- Email: [email protected]
Clinical evaluation of chemokine and enzymatic biomarkers of Gaucher disease
Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267 www.elsevier.com/locate/ybcmd
Clinical evaluation of chemokine and enzymatic biomarkers of Gaucher disease Patrick B. Deegan a,*, Mary Teresa Morana, Ian McFarlaneb, J. Paul Schofield a, Rolf G. Boot c, Johannes M.F.G. Aerts c, Timothy M. Cox a a
Department of Medicine, University of Cambridge, Box 157, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2QQ, UK b Department of Clinical Biochemistry, University of Cambridge, UK c Department of Biochemistry, Academic Medical Center, University of Amsterdam, The Netherlands Submitted 5 May 2005; revised 13 May 2005 (Communicated by E. Beutler, M.D., 13 May 2005)
Abstract Purpose: Gaucher disease is an exemplary orphan disorder. Enzyme replacement therapy with imiglucerase is effective, but very expensive. To improve the assessment of severity of disease and responses to this costly treatment, we have evaluated several enzymatic biomarkers and a newly-described chemokine. Subjects and methods: We studied 48 untreated adults with Type I Gaucher disease: 20 patients were studied after the introduction of enzyme replacement. Disease activity was monitored by serial measurement of platelet count, visceral volumes (spleen and liver) by magnetic resonance imaging, serum activities of total acid phosphatase, angiotensin-converting enzyme (ACE) and the lysosomal chitinase, chitotriosidase. Pulmonary and activation-regulated chemokine (PARC/CCL 18) was also determined in serum by ELISA. Results: Serum PARC concentrations were elevated 10 – 40-fold in patients with Gaucher disease compared with 67 healthy controls, without overlap ( P < 0.0001). Unlike chitotriosidase, PARC was detectable in all individuals. Serum PARC was a reliable indicator of splenic (R = 0.53, P < 0.01) and liver (R = 0.65, P < 0.01) volume and platelet count (R = 0.50, P < 0.01). In splenectomized patients and in patients with null alleles of the chitotriosidase gene, serum PARC concentration correlates with visceral volume and other biomarkers of disease activity. Unlike chitotriosidase, serum PARC concentrations showed unbiased covariation with splenic and platelet responsiveness to enzyme replacement. Conclusion: Serum PARC concentrations are correlated with visceral Gaucher disease and with key clinical responses to enzyme complementation. Determination of this chemokine is a facile and universally applicable method that permits objective monitoring of enzyme replacement therapy for patients with Gaucher disease. D 2005 Published by Elsevier Inc. Keywords: Gaucher; Biomarker; Chemokine; Evaluation; CCL18; Chitotriosidase
Introduction Gaucher disease is a leading orphan disorder: its treatment has been revolutionized by the introduction of enzyme replacement therapy (ERT) [1 –4]. A second-line oral substrate reduction treatment has also been licensed [5]. * Corresponding author. E-mail address: [email protected] (P.B. Deegan). 1079-9796/$ - see front matter D 2005 Published by Elsevier Inc. doi:10.1016/j.bcmd.2005.05.005
A major consequence of the introduction of treatment for very rare diseases is the high cost—an issue that applies particularly to Gaucher disease. This cost places a particular responsibility on physicians to prescribe in a cost-effective manner. To meet these goals, there is a need for new biomarkers of the disease that can be monitored simply and which reflect therapeutic responses. Gaucher disease (OMIM 230800) results from deficiency of glucocerebrosidase [6]. Pathological macrophages (Gaucher cells) engorged with undegraded glycosphingoli-
260
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
pids [7– 9] accumulate in liver, spleen and bone marrow, causing hepatosplenomegaly and thrombocytopenia. Bone marrow infiltration is associated with painful episodes of avascular necrosis. Episodic infarction of the spleen and liver followed by fibrotic scarring also occurs frequently in the untreated condition [10]. There is striking variation in the clinical manifestations of Gaucher disease, even in monozygotic twins homozygous for the mild N307S mutation [11]. The systemic manifestations of Gaucher disease are accompanied by evidence of inflammation with prominent immunoglobulin and cytokine responses [9,12,13]. Enzyme replacement therapy induces salutary responses in most Gaucher patients [14]. The response of patients to ERT varies and certain tissues appear to be more resistant to enzymatic complementation than others. Clearly established tissue injury and scarring are irreversible, so that although the treatment is beneficial for the systematic manifestations, bone disease may demonstrate a lesser response [14 – 16]. In an attempt to maximize cost-effectiveness, some clinicians prescribe doses of enzyme replacement therapy that vary according to disease severity and therapeutic responses [17 – 21]. No comparative trials have shown that individualized dosing strategies are more effective than fixed dose regimens. Reliable biomarkers of disease severity and of the responses to treatment are thus highly sought after for the evaluation of therapeutic responses, though any potential biomarker requires a systematic analysis of its clinical usefulness. An ideal biomarker should be greatly elevated in the disease and show no overlap between untreated patients and healthy subjects. The marker should reflect the total burden of disease. Its concentration or activity should change in response to treatment in parallel to key clinical endpoints. An ideal biomarker should not vary in response to factors unrelated to the disease, such as genetic variation in the population. Finally, an ideal surrogate biomarker should be estimated quickly, reliably and cheaply in easily accessible body fluids. Although several biomarkers are in widespread use for monitoring Gaucher disease [12,22] (tartrate-resistant acid phosphatase (TRAP) [23,24], angiotensin-converting enzyme (ACE) [25] and chitotriosidase [26]), these analytes vary widely in their ability to reflect disease activity and none meets ideal criteria. TRAP is neither specific for Gaucher disease nor greatly elevated [27]; the protein is unstable and is subject to wide analytical variability [28]. ACE activity is subject to variable expression related to a common genetic polymorphism [29] and decreased by the use of frequently-prescribed ACE inhibitors [30]. Chitotriosidase has achieved popularity for monitoring and is often elevated several hundred-fold in patients with active disease [26]. Chitotriosidase activity falls briskly on the introduction of ERT and rises again when treatment is stopped. However, analysis of chitotriosidase activity is technically complex, labor-intensive [31] and not standardized between laboratories. It is also subject to genetic
variation [32]: about one-third of the population is heterozygous for a null (insertion) variant and expresses approximately half the activity of chitotriosidase observed in those who are wild-type. Five percent of the population is homozygous for this null chitotriosidase variant and enzyme activity is absent, even in Gaucher disease. Recently, a novel chemokine, pulmonary and activationregulated chemokine (PARC/CCL18/macrophage inhibitory protein-4), has been described [33]. Although its function in humans is as yet unknown, this member of the human CCchemokine family is produced by macrophages and dendritic cells as a product of ‘‘alternative’’ activation [34], which has been implicated in chronic inflammation and fibrotic scarring. Through development of a Gaucher-specific splenic cDNA library, Moran and colleagues [35] demonstrated greatly increased expression of PARC mRNA in Gaucher spleen. Boot and co-workers have recently applied surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) to demonstrate marked elevation of PARC/CCL18 in plasma of patients with untreated Gaucher disease [36]. Plasma CCL18 concentrations, like chitotriosidase activity, decreased during the introduction of enzyme therapy. Since there is a need for reliable surrogate biomarkers of disease activity for the monitoring of treatment in Gaucher disease with expensive orphan drugs, we have conducted a clinical evaluation of PARC/CCL18 and enzymatic markers to identify the most informative biomarkers for clinical application.
Materials and methods Patients and study design The study was approved by the Local Regional Ethical Committee and all patients gave informed consent. To examine whether PARC concentration was a suitable biomarker of disease bulk and severity, we have compared this marker with other markers and with clinical measurements (including organ volumes) in 48 patients naı¨ve to ERT. The patients were selected on the basis of the availability of stored frozen serum from before and during ERT. PARC concentrations were also estimated in 67 control subjects (17 healthy laboratory controls and 50 outpatients screened for thyroid disease with normal TSH concentrations; 34 males and 37 females, age range 16 to 77 years; median 45 years) and 20 patients with several other lysosomal storage disorders. Of the 48 Gaucher disease patients, 21 were men and 27 women, age ranged from 12 to 81 years, 30 had an intact spleen and 18 were splenectomized. Age of presentation was 5 to 69 years (median 23 years) (not including two asymptomatic individuals). Of the 30 with an intact spleen, 25 had pre-ERT estimations of liver and spleen volume by magnetic resonance imaging (MRI). Of the 18 splenectom-
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
ized patients, 16 had pre-ERT estimation of liver volume by MRI. Among the 48 patients was a pair of monozygotic twins, homozygous for the null variant of chitotriosidase and for the N370S allele, but discordant for Gaucher disease activity [11]. The recognized Gaucher Severity Score Index (SSI) [37] was calculated on the basis of the clinical features at the time of the baseline assessment after casenote review. Severity Score Index ranged from 0 to 19 (median 7.5). We noted the prevalence of the following complications of Gaucher disease: avascular necrosis, fragility fracture, osteoporosis, liver disease, polyclonal gammopathy, clonal gammopathy and visceral infarction. We compared serial measurements of PARC and other biomarkers with clinical data in 20 patients (7 splenectomized and 13 with intact spleen) to examine their role as markers of response to treatment. Length of follow-up on ERT ranged from 1.2 to 10.6 years (median 5 years). Doses of imiglucerase (Cerezyme, Genzyme Corp.) ranged from 20 to 60 U/kg/month. Of these 20 patients, 19 received ERT from the outset and one patient, who had previously participated in a clinical trial of substrate reduction therapy [38], was started on ERT after a 1 year period without specific treatment. Three patients were homozygous for the null chitotriosidase allele and one heterozygous patient did not have increased chitotriosidase activity. The chitotriosidase activity of these four patients was not included in the analysis. In one chitotriosidase deficient patient, baseline imaging data were not available. Analytical methods Sample storage All serum samples for estimation of PARC were stored at 46-C until required. PARC concentration A sandwich ELISA for human PARC was employed for use in serum. The assay used a monoclonal anti-human PARC capture antibody, a biotinylated polyclonal antihuman PARC detection antibody and streptavidin-conjugated horseradish peroxidase (R&D Systems, Minneapolis, USA). Recombinant human PARC was used to generate a standard curve. Serial estimations in each patient were performed in a single batch. Organ volumes Volumes of spleen and liver were derived from MRI images (Fig. 1) as described by Rosenthal et al. [39]. Excess liver volume was derived by subtracting a notional ‘‘expected’’ liver volume (2% of body weight) from the observed liver volume. Platelet count Platelet count was expressed as a reciprocal for ease of display and analysis.
261
Chitotriosidase activity Chitotriosidase activity was estimated using the fluorimetric method of Hollak et al. [26]. Chitotriosidase genotyping was performed as previously described [32]. ACE and total acid phosphatase Angiotensin-converting enzyme activity was estimated using a commercial modification of a spectrophotometric assay (Sigma Diagnostics, St. Louis, MO, USA) [40]. Total acid phosphatase activity was estimated using a commercial modification of a photometic assay (Randox Laboratories, Crumlin, Co. Antrim, UK) [41]. Statistical methods To compare PARC in Gaucher patients and control subjects, the non-parametric Mann – Whitney U test was employed. The non-parametric Spearman Rank correlation was used to compare each blood marker with appropriate clinical measurements. The clinical measurements employed were as follows: for non-splenectomized patients, we used spleen volume, excess liver volume, combined splenic and excess liver volume and platelet count (expressed as a quotient of the upper end of the reference range); for splenectomized patients, we used excess liver volume. Stepwise multiple regression analysis was used to examine any associations between blood markers and each of the disease complications mentioned above. In the comparison of treatment responses, each marker was expressed as a percentage of its pre-treatment baseline value. Only values of markers obtained from blood taken within a week of the relevant clinical measurement were used. In each patient, each of the four biomarkers was compared with each on the clinical measurements. As both markers and clinical responses are subject to measurement error, we used Deming regression to compare these methods of assessing response to treatment [42]. The results of over 200 such regression analyses were summarized by calculating the mean and 95% confidence interval of the slope and intercept of each relationship between biomarker and clinical measurement in this patient population. All statistical analyses were performed using Analyse-it Software (Analyse-it Company, Leeds, UK) for Microsoft Excel.
Results Analysis of PARC Serum PARC was stable on storage and multiple freeze-thaw cycles. Concentrations obtained from serum and plasma were not different. Recovery of recombinant PARC in control serum was 85– 110%. Concen-
262
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
Fig. 1. Serial magnetic resonance images of a patient with Gaucher disease, taken in the transverse plane at the level of the renal vessels, in the course of enzyme replacement therapy. (A) Baseline assessment October 2000, splenic volume 6.21 L. (B) May 2001 splenic volume 3.17 L. (C) May 2003, splenic volume 1.47 L. (D) April 2004, splenic volume 1.15 L.
tration was linear on dilution over the range found in clinical samples. Within-batch coefficient of variation was 4%. PARC in Gaucher disease and control subjects Median PARC concentration in 48 pre-treatment Gaucher disease sera was 875 (25th percentile, 610; 75th percentile, 1250) ng/mL. Median PARC in 67 control subjects was 54 (25th percentile, 41; 75th percentile, 75) ng/mL. There was no overlap in PARC concentration between Gaucher and control samples ( P < 0.0001). Among the control samples, there was no association between PARC concentration and gender (1-way ANOVA). However, there was a significant association between PARC concentration and age (R 2 = 0.11, P = 0.006, n = 67), with a gradient indicating a rise of 0.7 ng/mL per year. This relationship was not observed among Gaucher patients.
PARC in other lysosomal storage disorders Plasma samples from 16 children and three adults with other lysosomal storage disorders were analyzed. Median PARC concentration in this group was 94 (25th percentile 75, 75th percentile 302) ng/mL. PARC concentration in four samples overlapped the Gaucher disease range: a Mannosidosis (n = 2) 357 and 401 ng/mL, Niemann-Pick Disease A 1698 ng/mL and Niemann-Pick Disease B 388 ng/mL. Measurement of PARC in patients harboring null chitotriosidase alleles The correlation between serum PARC concentration and chitotriosidase activity in 43 patients of different chitotriosidase genotype is depicted in Fig. 2. In wild-type and patients heterozygous for the inactivating mutation, there is a significant correlation between PARC and chitotriosidase
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
Fig. 2. Correlation between PARC concentration, determined by ELISA, and chitotriosidase activity, determined by fluorimetric assay, in 43 Gaucher disease patients, grouped according to chitotriosidase genotype (g wildtype: R 2 = 0.37, n = 26, . heterozygote: R 2 = 0.48, n = 14, r homozygous null: n = 3).
(R 2 = 0.37 and 0.48). The slopes of the regression lines differ by approximately a factor of two (4.5 and 1.7). Indices of visceral disease and thrombocytopenia The results of Spearman rank correlations between blood markers and clinical measures of Gaucher disease are shown in Table 1. Only PARC and chitotriosidase are significant predictors of clinical measurements in all cases. None of the markers correlated with the occurrence of the following complications of Gaucher disease: avascular necrosis, fragility fracture, osteoporosis, liver disease, polyclonal or clonal gammopathy. There was an association between PARC and the presence of visceral infarction, which was dependent upon organ volume. Of the blood markers, only ACE was associated with the Severity Score Index [37]. Correction for multiple comparisons [43] abrogated the significance of this relationship (Spearman, Rs = 0.31, P = 0.035, adjusted P = 0.14). PARC was elevated in all five patients with no detectable chitotriosidase activity. In monozygotic twins discordant for Gaucher disease activity [11] and homozygous for the N370S glucocerebrosidase mutation and for the inactivating chitotriosidase mutation, PARC concentration reflected disease activity. The proposita exhibited hepatosplenomegaly and thrombocytopenia requiring splenectomy at the age of 72 years. At the age of 80, she suffered multiple fragility fractures. PARC concentration at 81 years was 2074 ng/mL. The asymptomatic identical twin had no features of Gaucher disease despite equivalent glucocerebrosidase activity in leucocytes. Her PARC concentration on the same occasion was 893 ng/mL. Effects of treatment on clinical and laboratory parameters The effect of ERT on serum PARC concentration is shown in Fig. 3. We correlated the effects of treatment on
263
key clinical manifestations with biomarker responses in each patient at baseline and serially for at least 1 year. A summary of the four blood markers as response predictors is shown in Table 2. Chitotriosidase demonstrated a constant and proportional bias, when used as a marker of response of the following clinical parameters: (i) splenic volume, (ii) volume of spleen plus excess liver volume, (iii) reciprocal platelet count. Chitotriosidase activity decreased more rapidly than these clinically relevant measures. PARC determinations were unbiased in relation to these clinical disease responses. Chitotriosidase, on the other hand, reflected changes in liver volume more closely than other markers. Summarizing regression studies for individual patients as in Table 2 does not permit us to display the overall relationship in a graphic form, but does permit a valid statistical analysis. The relationships between spleen volume and the four markers in response to treatment when all the data points from all the patients are pooled are shown in Fig. 4 to allow the reader a visual impression of the data presented and analyzed in Table 2. In two splenectomized patients, initial thrombocytopenia reflected bone marrow failure. In these patients, decrease in
Table 1 Spearman rank correlation between clinical parameters of disease activity and biomarkers (pulmonary and activation-regulated chemokine (PARC), chitotriosidase, angiotensin-converting enzyme (ACE) and total acid phosphatase (TAP)) in Patients naı¨ve to enzyme replacement therapy Variables
Rhoa
P
n
Spleen volume PARC Chitotriosidase ACE TAP
0.525 0.573 0.472 0.546
0.01 0.0086 0.0206 0.0088
25 22 25 24
Splenic plus excess liver volume PARC 0.644 Chitotriosidase 0.586 ACE 0.612 TAP 0.525
0.0016 0.0072 0.0027 0.0118
25 22 25 24
Excess liver volume (intact spleen) PARC 0.649 Chitotriosidase 0.439 ACE 0.682 TAP 0.395
0.0015 0.0442 0.0008 NS
25 22 25 24
1/platelet count (intact spleen) PARC 0.499 Chitotriosidase 0.402 ACE 0.196 TAP 0.336
0.0083 0.0442 NS NS
29 26 29 28
Excess liver volume (splenectomized) PARC 0.668 Chitotriosidase 0.633 ACE 0.442 TAP 0.719
0.0097 0.0178 NS 0.0053
16 15 16 16
a
Coefficient of rank is represented by Rho.
264
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
Fig. 3. Serial concentration of PARC (as percentage of baseline concentration) in 17 patients with Gaucher disease. Time of follow-up is in years from the start of enzyme replacement therapy.
PARC concentration correlated with increase in platelet count (R 2 = 0.69 and 0.66, P = 0.01 and 0.03, n = 8 and 7, respectively). In all four patients without measurable chitotriosidase activity, and who were monitored serially during ERT, PARC concentration decreased in parallel to clinical parameters (data not shown). Table 2 Relationship between clinical measurements and blood markers (pulmonary and activation-regulated chemokine (PARC), chitotriosidase, angiotensinconverting enzyme (ACE) and total acid phosphatase (TAP)) in response to enzyme replacement therapy Slope
Intercept ( y, when x = 0)
Spleen volume (patients with intact spleen) PARC 1.03 (0.79 to 1.2) 0.6 ( 23 Chitotriosidase 1.39 (1.21 to 1.57) 40.9 ( 60 ACE 1.2 (1.01 to 1.39) 17.9 ( 38 TAP 1.24 (0.95 to 1.54) 24.5 ( 53 Excess liver volume (all patients) PARC 0.67 (0.5 to 0.85) Chitotriosidase 0.83 (0.69 to 0.97) ACE 0.74 (0.62 to 0.85) TAP 0.69 (0.55 to 0.83) Spleen volume plus PARC Chitotriosidase ACE TAP
32.7 5.6 21.2 23.3
to 24) to 22) to 2) to 4)
(20 to 45) ( 9 to 21) (9 to 33) (8 to 38)
excess liver volume (patients with intact spleen) 0.93 (0.72 to 1.14) 10.4 ( 7 to 27) 1.22 (1.1 to 1.34) 25.9 ( 39 to 13) 1.05 (0.94 to 1.17) 3.5 ( 15 to 8) 1.09 (0.85 to 1.33) 12.2 ( 36 to 12)
Platelet count (as reciprocal, in thrombocytopenic patients) PARC 1.16 (0.76 to 1.55) 17.6 ( 55 to 20) Chitotriosidase 1.66 (1.06 to 2.26) 70.2 ( 127 to 13) ACE 1.29 (0.82 to 1.76) 31 ( 76 to 14) TAP 1.38 (0.91 to 1.85) 42.9 ( 87 to 1)
n 10 9 10 10
17 15 17 16
10 9 10 10
13 11 12 13
Slope and intercept are derived from Deming regression in each patient individually, and summarized as mean and 95% confidence interval of the mean. n reflects number of evaluable patients.
Discussion In the last decade, a revolutionary treatment has transformed the lives of patients with Gaucher disease. Monitoring of response to enzyme replacement requires accurate, rapid and non-invasive monitoring of disease activity. Hitherto, clinical and biochemical markers, such as spleen size, platelet count and chitotriosidase activity, have each been inapplicable in certain groups of patients. Herein, we show that PARC has advantages over each of the markers in current use: (i) measurement of PARC is of value in splenectomized patients, where spleen size and platelet count are not informative, (ii) serum PARC reflects disease activity in patients homozygous for the null variant of chitotriosidase, (iii) serum PARC more closely reflects splenic volume and platelet count during treatment than does chitotriosidase and (iv) unlike measurement of chitotriosidase and acid phosphatase activity, enzyme-linked immunosorbent assay of serum PARC is straightforward and amenable to standardization. We have confirmed that PARC is greatly elevated in the serum of patients with Gaucher disease (10 –40 fold). PARC concentrations were increased in several other lysosomal storage disorders. Thus, PARC is not specific for Gaucher disease and further research is required to examine its expression in macrophage and other storage disorders. None of the markers (PARC, Chitotriosidase, ACE, Acid Phosphatase) was able to indicate the past occurrence of the complications of Gaucher disease, such as avascular necrosis, which lead to much long-term disability in this condition. This may be because these blood markers reflect only the current activity of disease. PARC was associated with infarction of the liver and spleen, though this association was indirect and dependent upon organ volume. There was no significant correlation between any of the blood markers and the clinical severity score published by
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
265
Fig. 4. Spleen volumes in 10 patients, plotted against (A) serum PARC concentration, (B) chitotriosidase activity, (C) angiotensin-converting enzyme activity and (D) total acid phosphatase activity in response to enzyme replacement therapy over a median follow-up of 5 years. All patient data are pooled and expressed as % of baseline value. The line of unity (x = y) is plotted for comparison. Each paired measurement (spleen volume/marker) was made within a 7 day period.
Zimran and co-workers. Boot, similarly, did not find an association between PARC and this index. Giraldo and coworkers, however, [44] did note a correlation between chitotriosidase and the same severity score, but did not examine other markers. This index has been discussed [7,45]; it appears not to distinguish between some severe aspects of the disease. It was not designed to reflect amelioration of the disorder by enzyme treatment. We believe that the lack of association between PARC and this severity index may be due to the fact that visceral bulk receives little weighting in the severity score. Changes in PARC concentration reflect changes in organ volume and platelet count without bias. In contrast, changes in chitotriosidase activity give a biased account of response of splenic volume and platelet count to treatment. Chitotriosidase activity declined in parallel with excess liver volume during treatment. However, in patients naı¨ve to ERT, the relationship between chitotriosidase and liver volume was weaker. The close association of chitotriosidase and liver volume in response to ERT may be indirect, as a consequence of the more rapid response of each of these features to ERT, when compared with changes in spleen volume or platelet count. Chitotriosidase activity is often elevated several hundred-fold in patients with Gaucher
disease, whereas spleen volume, PARC, ACE and Acid phosphatase are elevated 20 – 40-fold. Thus, it is not surprising that, in response to specific treatment, chitotriosidase falls at a greater rate than other markers and spleen volume. In comparing organ volumes and platelet count with markers at baseline, we used a rank-based test, thus obviating this discrepancy. In over 150 patients and control subjects, PARC was universally detectable and it appears that a constitutive deficiency of this potential biomarker is rare in humans. At present, insufficient data are available to comment on any potential effect of intercurrent illness or drugs on PARC concentration in patients with Gaucher disease. The relationship between PARC and chitotriosidase in patients stratified according to chitotriosidase genotype demonstrates the principal drawback of the latter marker—its genetic variability. The PARC assay exhibits pre-analytic and analytic robustness. The within-batch variation is acceptable over a wide range of concentrations, but between-batch variation is approximately 15%. Such variation can be overcome by the simple introduction of automated immunoassay platforms. We conclude that determination of the serum concentration of the newly identified human chemokine, PARC,
266
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
satisfies many of the requirements for a reliable biomarker for quantitative evaluation and monitoring of Gaucher disease. Our findings prompt further study into the potential role of PARC in predicting outcome and guiding therapy.
Acknowledgments Plasma samples from pediatric patients with lysosomal storage disorders were received by kind permission of Ms. Elizabeth Young, Institute of Child Health, London. We gratefully acknowledge an unrestricted donation from Actelion Inc. in support of research into biomarkers of Gaucher disease. We thank the members of the Gaucher’s Association and patients for their continuing support.
References [1] R.O. Brady, G.J. Murray, N.W. Barton, Modifying exogenous glucocerebrosidase for effective replacement therapy in Gaucher disease, J. Inherited Metab. Dis. 17 (4) (1994) 510 – 519. [2] R.O. Brady, N.W. Barton, Enzyme replacement therapy for Gaucher disease: critical investigations beyond demonstration of clinical efficacy, Biochem. Med. Metab. Biol. 52 (1) (1994) 1 – 9. [3] G.A. Grabowski, N.W. Barton, G. Pastores, et al., Enzyme therapy in type 1 Gaucher disease: comparative efficacy of mannose-terminated glucocerebrosidase from natural and recombinant sources, Ann. Intern. Med. 122 (1) (1995) 33 – 39. [4] N.W. Barton, R.O. Brady, J.M. Dambrosia, et al., Replacement therapy for inherited enzyme deficiency-macrophage-targeted glucocerebrosidase for Gaucher’s disease, N. Engl. J. Med. 324 (21) (1991) 1464 – 1470. [5] T. Cox, R. Lachmann, C. Hollak, et al., Novel oral treatment of Gaucher’s disease with N-butyldeoxynojirimycin (OGT 918) to decrease substrate biosynthesis, Lancet 355 (9214) (2000) 1481 – 1485. [6] R.O. Brady, Enzymatic abnormalities in diseases of sphingolipid metabolism, Clin. Chem. 13 (7) (1967) 565 – 577. [7] T.M. Cox, J.P. Schofield, Gaucher’s disease: clinical features and natural history, Bailliere’s Clin. Haematol. 10 (4) (1997) 657 – 689. [8] G.M. Pastores, Gaucher’s disease. Pathological features, Bailliere’s Clin. Haematol. 10 (4) (1997) 739 – 749. [9] E. Beutler, G. Grabowski, Gaucher disease, in: Scriver Cea (Ed.), The Metabolic and Molecular Bases of Inherited Disease, 8th edR, vol. III, McGraw Hill, New York, 2001, pp. 3635 – 3668. [10] R.H. Lachmann, D.G. Wight, D.J. Lomas, et al., Massive hepatic fibrosis in Gaucher’s disease: clinico-pathological and radiological features, QJM 93 (4) (2000) 237 – 244. [11] R.H. Lachmann, I.R. Grant, D. Halsall, T.M. Cox, Twin pairs showing discordance of phenotype in adult Gaucher’s disease, QJM 97 (4) (2004) 199 – 204. [12] J.M. Aerts, C.E. Hollak, Plasma and metabolic abnormalities in Gaucher’s disease, Bailliere’s Clin. Haematol. 10 (4) (1997) 691 – 709. [13] Y. Shoenfeld, L.A. Gallant, M. Shaklai, E. Livni, M. Djaldetti, J. Pinkhas, Gaucher’s disease: a disease with chronic stimulation of the immune system, Arch. Pathol. Lab. Med. 106 (8) (1982) 388 – 391. [14] N.J. Weinreb, J. Charrow, H.C. Andersson, et al., Effectiveness of enzyme replacement therapy in 1028 patients with type 1 Gaucher disease after 2 to 5 years of treatment: a report from the Gaucher Registry, Am. J. Med. 113 (2) (2002) 112 – 119. [15] D. Elstein, M. Itzchaki, H.J. Mankin, Skeletal involvement in Gaucher’s disease, Bailliere’s Clin. Haematol. 10 (4) (1997) 793 – 816.
[16] D. Elstein, I. Hadas-Halpern, M. Itzchaki, A. Lahad, A. Abrahamov, A. Zimran, Effect of low-dose enzyme replacement therapy on bones in Gaucher disease patients with severe skeletal involvement, Blood Cells, Mol. Dis. 22 (2) (1996) 104 – 111 (discussion 112 – 4). [17] G. Altarescu, R. Schiffmann, C.C. Parker, et al., Comparative efficacy of dose regimens in enzyme replacement therapy of type I Gaucher disease, Blood Cells, Mol. Dis. 26 (4) (2000) 285 – 290. [18] D. Elstein, A. Abrahamov, I. Hadas-Halpern, A. Meyer, A. Zimran, Low-dose low-frequency imiglucerase as a starting regimen of enzyme replacement therapy for patients with type I Gaucher disease, QJM 91 (7) (1998) 483 – 488. [19] C.E. Hollak, J.M. Aerts, M.H. van Oers, Treatment of Gaucher’s disease, N. Engl. J. Med. 328 (21) (1993) 1565 – 1566 (author reply 1567). [20] A. Zimran, C.E. Hollak, A. Abrahamov, M.H. van Oers, M. Kelly, E. Beutler, Home treatment with intravenous enzyme replacement therapy for Gaucher disease: an international collaborative study of 33 patients, Blood 82 (4) (1993) 1107 – 1109. [21] P.K. Mistry, S. Davies, A. Corfield, A.K. Dixon, T.M. Cox, Successful treatment of bone marrow failure in Gaucher’s disease with low-dose modified glucocerebrosidase, Q. J. Med. 83 (303) (1992) 541 – 546. [22] J.M. Aerts, C. Hollak, R. Boot, A. Groener, Biochemistry of glycosphingolipid storage disorders: implications for therapeutic intervention, Philos. Trans. R. Soc. London, B. Biol. Sci. 358 (1433) (2003) 905 – 914. [23] L.R. Tuchman, H. Suna, J. Carr, Elevation of serum acid phosphatatse in Gaucher’s disease, J. Mt. Sinai Hosp. 23 (1956) 227 – 230. [24] E. Beutler, W. Kuhl, F. Matsumoto, G. Pangalis, Acid hydrolases in leukocytes and platelets of normal subjects and in patients with Gaucher’s and Fabry’s disease, J. Exp. Med. 143 (4) (1976) 975 – 980. [25] J. Lieberman, E. Beutler, Elevation of serum angiotensin-converting enzyme in Gaucher’s disease, N. Engl. J. Med. 294 (26) (1976) 1442 – 1444. [26] C.E. Hollak, S. van Weely, M.H. van Oers, J.M. Aerts, Marked elevation of plasma chitotriosidase activity. A novel hallmark of Gaucher disease, J. Clin. Invest. 93 (3) (1994) 1288 – 1292. [27] H. Bull, P.G. Murray, D. Thomas, A.M. Fraser, P.N. Nelson, Acid phosphatases, Mol. Pathol. 55 (2) (2002) 65 – 72. [28] D.B. Robinson, R.H. Glew, Acid phosphatase in Gaucher’s disease, Clin. Chem. 26 (3) (1980) 371 – 382. [29] S.B. Harrap, H.R. Davidson, J.M. Connor, et al., The angiotensin I converting enzyme gene and predisposition to high blood pressure, Hypertension 21 (4) (1993) 455 – 460. [30] A.D. Struthers, G. Anderson, R.J. MacFadyen, C. Fraser, T.M. MacDonald, Non-adherence with ACE inhibitor treatment is common in heart failure and can be detected by routine serum ACE activity assays, Heart 82 (5) (1999) 584 – 588. [31] B. Aguilera, K. Ghauharali-van der Vlugt, M.T. Helmond, et al., Transglycosidase activity of chitotriosidase: improved enzymatic assay for the human macrophage chitinase, J. Biol. Chem. 278 (42) (2003) 40911 – 40916. [32] R.G. Boot, G.H. Renkema, M. Verhoek, et al., The human chitotriosidase gene. Nature of inherited enzyme deficiency, J. Biol. Chem. 273 (40) (1998) 25680 – 25685. [33] K. Hieshima, T. Imai, M. Baba, et al., A novel human CC chemokine PARC that is most homologous to macrophage-inflammatory protein1 alpha/LD78 alpha and chemotactic for T lymphocytes, but not for monocytes, J. Immunol. 159 (3) (1997) 1140 – 1149. [34] S. Gordon, Alternative activation of macrophages, Nat. Rev. Immunol. 3 (1) (2003) 23 – 35. [35] M.T. Moran, J.P. Schofield, A.R. Hayman, G.P. Shi, E. Young, T.M. Cox, Pathologic gene expression in Gaucher disease: up-regulation of cysteine proteinases including osteoclastic cathepsin K, Blood 96 (5) (2000) 1969 – 1978. [36] R.G. Boot, M. Verhoek, M. de Fost, et al., Marked elevation of the chemokine CCL18/PARC in Gaucher disease: a novel surrogate
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
[37]
[38]
[39] [40]
marker for assessing therapeutic intervention, Blood 103 (1) (2004) 33 – 39. A. Zimran, A. Kay, T. Gelbart, et al., Gaucher disease. Clinical, laboratory, radiologic, and genetic features of 53 patients, Medicine (Baltimore) 71 (6) (1992) 337 – 353. T.M. Cox, J.M. Aerts, G. Andria, et al., The role of the iminosugar Nbutyldeoxynojirimycin (miglustat) in the management of type I (nonneuronopathic) Gaucher disease: a position statement, J. Inherited Metab. Dis. 26 (6) (2003) 513 – 526. D.I. Rosenthal, N.W. Barton, K.A. McKusick, et al., Quantitative imaging of Gaucher disease, Radiology 185 (3) (1992) 841 – 845. B. Holmquist, P. Bunning, J.F. Riordan, A continuous spectrophotometric assay for angiotensin converting enzyme, Anal. Biochem. 95 (2) (1979) 540 – 548.
267
[41] G. Hillman, Continuous photometric measurement of prostate acid phosphatase activity, Z. Klin. Chem. Klin. Biochem. 9 (3) (1971) 273 – 274. [42] W. Deming, Statistical Adjustment of Data, Wiley and Sons, Inc., New York, 1938. [43] C.E. Bonferroni, Il calcolo delle assicurazioni su gruppi di teste, Studi in Onore del Professore Salvatore Ortu Carboni. Rome, Italy, 1935, pp. 13 – 60. [44] P. Giraldo, A. Cenarro, P. Alfonso, et al., Chitotriosidase genotype and plasma activity in patients type 1 Gaucher’s disease and their relatives (carriers and non carriers), Haematologica 86 (9) (2001) 977 – 984. [45] P.K. Mistry, A. Abrahamov, A practical approach to diagnosis and management of Gaucher’s disease, Bailliere’s Clin. Haematol. 10 (4) (1997) 817 – 838.
Clinical evaluation of chemokine and enzymatic biomarkers of Gaucher disease Patrick B. Deegan a,*, Mary Teresa Morana, Ian McFarlaneb, J. Paul Schofield a, Rolf G. Boot c, Johannes M.F.G. Aerts c, Timothy M. Cox a a
Department of Medicine, University of Cambridge, Box 157, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2QQ, UK b Department of Clinical Biochemistry, University of Cambridge, UK c Department of Biochemistry, Academic Medical Center, University of Amsterdam, The Netherlands Submitted 5 May 2005; revised 13 May 2005 (Communicated by E. Beutler, M.D., 13 May 2005)
Abstract Purpose: Gaucher disease is an exemplary orphan disorder. Enzyme replacement therapy with imiglucerase is effective, but very expensive. To improve the assessment of severity of disease and responses to this costly treatment, we have evaluated several enzymatic biomarkers and a newly-described chemokine. Subjects and methods: We studied 48 untreated adults with Type I Gaucher disease: 20 patients were studied after the introduction of enzyme replacement. Disease activity was monitored by serial measurement of platelet count, visceral volumes (spleen and liver) by magnetic resonance imaging, serum activities of total acid phosphatase, angiotensin-converting enzyme (ACE) and the lysosomal chitinase, chitotriosidase. Pulmonary and activation-regulated chemokine (PARC/CCL 18) was also determined in serum by ELISA. Results: Serum PARC concentrations were elevated 10 – 40-fold in patients with Gaucher disease compared with 67 healthy controls, without overlap ( P < 0.0001). Unlike chitotriosidase, PARC was detectable in all individuals. Serum PARC was a reliable indicator of splenic (R = 0.53, P < 0.01) and liver (R = 0.65, P < 0.01) volume and platelet count (R = 0.50, P < 0.01). In splenectomized patients and in patients with null alleles of the chitotriosidase gene, serum PARC concentration correlates with visceral volume and other biomarkers of disease activity. Unlike chitotriosidase, serum PARC concentrations showed unbiased covariation with splenic and platelet responsiveness to enzyme replacement. Conclusion: Serum PARC concentrations are correlated with visceral Gaucher disease and with key clinical responses to enzyme complementation. Determination of this chemokine is a facile and universally applicable method that permits objective monitoring of enzyme replacement therapy for patients with Gaucher disease. D 2005 Published by Elsevier Inc. Keywords: Gaucher; Biomarker; Chemokine; Evaluation; CCL18; Chitotriosidase
Introduction Gaucher disease is a leading orphan disorder: its treatment has been revolutionized by the introduction of enzyme replacement therapy (ERT) [1 –4]. A second-line oral substrate reduction treatment has also been licensed [5]. * Corresponding author. E-mail address: [email protected] (P.B. Deegan). 1079-9796/$ - see front matter D 2005 Published by Elsevier Inc. doi:10.1016/j.bcmd.2005.05.005
A major consequence of the introduction of treatment for very rare diseases is the high cost—an issue that applies particularly to Gaucher disease. This cost places a particular responsibility on physicians to prescribe in a cost-effective manner. To meet these goals, there is a need for new biomarkers of the disease that can be monitored simply and which reflect therapeutic responses. Gaucher disease (OMIM 230800) results from deficiency of glucocerebrosidase [6]. Pathological macrophages (Gaucher cells) engorged with undegraded glycosphingoli-
260
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
pids [7– 9] accumulate in liver, spleen and bone marrow, causing hepatosplenomegaly and thrombocytopenia. Bone marrow infiltration is associated with painful episodes of avascular necrosis. Episodic infarction of the spleen and liver followed by fibrotic scarring also occurs frequently in the untreated condition [10]. There is striking variation in the clinical manifestations of Gaucher disease, even in monozygotic twins homozygous for the mild N307S mutation [11]. The systemic manifestations of Gaucher disease are accompanied by evidence of inflammation with prominent immunoglobulin and cytokine responses [9,12,13]. Enzyme replacement therapy induces salutary responses in most Gaucher patients [14]. The response of patients to ERT varies and certain tissues appear to be more resistant to enzymatic complementation than others. Clearly established tissue injury and scarring are irreversible, so that although the treatment is beneficial for the systematic manifestations, bone disease may demonstrate a lesser response [14 – 16]. In an attempt to maximize cost-effectiveness, some clinicians prescribe doses of enzyme replacement therapy that vary according to disease severity and therapeutic responses [17 – 21]. No comparative trials have shown that individualized dosing strategies are more effective than fixed dose regimens. Reliable biomarkers of disease severity and of the responses to treatment are thus highly sought after for the evaluation of therapeutic responses, though any potential biomarker requires a systematic analysis of its clinical usefulness. An ideal biomarker should be greatly elevated in the disease and show no overlap between untreated patients and healthy subjects. The marker should reflect the total burden of disease. Its concentration or activity should change in response to treatment in parallel to key clinical endpoints. An ideal biomarker should not vary in response to factors unrelated to the disease, such as genetic variation in the population. Finally, an ideal surrogate biomarker should be estimated quickly, reliably and cheaply in easily accessible body fluids. Although several biomarkers are in widespread use for monitoring Gaucher disease [12,22] (tartrate-resistant acid phosphatase (TRAP) [23,24], angiotensin-converting enzyme (ACE) [25] and chitotriosidase [26]), these analytes vary widely in their ability to reflect disease activity and none meets ideal criteria. TRAP is neither specific for Gaucher disease nor greatly elevated [27]; the protein is unstable and is subject to wide analytical variability [28]. ACE activity is subject to variable expression related to a common genetic polymorphism [29] and decreased by the use of frequently-prescribed ACE inhibitors [30]. Chitotriosidase has achieved popularity for monitoring and is often elevated several hundred-fold in patients with active disease [26]. Chitotriosidase activity falls briskly on the introduction of ERT and rises again when treatment is stopped. However, analysis of chitotriosidase activity is technically complex, labor-intensive [31] and not standardized between laboratories. It is also subject to genetic
variation [32]: about one-third of the population is heterozygous for a null (insertion) variant and expresses approximately half the activity of chitotriosidase observed in those who are wild-type. Five percent of the population is homozygous for this null chitotriosidase variant and enzyme activity is absent, even in Gaucher disease. Recently, a novel chemokine, pulmonary and activationregulated chemokine (PARC/CCL18/macrophage inhibitory protein-4), has been described [33]. Although its function in humans is as yet unknown, this member of the human CCchemokine family is produced by macrophages and dendritic cells as a product of ‘‘alternative’’ activation [34], which has been implicated in chronic inflammation and fibrotic scarring. Through development of a Gaucher-specific splenic cDNA library, Moran and colleagues [35] demonstrated greatly increased expression of PARC mRNA in Gaucher spleen. Boot and co-workers have recently applied surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) to demonstrate marked elevation of PARC/CCL18 in plasma of patients with untreated Gaucher disease [36]. Plasma CCL18 concentrations, like chitotriosidase activity, decreased during the introduction of enzyme therapy. Since there is a need for reliable surrogate biomarkers of disease activity for the monitoring of treatment in Gaucher disease with expensive orphan drugs, we have conducted a clinical evaluation of PARC/CCL18 and enzymatic markers to identify the most informative biomarkers for clinical application.
Materials and methods Patients and study design The study was approved by the Local Regional Ethical Committee and all patients gave informed consent. To examine whether PARC concentration was a suitable biomarker of disease bulk and severity, we have compared this marker with other markers and with clinical measurements (including organ volumes) in 48 patients naı¨ve to ERT. The patients were selected on the basis of the availability of stored frozen serum from before and during ERT. PARC concentrations were also estimated in 67 control subjects (17 healthy laboratory controls and 50 outpatients screened for thyroid disease with normal TSH concentrations; 34 males and 37 females, age range 16 to 77 years; median 45 years) and 20 patients with several other lysosomal storage disorders. Of the 48 Gaucher disease patients, 21 were men and 27 women, age ranged from 12 to 81 years, 30 had an intact spleen and 18 were splenectomized. Age of presentation was 5 to 69 years (median 23 years) (not including two asymptomatic individuals). Of the 30 with an intact spleen, 25 had pre-ERT estimations of liver and spleen volume by magnetic resonance imaging (MRI). Of the 18 splenectom-
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
ized patients, 16 had pre-ERT estimation of liver volume by MRI. Among the 48 patients was a pair of monozygotic twins, homozygous for the null variant of chitotriosidase and for the N370S allele, but discordant for Gaucher disease activity [11]. The recognized Gaucher Severity Score Index (SSI) [37] was calculated on the basis of the clinical features at the time of the baseline assessment after casenote review. Severity Score Index ranged from 0 to 19 (median 7.5). We noted the prevalence of the following complications of Gaucher disease: avascular necrosis, fragility fracture, osteoporosis, liver disease, polyclonal gammopathy, clonal gammopathy and visceral infarction. We compared serial measurements of PARC and other biomarkers with clinical data in 20 patients (7 splenectomized and 13 with intact spleen) to examine their role as markers of response to treatment. Length of follow-up on ERT ranged from 1.2 to 10.6 years (median 5 years). Doses of imiglucerase (Cerezyme, Genzyme Corp.) ranged from 20 to 60 U/kg/month. Of these 20 patients, 19 received ERT from the outset and one patient, who had previously participated in a clinical trial of substrate reduction therapy [38], was started on ERT after a 1 year period without specific treatment. Three patients were homozygous for the null chitotriosidase allele and one heterozygous patient did not have increased chitotriosidase activity. The chitotriosidase activity of these four patients was not included in the analysis. In one chitotriosidase deficient patient, baseline imaging data were not available. Analytical methods Sample storage All serum samples for estimation of PARC were stored at 46-C until required. PARC concentration A sandwich ELISA for human PARC was employed for use in serum. The assay used a monoclonal anti-human PARC capture antibody, a biotinylated polyclonal antihuman PARC detection antibody and streptavidin-conjugated horseradish peroxidase (R&D Systems, Minneapolis, USA). Recombinant human PARC was used to generate a standard curve. Serial estimations in each patient were performed in a single batch. Organ volumes Volumes of spleen and liver were derived from MRI images (Fig. 1) as described by Rosenthal et al. [39]. Excess liver volume was derived by subtracting a notional ‘‘expected’’ liver volume (2% of body weight) from the observed liver volume. Platelet count Platelet count was expressed as a reciprocal for ease of display and analysis.
261
Chitotriosidase activity Chitotriosidase activity was estimated using the fluorimetric method of Hollak et al. [26]. Chitotriosidase genotyping was performed as previously described [32]. ACE and total acid phosphatase Angiotensin-converting enzyme activity was estimated using a commercial modification of a spectrophotometric assay (Sigma Diagnostics, St. Louis, MO, USA) [40]. Total acid phosphatase activity was estimated using a commercial modification of a photometic assay (Randox Laboratories, Crumlin, Co. Antrim, UK) [41]. Statistical methods To compare PARC in Gaucher patients and control subjects, the non-parametric Mann – Whitney U test was employed. The non-parametric Spearman Rank correlation was used to compare each blood marker with appropriate clinical measurements. The clinical measurements employed were as follows: for non-splenectomized patients, we used spleen volume, excess liver volume, combined splenic and excess liver volume and platelet count (expressed as a quotient of the upper end of the reference range); for splenectomized patients, we used excess liver volume. Stepwise multiple regression analysis was used to examine any associations between blood markers and each of the disease complications mentioned above. In the comparison of treatment responses, each marker was expressed as a percentage of its pre-treatment baseline value. Only values of markers obtained from blood taken within a week of the relevant clinical measurement were used. In each patient, each of the four biomarkers was compared with each on the clinical measurements. As both markers and clinical responses are subject to measurement error, we used Deming regression to compare these methods of assessing response to treatment [42]. The results of over 200 such regression analyses were summarized by calculating the mean and 95% confidence interval of the slope and intercept of each relationship between biomarker and clinical measurement in this patient population. All statistical analyses were performed using Analyse-it Software (Analyse-it Company, Leeds, UK) for Microsoft Excel.
Results Analysis of PARC Serum PARC was stable on storage and multiple freeze-thaw cycles. Concentrations obtained from serum and plasma were not different. Recovery of recombinant PARC in control serum was 85– 110%. Concen-
262
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
Fig. 1. Serial magnetic resonance images of a patient with Gaucher disease, taken in the transverse plane at the level of the renal vessels, in the course of enzyme replacement therapy. (A) Baseline assessment October 2000, splenic volume 6.21 L. (B) May 2001 splenic volume 3.17 L. (C) May 2003, splenic volume 1.47 L. (D) April 2004, splenic volume 1.15 L.
tration was linear on dilution over the range found in clinical samples. Within-batch coefficient of variation was 4%. PARC in Gaucher disease and control subjects Median PARC concentration in 48 pre-treatment Gaucher disease sera was 875 (25th percentile, 610; 75th percentile, 1250) ng/mL. Median PARC in 67 control subjects was 54 (25th percentile, 41; 75th percentile, 75) ng/mL. There was no overlap in PARC concentration between Gaucher and control samples ( P < 0.0001). Among the control samples, there was no association between PARC concentration and gender (1-way ANOVA). However, there was a significant association between PARC concentration and age (R 2 = 0.11, P = 0.006, n = 67), with a gradient indicating a rise of 0.7 ng/mL per year. This relationship was not observed among Gaucher patients.
PARC in other lysosomal storage disorders Plasma samples from 16 children and three adults with other lysosomal storage disorders were analyzed. Median PARC concentration in this group was 94 (25th percentile 75, 75th percentile 302) ng/mL. PARC concentration in four samples overlapped the Gaucher disease range: a Mannosidosis (n = 2) 357 and 401 ng/mL, Niemann-Pick Disease A 1698 ng/mL and Niemann-Pick Disease B 388 ng/mL. Measurement of PARC in patients harboring null chitotriosidase alleles The correlation between serum PARC concentration and chitotriosidase activity in 43 patients of different chitotriosidase genotype is depicted in Fig. 2. In wild-type and patients heterozygous for the inactivating mutation, there is a significant correlation between PARC and chitotriosidase
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
Fig. 2. Correlation between PARC concentration, determined by ELISA, and chitotriosidase activity, determined by fluorimetric assay, in 43 Gaucher disease patients, grouped according to chitotriosidase genotype (g wildtype: R 2 = 0.37, n = 26, . heterozygote: R 2 = 0.48, n = 14, r homozygous null: n = 3).
(R 2 = 0.37 and 0.48). The slopes of the regression lines differ by approximately a factor of two (4.5 and 1.7). Indices of visceral disease and thrombocytopenia The results of Spearman rank correlations between blood markers and clinical measures of Gaucher disease are shown in Table 1. Only PARC and chitotriosidase are significant predictors of clinical measurements in all cases. None of the markers correlated with the occurrence of the following complications of Gaucher disease: avascular necrosis, fragility fracture, osteoporosis, liver disease, polyclonal or clonal gammopathy. There was an association between PARC and the presence of visceral infarction, which was dependent upon organ volume. Of the blood markers, only ACE was associated with the Severity Score Index [37]. Correction for multiple comparisons [43] abrogated the significance of this relationship (Spearman, Rs = 0.31, P = 0.035, adjusted P = 0.14). PARC was elevated in all five patients with no detectable chitotriosidase activity. In monozygotic twins discordant for Gaucher disease activity [11] and homozygous for the N370S glucocerebrosidase mutation and for the inactivating chitotriosidase mutation, PARC concentration reflected disease activity. The proposita exhibited hepatosplenomegaly and thrombocytopenia requiring splenectomy at the age of 72 years. At the age of 80, she suffered multiple fragility fractures. PARC concentration at 81 years was 2074 ng/mL. The asymptomatic identical twin had no features of Gaucher disease despite equivalent glucocerebrosidase activity in leucocytes. Her PARC concentration on the same occasion was 893 ng/mL. Effects of treatment on clinical and laboratory parameters The effect of ERT on serum PARC concentration is shown in Fig. 3. We correlated the effects of treatment on
263
key clinical manifestations with biomarker responses in each patient at baseline and serially for at least 1 year. A summary of the four blood markers as response predictors is shown in Table 2. Chitotriosidase demonstrated a constant and proportional bias, when used as a marker of response of the following clinical parameters: (i) splenic volume, (ii) volume of spleen plus excess liver volume, (iii) reciprocal platelet count. Chitotriosidase activity decreased more rapidly than these clinically relevant measures. PARC determinations were unbiased in relation to these clinical disease responses. Chitotriosidase, on the other hand, reflected changes in liver volume more closely than other markers. Summarizing regression studies for individual patients as in Table 2 does not permit us to display the overall relationship in a graphic form, but does permit a valid statistical analysis. The relationships between spleen volume and the four markers in response to treatment when all the data points from all the patients are pooled are shown in Fig. 4 to allow the reader a visual impression of the data presented and analyzed in Table 2. In two splenectomized patients, initial thrombocytopenia reflected bone marrow failure. In these patients, decrease in
Table 1 Spearman rank correlation between clinical parameters of disease activity and biomarkers (pulmonary and activation-regulated chemokine (PARC), chitotriosidase, angiotensin-converting enzyme (ACE) and total acid phosphatase (TAP)) in Patients naı¨ve to enzyme replacement therapy Variables
Rhoa
P
n
Spleen volume PARC Chitotriosidase ACE TAP
0.525 0.573 0.472 0.546
0.01 0.0086 0.0206 0.0088
25 22 25 24
Splenic plus excess liver volume PARC 0.644 Chitotriosidase 0.586 ACE 0.612 TAP 0.525
0.0016 0.0072 0.0027 0.0118
25 22 25 24
Excess liver volume (intact spleen) PARC 0.649 Chitotriosidase 0.439 ACE 0.682 TAP 0.395
0.0015 0.0442 0.0008 NS
25 22 25 24
1/platelet count (intact spleen) PARC 0.499 Chitotriosidase 0.402 ACE 0.196 TAP 0.336
0.0083 0.0442 NS NS
29 26 29 28
Excess liver volume (splenectomized) PARC 0.668 Chitotriosidase 0.633 ACE 0.442 TAP 0.719
0.0097 0.0178 NS 0.0053
16 15 16 16
a
Coefficient of rank is represented by Rho.
264
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
Fig. 3. Serial concentration of PARC (as percentage of baseline concentration) in 17 patients with Gaucher disease. Time of follow-up is in years from the start of enzyme replacement therapy.
PARC concentration correlated with increase in platelet count (R 2 = 0.69 and 0.66, P = 0.01 and 0.03, n = 8 and 7, respectively). In all four patients without measurable chitotriosidase activity, and who were monitored serially during ERT, PARC concentration decreased in parallel to clinical parameters (data not shown). Table 2 Relationship between clinical measurements and blood markers (pulmonary and activation-regulated chemokine (PARC), chitotriosidase, angiotensinconverting enzyme (ACE) and total acid phosphatase (TAP)) in response to enzyme replacement therapy Slope
Intercept ( y, when x = 0)
Spleen volume (patients with intact spleen) PARC 1.03 (0.79 to 1.2) 0.6 ( 23 Chitotriosidase 1.39 (1.21 to 1.57) 40.9 ( 60 ACE 1.2 (1.01 to 1.39) 17.9 ( 38 TAP 1.24 (0.95 to 1.54) 24.5 ( 53 Excess liver volume (all patients) PARC 0.67 (0.5 to 0.85) Chitotriosidase 0.83 (0.69 to 0.97) ACE 0.74 (0.62 to 0.85) TAP 0.69 (0.55 to 0.83) Spleen volume plus PARC Chitotriosidase ACE TAP
32.7 5.6 21.2 23.3
to 24) to 22) to 2) to 4)
(20 to 45) ( 9 to 21) (9 to 33) (8 to 38)
excess liver volume (patients with intact spleen) 0.93 (0.72 to 1.14) 10.4 ( 7 to 27) 1.22 (1.1 to 1.34) 25.9 ( 39 to 13) 1.05 (0.94 to 1.17) 3.5 ( 15 to 8) 1.09 (0.85 to 1.33) 12.2 ( 36 to 12)
Platelet count (as reciprocal, in thrombocytopenic patients) PARC 1.16 (0.76 to 1.55) 17.6 ( 55 to 20) Chitotriosidase 1.66 (1.06 to 2.26) 70.2 ( 127 to 13) ACE 1.29 (0.82 to 1.76) 31 ( 76 to 14) TAP 1.38 (0.91 to 1.85) 42.9 ( 87 to 1)
n 10 9 10 10
17 15 17 16
10 9 10 10
13 11 12 13
Slope and intercept are derived from Deming regression in each patient individually, and summarized as mean and 95% confidence interval of the mean. n reflects number of evaluable patients.
Discussion In the last decade, a revolutionary treatment has transformed the lives of patients with Gaucher disease. Monitoring of response to enzyme replacement requires accurate, rapid and non-invasive monitoring of disease activity. Hitherto, clinical and biochemical markers, such as spleen size, platelet count and chitotriosidase activity, have each been inapplicable in certain groups of patients. Herein, we show that PARC has advantages over each of the markers in current use: (i) measurement of PARC is of value in splenectomized patients, where spleen size and platelet count are not informative, (ii) serum PARC reflects disease activity in patients homozygous for the null variant of chitotriosidase, (iii) serum PARC more closely reflects splenic volume and platelet count during treatment than does chitotriosidase and (iv) unlike measurement of chitotriosidase and acid phosphatase activity, enzyme-linked immunosorbent assay of serum PARC is straightforward and amenable to standardization. We have confirmed that PARC is greatly elevated in the serum of patients with Gaucher disease (10 –40 fold). PARC concentrations were increased in several other lysosomal storage disorders. Thus, PARC is not specific for Gaucher disease and further research is required to examine its expression in macrophage and other storage disorders. None of the markers (PARC, Chitotriosidase, ACE, Acid Phosphatase) was able to indicate the past occurrence of the complications of Gaucher disease, such as avascular necrosis, which lead to much long-term disability in this condition. This may be because these blood markers reflect only the current activity of disease. PARC was associated with infarction of the liver and spleen, though this association was indirect and dependent upon organ volume. There was no significant correlation between any of the blood markers and the clinical severity score published by
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
265
Fig. 4. Spleen volumes in 10 patients, plotted against (A) serum PARC concentration, (B) chitotriosidase activity, (C) angiotensin-converting enzyme activity and (D) total acid phosphatase activity in response to enzyme replacement therapy over a median follow-up of 5 years. All patient data are pooled and expressed as % of baseline value. The line of unity (x = y) is plotted for comparison. Each paired measurement (spleen volume/marker) was made within a 7 day period.
Zimran and co-workers. Boot, similarly, did not find an association between PARC and this index. Giraldo and coworkers, however, [44] did note a correlation between chitotriosidase and the same severity score, but did not examine other markers. This index has been discussed [7,45]; it appears not to distinguish between some severe aspects of the disease. It was not designed to reflect amelioration of the disorder by enzyme treatment. We believe that the lack of association between PARC and this severity index may be due to the fact that visceral bulk receives little weighting in the severity score. Changes in PARC concentration reflect changes in organ volume and platelet count without bias. In contrast, changes in chitotriosidase activity give a biased account of response of splenic volume and platelet count to treatment. Chitotriosidase activity declined in parallel with excess liver volume during treatment. However, in patients naı¨ve to ERT, the relationship between chitotriosidase and liver volume was weaker. The close association of chitotriosidase and liver volume in response to ERT may be indirect, as a consequence of the more rapid response of each of these features to ERT, when compared with changes in spleen volume or platelet count. Chitotriosidase activity is often elevated several hundred-fold in patients with Gaucher
disease, whereas spleen volume, PARC, ACE and Acid phosphatase are elevated 20 – 40-fold. Thus, it is not surprising that, in response to specific treatment, chitotriosidase falls at a greater rate than other markers and spleen volume. In comparing organ volumes and platelet count with markers at baseline, we used a rank-based test, thus obviating this discrepancy. In over 150 patients and control subjects, PARC was universally detectable and it appears that a constitutive deficiency of this potential biomarker is rare in humans. At present, insufficient data are available to comment on any potential effect of intercurrent illness or drugs on PARC concentration in patients with Gaucher disease. The relationship between PARC and chitotriosidase in patients stratified according to chitotriosidase genotype demonstrates the principal drawback of the latter marker—its genetic variability. The PARC assay exhibits pre-analytic and analytic robustness. The within-batch variation is acceptable over a wide range of concentrations, but between-batch variation is approximately 15%. Such variation can be overcome by the simple introduction of automated immunoassay platforms. We conclude that determination of the serum concentration of the newly identified human chemokine, PARC,
266
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
satisfies many of the requirements for a reliable biomarker for quantitative evaluation and monitoring of Gaucher disease. Our findings prompt further study into the potential role of PARC in predicting outcome and guiding therapy.
Acknowledgments Plasma samples from pediatric patients with lysosomal storage disorders were received by kind permission of Ms. Elizabeth Young, Institute of Child Health, London. We gratefully acknowledge an unrestricted donation from Actelion Inc. in support of research into biomarkers of Gaucher disease. We thank the members of the Gaucher’s Association and patients for their continuing support.
References [1] R.O. Brady, G.J. Murray, N.W. Barton, Modifying exogenous glucocerebrosidase for effective replacement therapy in Gaucher disease, J. Inherited Metab. Dis. 17 (4) (1994) 510 – 519. [2] R.O. Brady, N.W. Barton, Enzyme replacement therapy for Gaucher disease: critical investigations beyond demonstration of clinical efficacy, Biochem. Med. Metab. Biol. 52 (1) (1994) 1 – 9. [3] G.A. Grabowski, N.W. Barton, G. Pastores, et al., Enzyme therapy in type 1 Gaucher disease: comparative efficacy of mannose-terminated glucocerebrosidase from natural and recombinant sources, Ann. Intern. Med. 122 (1) (1995) 33 – 39. [4] N.W. Barton, R.O. Brady, J.M. Dambrosia, et al., Replacement therapy for inherited enzyme deficiency-macrophage-targeted glucocerebrosidase for Gaucher’s disease, N. Engl. J. Med. 324 (21) (1991) 1464 – 1470. [5] T. Cox, R. Lachmann, C. Hollak, et al., Novel oral treatment of Gaucher’s disease with N-butyldeoxynojirimycin (OGT 918) to decrease substrate biosynthesis, Lancet 355 (9214) (2000) 1481 – 1485. [6] R.O. Brady, Enzymatic abnormalities in diseases of sphingolipid metabolism, Clin. Chem. 13 (7) (1967) 565 – 577. [7] T.M. Cox, J.P. Schofield, Gaucher’s disease: clinical features and natural history, Bailliere’s Clin. Haematol. 10 (4) (1997) 657 – 689. [8] G.M. Pastores, Gaucher’s disease. Pathological features, Bailliere’s Clin. Haematol. 10 (4) (1997) 739 – 749. [9] E. Beutler, G. Grabowski, Gaucher disease, in: Scriver Cea (Ed.), The Metabolic and Molecular Bases of Inherited Disease, 8th edR, vol. III, McGraw Hill, New York, 2001, pp. 3635 – 3668. [10] R.H. Lachmann, D.G. Wight, D.J. Lomas, et al., Massive hepatic fibrosis in Gaucher’s disease: clinico-pathological and radiological features, QJM 93 (4) (2000) 237 – 244. [11] R.H. Lachmann, I.R. Grant, D. Halsall, T.M. Cox, Twin pairs showing discordance of phenotype in adult Gaucher’s disease, QJM 97 (4) (2004) 199 – 204. [12] J.M. Aerts, C.E. Hollak, Plasma and metabolic abnormalities in Gaucher’s disease, Bailliere’s Clin. Haematol. 10 (4) (1997) 691 – 709. [13] Y. Shoenfeld, L.A. Gallant, M. Shaklai, E. Livni, M. Djaldetti, J. Pinkhas, Gaucher’s disease: a disease with chronic stimulation of the immune system, Arch. Pathol. Lab. Med. 106 (8) (1982) 388 – 391. [14] N.J. Weinreb, J. Charrow, H.C. Andersson, et al., Effectiveness of enzyme replacement therapy in 1028 patients with type 1 Gaucher disease after 2 to 5 years of treatment: a report from the Gaucher Registry, Am. J. Med. 113 (2) (2002) 112 – 119. [15] D. Elstein, M. Itzchaki, H.J. Mankin, Skeletal involvement in Gaucher’s disease, Bailliere’s Clin. Haematol. 10 (4) (1997) 793 – 816.
[16] D. Elstein, I. Hadas-Halpern, M. Itzchaki, A. Lahad, A. Abrahamov, A. Zimran, Effect of low-dose enzyme replacement therapy on bones in Gaucher disease patients with severe skeletal involvement, Blood Cells, Mol. Dis. 22 (2) (1996) 104 – 111 (discussion 112 – 4). [17] G. Altarescu, R. Schiffmann, C.C. Parker, et al., Comparative efficacy of dose regimens in enzyme replacement therapy of type I Gaucher disease, Blood Cells, Mol. Dis. 26 (4) (2000) 285 – 290. [18] D. Elstein, A. Abrahamov, I. Hadas-Halpern, A. Meyer, A. Zimran, Low-dose low-frequency imiglucerase as a starting regimen of enzyme replacement therapy for patients with type I Gaucher disease, QJM 91 (7) (1998) 483 – 488. [19] C.E. Hollak, J.M. Aerts, M.H. van Oers, Treatment of Gaucher’s disease, N. Engl. J. Med. 328 (21) (1993) 1565 – 1566 (author reply 1567). [20] A. Zimran, C.E. Hollak, A. Abrahamov, M.H. van Oers, M. Kelly, E. Beutler, Home treatment with intravenous enzyme replacement therapy for Gaucher disease: an international collaborative study of 33 patients, Blood 82 (4) (1993) 1107 – 1109. [21] P.K. Mistry, S. Davies, A. Corfield, A.K. Dixon, T.M. Cox, Successful treatment of bone marrow failure in Gaucher’s disease with low-dose modified glucocerebrosidase, Q. J. Med. 83 (303) (1992) 541 – 546. [22] J.M. Aerts, C. Hollak, R. Boot, A. Groener, Biochemistry of glycosphingolipid storage disorders: implications for therapeutic intervention, Philos. Trans. R. Soc. London, B. Biol. Sci. 358 (1433) (2003) 905 – 914. [23] L.R. Tuchman, H. Suna, J. Carr, Elevation of serum acid phosphatatse in Gaucher’s disease, J. Mt. Sinai Hosp. 23 (1956) 227 – 230. [24] E. Beutler, W. Kuhl, F. Matsumoto, G. Pangalis, Acid hydrolases in leukocytes and platelets of normal subjects and in patients with Gaucher’s and Fabry’s disease, J. Exp. Med. 143 (4) (1976) 975 – 980. [25] J. Lieberman, E. Beutler, Elevation of serum angiotensin-converting enzyme in Gaucher’s disease, N. Engl. J. Med. 294 (26) (1976) 1442 – 1444. [26] C.E. Hollak, S. van Weely, M.H. van Oers, J.M. Aerts, Marked elevation of plasma chitotriosidase activity. A novel hallmark of Gaucher disease, J. Clin. Invest. 93 (3) (1994) 1288 – 1292. [27] H. Bull, P.G. Murray, D. Thomas, A.M. Fraser, P.N. Nelson, Acid phosphatases, Mol. Pathol. 55 (2) (2002) 65 – 72. [28] D.B. Robinson, R.H. Glew, Acid phosphatase in Gaucher’s disease, Clin. Chem. 26 (3) (1980) 371 – 382. [29] S.B. Harrap, H.R. Davidson, J.M. Connor, et al., The angiotensin I converting enzyme gene and predisposition to high blood pressure, Hypertension 21 (4) (1993) 455 – 460. [30] A.D. Struthers, G. Anderson, R.J. MacFadyen, C. Fraser, T.M. MacDonald, Non-adherence with ACE inhibitor treatment is common in heart failure and can be detected by routine serum ACE activity assays, Heart 82 (5) (1999) 584 – 588. [31] B. Aguilera, K. Ghauharali-van der Vlugt, M.T. Helmond, et al., Transglycosidase activity of chitotriosidase: improved enzymatic assay for the human macrophage chitinase, J. Biol. Chem. 278 (42) (2003) 40911 – 40916. [32] R.G. Boot, G.H. Renkema, M. Verhoek, et al., The human chitotriosidase gene. Nature of inherited enzyme deficiency, J. Biol. Chem. 273 (40) (1998) 25680 – 25685. [33] K. Hieshima, T. Imai, M. Baba, et al., A novel human CC chemokine PARC that is most homologous to macrophage-inflammatory protein1 alpha/LD78 alpha and chemotactic for T lymphocytes, but not for monocytes, J. Immunol. 159 (3) (1997) 1140 – 1149. [34] S. Gordon, Alternative activation of macrophages, Nat. Rev. Immunol. 3 (1) (2003) 23 – 35. [35] M.T. Moran, J.P. Schofield, A.R. Hayman, G.P. Shi, E. Young, T.M. Cox, Pathologic gene expression in Gaucher disease: up-regulation of cysteine proteinases including osteoclastic cathepsin K, Blood 96 (5) (2000) 1969 – 1978. [36] R.G. Boot, M. Verhoek, M. de Fost, et al., Marked elevation of the chemokine CCL18/PARC in Gaucher disease: a novel surrogate
P.B. Deegan et al. / Blood Cells, Molecules, and Diseases 35 (2005) 259 – 267
[37]
[38]
[39] [40]
marker for assessing therapeutic intervention, Blood 103 (1) (2004) 33 – 39. A. Zimran, A. Kay, T. Gelbart, et al., Gaucher disease. Clinical, laboratory, radiologic, and genetic features of 53 patients, Medicine (Baltimore) 71 (6) (1992) 337 – 353. T.M. Cox, J.M. Aerts, G. Andria, et al., The role of the iminosugar Nbutyldeoxynojirimycin (miglustat) in the management of type I (nonneuronopathic) Gaucher disease: a position statement, J. Inherited Metab. Dis. 26 (6) (2003) 513 – 526. D.I. Rosenthal, N.W. Barton, K.A. McKusick, et al., Quantitative imaging of Gaucher disease, Radiology 185 (3) (1992) 841 – 845. B. Holmquist, P. Bunning, J.F. Riordan, A continuous spectrophotometric assay for angiotensin converting enzyme, Anal. Biochem. 95 (2) (1979) 540 – 548.
267
[41] G. Hillman, Continuous photometric measurement of prostate acid phosphatase activity, Z. Klin. Chem. Klin. Biochem. 9 (3) (1971) 273 – 274. [42] W. Deming, Statistical Adjustment of Data, Wiley and Sons, Inc., New York, 1938. [43] C.E. Bonferroni, Il calcolo delle assicurazioni su gruppi di teste, Studi in Onore del Professore Salvatore Ortu Carboni. Rome, Italy, 1935, pp. 13 – 60. [44] P. Giraldo, A. Cenarro, P. Alfonso, et al., Chitotriosidase genotype and plasma activity in patients type 1 Gaucher’s disease and their relatives (carriers and non carriers), Haematologica 86 (9) (2001) 977 – 984. [45] P.K. Mistry, A. Abrahamov, A practical approach to diagnosis and management of Gaucher’s disease, Bailliere’s Clin. Haematol. 10 (4) (1997) 817 – 838.