Fabry or not Fabry – a question of ascertainment

Fabry disease, an X-linked sphingolipidosis,¹ is the most common lysosomal storage disease in Norway.² Affected individuals may be offered intravenous enzyme replacement therapy, every second week at a cost of about € 200 000 per year for an average adult. To prevent organ damage, early treatment is probably the best. Accordingly, Fabry disease screening of patients with renal failure, cardiomyopathy, stroke and small-fiber neuropathy has been advocated.³ If Fabry disease is detected, the family can be offered diagnostic follow-up and treatment of mutation carriers.

Recently, the diagnosis of Fabry disease in both transplanted kidneys from a 16-year-old male traffic victim was published.⁴ Kidney graft biopsy samples at the time of transplantation and at 3 and 12 years later showed large and constant amounts of Fabry-specific lipid deposits, mainly in the podocytes. Furthermore, the alpha-galactosidase A activity in a donor plasma sample at the time of transplantation was 0.3 nkat/l (normal range 2.3–9.9), that is, the same level as in classical Fabry disease males.⁴ Fourteen years later (this study), a likely missense mutation of an evolutionary conserved isoleucine residue, c.593C>T [p.I198T], was found in GLA, the alpha-galactosidase A gene, in an archived blood sample (Table 1). This amino acid substitution has a well-established pathogenic potential and constitutes about 1% of all missense changes in the Human Gene Mutation Database (http://www.hgmd.org). Retrospectively, there were no indications of clinical disease, that is, no history of pain in hands or feet, and no records of poor sweating or unexplained fever episodes. However, it is well known that clinical symptoms of Fabry disease can be minimal or remain unnoticed at young age, also in males.

Table 1. Findings in female carriers of the GLA missense change.

	Sister 1 (donor's mother)	Sister 2	Sister 3	Sister 4
GLA c.593C>T [p.Ile198T]	+	+	+	+
Blood α-gal A activity (μkat/kg prot, ref. 22–36)	2.2	3.4	6.3	13.4
Plasma GL-3 (μmol/l, ref. 1.6–3.3)	3.9	2.9	2.3	3.0
Urine GL-3 (ref. <10 μmol/mol creat)	11	0.26	0.37	0.16
GFR (ml/min/1.73 m2)	89	79	109	105
U-microalbuminuria (mg/nmol creat, ref. < 2.5)	0.6–0.7	0.6–1.0	1.1–1.7	0.8–1.2
EKG	N	N	N	N
LV mass (g/m2.7, ref. 18–44)	60	37	38	38
Kidney biopsy	n.d.	n.d.	N	N
Cerebral MRI with MR–angio	∼N	N	∼N	∼N
Skin angiokeratomas	none	none	none	none
Eye examination (eg cornea verticillata)	N	N	N	N
Audiometry	N	N	N	N
Thermotest (for thin-fiber neuropathy)	N	N	∼N	∼N
Offspring	2 boys	2 boys/1 girl	3 boys	2 boys/1 girl
Other findings	Hypertension	Healthy	Healthy	Healthy

Abbreviations: +, ‘present' N, normal; ∼N, nearly normal (minimal changes); n.d., not done; ref., reference interval.

Thirteen years after the kidney transplantation, a diagnostic follow-up of the donor's family was initiated. The boy's mother and the mother's three sisters, now 54–64 years old, were diagnosed with from 10 to 61% of lower normal enzyme activity levels in blood leukocytes (from 2.2–13.4 μkat/kg protein in the presence of N-acetylgalactosamine inhibitor, normal range 23–38), as commonly found in Fabry heterozygotes (Table 1). His mother also had elevated plasma and urine globotriaosylceramide (GL-3) levels. However, as in her son, Fabry-related symptoms were lacking, and our baseline Fabry investigations gave normal findings (Table 1). Diagnostic kidney biopsies, performed in two of the mother's sisters, revealed completely normal findings in one and only a single cell (podocyte) with myelin figures in the other. All the sisters' 11 children (9 boys and 2 girls) were healthy and asymptomatic.

The index patient certainly fulfilled major diagnostic criteria for a Fabry disease diagnosis – but would he have become clinically affected? The family history suggests otherwise: the maternal grandmother was GLA c.593C>T-negative, and the maternal grandfather had died at an age of 91 years after a life as a healthy man. Even though confined germline mosaicism for a GLA mutation could be an explanation for his lack of symptoms,⁵ we find it more likely that this previously unknown GLA missense change is non-penetrant. The major reason for this is the normal kidney biopsies in two of his daughters. Characteristic morphological changes in the kidneys are unequivocal findings in both male and female Fabry patients, from early childhood and onward.⁶ In addition, the a priori likelihood that nine grown-up sons of female GLA mutation carriers have always been completely asymptomatic, if there really is Fabry disease in the family, is <1%.

The prevalence of such biochemically true-positive but clinically false-positive alleles in metabolic enzymes is unknown. We suggest calling them as fringe alleles. Such alleles are most likely found upon screening of low-risk group, like in general newborn screening. In Fabry disease, pharmaceutical companies advocate low threshold for testing and family follow-up. This family exemplifies that diagnostic skepticism may be warranted if clinical support of a metabolic condition is lacking. In the era of expensive orphan drug therapy for an increasing number of metabolic disorders, this letter is meant as a word of caution against the prevailing view that is reminiscent of a Robbie Williams hit: Let me ascertain you!

The authors declare no conflict of interest.