Blood spot versus plasma chitotriosidase: A systematic clinical comparison

Clinical Biochemistry 47 (2014) 38–43

Contents lists available at ScienceDirect

Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem

Blood spot versus plasma chitotriosidase: A systematic clinical comparison Mohamed A. Elmonem a,⁎, Dalia I. Ramadan a, Marianne S.M. Issac a, Laila A. Selim b, Sara M. Elkateb a a b

Department of Clinical and Chemical Pathology, Faculty of Medicine, Cairo University, Cairo, Egypt Department of Pediatrics, Faculty of Medicine, Cairo University, Cairo, Egypt

a r t i c l e

i n f o

Article history: Received 23 June 2013 Received in revised form 18 September 2013 Accepted 20 October 2013 Available online 29 October 2013 Keywords: Chitotriosidase Dried blood spot Lysosomal storage disorder Clinical agreement

a b s t r a c t Objectives: This study aimed to evaluate the agreement between blood spot and plasma chitotriosidase using the economic substrate 4-methylumbelliferyl-β-D-N,N′,N″-triacetylchitotrioside, and to investigate the utility of the blood spot assay for the wide scale screening for lysosomal storage disorders among the clinically suspected. Design and methods: Blinded blood spot samples were compared with the corresponding plasma levels in 199 children (56 with confirmed diagnoses of ten different lysosomal storage disorders, 73 normal controls and 70 pathological controls). Several performance criteria (limit of detection, linearity, within-run and dayto-day precision and sample stability) were also evaluated. Results: Plasma assay performed better by most criteria; however, blood spot performance was quite satisfactory. Quantitative values of the two methods can't be used interchangeably based on their 95% limits of agreement. Diagnostic sensitivity and specificity derived from ROC curves were 75.0 and 85.3% for the plasma assay and 71.4 and 79.0% for the blood spot assay, respectively. Cohen's kappa was 0.72 (95% CI: 0.616–0.821) denoting a good categorical agreement between the two methods. Conclusion: The clinical use of blood spot chitotriosidase for the screening of lysosomal storage disorders can be quite practical, provided proper cut-off values are determined for each lab. © 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.

Introduction The chitotriosidase enzyme was the first discovered chitinase in humans. It was markedly elevated in plasma of patients with Gaucher's disease, a relatively common lysosomal storage disorder (LSD) [1]. Being produced exclusively by activated macrophages and polymorphonuclear leucocytes raised suspicion of its involvement in innate physiological immunity against chitin containing pathogens [2]. After initial discovery, its elevation was detected in many other LSDs [3–5], and many other non-lysosomal disorders; e.g. coronary heart disease [6], multiple sclerosis [7], β-thalassemia [8] and different types of infections [9,10]. Chitotriosidase activity in plasma is also established as a therapeutic monitor for enzyme replacement therapy in Gaucher's patients [11] and is under consideration to monitor therapy in other diseases such as Fabry [12], nephropathic cystinosis [13] and sarcoidosis [14]. Abbreviations: DBS, dried blood spot; GM1, gangliosidosis M1; LSDs, lysosomal storage disorders; MLD, metachromatic leukodystrophy; MPS, mucopolysaccharidosis; 4-MU, 4methylumbelliferone; 4-MU-C3, 4-methylumbelliferyl-β-D-N,N′,N″-triacetylchitotrioside; 4-MU-dC2, 4-methylumbelliferyl-deoxychitobiose. ⁎ Corresponding author at: Inherited Metabolic Disorder Laboratory (IMDL), Department of Clinical and Chemical Pathology, Faculty of Medicine, Cairo University, 2 Ali Pasha Ibrahim Street, Center of Social and Preventive Medicine, Room 409, Monira, Cairo 11628, Egypt. E-mail address: [email protected] (M.A. Elmonem).

Dried blood spot (DBS) on filter paper is a minimally invasive method for obtaining blood samples. In comparison with venipuncture, it is relatively easy and has a much lower cost for sample collection, transport and storage [15]. This is especially important in developing countries where specialized metabolic labs are rare and financial aspect is always an obstacle. A method for assaying chitotriosidase activity in blood spots was first described using the substrate 4-methylumbelliferyl-β-DN,N′,N″-triacetylchitotrioside (4-MU-C3) [16]. Another substrate; 4-methylumbelliferyl-deoxychitobiose (4-MU-dC2), was later developed to avoid a flaw in the former one whereby, at optimal substrate concentration, glycolytic enzyme activity on 4-MU-C3 is hindered by its transglycosidase activity; the new substrate is not susceptible to this effect [17]. However, so far, 4-MU-dC2 is extremely expensive and not easily available; thus its utility in clinical practice is greatly minimized. A limited comparison between blood spot and plasma chitotriosidase was previously conducted with 14 Gauchers patients and 12 healthy control subjects [18]. However, to our knowledge, no systematic comparison has been made. In this study, we tried to draw a full comparison between the two sample types by evaluating several performance criteria and a clinical comparison in 199 children, including some with LSDs as well as pathological and normal controls, seeking the level of agreement between the two methods with the clinically available substrate 4-MU-C3.

0009-9120/$ – see front matter © 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clinbiochem.2013.10.024

M.A. Elmonem et al. / Clinical Biochemistry 47 (2014) 38–43

Materials and methods Subjects Subjects included in this study were recruited from children attending different outpatient clinics at the Center of Social and Preventive Medicine (CSPM) and Abou Alreesh Children Hospital, Cairo University, Egypt. Over the period from December 2009 till August 2012, a total of 199 subjects were recruited: 56 children (39 males, 5.60 ± 3.55 years) with confirmed diagnoses of 10 different LSDs, diagnosed at the Inherited Metabolic Disorder Laboratory (IMDL) at the CSPM, and 143 matched control children; 73 (45 males, 5.30 ± 3.60 years) with minor nonrelated complaints and normal routine laboratory results were considered as normal controls and 70 (44 males, 5.60 ± 4.40 years) with preliminary symptoms that were confusing with some LSDs, but diagnosed otherwise. The latter group was considered as pathological controls and was further divided according to major complaint into three subgroups: 31 with neurological manifestations, 22 with hepatosplenomegaly and 17 with idiopathic cardiomyopathy after excluding cardiomyopathycausing LSDs (Table 1). Another 23 LSD patients, diagnosed at our laboratory during the study period, were excluded due to unavailability of either plasma or blood spot samples. The study was approved by the institutional review board at Cairo University Children's Hospital, and written informed consents were obtained from parents of patients. Samples One to two milliliter of blood was collected from each subject, mixed appropriately with EDTA, and then 50 μL aliquots were spotted on Whatman 903 filter paper (Whatman Inc. USA), while the remainder was centrifuged for plasma collection. Blood spot cards were left at room temperature until fully dried, then put in tightly sealed aluminum bags and kept at −80 °C together with plasma samples till time of analysis. Blood spots were later coded and blinded from the operator who also performed the plasma assay. Tests were assayed in duplicate for both sample types.

39

Chitotriosidase in blood spots was measured according to Chamoles et al. [16]. A 3.2 mm dried blood spot punch was mixed with 20 μL of 0.19 mmol/L of 4-MU-C3 in distilled water and 20 μL of 0.5 mol/L acetate buffer, pH 5.0. Incubation was performed at 37 °C for 30 min and the reaction was stopped with 300 μL of 0.5 mol/L Carbonate/Bicarbonate buffer, pH 10.7. Fluorescence activity was measured by FP 6200, Jasco, Tokyo, Japan; Excitation: 365 nm, Emission: 448 nm. Enzyme activities were calculated based on a calibration curve of 4-methylumbelliferone (4-MU) for each assay. Plasma samples exceeding determined linearity were diluted in Citrate/Phosphate buffer; while blood spot dilution was performed by manipulating the acetate buffer volume. Performance criteria Limit of detection Determination of the Limit of detection was conducted through assaying 5 blank samples and 5 low-level samples in duplicate over 5 analytical days following the non-parametric approach [19]. Blank samples were taken from suspected enzyme deficient individuals, and low-level samples were below 10 nmol/mL/h for both sample types. Linearity Linearity curves were constructed by serial dilution of both sample types obtained from a Gaucher's patient and plotting the measured against the expected concentration of each dilution. All levels were tested in duplicate and the average was taken. Linearity experiments were repeated twice on two separate days. Precision Within-run and day-to-day precision studies were performed for both assays through the analysis of 3 concentration levels (Level 1: normal, Level 2: mildly elevated and Level 3: markedly elevated). Each level was assayed in duplicate for 20 analytical days over a 2 month period. Within-run and day-to-day precision values were evaluated based on the duplicates and first replicates of each day, respectively [20].

Chitotriosidase assays Chitotriosidase in plasma was measured based on the method of Hollak et al. [1]. Ten microliter of plasma was mixed with 100 μL of 0.022 mmol/L 4-MU-C3 (Sigma) in Citrate/Phosphate buffer, 0.1/0.2 mol/L, pH 5.2 and incubated at 37 °C for 15 min. The reaction was stopped with 2 mL of 0.5 mol/L Carbonate/Bicarbonate buffer, pH 10.7.

Sample stability Evaluation of sample stability was done by aliquoting plasma and blood spots of a normal control. Aliquots were preserved at different temperatures: 37 °C, room temperature (RT: 20–25 °C), 4 °C and −80 °C, and measured repeatedly in duplicate over a 4 month period. Any plasma or blood spot sample crossing the limits determined by its base line level ±2 standard deviations of the corresponding sample in

Table 1 Study subjects and levels of chitotriosidase enzyme assays. n

Plasma chitotriosidase ª

Normal controls Pathological controls Neurological Hepatosplenomegaly Cardiomyopathy Patients Gaucher Maroteaux-Lamy (MPS VI) Metachromatic leucodystrophy (MLD) Hurler–Scheie (MPS I) Morquio (MPS IV) Pompe San Filippo (MPS III) Fabry Niemann–Pick A/B Gangliosidosis M1 (GM1) ª

73 70 31 22 17 56 21 8 5 5 5 4 4 2 1 1

Blood spot chitotriosidase

(High/total)

Median (range), nmol/mL plasma/h

(High/total)ª

Median (range), nmol/mL blood/h

3/73 18/70 6/31 10/22 2/17 42/56 21/21 4/8 4/5 2/5 4/5 2/4 1/4 2/2 1/1 1/1

13.2 (0–72.3) 15.6 (0–1059) 13.6 (0–92.4) 32.0 (0.9–1059) 7.0 (0–45.3) 116 (0–8156) 2786 (1235–8156) 30.5 (0–118) 147 (18.1–277) 9.7 (8.8–38.4) 63.5 (7.8–115) 31.8 (0–129) 8.8 (0.2–67.2) 56.5, 53.1 837 172

6/73 24/70 7/31 13/22 4/17 40/56 21/21 4/8 4/5 1/5 4/5 1/4 1/4 2/2 1/1 1/1

9.7 (0–51.7) 16.1 (0–974) 14.9 (0–57.2) 39.5 (0–974) 9.0 (0–94.3) 64.3 (0–4410) 1584 (476–4410) 15.6 (0–62.7) 35.1 (21.5–139) 8.0 (4.5–31.1) 27.2 (12.7–102) 11.5 (0–25.7) 1.9 (0–23.6) 39.3, 29.3 675 90.5

Chitotriosidase (high/total) was based on cut-off values determined by optimal sensitivity and specificity derived from ROC curves.

40

M.A. Elmonem et al. / Clinical Biochemistry 47 (2014) 38–43

the day-to-day precision profile for at least two successive measurements was considered unstable.

Table 2 Precision profile of plasma and blood spot chitotriosidase. Level

Mean (nmol/mL/h)

Measures of agreement The 95% limits of agreement were used to evaluate the agreement of paired quantitative data between plasma and blood spot chitotriosidase. Logarithmic transformation of data was applied as advised by Bland and Altman when the analyte concentration range is extremely wide [21]. Within-method variability based on duplicate measurements of each data point was considered [22] and data points below detection limits in both assays were omitted to avoid erroneous log values. We are aware that our group of patients is not representative of all lysosomal disorders; however, receiver operating characteristic (ROC) curves were created for both sample types for the total LSD data and also separately for Gaucher's patients, just to establish the comparison on a more clinical ground. Cohen's kappa is another measure that was used to evaluate the categorical or clinical agreement between the two methods. Statistical analysis Performance criteria for each method and the 95% limits of agreement were estimated using Excel spreadsheets (Microsoft Office; 2007), while P values (Mann–Whitney U test), ROC curves and Cohen's Kappa were evaluated by WINPEPI software (version 11.9) [23]. The significance level was set at P b 0.05 based on a 2-sided test.

Plasma

Blood spot

Level 1 Level 2 Level 3 Level 1 Level 2 Level 3

14.1 72.7 2197 10.1 40.5 1724

CV% Within-run

Day-to-day

6.7% 4.4% 3.6% 9.0% 8.3% 9.9%

11.9% 8.7% 8.9% 12.6% 9.2% 13.1%

Sample stability Plasma chitotriosidase activity decreased rapidly at 37 °C, was stable for almost 30 days at RT and then decreased, and remained stable over the four month period at 4 °C and at −80 °C (Fig. 1A). Regarding blood spots, activity was maintained for ten days at 37 °C and then increased beyond the allowable limit. At RT and 4 °C, activity was stable for 33 and 40 days, respectively, and then increased. At −80 °C, activity was preserved for the whole study period (Fig. 1B). After crossing the allowable limit at above zero temperatures, blood spot values remained fluctuating around a 50% increase from their original base line for the remainder of the four months.

Sample size The sample size calculation was mainly targeting the agreement between the two methods. For quantitative agreement, a sample size of 128 was enough for a reasonable confidence interval (CI) to the 95% limits of agreement (±0.3 standard deviations of difference) [21]. Considering categorical agreement (Cohen's Kappa), a sample size of 187 was needed to ensure a 95% CI width less than 0.2 assuming Kappa is 0.8, the frequency of positives is 25% and the significance level is 5% [23]. A total of 199 subjects were recruited. Results Performance criteria Limit of detection Calculated values for limit of detection were very comparable between the two methods: 2.7 and 2.0 nmol/mL/h for the plasma and blood spot assays, respectively. Similarly, the determined limits of quantification were 3.6 nmol/mL/h for plasma and 3.5 nmol/mL/h for blood spots. At these levels; coefficients of variation (CVs%) based on a five-day experiment were 13.6% and 14.1%, respectively. Linearity Plasma offered a considerably better linearity limit (500 nmol/mL/h), compared to blood spots (150 nmol/mL/h). Actually, two approaches for diluting blood spots were evaluated. In the first, dilution was done during reaction by manipulating the volume of acetate buffer, and in the second it was done before reaction (the blood spot was incubated with acetate buffer for 30 min, the eluted fluid was diluted serially, and then incubated with the substrate for another 30 min). The first approach was adopted for all blood spot dilutions thereafter as it provided better sensitivity, linearity and shorter reaction time. Precision The plasma assay performed better regarding both the within-run and day-to-day precision for all the three levels; however, the obtained blood spot CVs% were also acceptable (Table 2).

Fig. 1. Graphic representation of sample stability study for plasma and blood spot chitotriosidase assays. Averages of duplicates were taken for each day, and both sample types belonged to the same control subject. A: Plasma, B: Blood spot, RT: Room temperature, SD: Standard deviation of the corresponding sample in the day-to-day precision profile.

M.A. Elmonem et al. / Clinical Biochemistry 47 (2014) 38–43

41

Fig. 2. Box and Whisker plot of chitotriosidase plasma and blood spot results of lysosomal patients compared to normal and pathological controls. Averages of duplicates were used for both sample types. PL: Plasma, BS: Blood spot, HSM: Hepatosplenomegaly, LSDs: Lysosomal storage disorders.

Results of chitotriosidase enzyme assays in different groups Fig. 2 provides a good idea about levels of chitotriosidase in different groups for both sample types. The 97.5th percentiles of normal controls for plasma and blood spot assays were 34 and 26 nmol/mL/h, respectively. The results of pathological controls with neurological manifestations and cardiomyopathy were insignificantly different from normal controls: P = 0.84 and 0.21, respectively, for the plasma assay, and P = 0.23 and 0.87, respectively, for the blood spot assay. However, the subgroup with hepatosplenomegaly was different: P b 0.001 for both assays. This subgroup included five patients ranging from 197 to 1059 nmol/mL plasma/h and from 102 to 974 nmol/mL blood/h. All five patients had a diagnosis of β-thalassemia. Gaucher's patients (21/21) gave results beyond the limits of linearity in both assays. After proper dilution, plasma levels ranged from 1235 to 8156 nmol/mL/h and blood spot levels ranged from 476 to 4410 nmol/mL/h (Fig. 2). When compared to normal and pathological controls, the P value was b 0.001 for both sample types. The highest level among the other 35 LSDs was attained by a Niemann–Pick patient, 837 and 675 nmol/mL/h in the plasma and blood spot assays, respectively. Plasma and blood spot samples from other LSD groups were significantly different from the corresponding normal controls: P b 0.001; however, in plasma other LSD group showed only a marginally significant difference from pathological controls: P = 0.044, while in the corresponding blood spot samples there was no significant difference: P = 0.49.

Agreement of quantitative data (95% limits of agreement) The mean differences between log values for normal controls, pathological controls and patients were 0.11, 0.02 and 0.19, respectively and 95% of differences lay between (−0.39 and 0.62), (−0.49 and 0.52) and (−0.41 and 0.8) for the three groups, respectively (Fig. 3A, B and C). When considering the whole data the mean difference was 0.116 and 95% of differences lay between −0.43 and 0.67. Obtaining the antilogs of the last three values we get 1.32, 0.37 and 4.6, which means that the average plasma to blood spot ratio was 1.32 and for 95% of

Fig. 3. Bland and Altman plots for Log values of plasma and blood spot chitotriosidase. A: normal controls, B: pathological controls, C: LSD patients. Data points below detection limits in both assays were omitted. Dashed line represents mean difference and dotted lines represent upper and lower 95% limits of agreement.

42

M.A. Elmonem et al. / Clinical Biochemistry 47 (2014) 38–43

Discussion

Fig. 4. ROC curves of plasma and blood spot chitotriosidase. A: LSDs versus normal and pathological controls, B: Gaucher's patients versus normal and pathological controls, dashed line represents AUC = 50%.

study subjects the plasma level was between 0.37 and 4.6 times the blood spot level, which is totally unacceptable.

Clinical agreement ROC curves The area under ROC curve (AUC) was higher in the plasma assay: 81.9% (95% CI: 73.9–89.8%), than in the blood spot assay: 76.3% (95% CI: 68.0–84.7%) (Fig. 4A). The difference was significant if the limit of acceptable change in AUC was set at 5%. Optimal cut-off points were 33 nmol/mL plasma/h and 23 nmol/mL blood/h, which were very comparable to the cut-off levels based on plasma and blood spot normal controls: 34 and 26 nmol/mL/h, respectively. Optimal sensitivity and specificity at the chosen cut-off points were 75.0 and 85.3% for the plasma assay and 71.4 and 79.0% for the blood spot assay, respectively. ROC curves were also created for the 21 Gaucher's patients against normal and pathological controls. Plasma chitotriosidase AUC was 99.8%, while blood spot AUC was 99.6% (Fig. 4B).

Cohen's kappa Based on Cohen's kappa, the clinical agreement between plasma and blood spot chitotriosidase is considered to be good (Kappa = 0.720, 95% CI: 0.616–0.821).

In the current study, we were seeking the level of agreement between blood spot and plasma chitotriosidase using 4-MU-C3 as a substrate. Our results show that blood spot chitotriosidase can be a dependable diagnostic tool, with satisfactory behavior under different performance criteria and good clinical agreement with corresponding plasma levels. Although the precision using blood spots was inferior to plasma, it was quite acceptable for a blood spot method, as all CVs% fell below 15%, which is our predetermined limit of acceptability. The calculated limit of detection for blood spots was quite acceptable, looking at the wide range of the analytical spectrum. The blood spot linearity limit was considerably lower when compared to plasma; however, proper dilution of the blinded blood spot samples gave relatively high concentrations comparable to the plasma samples for all 21 Gaucher's patients and for the highest results in the other groups as well. AUC for blood spot chitotriosidase was significantly inferior to plasma when considering the whole data; however, when ROC curves were created for Gaucher's disease only, plasma and blood spot chitotriosidase AUCs were almost the same. A kappa equal to 0.72 is considered substantial agreement, which implies that both methods are of approximate clinical value. The elevation of blood spot chitotriosidase activity over time in the sample stability study was unexpected, however, this might be explained by the fact that chitotriosidase is present in considerable amounts in blood macrophages and polymorphs. Over time, there might be a slow release of the enzyme from the still intact cells or lysosomal vesicles in blood spots, a process that is suppressed at subzero temperatures. The enzyme is also apparently extremely stable in the blood spot matrix, as it maintained its initial elevation for months. Of course this phenomenon has to be studied more elaborately with different concentration levels and for longer periods. Chamoles et al. [16] conducted a stability study for chitotriosidase in blood spots together with β-glucosidase and acid sphingomyelinase, and reported no significant change in the enzyme activities after storage for 21 days at −20, 4 or 25 °C; however, no specific details were given about chitotriosidase stability at higher temperatures or for longer periods. Rodrigues et al. [18] reported poor correlation between blood spot and plasma chitotriosidase in 14 Gaucher's patients using 4-MUC3 in a microplate assay (r = 0.67), while correlation in the same patients using 4-MU-dC2 was much better (r = 0.89). Actually, 4MU-C3 achieved very low values for blood spot chitotriosidase in their study (b 130 nmol/mL/h). We cannot speculate as to the exact cause of these low values; however, several studies using 4-MU-C3 have reported much higher blood spot chitotriosidase levels, especially for Gaucher's patients [16,24,25]. Due to its resistance to transglycosylation, the performance of 4MU-dC2 has been found to be superior in multiple studies [17,26,27]; however, the economic factor, so far, has prevented its clinical spread and the reasonable performance of 4-MU-C3 has maintained its use. From a rapid screening for original research articles in English indexed in PubMed in 2011 and 2012, among a total of 42 studies reporting the methodology for chitotriosidase, only four used 4-MU-dC2, while 38 used 4-MU-C3. Chamoles et al. [16] reported non-linear correlation between chitotriosidase activity in blood spots and in purified leucocytes; this was explained by the different enzyme concentrations present in plasma and blood cells. The same reasoning applies to the obvious disagreement in our quantitative data between blood spots and plasma, demonstrated by the 95% limits of agreement. This means that the numerical values of chitotriosidase from the two methods cannot be used interchangeably, especially for therapeutic monitoring purposes; however, it doesn't mean that either method can't be used in isolation. Of course, this needs further evaluation in follow up patients.

M.A. Elmonem et al. / Clinical Biochemistry 47 (2014) 38–43

Patients with hepatosplenomegaly had significantly different chitotriosidase values from normal controls and other types of pathological controls in both plasma and blood spots. Patients with β-thalassemia are the most distinguished among this group mainly because of iron overload inside macrophages which triggers chitotriosidase release [8]; however, the identification of hemolytic disorders is quiet easy with standard hematological laboratory procedures in contrast to patients with unexplained neurological manifestations or idiopathic cardiomyopathy who may remain undiagnosed for years. It is well known that about 6% of Caucasian populations are deficient for chitotriosidase activity, being homozygous for a 24-base pair duplication mutation at exon ten of chitotriosidase gene (CHIT1) [28]. Our primary data shows complete enzyme deficiency in around 7% of the studied subjects; however, the clinical utility of the test is minimally affected. Various studies have reported extremely different cut-off values for chitotriosidase based on normal populations. It is better that all cut-off values are set inside each lab due to the considerable inter-laboratory variation. Besides, cut-off values shouldn't be dependent on normal controls only. Pathological controls and patients must be considered. A disease-specific cut-off value can be determined for a disorder when sufficient patients are available and if clinical utility could be proven; for example, based on our data, a plasma cut-off limit of 1000 nmol/mL/h offered 100% sensitivity and 99% specificity for Gaucher's patients, while a blood spot cut-off limit of 400 nmol/mL/h offered 100% sensitivity and 97% specificity. Egypt is a big country with a relatively hot climate, especially in the summer and for the application of a wide scale clinical screening for LSDs, samples have to travel, sometimes for days to reach its destination. Blood spot samples are very practical in this aspect, as they are more stable at higher temperatures as evident in our study, they don't need special devices for preparation after sampling and they can be easily delivered through normal postal services. Screening for chitotriosidase activity among the clinically suspected could be the first step before implementing the screening for more specific lysosomal enzymes in blood spots. In conclusion, blood spot sampling offers many advantages over traditional venous sampling. The level of agreement between blood spot and plasma chitotriosidase using the 4-MU-C3 substrate was found to be acceptable. Blood spot chitotriosidase is a clinically valid test. Role of funding source This work was supported by the Egyptian Science and Technology Development Fund (STDF), project No.526. References [1] Hollak CEM, Weely SV, Van Oers MHJ, Aertes JMFG. Marked elevation of plasma chitotriosidase activity. A novel hallmark of Gaucher disease. J Clin Invest 1994;93:1288–92. [2] van Eijk M, van Roomen CP, Renkema GH, Bussink AP, Andrews L, Blommaart EF, et al. Characterization of human phagocyte-derived chitotriosidase, a component of innate immunity. Int Immunol 2005;17:1505–12. [3] Guo Y, He W, Boer AM, Wevers RA, de Bruijn AM, Groener JE, et al. Elevated plasma chitotriosidase activity in various lysosomal storage disorders. J Inherit Metab Dis 1995;18:717–22.

43

[4] Michelakakis H, Dimitriou E, Labadariadis E. The expanding spectrum of disorders with elevated plasma chitotriosidase activity: an update. J Inherit Metab Dis 2004;27:705–6. [5] Isman F, Natowicz MR. Plasma chitotriosidase in lysosomal storage diseases. Clin Chim Acta 2008;387:165–7. [6] Karadag B, Kucur M, Isman FK, Hacibekiroglu M, Vural VA. Serum chitotriosidase activity in patients with coronary artery disease. Circ J 2008;72:71–5. [7] Verbeek MM, Notting EA, Faas B, Jongen PJH. Increased cerebrospinal fluid chitotriosidase index in patients with multiple sclerosis. Acta Neurol Scand 2010;121:309–14. [8] Barone R, Di Gregorio F, Romeo MA, Schilirò G, Pavone L. Plasma chitotriosidase activity in patients with β-thalassemia. Blood Cells Mol Dis 1999;25:1–8. [9] Barone R, Simporè J, Malaguarnera L, Pignatelli S, Musumeci S. Plasma chitotriosidase activity in acute Plasmodium falciparum malaria. Clin Chim Acta 2003;331:79–85. [10] Labadaridis I, Dimitriou E, Theodorakis M. Chitotriosidase in neonates with fungal and bacterial infections. Arch Dis Child Fetal Neonatal Ed 2005;90:531–2. [11] Hollak CE, Maas M, Aerts JMFG. Clinically relevant therapeutic endpoints in type 1 Gaucher disease. J Inherit Metab Dis 2001;24:97–105. [12] Vedder AC, Cox-Brinkman J, Hollak CEM, Linthorst GE, Groener JEM, Helmond MTJ, et al. Plasma chitotriosidase in male fabry patients: a marker for monitoring lipidladen macrophages and their correction by enzyme replacement therapy. Mol Genet Metab 2006;89:239–44. [13] Xaidara A, Karavitakis EM, Kosma K, Emma F, Dimitriou E, Michelakakis H. Chitotriosidase plasma activity in nephropathic cystinosis. J Inherit Metab Dis 2009. http://dx.doi.org/10.1007/s10545-009-1118-8. [14] Grosso S, Margollicci MA, Bargagli E, Buccoliero QR, Perrone A, Galimberti D, et al. Serum levels of chitotriosidase as a marker of disease activity and clinical stage in sarcoidosis. Scand J Clin Lab Invest 2004;64:57–62. [15] McDade TW, Williams S, Snodgrass JJ. What a drop can do: dried blood spot as a minimally invasive method for integrating biomarkers into population based research. Demography 2007;44:899–925. [16] Chamoles NA, Blanko MB, Gaggioli D, Casentini C. Gaucher and Niemann–Pick diseases: enzymatic diagnosis in dried blood spots on filter paper: retrospective diagnosis in newborn screening cards. Clin Chim Acta 2002;317:191–7. [17] Aguilera B, van der Vlugt KG, Helmond MTJ, Out JMM, Donker-Koopman WE, Groener JEM, et al. Transglycosidase activity of chitotriosidase, improved enzymatic assay for the human macrophage chitinase. J Biol Chem 2003;278:40991–6. [18] Rodrigues MDB, Oliveira AC, Muller KB, Martins AM, D'Almeida V. Chitotrioside determination in plasma and in dried blood spots: a comparison using two different substrates in a microplate assay. Clin Chim Acta 2009;406:86–8. [19] Linnet K, Kondratovich M. Partly nonparametric approach for determining the limit of detection. Clin Chem 2004;50:732–40. [20] Linnet K, Boyd JC. Selection and analytical evaluation of methods with statistical techniques. In: Burtis CA, Ashwood ER, Bruns DE, editors. Tietz textbook of clinical chemistry and molecular diagnostics. 4th ed. St Louis, Missouri, USA: Elsevier Inc; 2006. p. 353–407. [21] Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307–10. [22] Bland JM, Altman DG. Measuring agreement in method comparison studies. Stat Methods Med Res 1999;8:135–60. [23] Abramson JH. WINPEPI updated: computer programs for epidemiologists, and their teaching potential. Epidemiol Perspect Innov 2011;8:1. [24] Civallero G, Michelin K, de Mari J, Viapiana M, Burin M, Coelho JC, et al. Twelve different enzyme assays on dried-blood filter paper samples for detection of patients with selected inherited lysosomal storage diseases. Clin Chim Acta 2006;372:98–102. [25] Goldim MP, Garcia Cda S, de Castilhos CD, Daitx VV, Mezzalira J, Breier AC, et al. Screening of high-risk Gaucher disease patients in Brazil using miniaturized dried blood spots and leukocyte techniques. Gene 2012;508:197–8. [26] Schoonhoven A, Rudensky B, Elstein D, Zimran A, Hollak CE, Groener JE, et al. Monitoring of Gaucher patients with a novel chitotriosidase assay. Clin Chim Acta 2007;381:136–9. [27] Bussink AP, Verhoek M, Vreede J, Ghauharali-van der Vlugt K, Donker-Koopman WE, Sprenger RR, Hollak CE, Aerts JMFG, Boot RG. Common G102S polymorphism in chitotriosidase differentially affects activity towards 4-methylumbelliferyl substrates. FEBS J 2009;276:5678–88. [28] Boot RG, Renkema GH, Verhoek M, Strijland A, Bliek J, de Meulemeester TM, et al. The human chitotriosidase gene. Nature of inherited enzyme deficiency. J Biol Chem 1998;273:25680–5.