Jogc-aug_07.vp

DRUGS IN PREGNANCY
DRUGS IN PREGNANCY
MOTHERISK ROUNDS
Fetal Safety of Letrozole and Clomiphene
Citrate for Ovulation Induction
Rachel Forman, MD, FRCSC,1 Simmerpal Gill, MSc,2 Myla Moretti, MSc,2 Togas Tulandi, MD CM,3
Gideon Koren, MD, FRCPC,2 Robert Casper, MD, FRCSC1
1Toronto Centre for Advancing Reproduction Technology, Toronto ON
2Motherisk Program, The Hospital for Sick Children, Toronto ON
3Department of Obstetrics and Gynecology, McGill University, Montreal QC
estrogen secretion is restored (after clearance of letrozole).8 J Obstet Gynaecol Can 2007;29(8):668–671 In its ability to induce ovulation, letrozole compares favour-ably to clomiphene citrate, which has been the first linetreatment for ovulatory disorders for more than 40 years, INTRODUCTION
and it has emerged as an alternative to CC for ovulationinduction. The clinical outcome of early pregnancies Letrozoleisathird-generationselectivearomataseinhibi- achieved through the use of letrozole for ovulation induc- tor that blocks the rate-limiting step in the production tion or controlled ovarian hyperstimulation for intrauterine of estrogen from androstenedione and testosterone sub- insemination was reported in 2005 in a cohort study.9 The strates. Letrozole is approved in Canada for use in the treat- outcomes of pregnancies achieved through letrozole and ment of postmenopausal women with breast cancer.1 other ovarian stimulation regimens (CC) were compared Letrozole has no significant active metabolites. It is com- with a control group composed of women who had con- pletely absorbed after oral administration and has a mean ceived spontaneously. Pregnancies conceived after use of terminal half-life of approximately 45 hours (range letrozole were associated with rates of miscarriage and 30–60 hours). It is cleared from the systemic circulation ectopic pregnancy that were comparable to rates associated with all other pregnancies, including spontaneous concep- In the late 1990s, aromatase inhibitors, including letrozole, tions. Letrozole use was associated with a significantly began to be used to induce ovulation by being administered lower multiple gestation rate than use of CC.9 It seems in the early part of the menstrual cycle.2–4 Estrogen produc- unlikely that there would be significant exposure of the tion from all sources is blocked by inhibiting aromatization, embryo to letrozole, as the short half-life of letrozole and releasing the hypothalamic-pituitary axis from estrogenic the timing of administration in the early follicular phase negative feedback and resulting in increased gonadotropin should result in clearance of the drug before implantation secretion and ovarian follicular stimulation.5,6 In the ovary, aromatase inhibitors increase follicular sensitivity to FSH,as there is an accumulation of intraovarian androgens.7 Concern was raised at the 2005 Annual Meeting of theAmerican Society for Reproductive Medicine about the At the level of the endometrium, estrogen receptors may be safety of the fetus in mothers who used letrozole.10 One upregulated, resulting in rapid endometrial growth once hundred fifty babies from 130 pregnancies conceived afterthe use of letrozole were compared with 36 000 babies con- Key Words: Letrozole, clomiphene citrate, pregnancy,
ceived spontaneously and born to women at low risk in a community hospital. Although there was no difference inthe overall rate of congenital anomalies between the two l AUGUST JOGC AOÛT 2007
Fetal Safety of Letrozole and Clomiphene Citrate for Ovulation Induction Table 1. Comparing the effects on pregnancy outcomes of letrozole and clomiphene citrate with controls
*data from multiple births (i.e., twins, triplets) are excluded from each group (Motherisk controls, letrozole and CC) in this analysis.
HSingleton only dataIP < 0.05 compared with Motherisk controls groups, the authors reported that the incidence of cardiac Technology. The data recorded were maternal age at birth, and bone anomalies was higher in the letrozole group than gender of offspring, gestational age at birth, birth weight, in the control group. There were numerous concerns and congenital malformations. Each woman in the letrozole regarding the methodology of this study: the small size of group was matched by age with a control from the the letrozole group, the choice of a control group that Motherisk database. All Motherisk controls conceived would have a lower risk of pregnancy complications and spontaneously. In each group, data were analyzed with and congenital malformations than an infertile population, and without exclusion of multiples, and centiles for birthweight the under-representation of congenital anomalies in the adjusted for GA were calculated for all available data using control group (noting that any babies identified as abnormal on prenatal ultrasound would be delivered at a tertiary carehospital rather than a community hospital).
Statistical Analysis
The present study was designed to compare the risks to the Data were analyzed using GraphPad InStat 3.05 (GraphPad fetus of letrozole and CC and to compare them with the Software Inc, San Diego, CA). The Kolmogorov and risks to the fetus in the general obstetrical population. The Smirnov method was used to test the normality of data dis- primary objective was to compare the malformation rates in tribution, specifically for maternal age, gestational age, and the offspring of women who conceived using letrozole, for birth weight including and excluding multiple births. If women who conceived spontaneously (age-matched con- the data passed the normality test, then a one-way ANOVA trols), and women who conceived using CC (dis- was conducted; if not, then the Kruskal-Wallis one way ease-matched controls). The secondary objective was to ANOVA on Ranks test (a nonparametric ANOVA) was compare other pregnancy outcomes (birth weight and ges- used for analysis. For any statistically significant differences tational age at birth) among the three groups.
found in these data, Dunn’s Multiple Comparison test wasused. The chi-square test was used to compare time of delivery and incidence of malformations between groups.
Unpaired t tests were used to compare the mean centiles We reviewed the records of women who had delivered after using either letrozole or CC for ovulation induction duringtreatment at the McGill Reproductive Centre in Montreal or the Toronto Centre for Advanced Reproductive In this retrospective multicentre study, we analyzed datafrom 94 women who conceived using letrozole, 242 women ABBREVIATIONS
who conceived using CC, and 94 women who conceived spontaneously. In the letrozole group, 112 babies were born (including 14 sets of twins and 2 sets of triplets). In the CC group, 271 babies were born (including 27 sets of twins and 2 sets of triplets). There were no multiple births in the group who conceived spontaneously (Motherisk control group).
When the letrozole group was compared with the CC and AUGUST JOGC AOÛT 2007 l
DRUGS IN PREGNANCY
the Motherisk control group, there were no statistically sig- Table 2. comparing the birthweight adjusted for
nificant differences in the median maternal age at time of gestational age between letrozole, disease-matched
delivery (33 years), GA at birth (38.5, 39, and 38.4 weeks, (CC), and age-matched (Motherisk) controls*
respectively), or the rate of malformations (0%, 3.2%, and 2.6%, respectively) (Table 1). The median birth weight of babies in the letrozole group was not significantly differentfrom that of babies in the Motherisk control group, either when multiples were included in the letrozole group (3220 vs. 3320 g) or when multiples were excluded (3538 vs. 3391 g).
In contrast, the median birth weight of all babies born to women who conceived using CC was significantly reduced compared to the controls (3240 vs. 3320 g, P < 0.01).
Further analysis was performed to determine if there were *Data from multiple births (i.e., twins, triplets) are excluded from each group differences between groups in median birth weight adjusted (Motherisk controls, letrozole, and CC) in this analysis for GA and with multiples excluded. Mean centiles for HData for gestational age and/or birthweight were not available for all women babies in the letrozole group did not differ significantly I P < 0.05 compared with Motherisk controls from the controls (54.8 vs. 61.7, P > 0.05), and the meancentiles for babies in the CC group were lower (37.2,P < 0.0001) (Table 2).
control for possible effects of litter of origin on male sexualfunction.
DISCUSSION
Because CC has been used for ovulation induction for more The present matched control study provides evidence that than 40 years, there is significantly more published informa- letrozole is not a human teratogen. Instead, it suggests that tion on congenital anomaly rates in human and animal off- use of CC may result in small for gestational age infants. A spring following use of CC than use of letrozole. In human pregnancies conceived after use of CC for induction of ovu- addressed some of the methodological issues in the 2005 lation, the reported overall rates of major or minor malfor- American Society for Reproductive Medicine abstract.10 mations have not been significantly different from those This new study included a much larger number of women observed in spontaneously conceived pregnancies.14,15 An who had used letrozole to conceive. In addition, the popu- association between maternal use of CC and coarctation of lations being compared were both composed of patients the aorta (odds ratio 4.5; 99% CI 1.0–19.9) was observed in with infertility. The study did not find any increase in the a case–control study that included 126 children. This asso- rates of major malformations in babies conceived after ciation was not seen in a subsequent case–control study of letrozole treatment.12 The present study provides support 83 infants with conotruncal cardiac defects.16 A possible for these findings. We identified no major congenital mal- association between the use of CC for ovulation induction formations in the letrozole group, and the number of major and an increased incidence of neural tube defects in the off- malformations in the CC group was not statistically signifi- spring has been debated in the literature. Proponents of the cantly different from those in the Motherisk control group.
association do agree, however, that any increased risk asso- Aside from the retrospective Canadian study described ciated with CC is not large.17 A small number of reports above,12 there are no published reports of congenital anom- have described an association between use of CC or other alies in human offspring delivered after the use of letrozole.
agents that may alter hormone levels in early pregnancy and When letrozole was given to pregnant rats at 1% of the dose an increased incidence of neuroendocrine tumours, includ- used in humans for ovulation induction, fetal anomalies ing neuroblastoma.18–20 In other reports acardic twins have involving the kidney and ureter, and incomplete skeletal been reported twice in pregnancies when CC was used for ossification were seen. Treatment of pregnant rats with ovulation induction.12 The risk of craniosynostosis was letrozole 1 mg/kg/day on gestation days 21 and 22 resulted increased in a small population of infants whose mothers in altered sexual function in male offspring.13 Although had used CC for ovulation induction,21,22 although the these male offspring underwent puberty at the normal age authors of these reports indicated that the results were not and had normal body and testicular weights, there was a 24% decrease in rates of pregnancy when they were mated Our present comparison of the outcome of babies con- with normal females. Sexual activity was reduced and there ceived after the use of letrozole with matched controls con- was a mild decrease in testicular spermatid number. The ceived after use of CC and a control group of babies whose conclusions of this study are limited, however, by failure to mothers conceived spontaneously found that the birth l AUGUST JOGC AOÛT 2007
Fetal Safety of Letrozole and Clomiphene Citrate for Ovulation Induction weight of babies in the CC group was significantly lower 3. Mitwally MFM, Casper RF. Aromatase inhibitors of the treatment of infertility.
Expert Opin Investig Drugs 2003;12:353–71.
than the birth weight of babies in both the letrozole group 4. Mitwally MFM, Casper RF. Review: aromatase inhibitors for ovulation and the control group, even after controlling for maternal induction. J Clin Endocrinol Metab 2006;91:760–71.
age and GA at birth, and excluding multiple gestations.
5. Naftolin F, MacLusky NJ, Leranth CZ Sakamoto HS, Garcia-Segura LM. The Although this observation has not been previously reported cellular effects of estrogens on neuroendocrine tissues. J Steroid Biochem1988;30:195–207.
in human studies, there is support from a small amount of 6. Mason AJ, Berkemeier LM, Schmelzer CH, Schwall RH. Activin B: precursor animal data. In some reports, CC treatment during preg- sequences, genomic structure and in vitro activities. Mol Endocrinol nancy was associated with decreased fetal growth.
Decreased implantation rates and increased rates of fetal 7. Weil SJ, Vendola K, Zhou J Adesanya OO, Wang J, Okafor J. Androgen growth retardation and exencephaly were observed among receptor gene expression in the primate ovary: cellular localization, regulation,and functional correlations. J Clin Endocrinol Metab 1998;83:2479–85.
offspring of mice treated with CC just before ovulation in 8. Fatemi HM, Kolibianakis E, Tournaye H Camus M, Van Steirteghem AC, doses similar to those used in humans.23 Blastocyst transfer Devroey P. Clomiphene citrate versus Letrozole for ovarian stimulation: a pilot experiments in mice indicate that the preovulatory adminis- study. Reprod Biomed Online 2003;7:543–6.
tration of CC impairs uterine function, which subsequently 9. Mitwally MF, Biljan MM and Casper RF. Pregnancy outcome after the use of an aromatase inhibitor for ovarian stimulation. Am J Obstet Gynecol reduces embryonic growth and development.23 A possible mechanism for the observed intrauterine growth restriction 10. Biljan MM, Hemmings R, Brassard N. The outcome of 150 babies following the in animals and humans when CC is used for ovulation treatment with Letrozole or Letrozole and gonadotropins. Fertil Steril 2005;84 induction is the long half-life of the CC isomers (elimina- tion time 5–7 days) and the possible presence of anti- 11. Oken E, Kleinman K, Rich-Edwards J, Gillman M. A nearly continuous measure estrogenic effects in the endometrium and uterus causing of birth weight for gestational age using a United States national reference. BMCPediatrics 2003;3:1–10.
reduced blood flow or affecting embryogenesis.
12. Tulandi T, Martin J, Al-Fadhli R, Kabli N, Forman R, Hitkari J, et al. Congenital Despite the use of matched controls, it remains a possibility malformations among 911 newborns conceived after infertility treatment with that the women in the two treatment groups were given CC Letrozole or clomiphene citrate. Fertil Steril 2006;85:1761–5.
or letrozole based on different characteristics and therefore 13. Gerardin DC, Pereira OC. Reproductive changes in male rats treated perinatally with an aromatase inhibitor. Pharmacol Biochem Behav 2002;71:301–5.
were not directly comparable. After reviewing the prescrib-ing practices of the clinics involved in this study, we do not 14. Kurachi K, Aono T, Minagawa J, Miyake A. Congenital malformations of newborn infants after clomiphene-induced ovulation. Fertil Steril 1983;40:187–9.
believe this to be the case” Some clinics have a preference 15. Somani RA, Moretti M, Katz D, Chu A, Koren C, Ito S. Pregnancy outcome for one medication and treat all patients with this medica- following gestational exposure to clomiphene citrate: a prospective controlled tion first unless a drug allergy or adverse reaction has been reported. In some instances, patients are treated with CC 16. Loffredo CA, Ferencz C, Rubin JD, Magee CA. A comparative epidemiologic first and are treated with letrozole if CC is unsuccessful.
evaluation for hypoplastic left heart syndrome, aortic stenosis, and coarctation of Occasionally, a patient who conceived after use of letrozole or CC will wish to use this medication when trying to 17. Wu YE, Croan LA, Henning L, Najjar DV, Schembri M, Croughan MS. Potential association between infertility and spinal neural tube defects. Birth Defects Res become pregnant again. Letrozole and CC are comparable A Clin Mol Teratol 2006;76:718–22.
in price, so the groups would not likely differ on the basis of 18. Mandel M, Toren A, Rechavi G Dor J, Ben-Bassat I, Neumann Y. Hormonal treatment in pregnancy: a possible risk factor for neuroblastoma. Med PediatrOncol 1994;23:133–5.
CONCLUSION
19. White L, Giri N, Vowels MR, Lancaster PAL. Neuroectodermal tumours in children born after assisted conception. Lancet 1989;336:1577.
The use of letrozole for ovulation induction does notappear to increase the risk of congenital malformations and 20. Kobayashi N, Matsui I, Tanimura M, Nagahara N, Akatsuka J, Hirayama T.
Childhood neuroectodermal tumours and malignant lymphoma after maternal does not affect birth weight. The results of the present ovulation induction. Lancet 1991;338:955.
study suggest use of CC increases the incidence of small for 21. Haring DA, Cornel MC, Van Der Linden JC, van Vugt JM, Kwee ML. Acardius acephalus after induced abortion: A case report. Teratology 1993;47:257–62.
22. Reefhuis J, Honein MA, Shaw GM, Romitti PA. Fertility treatments and REFERENCES
Craniosynostosis: California, Georgia and Iowa, 1993–1997. Pediatrics2003;111:1163–6.
1. Cole PA, Robinson CH. Mechanism and inhibition of cytochrome P-450 aromatase. J Med Chem 1999;33:2933–44.
23. Dziadek M. Preovulatory administration of clomiphene citrate to mice causes 2. Mitwally MFM, Casper RF. Aromatase inhibition of ovarian stimulation: future fetal growth retardation and neural tube defects (exencephaly) by an indirect avenues for infertility management. Curr Opin Obstet Gynecol 2002;14:225–63.
maternal effect. Teratology 1993;47:263–73.
AUGUST JOGC AOÛT 2007 l

Source: http://www.galaxypharma.org/images/200708_DrugsinPregnancy_1.pdf