To the Editor:
We are writing in response to a recent case report by Gubbels and colleagues (1) describing a woman with classic galactosemia who became pregnant shortly after a low anti-mullerian hormone (AMH) blood measurement. The authors contrast their case with results published by our group from a study of 35 galactosemic girls and women (2). We demonstrated that 32/35 of these individuals had low to undetectable AMH, and noted the similarity of this frequency with rates of clinical ovarian insufficiency reported for this patient population. Gubbels and colleagues’ stated objective was “to draw attention to the limited predictive value of ovarian reserve tests” in galactosemics, and they concluded, “Commonly used ovarian function and reserve tests seem to have no significance.”
While the patient they described is surely of interest, their conclusions are problematic. A few points:
1. AMH is made by a subset of maturing follicles. Low to undetectable AMH does not mean a woman is infertile, but that she has ovarian insufficiency and may be less likely to conceive than other women her same age demonstrating normal levels of AMH.
2. The young woman is reported as having “undetectable AMH,” although her level is listed as <0.1µg/L. The lower limit of detection in our study was 0.01µg/L, or ten-fold lower.
3. The patient described in the case had galactose 1-phosphate uridyltransferase (GALT) activity measured at 0.82-1.23 µmol/h/g Hb; the reference level was 4.8-30 µmol/h/g Hb. While the authors state that “the residual activity is too low to have any clinical relevance,” we respectfully disagree. This level of GALT activity is well above that typically seen in patients with classic galactosemia and may be of great clinical significance.
4. Gubbels and colleagues suggest that “malfunction of gonadotropins” may underlie the ovarian insufficiency associated with galactosemia. We tested this hypothesis (Figure 4 ); FSH bioactivity in samples from 10 galactosemic patients failed to show a statistically-significant difference compared to the FSH bioactivity in samples from 8 controls.
In summary, we wish to stress that while the conclusions drawn by Gubbels and colleagues may be at odds with our paper, the results of the case are not. Further, both reports agree that, absent surgical removal of the ovaries or uterus, one should never tell a young woman with classic galactosemia that she is absolutely infertile. We all still have much to learn about the causes, natural history and best predictors of ovarian insufficiency in galactosemia.
Judith L. Fridovich-Keil, Ph.D.a
Rebecca D. Sanders, B.A.b
Jessica B. Spencer, M.D., M.Sc.c
Michael P. Epstein, Ph.D.a
Joyce W. Lustbader, M.D.d
Pratibhasri A. Vardhana, M.D.d
aDepartment of Human Genetics
bGraduate Program in Biochemistry, Cell and Developmental Biology
cDepartment of Gynecology and Obstetrics
Emory University School of Medicine
dDepartment of Obstetrics and Gynecology
Columbia University College of Physicians & Surgeons
New York, New York
1. Gubbels CS, Kuppens SM, Bakker JA, Konings CJ, Wodzig KW, de Sain-van der Velden MG, Menheere PP, Rubio-Gozalbo ME. Pregnancy in classic galactosemia despite undetectable anti-Müllerian hormone. Fertil Steril April 2009 [Epub ahead of print]
2. Sanders RD, Spencer JB, Epstein MP, Pollak SV, Vardhana PA, Lustbader JW, Fridovich-Keil JL. Biomarkers of ovarian function in girls and women with classic galactosemia. Fertil Steril. 2008 Aug 4. [Epub ahead of print]
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2009.06.016
The Authors Respond:
We thank Prof. Fridovich-Keil’s group for their comments and take this opportunity to respond to them.
1. “AMH is made by a subset of maturing follicles. Low to undetectable AMH does not mean a woman is infertile, but that she may be less likely to conceive.” We completely agree with this. A low AMH only indicates that there is a lower maturation rate. Whether this is due to fewer primordial follicles, a maturation arrest, or both, remains to be elucidated.
2. The exact level of AMH measured in this patient was 0.018 µg/L. However, since the minimal detectable concentration (MDC, defined as the lowest concentration that can be detected with 95% confidence interval) of this assay was 0.025 µg/L, the concentration in this sample was below the detection limit. The detection limit of 0.01 µg/L, as described by Fridovich-Keil and colleagues in their article (1) was probably also below the calculated MDC in their assays. In fact, we use the same (commercially available) assay and have calculated the MDC of our runs, which varied between 0.018 and 0.039 µg/L. The value of 0.01 µg/L should not have any analytical significance.
3. The level of GALT activity in the described patient is 0.70-1.23 µmol/h/g Hb during pregnancy [in animal models it has been seen that GALT activity is upregulated during pregnancy (2)]. When calculating the percentage of normal activity, we use our non-carrier reference population, leaving our patient with levels of GALT activity of 2.7-4.7% during pregnancy. Measurement of GALT activity in the neonatal period was 0.58 mmol P/mol Hb/min, corresponding to 0.5 µmol/h/g Hb. This was, however, measured after a blood transfusion, and her GALT activity at this point must have been lower. The patient’s mutation, medical history (including neonatal toxic syndrome, cognitive impairment, premature ovarian insufficiency and diminished bone mineral density) and metabolite levels are in line with a severe classic galactosemia phenotype.
4. Prof. Fridovich-Keil’s group elegantly showed that FSH bioactivity in vitro was within reference range or even higher in a subset of patients (1), in line with earlier observations (3). This does not necessarily exclude FSH malfunction. The FSH signal intensity and duration is regulated through the glycan-dependent isoform pattern (4) which might be affected in this disease (5). Possible FSH-receptor glycan abnormality and subsequent malfunction is not ruled out by the described reactions to exogenous FSH (3, 6), as supraphysiological concentrations are administered. Data regarding isoform patterns in these patients or FSH-FSH receptor interaction are limited. Despite FSH bioactivity being normal, these other mechanisms need to be explored.
In conclusion, we stand by our earlier statement that AMH, as well as the other described ovarian reserve tests, may not be the best predictive factors for pregnancy in these women. If there is a maturation arrest in this disease, AMH will be low despite the presence of primordial follicles. As the etiology is not well understood in classic galactosemia, the value of these tests might differ from that in the general population.
Cynthia S Gubbels, M.D.a
Paul P Menheere, Ph.D.b
M. Estela Rubio-Gozalbo, M.D., Ph.D.a,c
aDepartment of Pediatrics
bDepartment of Clinical Chemistry
cLaboratory of Inherited Metabolic Diseases
Maastricht University Medical Center
Maastricht, The Netherlands
1. Sanders RD, Spencer JB, Epstein MP, Pollak SV, Vardhana PA, Lustbader JW et al. Biomarkers of ovarian function in girls and women with classic galactosemia. Fertil Steril;Advance Access published Aug 4, 2008.
2. Rogers SR, Bovee BW, Saunders SL, Segal S. Activity of hepatic galactose-metabolizing enzymes in the pregnant rat and fetus. Pediatr Res 1989;25:161-6.
3. Kaufman FR, Kogut MD, Donnell GN, Goebelsmann U, March C, Koch R. Hypergonadotropic hypogonadism in female patients with galactosemia. N Engl J Med 1981;304:994-8.
4. Ulloa-Aguirre A, Midgley AR, Jr., Beitins IZ, Padmanabhan V. Follicle-stimulating isohormones: characterization and physiological relevance. Endocr Rev 1995;16:765-87.
5. Prestoz LL, Couto AS, Shin YS, Petry KG. Altered follicle stimulating hormone isoforms in female galactosaemia patients. Eur J Pediatr 1997;156:116-20.
6. Menezo YJ, Lescaille M, Nicollet B, Servy EJ. Pregnancy and delivery after stimulation with rFSH of a galatosemia patient suffering hypergonadotropic hypogonadism: case report. J Assist Reprod Genet 2004;21:89-90.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2009.06.017