Mayo Clinic, Rochester, MN, United States a r t i c l e i n f o Article history: Received 31 December 2015 Received in revised form 15 February 2016 Accepted 16 February 2016 Available online xxxx Keywords: Allergic disease Mast cell disorder Arachidonic acids Prostaglandin Histamine Tryptase Mass spectrometry Method development a b s t r a c t Objectives: Systemic mastocytosis (SM) is a disorder characterized by the excessive accumulation of clonally derived mast cells in various tissues. When triggered, mast cells release large amounts of histamine, prostaglandins and leukotrienes. Leukotriene E4 (LTE4) is the primary stable metabolite of total cysteinyl leukotrienes. We hypothesized that secretion of LTE4 would be increased in SM and could be used alone or in combination with current urinary biomarkers to optimize screening for SM. Design and methods: LTE4 was measured by liquid chromatography followed by tandem mass spectrometry (LC–MS/MS). Analytical assay validation was performed using residual urine specimens. LTE4 results were normalized to urine creatinine for clinical use. Reference interval was established using a healthy volunteer cohort. Clinical sensitivity and specificity for SM detection were determined by measuring urinary biomarkers (LTE4, Nmethyl histamine [NMH] and 11β-prostaglandin F2α [BPG]) in a cohort of 409 patients referred to allergy specialists, 66 (16%) of which were diagnosed with SM. Results: Urinary LTE4 measurement was accurate, precise and linear across a range of 31–3020 pg/mL. The 95th percentile of the reference interval population was b 104 pg/mg creatinine. Median urine LTE4 concentrations were significantly higher among patients with SM (97 pg/mg cr. vs. 50 pg/mg cr.; p b 0.01). Elevated urinary LTE4 was 48% sensitive and 84% specific for SM. Clinical sensitivity was 53% for BPG (N 1000 ng/mL) and 71% for NMH (N 200 ng/mL). Incorporating all three urinary metabolites improved the SM diagnostic sensitivity to 97%, with minimal change in specificity. Conclusions: We have developed a sensitive and precise LC–MS/MS assay for quantitation of LTE4 in urine. Incorporating LTE4 into a panel including BPG and NMH provides a much-needed screening tool for a complicated disease with non-specific symptoms and invasive confirmatory testing. © 2016 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. 1. Introduction Systemic mastocytosis (SM) is a disorder characterized by the excessive accumulation of clonally derived mast cells in various tissues. When triggered, mast cells release large amounts of tryptase, histamine, prostaglandins and leukotrienes. This release of signal molecules causes intermittent “spells” with varying symptoms that may include itching, flushing, lightheadedness, tachycardia, gastrointestinal distress, or even loss of consciousness. Diagnostic criteria established by the World Health Organization (WHO) recommend a bone marrow biopsy, specialized cytology studies or genetic testing [1]. Urine concentrations of N-methyl histamine (NMH) and 11-beta prostaglandin F2α (BPG), which are metabolites of mast cell derived histamine and prostaglandin, have been used to aid in screening and reduce unnecessary biopsies [2–4]. Cysteinyl leukotrienes are another ⁎ Corresponding author at: 200 First St. SW, Rochester, MN 55905, United States. E-mail address: meeusen.jeffrey@mayo.edu (J.W. Meeusen). class of mast cell secretory molecules and potent inflammatory mediators. Leukotriene E4 (LTE4) is the primary stable metabolite of total cysteinyl leukotrienes [5]. Concentrations of LTE4 are low in circulation but accumulate in the urine. We hypothesized that urinary LTE4 could be used alone or in combination with NMH and BPG to optimize screening for SM. Here we describe a novel (LC–MS/MS) assay to accurately and precisely quantitate LTE4 in urine and outline its clinical utility in SM screening. 2. Patients and methods 2.1. Study populations All patient data were accessed in compliance with the Institutional Review Board. A normal reference population of 128 apparently healthy donors (64 men and 64 women) with a mean age of 43 (SD 14) years was recruited. Patients were self-reported as not having allergic disease or immunologic disorders and had not taken any antihistamines, non- http://dx.doi.org/10.1016/j.clinbiochem.2016.02.007 0009-9120/© 2016 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Please cite this article as: A.J. Lueke, et al., Analytical and clinical validation of an LC–MS/MS method for urine leukotriene E4: A marker of systemic mastocytosis, Clin Biochem (2016), http://dx.doi.org/10.1016/j.clinbiochem.2016.02.007 2 A.J. Lueke et al. / Clinical Biochemistry xxx (2016) xxx–xxx steroidal anti-inflammatories, cyclooxygenase inhibitors, or 5lipoxygenase inhibitors within 2 weeks. Clinical performance was determined among a retrospective cohort of consecutive patients with clinically ordered 24-hour urine NMH measurement. Patients b 18 years of age, solid organ transplant recipients or currently pregnant were excluded. A total of 409 patients were enrolled between April 26th, 2012 and March 21st, 2013. An allergy specialist evaluated all patients and final diagnoses were adjudicated by a chart review. The WHO criteria for the diagnosis of SM include the presence of the major criterion (the presence of multifocal dense infiltrates of mast cells (MC) in tryptase-stained biopsy sections of the bone marrow or of another extracutaneous organ) plus one minor criterion (more than 25% of MC show abnormal morphology; the presence of KIT Asp816Val mutation, abnormal mast cell phenotype indicated by the presence of CD25 on MC; serum total tryptase N20 ng/mL). Alternatively, the presence of three minor criteria will satisfy the requirements for the diagnosis [6]. Detailed diagnoses and medication usage are included in Supplemental Table 1. 2.2. LTE4 method D5-labeled internal standard (D5-LTE4) was added to waste urine specimens, controls, and standards, followed by the addition of acetonitrile (ACN) to precipitate any excess salts. Precipitate was removed using positive pressure filtration through a 0.2 micron 96 well PTFE filter plate prior to injection of 45 μL via a CTC Pal autosampler and onto a Turboflow MAX mixed-mode anion exchange column (0.5 × 50 mm, Thermo Fisher Scientific). Following elution with Methanol(MeOH)/ ammonium hydroxide, LTE4 was further chromatographically separated on a C8 2.5-μm analytical column (Waters Xbridge, 2.1 × 50 mm, 60 °C) using a H2O/MeOH mobile phase with ammonium hydroxide as the modifier. LTE4 was monitored in negative MRM mode (AB Sciex API 5000 MS/MS). Detailed parameters are listed in Supplemental Table 2. The total analysis time was 10.3 min. All LTE4 concentrations were normalized to creatinine (enzymatic method, Roche Diagnostics). SI unit conversions are 0.25 pg LTE4/mg creatinine = 1.0 pmol LTE4/mmol creatinine, 0.90 ng NMH/mg creatinine = 1.0 nmol/mmol creatinine, and 2.82 ng BPG/24 h = 1.0 pmol BPG/24 h. 2.3. NMH and BPG analytical methods NMH was isolated from urine specimens by solid phase extraction and measured by LC–MS/MS using a stable isotope labeled internal D3-N-methyl histamine. BPG was measured by competitive ELISA (Cayman Chemicals, Ann Arbor, MI). The method uses an acetylcholinesterase-linked BPG tracer, which competes with patient BPG for a limited number of rabbit-antibody binding sites. The unbound BPG is then washed away, and the remaining BPG tracer signal is inversely proportional to the concentration of BPG in the urine sample. 2.4. Method validation Accuracy of the clinical LC–MS/MS method for LTE4 was determined by spiking pooled urine with known amounts of purified LTE4 between 100 to 2000 pg/mL. Average recovery was 111% (range 99–120%). Assay precision was determined by repeat analysis of three urine pools with average LTE4 concentrations of 44, 445 and 1378 pg/mL over 20 days. Analytical sensitivity was determined by repeat injection (n = 20) of a charcoal stripped urine pool and a mobile phase blank. Linearity was assessed by mixing urine specimens with high and low LTE4 concentrations. Three mixing studies were performed on three different days for a total of nine unique experiments (Fig. 1). Specimen stability was determined by measuring 10 specimens immediately after collection and at 1, 3, 7, 14, 30, and 90 days stored ambient, refrigerated, or frozen Fig. 1. Analytical performance of a novel liquid chromatography tandem mass spectrometry (LC–MS/MS) method for urinary leukotriene E4 measurement. A) Linearity was demonstrated by mixing urine samples with high and low concentrations of LTE4 (slope 0.9929; R2 0.9985). B) LTE4 values were measured in 128 healthy donors (67 males, closed circles; 61 females, open circles); the 95th percentile value was 104 pg/mg cr. (dotted line). (−20 °C). Analytical specificity was assessed by spiking a urine sample with 11-trans LTE4. 2.5. Statistical analyses Statistical analyses were performed using JMP software (SAS Inc.; Cary, NC). Relationships of the appropriate percentiles with age and sex were evaluated using quantile regression by minimizing an asymmetrically weighted sum of absolute errors; 95th percentile was established using a smoothed empirical likelihood quantile regression. Receiver operator characteristic (ROC) curve analysis was used to identify the optimal diagnostic cutoff and the discrimination c-statistic for each biomarker. Linearity was assessed by linear regression and Passing–Bablok regression. Significant relationships were defined as a pvalue b0.05. Please cite this article as: A.J. Lueke, et al., Analytical and clinical validation of an LC–MS/MS method for urine leukotriene E4: A marker of systemic mastocytosis, Clin Biochem (2016), http://dx.doi.org/10.1016/j.clinbiochem.2016.02.007 A.J. Lueke et al. / Clinical Biochemistry xxx (2016) xxx–xxx 3 3. Results 3.3. LTE4 as a clinical biomarker of SM 3.1. LTE4 analytical method accuracy, precision, linearity, and stability The clinical utility of urinary LTE4 as a biomarker of SM was determined in a cohort of 409 patients referred for allergic disease evaluations. The median LTE4 concentration in the entire cohort was 55 pg/mg creatinine (IQR 33–90). Chart review identified 66 patients (16%) as being positive for SM. The most common alternate diagnoses included spells of unknown etiology (21%), idiopathic angioedema (11%), chronic urticaria (9%), idiopathic anaphylaxis (7%), reaction to an identified allergen (6%), irritable bowel syndrome (4%), and eosinophilia (3%; Supplemental Table 1). The median urine LTE4 among patients with SM was 97 pg/mg creatinine, significantly higher than patients without SM (50 pg/mg creatinine; p b 0.01; Table 2). The only other diagnoses with elevated median LTE4 were eosinophilia (119 pg/mg cr.; n = 12), monoclonal mast cell activation syndrome (119 pg/mg cr.; n = 3), and aleukemic mast cell leukemia (630 pg/mg cr.; n = 2); however the small number of patients with these diagnoses prevented statistical comparisons. NMH and BPG were also significantly elevated among SM patients (Table 2). Receiver-operating-characteristic (ROC) analysis identified an optimal urine LTE4 cutoff of 84 pg/mg creatinine with an area under the curve (AUC) of 0.705 (Table 2). ROC curve analysis for NMH and BPG in this population identified AUCs of 0.853 and 0.649, respectively. The reference interval cutoff of 104 pg/mg creatinine for LTE4 was 48% sensitive and 84% specific for a diagnosis of SM. Clinical sensitivity was 45% for BPG (N 1000 ng/mL) and 73% for NMH (N 200 ng/mL). Sensitivity improved to 86% with a specificity of 68% when BPG and NMH were both considered. Adding LTE4 improved the SM diagnostic sensitivity to 97% with a specificity of 61%. The method was accurate and linear between 31 and 3154 pg/mL LTE4 (slope 0.9929; R2 0.9985), and imprecision was b 10% (Table 1). Specimens were stable 24 h at room temperature, 7 days at 4 °C, and 30 days at −20 °C. Samples were unaffected by use of toluene, acetic acid, boric acid or sodium carbonate 24 hour collection preservatives. The limit of detection was determined to be 2 pg/mL and the limit of quantitation was 8 pg/mL. Spiking with 2000 pg/mL 11-trans-LTE4 increased measured LTE4 concentrations by 1730 pg/mL. 3.2. Urine LTE4 values in a healthy population The distribution of urinary LTE4 concentrations was compared between 46 paired 24-h and random urine collections. No significant difference in LTE4 concentration was observed between the random (mean ± SD; 99 ± 61 pg/mL) and 24-h (109 ± 67 pg/mL) urine samples. LTE4 values observed in a healthy population (n = 128) were determined using a random urine collection and values normalized to creatinine. Urinary LTE4 concentrations were not significantly altered by age or gender (Fig. 1). The median LTE4 value was 42 pg/mg creatinine (IQR 31–62) and the 95th percentile was 104 pg/mg creatinine (95CI 63–145). The 95th percentile values for NMH and BPG in the same cohort were 194 ng/mg creatinine (95CI 164–224) and 1118 pg/mg creatinine (95CI 892–1342), respectively. 4. Discussion Table 1 Analytical performance characteristics of the LTE4 LC–MS/MS method. Precision Mean (pg/mL)/SD/%CV Within run Pool 1 Pool 2 Pool 3 41 631 1452 Accuracy Between run 2.0 21 38 5.0% 3.4% 2.6% 44 445 1380 3.3 31 112 7.6% 6.9% 8.2% Recovery; mean (SD) Range Slope; R2 111% (9%) 100–2000 pg/mL 1.128; 0.994 2 Linearity Range (pg/mL) Slope; R % difference; mean (SD) Mixing study 1 Mixing study 2 Mixing study 3 33–3154 53–2630 31–3020 0.979; 0.999 1.006; 0.998 0.999; 0.999 −1.2% (2.9%) −2.6% (4.4%) −1.2% (4.5%) Analytical sensitivity Mean SD %CV Limit of quantitation 8.2 0.9 11% Stability; % difference (mean ± SD) 1 day 3 days 7 days 90 days Ambient Refrigerate Frozen −3.7 ± 11% −0.8 ± 7.6% −2.6 ± 12% −15 ± 18% 4.0 ± 15% 8.4 ± 17% −38 ± 17% 0.5 ± 14.7% 7.0 ± 19% n.p. n.p. −1.8 ± 17% 24 h collection preservative 1 day Slope; R2 (mean difference) 7 days Slope; R2 (mean difference) Toluene Acetic acid Boric acid Sodium carbonate 0.926; 0.962 (−3.3 pg/mL) 0.959; 0.984 (−7.1 pg/mL) 1.069; 0.982 (−0.7 pg/mL) 1.066; 0.984 (1.7 pg/mL) 0.911; 0.952 (−2.9 pg/mL) 0.962; 0.988 (−6.2 pg/mL) 0.914; 0.978 (−2.0 pg/mL) 1.011; 0.988 (1.5 pg/mL) Reference interval N Median (95CI) 95th percentile (95CI) Overall Female Male 128 61 67 42 pg/mg cr. (38–48) 47 pg/mg cr. (41–57) 39 pg/mg cr. (34–45) 104 pg/mg cr. (85–145) 109 pg/mg cr. (86–158) 102 pg/mg cr. (74–153) To convert LTE4 from pg/mL to pmol/L multiply by 2.27; to convert from pg/mg cr. to pmol/mmol cr. multiply by 0.25. We have developed a high-throughput LC–MS/MS method to measure urinary LTE4 with suitable accuracy, precision, and stability for clinical applications. Using a population of apparently healthy normal volunteers and normalizing the analyte concentration to concentration of creatinine we established a urinary LTE4 reference value of 104 pg/mg creatinine and confirmed previously established reference ranges of 200 ng/mg creatinine (NMH) and 1000 pg/24 h (BPG). Furthermore, in a cohort of N400 patients presenting with suspicion of mast cell disease, elevated urinary LTE4 was highly specific at identifying patients with SM. SM is a rare and complicated disease that manifests with a variety of non-specific symptoms and sequelae depending on the site of mast cell accumulation [7,8]. Furthermore, SM signs and symptoms may be indolent or intermittent depending on the activity of the aberrant mast cells [9,10]. A weakness of the current study is lack of serum tryptase data. Mast cells store histamine and tryptase within intracellular granules that are released on activation. Conversely, arachidonic acid metabolites such as leukotrienes and prostaglandins are synthesized de novo by mast cells. Thus, urine NMH and serum tryptase may be indicators of activation, while LTE4 and BPG may allow for better estimation of overall mast-cell burden. Elevations of urinary leukotrienes among patients with SM have been previously reported in small patient cohorts (n = 9 to 25) using immunoassays. Leukotriene E5, 11-trans-LTE4, and N-acetyl-LTE4 have all been reported as interfering compounds that cross-react to varying degrees with antibodies used in these methods [11,12]. Of these, only 11-trans-LTE4 has a mass identical to LTE4, and therefore is measured as LTE4 by LC–MS/MS. However, it has been shown that the biological activities of these two LTE4 isoforms are similar [13]. In previous reports, and in our study, there was an overlap of urine LTE4 concentrations between cases and controls. A reference interval has not been previously described and it is worth noting that the median LTE4 concentration among patients was within the reference interval despite exclusion of patients taking 5-lipoxygenase inhibitors. Please cite this article as: A.J. Lueke, et al., Analytical and clinical validation of an LC–MS/MS method for urine leukotriene E4: A marker of systemic mastocytosis, Clin Biochem (2016), http://dx.doi.org/10.1016/j.clinbiochem.2016.02.007 4 A.J. Lueke et al. / Clinical Biochemistry xxx (2016) xxx–xxx Table 2 Clinical performance of urinary LTE4 and other urine biomarkers among 409 allergy patients presenting with symptoms suggestive of systemic mastocytosis (SM). Systemic mastocytosis Biomarker No (n = 343) Yes (n = 66) p-Value Sensitivity, % (95CI) Specificity, % (95CI) PPV, % (95CI) NPV, % (95CI) c-Statistic LTE4, pg/mg cr.; median (IQR) BPG, pg/24 h; median (IQR) NMH, ng/mg cr.; median (IQR) NMH or BPG Any marker 50 (30–80) 624 (358–987) 124 (100–163) N/A N/A 97 (53–177) 1046 (469–2711) 290 (175–527) N/A N/A b0.01 b0.01 b0.01 N/A N/A 48 (37–60) 45 (32–55) 73 (59–80) 86 (76–93) 97 (89–99) 84 (80–87) 79 (73–82) 88 (86–92) 68 (63–73) 61 (56–66) 36 (27–47) 29 (22–38) 59 (53–74) 34 (28–42) 32 (26–39) 87 (85–92) 88 (85–92) 94 (91–96) 96 (93–98) 99 (97–99.9) 0.705 0.649 0.853 0.857 0.901 Clinical sensitivity, specificity, PPV, and NPV were determined using established normal 95th percentile cutoffs: LTE4 104 pg/mg creatinine (26 pmol/mmol creatinine), BPG 1000 pg/24 h (2.82 pmol/24 h), NMH 200 ng/mg creatinine (180 nmol/mmol creatinine). PPV = positive predictive value; NPV = negative predictive value; c-statistic = ROC curve area under the curve. A growing number of reports on the efficacy of leukotriene inhibitors, such as zileuton, add support to the pathological role of LTE4 in SM [14–16]. Clinical data on the presence of elevated urinary LTE4 may not only identify a screening marker for SM but may also justify therapies beyond anti-histamines in afflicted patients. 5. Conclusions In conclusion, we have developed a sensitive and precise LC–MS/MS assay for quantitation of urinary LTE4. Incorporating LTE4 into a panel including BPG and NMH provides a powerful screening tool for systemic mastocytosis, a disease of protean symptoms that is commonly overlooked due to its rarity. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2016.02.007. References [1] P. Valent, L. Escribano, S. Broesby-Olsen, K. Hartmann, C. Grattan, K. Brockow, et al., Proposed diagnostic algorithm for patients with suspected mastocytosis: a proposal of the european competence network on mastocytosis, Allergy 69 (2014) 1267–1274. [2] M. Castells, K.F. Austen, Mastocytosis: mediator-related signs and symptoms, Int. Arch. Allergy Immunol. 127 (2002) 147–152. [3] J.H. Butterfield, C.R. Weiler, Prevention of mast cell activation disorder-associated clinical sequelae of excessive prostaglandin D(2) production, Int. Arch. Allergy Immunol. 147 (2008) 338–343. [4] P. Valent, M. Arock, P. Bonadonna, K. Brockow, S. Broesby-Olsen, L. Escribano, et al., European competence network on mastocytosis (ecnm): 10-year jubilee, update, and future perspectives, Wien. Klin. Wochenschr. 124 (2012) 807–814. [5] N. Rabinovitch, Urinary leukotriene e4, Immunol. Allergy Clin. N. Am. 27 (2007) 651–664 (vii). [6] J. Gotlib, A. Pardanani, C. Akin, A. Reiter, T. George, O. Hermine, et al., International Working Group-Myeloproliferative Neoplasms Research and Treatment (IWGMRT) & European Competence Network on Mastocytosis (ECNM) consensus response criteria in advanced systemic mastocytosis, Blood 121 (2013) 2393–2401. [7] P. Bonadonna, C. Lombardo, R. Zanotti, Mastocytosis and allergic diseases, J. Investig. Allergol. Clin. Immunol. 24 (2014) 288–297 (quiz 3 p preceding 97). [8] C. Akin, P. Valent, Diagnostic criteria and classification of mastocytosis in 2014, Immunol. Allergy Clin. N. Am. 34 (2014) 207–218. [9] P. Valent, Mastocytosis: a paradigmatic example of a rare disease with complex biology and pathology, Am. J. Cancer Res. 3 (2013) 159–172. [10] M. Arock, C. Akin, O. Hermine, P. Valent, Current treatment options in patients with mastocytosis: status in 2015 and future perspectives, Eur. J. Haematol. 94 (2015) 474–490. [11] J.H. Butterfield, Increased leukotriene e4 excretion in systemic mastocytosis, Prostaglandins Other Lipid Mediat. 92 (2010) 73–76. [12] M. Raithel, Y. Zopf, S. Kimpel, A. Naegel, G.J. Molderings, F. Buchwald, et al., The measurement of leukotrienes in urine as diagnostic option in systemic mastocytosis, J. Physiol. Pharmacol. 62 (2011) 469–472. [13] K. Bernstrom, S. Hammarstrom, Metabolism of leukotriene d by porcine kidney, J. Biol. Chem. 256 (1981) 9579–9582. [14] J. Tolar, W.D. Tope, J.P. Neglia, Leukotriene-receptor inhibition for the treatment of systemic mastocytosis, N. Engl. J. Med. 350 (2004) 735–736. [15] P.J. Turner, A.S. Kemp, M. Rogers, S. Mehr, Refractory symptoms successfully treated with leukotriene inhibition in a child with systemic mastocytosis, Pediatr. Dermatol. 29 (2012) 222–223. [16] T. Gulen, H. Hagglund, S.E. Dahlen, B. Sander, B. Dahlen, G. Nilsson, Flushing, fatigue, and recurrent anaphylaxis: a delayed diagnosis of mastocytosis, Lancet 383 (2014) 1608. Please cite this article as: A.J. Lueke, et al., Analytical and clinical validation of an LC–MS/MS method for urine leukotriene E4: A marker of systemic mastocytosis, Clin Biochem (2016), http://dx.doi.org/10.1016/j.clinbiochem.2016.02.007