To the Editor:
Yuko Hayashi and coauthors confirm single nucleotide polymorphism (SNP) rs1050606 (SNP5) of the nontranslated region of the ANXA5 gene as a risk factor for recurrent miscarriage (RM) in Japanese patients (1). They also conclude that the risk allele has no influence on pregnancy outcome from 79 abortions that occurred from November 2012 to February 2013 in a group of 264 Japanese women with RM history in Nagoya City University Hospital. Almost two-thirds of the 79 aborted fetuses, or ~ 64% had abnormal karyotypes, which leaves only n = 17 cases possibly influenced by thrombophilia through carriage of common ANXA5 SNPs/haplotypes, among other factors. Moreover, since no data about the timing of miscarriages are shared, but the majority of abortions in the patient group are apparently due to chromosomal aberrations, it is rather likely this is an early (predominantly before 10th week of gestation) RM group. It is certainly very plausible that live birth rates in such a preselected cohort would not be notably influenced by carriage of ANXA5 SNPs/haplotypes, as M2/ANXA5 for example should not be significant as a risk factor before weeks 10 to 12 of gestation (2).
Carriage rates of SNP5 in Japanese RM patients and control subjects have been established in a previous study (3), and they were 39.5% and 25%, respectively. This is based on estimated minor allelic frequencies (MAF) for SNP5 at 0.228 for patients and 0.130 for controls. The last value is in accordance with the Centre d’Etudes du Polymorphisme Humaine (CEPH) reference for rs1050606 in the Japanese population. While Hayashi et al. confirm MAF of 0.228 for their patient cohort, they come up with MAF of 0.192 for their control subjects, which is almost 1.5 times higher than previously stipulated. Thus, they can only justify the role of SNP5 as a risk factor and claim statistical significance when pooling together their own data with the previously published (3) reference (“merged data”).
In addition, they conclude no risk role for the M2 haplotype (SNPs rs112782763, rs28717001, rs28651243, and rs113588187), currently confirmed as an RM predisposition factor in European populations (2) and previously described as an RM risk factor in the Japanese population (3). This conclusion, again, is based on estimated MAFs of 0.115 (0.107, ref. 3) for RM patients and 0.102 for their control subjects. The last number, resulting in a carriage rate of 18% for the M2 haplotype is quite high, even for European control populations (at 15%, ref. 2), and almost twice as high as the previously estimated MAF of 0.055 for M2 in Japanese controls, corresponding to a carriage rate of 11% (3).
Since selection of true random healthy controls in a population is a rather difficult task, we cannot agree with the results presented in this study based on MAFs calculated for ANXA5 nontranslated region SNPs with values that are quite different from those previously reported. A recent study on Japanese population structure (4) concludes for example that “the inclusion of different proportions of individuals from different regions of Japan in case and control groups can lead to an inflated rate of false-positive results when the sample sizes are large.”
While discussing the role of SNP5 as an RM risk factor, Hayashi et al. also state that this SNP is not associated with the M2 haplotype. We observed this to be the case for European populations too, where SNP5 does not present as an RM predisposition (5).
Arseni Markoff, Ph.D., Institute of Medical Biochemistry, ZMBE
Nadja Bogdanova, M.D., Ph.D., Institute of Human Genetics, UKM and University of Muenster
1. Hayashi Y, Sasaki H, Suzuki S, Nishiyama T, Kitaori T, Mizutani E, et al. Genotyping analyses for polymorphisms of ANXA5 gene in patients with recurrent pregnancy loss. Fertil Steril 2013 doi:pii: S0015-0282(13)00719-X. 10.1016/j.fertnstert.2013.06.020.
2. Tüttelmann F, Ivanov P, Dietzel C, Sofroniou A, Tsvyatkovska TM, Komsa-Penkova RS, et al. Further insights into the role of the annexin A5 M2 haplotype as recurrent pregnancy loss factor, assessing timing of miscarriage and partner risk. Fertil Steril 2013 doi:pii: S0015-0282(13)00746-2. 10.1016/j.fertnstert.2013.06.046.
3. Miyamura H, Nishizawa H, Ota S, Suzuki M, Inagaki A, Egusa H, et al. Polymorphisms in the annexin A5 gene promoter in Japanese women with recurrent pregnancy loss. Mol Hum Reprod 2011;17:447-52.
4. Yamaguchi-Kabata Y, Nakazono K, Takahashi A, Saito S, Hosono N, Kubo M, et al. Japanese population structure, based on SNP genotypes from 7003 individuals compared to other ethnic groups: effects on population-based association studies. Am J Hum Genet 2008;83:445-56.
5. Rogenhofer N, Engels L, Bogdanova N, Tüttelmann F, Thaler CJ, Markoff A. Independent association of the M2/ANXA5 haplotype with recurrent pregnancy loss (RPL) in PCOS patients. Metabolism 2013;62:1057-60.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2013.10.016
The authors respond:
We do not deny that ANXA5 single nucleotide polymorphisms (SNPs) were risk alleles. Variations with the ANXA5 gene upstream region, especially SNP5, were confirmed to be risk factors of recurrent pregnancy loss (RPL), but not recurrent miscarriage, in our cross-sectional study (1). The reason why significant difference of the other SNPs except SNP5 could not be obtained in our study might be that the mean number of previous pregnancy losses was less than that in Miyamura’s study (2). The abnormal rate of embryonic (fetal) karyotype depends on the women’s age and the number of previous miscarriages. Over half of abnormalities were found in less than four miscarriages (3).
The presence or absence of the ANXA5 risk allele did not have any predictive effect for subsequent pregnancy outcome in unselected patients with RPL. Our patients with a history of two or more RPLs were not selected from patients who visited our hospital for examination. Recently, we found that an abnormal embryonic (fetal) karyotype was the most frequent cause of RPL, accounting for as high as 41% of the cases (4). The prevalence of truly unexplained causes, with normal karyotype, was only 25%. Thus, the predictive value of ANXA5 SNPs for subsequent pregnancy outcome might be found if the study was conducted in patients with five or more miscarriages or only miscarriages with normal embryonic (fetal) karyotype. Such patients are rare.
Furthermore, our controls had a history of live birth and no history of miscarriage, which is different from the general population. In the general population, 15% and 10%–15% of women, respectively, experience miscarriage and infertility. Our control women were more appropriate for controls against RPL.
Risk alleles have shown a relatively small odds ratio around 2.0 in the previous study (2). Our cohort study suggested that risk alleles with a small odds ratio had little influence on subsequent live birth. The subsequent live birth rate was 84.0% and 84.3% in patients with and without the risk allele of SNP5, after the exclusion of cases with an abnormal embryonic (fetal) karyotype. The combination of many kinds of risk alleles might cause the susceptibility to RPL. In fact, some patients received anticoagulant treatment because of risk alleles of ANXA5 in Japan, though the randomized controlled trial concluded that there was no effect of combined heparin and aspirin treatment for unexplained recurrent miscarriage (5). Our study was very important for clinicians and patients because it proved that patients with a risk allele could give live birth without medication.
Yuko Hayashi, M.D. and Mayumi Sugiura-Ogasawara, M.D.
Department of Obstetrics and Gynecology, Nagoya City University, Graduate School of Medical Sciences, Nagoya, Aichi, Japan
1. Hayashi Y, Sasaki H, Suzuki S, Nishiyama T, Kitaori T, Mizutani E, et al. Genotyping analyses for polymorphisms of ANXA5 gene in patients with recurrent pregnancy loss. Fertil Steril, in press.
2. Miyamura H, Nishizawa H, Ota S, Suzuki M, Inagaki A, Egusa H, et al. Polymorphisms in the annexin A5 gene promoter in Japanese women with recurrent pregnancy loss. Mol Hum Reprod 2011;17:447-52.
3. Ogasawara M, Aoki K, Okada S, Suzumori K. Embryonic karyotype of abortuses in relation to the number of previous miscarriages. Fertil Steril 2000;73:300-4.
4. Sugiura-Ogasawara M, Ozaki Y, Katano K, Suzumori N, Kitaori T, Mizutani E. Abnormal embryonic karyotype is the most frequent cause of recurrent miscarriage. Hum Reprod 2012;27:2297-303.
5. Kaandrop SP, Goddijn M, van der Post JA, Hutten BA, Verhoeve HR, Hamulyak K, et al. Aspirin plus heparin or aspirin alone in women with recurrent miscarriage. N Engl J Med 2010;362:1586-96.
Published online in Fertility and Sterility doi:10.1016/j.fertnstert.2013.10.018