Clin Exp Nephrol (2016) 20:284–293 DOI 10.1007/s10157-015-1146-7
ORIGINAL ARTICLE
Prevalence of Fabry disease in dialysis patients: Japan Fabry disease screening study (J-FAST) Osamu Saito1 • Eiji Kusano2 • Tetsu Akimoto1 • Yasushi Asano3 • Teruo Kitagawa4 Ken Suzuki4 • Nobuyuki Ishige4 • Takashi Akiba5 • Akira Saito6 • Eiji Ishimura7 • Motoshi Hattori8 • Akira Hishida9 • Chu Guili10 • Hiroki Maruyama10 • Masahisa Kobayashi11 • Touya Ohashi12 • Ichiro Matsuda13 • Yoshikatsu Eto14
•
Received: 30 March 2015 / Accepted: 11 July 2015 / Published online: 22 July 2015 Ó Japanese Society of Nephrology 2015
Abstract Background In Fabry disease, progressive glycolipid accumulation leads to damage in kidney and other organs. This study was designed to determine the prevalence rate of Fabry disease in Japanese dialysis patients. Methods All dialysis patients agreeing to Japan Fabry disease screening study (J-FAST) with informed consent were selected except for Fabry disease. The screening was performed by a method of measuring plasma and/or leukocytes lysosomal a-galactosidase A protein level and a-galactosidase A activity. If positive, genetic analysis was carried out upon patient’s agreement. Results J-FAST dealt with 8547 patients (male 5408, female 3139). At the tertiary examination, 26 out of 8547 patients were found to be positive. Six out of 26 patients could not accept genetic analysis because of death. Remaining 20 patients agreed with genetic analysis; then 2 patients (male 2, female 0) had a variation of the a-Gal
gene and 11 patients showed E66Q variations. Therefore, the frequency of Fabry disease in J-FAST was 0.04 % (2/ 5408) in males and 0 % (0/3139) in females, and then 0.02 % (2/8547) in all patients. The presumptive clinical diagnoses of end-stage kidney disease (ESKD) were 10 chronic glomerulonephritis, 7 diabetic nephropathy, 3 unknown etiology, 3 nephrosclerosis, 1 gouty nephropathy, 1 autosomal dominant polycystic kidney disease and 1 renal tuberculosis among 26 tertiary positive patients. Two male Fabry patients were initially diagnosed as nephrosclerosis and chronic glomerulonephritis. Conclusions The prevalence rate of Fabry disease in J-FAST was 0.02 %. Moreover, Fabry disease could not be ruled out as the clinical diagnosis of ESKD. Keywords A-galactosidase activity Fabry disease Dialysis patient Screening X-linked recessive inheritance E66Q variation
& Eiji Kusano [email protected]
8
Department of Pediatric Nephrology, Tokyo Women’s Medical University, School of Medicine, Tokyo, Japan
1
9
Yaizu City Hospital, Shizuoka, Japan
10
Department of Clinical Nephroscience, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan
Division of Nephrology, Department of Medicine, Jichi Medical University, Tochigi, Japan
2
Division of Nephrology, Department of Medicine, Japan Community Health Care Organization Utsunomiya Hospital, Minamitakasago-cho 11-17, Tochigi 321-0143, Japan
11
3
Department of Medicine, Koga Red Cross Hospital, Ibaraki, Japan
Department of Pediatrics, The Jikei University School of Medicine, Tokyo, Japan
12
Department of Pediatrics, Institute for DNA Medicine, The Jikei University School of Medicine, Tokyo, Japan
13
Health Sciences University of Hokkaido, Hokkaido, Japan
14
Department of Genetics and Genome Science, The Jikei University School of Medicine, Tokyo, Japan
4
Tokyo Health Service Association, Tokyo, Japan
5
Department of Blood Purification, Kidney Center, Tokyo Women’s Medical University, Tokyo, Japan
6
Yokohama Dai-ichi Hospital, Kanagawa, Japan
7
Department of Nephrology, Osaka City University Graduate School of Medicine, Osaka, Japan
123
Clin Exp Nephrol (2016) 20:284–293
Introduction Fabry disease is a lysosomal disease, which is characterized by a lack of a-galactosidase A (a-Gal) activity in lysosomes [1], and causes globotriaosylceramide (GL-3) accumulation in organelles and organs [2]. This may disturb the function of organelles and organs, such as the vascular endothelium that may induce end-stage kidney disease (ESKD), cerebral infarction and myocardial infarction. As for the kidney disease, there are cases that demonstrate severe problems ranging from renal failure to the need to undergo dialysis [3]. Since 2004, an enzyme replacement therapy of a-Gal has also come to be covered by health insurance in Japan and its early treatment and early diagnostics have raised the possibility of preventing the risk of death from cardiac infarction or cerebral infarction [4, 5]. There are reports on the frequency of Fabry diseaseinduced dialysis patients, but these were not necessarily examined using a sufficient number of dialysis patients [6– 23]. With respect to the clinical diagnosis of ESKD in dialysis patients, the frequency of diabetic nephropathy, chronic glomerulonephritis, nephrosclerosis, etc., is high; however, cases in which the clinical diagnosis of ESKD is unknown are also frequent at about 7.8 % in Japan [24]. Accordingly, these unknown clinical diagnoses may include renal failure caused by Fabry disease. Moreover, there are also several cases that have not undergone a renal biopsy among people with well-known diseases, meaning that there are some cases in which the presence of Fabry disease cannot always be ruled out. Therefore, J-FAST (Japan Fabry disease screening study) was established and the screening of patients on dialysis throughout Japan was initiated from 2007.
Subjects and methods Subjects All patients undergoing dialysis for the epidemiological investigation throughout Japan were selected in this study. In addition, people undergoing pediatric renal transplant operations who had been treated with dialysis in the past were also included. Moreover, dialysis patients who agreed to J-FAST were included as subjects regardless of the clinical diagnosis of ESKD, age or sex except for Fabry disease. The examination period was 4 years and 5 months from December 2007 to May 2012. We obtained approval from the ethics committee at Jichi Medical University (EKI 10-55), and for small clinics without an ethics committee, Jichi Medical University
285
conducted an investigation by procuration. Invitations for gathering participants for J-FAST were sent to 2497 dialysis facilities all over the country. One-hundred and seventy-seven facilities applied for registering to participate; however, only 127 dialysis facilities were actually approved by the ethics committee. The screening was performed by a method consisting of 3 steps. In each of the primary, secondary and tertiary examinations, blood was collected from patients who were explained the meaning of J-FAST by the doctor at the dialysis facilities and were screened after agreeing to it. Finally, the patients who were suspected as having Fabry disease and underwent genetic testing have previously provided informed consent. Also, plasma levels of globotriaosylsphingosine (Lyso-Gb3) in those patients were measured for the confirmation of Fabry disease. Methods Lysosomal a-galactosidase A (La-Gal) protein measurement For the primary examination, the amount of plasma lysosomal a-galactosidase A (La-Gal) protein was measured. We conducted the plasma La-Gal protein measurement by the ELISA method because there is a good correlation between La-Gal protein and a-Gal activity as shown by Kitagawa et al. [25]. The patients showing lower values, which were equal to or less than the cutoff value for each measured value, underwent the next examination under suspicion of having Fabry disease. Methods of a-galactosidase A (a-Gal) activity measurement in plasma and white blood cells Plasma and white blood cells were used as specimens for measuring a-Gal activity. Briefly, 3–5 ml of blood was provided as a sample in a heparin-added blood-collecting vessel, while blood plasma was collected by carrying out centrifugation for 5 min and 1500 rpm. The blood plasma was stored at -20 °C and sent to a testing facility by a ‘‘refrigerated courier service’’. 10–50 lL of plasma was used for each specimen when actually measuring the a-Gal activity. As for white blood cells, 10 ml of whole blood was provided as a sample into a heparin-added blood-collecting vessel and was inversed and mixed slowly to prevent blood coagulation; then, this was stored at 4 °C and sent to a testing facility by a ‘‘refrigerated courier service’’. In addition, the protein content was also measured using a pyrogallol red method and the enzyme activity was displayed as the specific activity per protein content.
123
286
The measurement of a-Gal activity was conducted by the method according to Desnick et al. and Mayes et al. [26, 27]. The patients showing lower values which were equal to or less than the cutoff value for each measured value underwent genetic analysis under suspicion of having Fabry disease. Setting the cutoff value used in this screening The plasma cutoff value in this screening was set with different value in male and female. Regarding male patients, the cutoff was set at: plasma La-Gal protein level 12.0 ng/mL and plasma a-Gal activity level 1.5 nmol/mLh with the 5 percentile as the standard for positive to pick up kidney subtype and heart subtype Fabry diseases in addition to classical types. In the case of females, the reference range was expanded because patients were observed to overlap the normal range; therefore, the 10 percentile was determined as the positive standard and provisionally, the cutoff was set at: protein level 13.0 ng/mL and activity level 2.0 nmol/mLh. On the other hand, the 5 percentile of white cell blood a-Gal activity was determined as the criterion in the same manner as the plasma and the cutoff value was set at 37.5 nmol/mg protein provisionally in male and female. Genetic analysis Genomic DNA was extracted from leukocytes using blood and cell culture DNA midi kitÒ (Quiagen, Jilden, Germany). Each exon and flanking intron sequence of the GLA gene was amplified by PCR using AmpliTaq gold 360 master mixÒ (Applied Biosystems, Foster city, CA, USA), and directly sequenced using the BigDye Terminator Kit, version 3.1Ò (Applied Biosystems, Foster city, CA, USA) [28]. Measurement of plasma globotriaosylsphingosine (Lyso-Gb3) Lyso-Gb3 was extracted from plasma as described [23]. Briefly, the O-phthaldialdehyde-derivatized lyso-Gb3 was separated by high-performance liquid chromatography (Jasco Co., Hachioji, Tokyo, Japan) with a Unison UK-C18 reversed-phase column (15034.6, 3 mm; Imtakt Co., Kyoto, Japan), and quantified by fluorescence detection. For quantification, authentic lyso-Gb3 (Sigma-Aldrich Co., St. Louis, MO) was added to normal plasma. Statistical analysis Values are shown as the mean ± SD for normally distributed data and as the median (minimum, maximum) for
123
Clin Exp Nephrol (2016) 20:284–293
non-normal distributions, unless otherwise noted. Comparisons for value of La-Gal protein and a-Gal activity for each group were assessed by Tukey–Kramer parametric analysis. All data were compared to the data of 26 wellestablished Fabry disease patients (male 15, female 11) (Ohashi’s unpublished data). All analyses were conducted using JMPÒ ver. 10 (SAS Institute, Cary, NC, USA) software with p \ 0.05 considered to be significant.
Results Clinical feature of J-FAST dialysis patients The primary examination was carried out on 8547 cases of patients (male 5408, female 3139, age 62.9 ± 13.0, range 5–98 years) at 104 facilities. Clinical profile of J-FAST dialysis patients is shown in Table 1. Although composition of the clinical diagnosis of ESKD seems to be almost same with those of registry of Japanese society of dialysis therapy (JSDT), patients with diabetic nephropathy are somewhat less than those of JSDT [24]. Screening for a-Gal deficiency Flow diagram is shown in Fig. 1. In primary measurement of La-Gal protein, 575 cases (male 291, female 284) at 90 facilities were found to be positive. In secondary measurement of plasma a-Gal activity, 196 patients were found to be positive (male 95, female 101) at 69 facilities. They were asked for re-collection of their blood sampling and all 196 patients were measured (MS/MS method) for a-Gal activity in leukocytes. As a result of the tertiary examination, 26 patients were found to be positive (male 11, female 15) and the overall positive rate in dialysis patients for screening was about 0.3 %. Genetic analysis As shown in Table 2, among 26 tertiary positive patients, 6 patients (male 1, female 5) did not accept genetic analysis because of death. Therefore, remaining 20 patients (male 10, female 10) received genetic analysis, then 2 patients (male 2, female 0) have manifested a variation of the a-Gal gene (R112C anomaly), namely Fabry disease, and 11 patients (male 6, female 5) showed E66Q variations and 1 female showed point mutation on intron 6. Other 6 patients (male 2, female 4) had no variations denying Fabry disease. All patients with gene mutation were performed measurement of plasma lyso-Gb3. Only 2 patients with R112C mutation showed high level of lyso-Gb3. Otherwise, other patients with E66Q and intron 6 mutation indicated normal level of lyso-Gb3. Accordingly, the frequency of Fabry
Clin Exp Nephrol (2016) 20:284–293
287
disease in J-FAST study was 0.04 % (2/5408) in males and 0 % (0/3139) in females. Therefore, the prevalence rate of dialysis patients with Fabry disease in J-FAST study was 0.02 % (2/8547). As for E66Q variation, 11 patients (6 males, 5 females) showed this variation. The frequency of E66Q variation in J-FAST study was 0.11 % in male (6/ Table 1 Clinical profile of J-FAST dialysis patients N
8547
Sex (male/female)
5408/3139
Age (years) (range)
62.9 ± 13.0 (5–98)
Clinical diagnosis of ESKD Chronic glomerulonephritis
Clinical diagnosis of ESKD of positive subjects upon tertiary examination
3240 (37.9 %)
Diabetic nephropathy
2475 (29.0 %)
Nephrosclerosis
665 (7.8 %)
ADPKD
329 (3.8 %)
Chronic pyelonephritis
80 (0.9 %)
RPGN
64 (0.7 %)
LN
53 (0.6 %)
Others
518 (6.1 %)
Unknown
1123 (13.1 %)
ESKD end-stage kidney disease, ADPKD autosomal dominant polycystic kidney disease, RPGN rapidly progressive glomerulonephritis, LN lupus nephritis
Fig. 1 Flow diagram of J-FAST and total outcome of screening for male and female. Two male cases found to be Fabry disease, and no one of female indicated to be Fabry heterozygote. Six of males and 5 of females found as an E66Q variant. The prevalence rate of dialysis patients with Fabry disease in J-FAST study was 0.02 % (2/8547)
5408) and 0.16 % in female (5/3139). The prevalence rate of E66Q variation was a total of 0.13 %. In J-FAST, the sensitivity of biochemical screening test can be estimated to be 1 (8/8) in male and 1 (5/5) in female. The specificity was 0.9994 (5397/5400) in male and 0.9968 (3124/3134) in female. Therefore, the false-positive rate was 0.05 % (3/5399) in male and 0.32 % (10/3134) in female. For these results, we assumed that the false-negative rate could be estimated to be 0 % in both male and female, because confirmation of genetic analysis was impossible in all biochemical negative patients.
As shown in Table 2, 26 patients who were suspected of Fabry disease upon tertiary examination, the recorded clinical diagnosis of ESKD was 10 patients with chronic glomerulonephritis including IgA nephropathy, 7 patients with diabetic nephropathy, 3 patients with unknown etiology, 3 patients with nephrosclerosis, 1 patient with gouty nephropathy, 1 patient with a polycystic kidney, and 1 patient with renal tuberculosis. Considering the circumstances mentioned above, Fabry disease cannot be ruled out as the clinical diagnosis of ESKD. Actually in the present
8547 plasma samples from hemodialysis patients were obtained from 104 hospitals and clinics in Japan
5,408 male cases Plasma α-gal A protein was measured.
3,139 female cases Plasma α-gal A protein was measured.
Primary examination 291 cases Plasma α-gal A activity was < 5 percentile of the controls, was requested to reconfirm the result.
284 cases Plasma α-gal A activity was < 10 percentile of the controls, was requested to reconfirm the result.
95 cases If α-gal A activity was
ORIGINAL ARTICLE
Prevalence of Fabry disease in dialysis patients: Japan Fabry disease screening study (J-FAST) Osamu Saito1 • Eiji Kusano2 • Tetsu Akimoto1 • Yasushi Asano3 • Teruo Kitagawa4 Ken Suzuki4 • Nobuyuki Ishige4 • Takashi Akiba5 • Akira Saito6 • Eiji Ishimura7 • Motoshi Hattori8 • Akira Hishida9 • Chu Guili10 • Hiroki Maruyama10 • Masahisa Kobayashi11 • Touya Ohashi12 • Ichiro Matsuda13 • Yoshikatsu Eto14
•
Received: 30 March 2015 / Accepted: 11 July 2015 / Published online: 22 July 2015 Ó Japanese Society of Nephrology 2015
Abstract Background In Fabry disease, progressive glycolipid accumulation leads to damage in kidney and other organs. This study was designed to determine the prevalence rate of Fabry disease in Japanese dialysis patients. Methods All dialysis patients agreeing to Japan Fabry disease screening study (J-FAST) with informed consent were selected except for Fabry disease. The screening was performed by a method of measuring plasma and/or leukocytes lysosomal a-galactosidase A protein level and a-galactosidase A activity. If positive, genetic analysis was carried out upon patient’s agreement. Results J-FAST dealt with 8547 patients (male 5408, female 3139). At the tertiary examination, 26 out of 8547 patients were found to be positive. Six out of 26 patients could not accept genetic analysis because of death. Remaining 20 patients agreed with genetic analysis; then 2 patients (male 2, female 0) had a variation of the a-Gal
gene and 11 patients showed E66Q variations. Therefore, the frequency of Fabry disease in J-FAST was 0.04 % (2/ 5408) in males and 0 % (0/3139) in females, and then 0.02 % (2/8547) in all patients. The presumptive clinical diagnoses of end-stage kidney disease (ESKD) were 10 chronic glomerulonephritis, 7 diabetic nephropathy, 3 unknown etiology, 3 nephrosclerosis, 1 gouty nephropathy, 1 autosomal dominant polycystic kidney disease and 1 renal tuberculosis among 26 tertiary positive patients. Two male Fabry patients were initially diagnosed as nephrosclerosis and chronic glomerulonephritis. Conclusions The prevalence rate of Fabry disease in J-FAST was 0.02 %. Moreover, Fabry disease could not be ruled out as the clinical diagnosis of ESKD. Keywords A-galactosidase activity Fabry disease Dialysis patient Screening X-linked recessive inheritance E66Q variation
& Eiji Kusano [email protected]
8
Department of Pediatric Nephrology, Tokyo Women’s Medical University, School of Medicine, Tokyo, Japan
1
9
Yaizu City Hospital, Shizuoka, Japan
10
Department of Clinical Nephroscience, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan
Division of Nephrology, Department of Medicine, Jichi Medical University, Tochigi, Japan
2
Division of Nephrology, Department of Medicine, Japan Community Health Care Organization Utsunomiya Hospital, Minamitakasago-cho 11-17, Tochigi 321-0143, Japan
11
3
Department of Medicine, Koga Red Cross Hospital, Ibaraki, Japan
Department of Pediatrics, The Jikei University School of Medicine, Tokyo, Japan
12
Department of Pediatrics, Institute for DNA Medicine, The Jikei University School of Medicine, Tokyo, Japan
13
Health Sciences University of Hokkaido, Hokkaido, Japan
14
Department of Genetics and Genome Science, The Jikei University School of Medicine, Tokyo, Japan
4
Tokyo Health Service Association, Tokyo, Japan
5
Department of Blood Purification, Kidney Center, Tokyo Women’s Medical University, Tokyo, Japan
6
Yokohama Dai-ichi Hospital, Kanagawa, Japan
7
Department of Nephrology, Osaka City University Graduate School of Medicine, Osaka, Japan
123
Clin Exp Nephrol (2016) 20:284–293
Introduction Fabry disease is a lysosomal disease, which is characterized by a lack of a-galactosidase A (a-Gal) activity in lysosomes [1], and causes globotriaosylceramide (GL-3) accumulation in organelles and organs [2]. This may disturb the function of organelles and organs, such as the vascular endothelium that may induce end-stage kidney disease (ESKD), cerebral infarction and myocardial infarction. As for the kidney disease, there are cases that demonstrate severe problems ranging from renal failure to the need to undergo dialysis [3]. Since 2004, an enzyme replacement therapy of a-Gal has also come to be covered by health insurance in Japan and its early treatment and early diagnostics have raised the possibility of preventing the risk of death from cardiac infarction or cerebral infarction [4, 5]. There are reports on the frequency of Fabry diseaseinduced dialysis patients, but these were not necessarily examined using a sufficient number of dialysis patients [6– 23]. With respect to the clinical diagnosis of ESKD in dialysis patients, the frequency of diabetic nephropathy, chronic glomerulonephritis, nephrosclerosis, etc., is high; however, cases in which the clinical diagnosis of ESKD is unknown are also frequent at about 7.8 % in Japan [24]. Accordingly, these unknown clinical diagnoses may include renal failure caused by Fabry disease. Moreover, there are also several cases that have not undergone a renal biopsy among people with well-known diseases, meaning that there are some cases in which the presence of Fabry disease cannot always be ruled out. Therefore, J-FAST (Japan Fabry disease screening study) was established and the screening of patients on dialysis throughout Japan was initiated from 2007.
Subjects and methods Subjects All patients undergoing dialysis for the epidemiological investigation throughout Japan were selected in this study. In addition, people undergoing pediatric renal transplant operations who had been treated with dialysis in the past were also included. Moreover, dialysis patients who agreed to J-FAST were included as subjects regardless of the clinical diagnosis of ESKD, age or sex except for Fabry disease. The examination period was 4 years and 5 months from December 2007 to May 2012. We obtained approval from the ethics committee at Jichi Medical University (EKI 10-55), and for small clinics without an ethics committee, Jichi Medical University
285
conducted an investigation by procuration. Invitations for gathering participants for J-FAST were sent to 2497 dialysis facilities all over the country. One-hundred and seventy-seven facilities applied for registering to participate; however, only 127 dialysis facilities were actually approved by the ethics committee. The screening was performed by a method consisting of 3 steps. In each of the primary, secondary and tertiary examinations, blood was collected from patients who were explained the meaning of J-FAST by the doctor at the dialysis facilities and were screened after agreeing to it. Finally, the patients who were suspected as having Fabry disease and underwent genetic testing have previously provided informed consent. Also, plasma levels of globotriaosylsphingosine (Lyso-Gb3) in those patients were measured for the confirmation of Fabry disease. Methods Lysosomal a-galactosidase A (La-Gal) protein measurement For the primary examination, the amount of plasma lysosomal a-galactosidase A (La-Gal) protein was measured. We conducted the plasma La-Gal protein measurement by the ELISA method because there is a good correlation between La-Gal protein and a-Gal activity as shown by Kitagawa et al. [25]. The patients showing lower values, which were equal to or less than the cutoff value for each measured value, underwent the next examination under suspicion of having Fabry disease. Methods of a-galactosidase A (a-Gal) activity measurement in plasma and white blood cells Plasma and white blood cells were used as specimens for measuring a-Gal activity. Briefly, 3–5 ml of blood was provided as a sample in a heparin-added blood-collecting vessel, while blood plasma was collected by carrying out centrifugation for 5 min and 1500 rpm. The blood plasma was stored at -20 °C and sent to a testing facility by a ‘‘refrigerated courier service’’. 10–50 lL of plasma was used for each specimen when actually measuring the a-Gal activity. As for white blood cells, 10 ml of whole blood was provided as a sample into a heparin-added blood-collecting vessel and was inversed and mixed slowly to prevent blood coagulation; then, this was stored at 4 °C and sent to a testing facility by a ‘‘refrigerated courier service’’. In addition, the protein content was also measured using a pyrogallol red method and the enzyme activity was displayed as the specific activity per protein content.
123
286
The measurement of a-Gal activity was conducted by the method according to Desnick et al. and Mayes et al. [26, 27]. The patients showing lower values which were equal to or less than the cutoff value for each measured value underwent genetic analysis under suspicion of having Fabry disease. Setting the cutoff value used in this screening The plasma cutoff value in this screening was set with different value in male and female. Regarding male patients, the cutoff was set at: plasma La-Gal protein level 12.0 ng/mL and plasma a-Gal activity level 1.5 nmol/mLh with the 5 percentile as the standard for positive to pick up kidney subtype and heart subtype Fabry diseases in addition to classical types. In the case of females, the reference range was expanded because patients were observed to overlap the normal range; therefore, the 10 percentile was determined as the positive standard and provisionally, the cutoff was set at: protein level 13.0 ng/mL and activity level 2.0 nmol/mLh. On the other hand, the 5 percentile of white cell blood a-Gal activity was determined as the criterion in the same manner as the plasma and the cutoff value was set at 37.5 nmol/mg protein provisionally in male and female. Genetic analysis Genomic DNA was extracted from leukocytes using blood and cell culture DNA midi kitÒ (Quiagen, Jilden, Germany). Each exon and flanking intron sequence of the GLA gene was amplified by PCR using AmpliTaq gold 360 master mixÒ (Applied Biosystems, Foster city, CA, USA), and directly sequenced using the BigDye Terminator Kit, version 3.1Ò (Applied Biosystems, Foster city, CA, USA) [28]. Measurement of plasma globotriaosylsphingosine (Lyso-Gb3) Lyso-Gb3 was extracted from plasma as described [23]. Briefly, the O-phthaldialdehyde-derivatized lyso-Gb3 was separated by high-performance liquid chromatography (Jasco Co., Hachioji, Tokyo, Japan) with a Unison UK-C18 reversed-phase column (15034.6, 3 mm; Imtakt Co., Kyoto, Japan), and quantified by fluorescence detection. For quantification, authentic lyso-Gb3 (Sigma-Aldrich Co., St. Louis, MO) was added to normal plasma. Statistical analysis Values are shown as the mean ± SD for normally distributed data and as the median (minimum, maximum) for
123
Clin Exp Nephrol (2016) 20:284–293
non-normal distributions, unless otherwise noted. Comparisons for value of La-Gal protein and a-Gal activity for each group were assessed by Tukey–Kramer parametric analysis. All data were compared to the data of 26 wellestablished Fabry disease patients (male 15, female 11) (Ohashi’s unpublished data). All analyses were conducted using JMPÒ ver. 10 (SAS Institute, Cary, NC, USA) software with p \ 0.05 considered to be significant.
Results Clinical feature of J-FAST dialysis patients The primary examination was carried out on 8547 cases of patients (male 5408, female 3139, age 62.9 ± 13.0, range 5–98 years) at 104 facilities. Clinical profile of J-FAST dialysis patients is shown in Table 1. Although composition of the clinical diagnosis of ESKD seems to be almost same with those of registry of Japanese society of dialysis therapy (JSDT), patients with diabetic nephropathy are somewhat less than those of JSDT [24]. Screening for a-Gal deficiency Flow diagram is shown in Fig. 1. In primary measurement of La-Gal protein, 575 cases (male 291, female 284) at 90 facilities were found to be positive. In secondary measurement of plasma a-Gal activity, 196 patients were found to be positive (male 95, female 101) at 69 facilities. They were asked for re-collection of their blood sampling and all 196 patients were measured (MS/MS method) for a-Gal activity in leukocytes. As a result of the tertiary examination, 26 patients were found to be positive (male 11, female 15) and the overall positive rate in dialysis patients for screening was about 0.3 %. Genetic analysis As shown in Table 2, among 26 tertiary positive patients, 6 patients (male 1, female 5) did not accept genetic analysis because of death. Therefore, remaining 20 patients (male 10, female 10) received genetic analysis, then 2 patients (male 2, female 0) have manifested a variation of the a-Gal gene (R112C anomaly), namely Fabry disease, and 11 patients (male 6, female 5) showed E66Q variations and 1 female showed point mutation on intron 6. Other 6 patients (male 2, female 4) had no variations denying Fabry disease. All patients with gene mutation were performed measurement of plasma lyso-Gb3. Only 2 patients with R112C mutation showed high level of lyso-Gb3. Otherwise, other patients with E66Q and intron 6 mutation indicated normal level of lyso-Gb3. Accordingly, the frequency of Fabry
Clin Exp Nephrol (2016) 20:284–293
287
disease in J-FAST study was 0.04 % (2/5408) in males and 0 % (0/3139) in females. Therefore, the prevalence rate of dialysis patients with Fabry disease in J-FAST study was 0.02 % (2/8547). As for E66Q variation, 11 patients (6 males, 5 females) showed this variation. The frequency of E66Q variation in J-FAST study was 0.11 % in male (6/ Table 1 Clinical profile of J-FAST dialysis patients N
8547
Sex (male/female)
5408/3139
Age (years) (range)
62.9 ± 13.0 (5–98)
Clinical diagnosis of ESKD Chronic glomerulonephritis
Clinical diagnosis of ESKD of positive subjects upon tertiary examination
3240 (37.9 %)
Diabetic nephropathy
2475 (29.0 %)
Nephrosclerosis
665 (7.8 %)
ADPKD
329 (3.8 %)
Chronic pyelonephritis
80 (0.9 %)
RPGN
64 (0.7 %)
LN
53 (0.6 %)
Others
518 (6.1 %)
Unknown
1123 (13.1 %)
ESKD end-stage kidney disease, ADPKD autosomal dominant polycystic kidney disease, RPGN rapidly progressive glomerulonephritis, LN lupus nephritis
Fig. 1 Flow diagram of J-FAST and total outcome of screening for male and female. Two male cases found to be Fabry disease, and no one of female indicated to be Fabry heterozygote. Six of males and 5 of females found as an E66Q variant. The prevalence rate of dialysis patients with Fabry disease in J-FAST study was 0.02 % (2/8547)
5408) and 0.16 % in female (5/3139). The prevalence rate of E66Q variation was a total of 0.13 %. In J-FAST, the sensitivity of biochemical screening test can be estimated to be 1 (8/8) in male and 1 (5/5) in female. The specificity was 0.9994 (5397/5400) in male and 0.9968 (3124/3134) in female. Therefore, the false-positive rate was 0.05 % (3/5399) in male and 0.32 % (10/3134) in female. For these results, we assumed that the false-negative rate could be estimated to be 0 % in both male and female, because confirmation of genetic analysis was impossible in all biochemical negative patients.
As shown in Table 2, 26 patients who were suspected of Fabry disease upon tertiary examination, the recorded clinical diagnosis of ESKD was 10 patients with chronic glomerulonephritis including IgA nephropathy, 7 patients with diabetic nephropathy, 3 patients with unknown etiology, 3 patients with nephrosclerosis, 1 patient with gouty nephropathy, 1 patient with a polycystic kidney, and 1 patient with renal tuberculosis. Considering the circumstances mentioned above, Fabry disease cannot be ruled out as the clinical diagnosis of ESKD. Actually in the present
8547 plasma samples from hemodialysis patients were obtained from 104 hospitals and clinics in Japan
5,408 male cases Plasma α-gal A protein was measured.
3,139 female cases Plasma α-gal A protein was measured.
Primary examination 291 cases Plasma α-gal A activity was < 5 percentile of the controls, was requested to reconfirm the result.
284 cases Plasma α-gal A activity was < 10 percentile of the controls, was requested to reconfirm the result.
95 cases If α-gal A activity was
