Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES)

+Model

ARTICLE IN PRESS

THERAP-12; No. of Pages 10 Therapie (2016) xxx, xxx—xxx

Available online at

ScienceDirect www.sciencedirect.com

PHARMACOEPIDEMIOLOGY

Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES) Effets indésirables chez les enfants de femmes exposées au diéthylstilbestrol (DES) in utero Michel Tournaire a,b,i,∗, Sylvie Epelboin c,d, Emmanuel Devouche b,e, Géraldine Viot b,f, Jérôme Le Bidois b,g, Anne Cabau a, Annabel Dunbavand h, Anne Levadou i a

Assistance publique—Hôpitaux de Paris, 75012 Paris, France Descartes University, 75006 Paris, France c Hôpital Bichat—Claude-Bernard, Assistance publique—Hôpitaux de Paris, 75018 Paris, France d Diderot University, 75013 Paris, France e Laboratoire psychopathologie et processus de santé, Research Unit 4057, 92100 Boulogne, France f Hôpital Cochin, Assistance publique—Hôpitaux de Paris, 75014 Paris, France g Pediatric Cardiology Unit, hôpital Necker, Assistance publique—Hôpitaux de Paris, 75015 Paris, France h Mutualité franc¸aise, 75015 Paris, France i Association Réseau DES France, 40000 Mont-de-Marsan, France b

Received 21 October 2015; accepted 18 January 2016

KEYWORDS Diethylstilbestrol; DES; Estrogens; Birth defects; Cerebral palsy;

∗

Summary Objective. — Exposure to diethylstilbestrol (DES) in utero is associated with adverse health effects, including genital anomalies in women and men, and cancers in women. Animal studies showed birth defects and tumors in the offspring of DES exposed mice, revealing transgenerational transmission of DES effects. In humans, birth defects, such as hypospadias were observed in children of prenatally exposed women. The aim of this research was to further assess the health effects in children of prenatally exposed women.

Corresponding author. Réseau DES France, 1052, rue de la Ferme-du-Carboué, 40000 Mont-de-Marsan, France. E-mail address: [email protected] (M. Tournaire).

http://dx.doi.org/10.1016/j.therap.2016.01.006 0040-5957/© 2016 Société franc ¸aise de pharmacologie et de thérapeutique. Published by Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

+Model THERAP-12; No. of Pages 10

2

ARTICLE IN PRESS M. Tournaire et al.

Cancer; Epigenetic alterations

Methods. — In a retrospective cohort study, the reports of women exposed to DES in utero on their 4409 children were compared with those of unexposed women on their 6203 children. Comparisons used odd ratios (OR) between children of exposed and unexposed women and standardized incidence rate (SIR) with the general population. These cohorts were recruited on a voluntary basis to answer questionnaires. Results. — There was a global increase of defects in children born to exposed women when compared with those born to unexposed (OR 2.29, 95% CI: 1.80—2.79, P < 0.001) and with the general population (SIR 2.39, 95% CI: 2.11—2.68). Increased defects were observed in male genital tract, esophagus, lip or palate, musculoskeletal and circulatory systems. For female genital tract anomalies, there was no significant increase. However, this cohort being relatively young, further follow-up is needed. An increase of cerebral palsy was revealed. The incidence of cancers was not increased, in particular for breast, uterus and ovary. Conclusion. — Our results confirmed a transgenerational transmission of defects in male genital tract. With caution due to possible bias associated with this method, our data suggest an increase of defects for esophagus, lip or palate, musculoskeletal and circulatory system in children of exposed women. © 2016 Société franc ¸aise de pharmacologie et de thérapeutique. Published by Elsevier Masson SAS. All rights reserved.

MOTS CLÉS Diéthylstilbestrol ; DES ; Estrogènes ; Malformations congénitales ; Infirmes moteurs cérébraux ; Cancer ; Modifications épigénétiques

Résumé Objectif. — L’exposition au diéthylstilbestrol (DES) in utero s’accompagne d’effets indésirables, en particulier d’anomalies de l’appareil génital chez la femme et chez l’homme, et de cancers chez la femme. Des études chez l’animal ont observé des malformations et des tumeurs chez les descendants de souris exposées au DES in utero, démontrant une transmission transgénérationnelle d’effets du DES. Dans l’espèce humaine, des malformations congénitales, telles que l’hypospadias ont été observées chez les enfants de femmes exposées in utero. L’objectif de cette recherche était de réévaluer les effets indésirables chez les enfants des femmes exposées au DES in utero. Méthodes. — Dans une étude de cohorte rétrospective, les informations recueillies auprès de femmes exposées au DES in utero sur leurs 4409 enfants ont été comparées avec celles de femmes non exposées sur leurs 6203 enfants. Les comparaisons ont utilisé les odds ratios (OR) entre les enfants des femmes exposées et non exposées, et les taux d’incidence normalisés (SIR) avec la population générale. Ces groupes ont été recrutés sur la base du volontariat pour répondre à des questionnaires. Résultats. — Il y avait une augmentation globale des malformations congénitales chez les enfants des femmes exposées comparés à ceux des femmes non exposées (OR 2,29 ; IC 95 % : 1,80—2,79 ; p < 0,001) et à la population générale (SIR 2,39 ; IC 95 % : 2,11—2,68). Une augmentation des malformations a été observée pour l’appareil génital masculin, l’œsophage, les fentes labiales ou palatines, l’appareil musculo-squelettique et le système circulatoire. Il n’y avait pas d’élévation significative d’anomalies de l’appareil génital féminin. Cependant, cette cohorte étant relativement jeune, la surveillance doit être poursuivie. Une augmentation des infirmités motrices cérébrales a été révélée. L’incidence des cancers n’était pas élevée, en particulier pour le sein, l’utérus et l’ovaire. Conclusion. — Nos résultats confirment la transmission transgénérationnelle de malformations de l’appareil génital masculin. Avec prudence, en raison de possibles biais liés à cette méthode, nos résultats suggèrent une augmentation des malformations de l’œsophage, des fentes labiales ou palatines, des anomalies de l’appareil musculo-squelettique et du système circulatoire chez les enfants de femmes exposées. © 2016 Société franc ¸aise de pharmacologie et de thérapeutique. Publié par Elsevier Masson SAS. Tous droits réservés.

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

+Model THERAP-12; No. of Pages 10

ARTICLE IN PRESS

Adverse health effects in children of women exposed in utero to DES

Abbreviations

Methods

ANSM

Participants

French National Agency for the Safety of Medicines and Health Products (Agence nationale de sécurité du médicament et des produits de santé) CCA clear-cell adenocarcinomas of vagina or cervix CI confidence interval ICD-10 international classification of disease No. 10 diethylstilbestrol DES Eurocat European Concerted Action on Congenital Anomalies and Twins Inserm French National Institute of Health and Medical Research (Institut national de la santé et de la recherche médicale) NCI National Cancer Institute OR odds ratio SIR standardised incidence rate

Introduction Diethylstilbestrol (DES) is a synthetic estrogen prescribed for pregnant women to prevent miscarriages and other complications of pregnancy. Adverse health effects have been described for women exposed to DES in utero: reproductive tract anomalies, infertility, pregnancy complications, such as miscarriage, ectopic pregnancy and premature birth, clear-cell adenocarcinomas of vagina or cervix (CCA) and increase of breast cancer [1—9]. Genital anomalies were observed in prenatally exposed men [10]. Other studies suggested an increased risk of infertility in men and testicular cancer [11,12]. Several million persons have been exposed in utero worldwide, including approximately 80,000 women and 80,000 men in France [8,13,14]. The children, females or males of prenatally exposed women compose the third generation. In animals, reproductive tract tissues of female mice exposed in utero showed persistent epigenetic alterations with changes in genes expression [15—17]. Some studies indicated a higher frequency of birth defects and tumors in female and male of third generation in mice [18—20]. These studies raise question of possible transgenerational transmission of epigenetic alterations in humans. An increase of overall birth defects was observed in women and men of the third generation [21]. A rise of hypospadias has also been described in this third generation [22—27]. A Dutch study revealed a significant increase in esophageal atresia and tracheoesophageal fistula [28]. However, the increase of this pathology described in an American report was interpreted as non-significant [21]. Moreover, a research raised questions about an increase of heart anomalies in girls of third generation [21]. Another type of health effect can be suspected in these third generation children, an increase of cerebral palsy associated with a higher incidence of premature births [29,30]. Finally, research on tumors revealed no overall increase of cancer in this generation. However, the incidence of ovarian cancers was higher than expected [31,32]. The objective of this report is to evaluate three adverse health effects in children of third generation, born to women exposed to DES in utero, compared with an unexposed cohort: birth defects, cerebral palsy and cancers.

3

This retrospective cohort study was conceived following the initiative of a patients association, Réseau DES France, funded by the French Medicines Agency, the ‘‘Agence nationale de la sécurité des médicaments et des produits de santé’’ (ANSM), and supported by a national health insurance, the ‘‘Mutualité franc¸aise’’. This investigation was retrospective, due to the absence of any sufficient prospective cohort of DES exposed subjects available in France. Questionnaires were designed to assess adverse effects of DES exposure in France for the three generations: first, women treated by DES during their pregnancies; second, women and men exposed in utero; third, daughters and sons of in utero exposed individuals. A questionnaire was also sent to a cohort of women unexposed in utero that constituted a comparison population with in utero exposed women. These unexposed women were defined by their date of birth, 1950 to 1977, corresponding to ages of DES exposure in France. The questionnaires were diffused nationwide on a voluntary basis through DES patients associations, health insurance, media (generalist and medical), family and acquaintances. In decreasing percentage, exposed women were recruited from media 44%, patients associations 21%, family 10%, miscellaneous 8%, and unknown 17%. For unexposed women, these percentages were acquaintances 43%, media 21%, family 13%, miscellaneous 12% and unknown 11% [9]. Answers to questionnaires were returned through Internet or postal mail from April 2013 until September 2013. A first study reported cancer risk in women exposed in utero [9]. For this study, we used two questionnaires, one from women exposed in utero and one from the comparison cohort of unexposed women, to assess information on their children [33]. After exclusion of incomplete answers to questionnaires or uncertain DES exposure, these cohorts were composed of 3436 exposed women having given birth to 4409 children (2228 girls and 2181 boys) and 3256 unexposed women with 6203 children (3054 girls and 3149 boys) (Fig. 1). The participants’ characteristics were detailed in a previous publication, using the same cohorts of women [9]. The comparison between exposed and unexposed cohorts showed only one important difference: an overrepresentation of children born from 1965 to 1972 for the exposed cohort, due to the peak of prescription of DES in France during this period. For this reason, ORs were adjusted on year of birth.

Recruitment bias We acknowledge that with a retrospective method based on questionnaires, recruitment bias may be present and difficult to quantify. However, two observations could be in agreement with limited bias in this study. First, the average number of questions answered by questionnaire was approximately fifty. If some items may incline subjects to complete a questionnaire, such a bias is unlikely to influence numerous other answers. Second, the increased risk of congenital defects, such as esophageal atresia, revealed in

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

+Model THERAP-12; No. of Pages 10

ARTICLE IN PRESS

4

M. Tournaire et al. • certain and supported by medical documents; • certain without document; • probable. To reduce the selection bias, women answering that their exposure was ‘‘probable’’ were excluded. The only eligible subjects considering exposure as certain consisted of two groups: women having documented exposure (25%) and women providing consistent information (75%), including clear recollection of their mother having taken DES and reproductive tract anomalies, such as typical cervical anomalies or T-shaped uterus known to be associated with DES exposure.

Unexposed cohort This cohort was recruited on a voluntary basis, from declaration of women on the absence of exposure.

Diagnosis of adverse health effects Birth defects Figure 1. Population according to reports of women prenatally exposed and unexposed to DES.

two medical publications, could not be identified by DES families in France as an established DES adverse effect [21,28].

Reporting bias Self-report is commonly considered as a weak point when obtaining information. In fact, the reliability of the data appeared to vary according to several factors, including category of health problem and quality of the answers depending on the formulation of questions. An example of the first point was the low report of uterine anomalies by women on their daughters. This can be interpreted as an under-report from the mother for genital anomalies that are not obvious in girls unlike cleft palate, club foot, or genital anomalies in boys, as observed by Titus-Ernstoff et al. [21]. The second point can be illustrated by heart defects. An increase of these anomalies having been suspected in a previous study, we designed detailed open-ended questions for diagnostic but also treatment and progress [21]. To obtain optimal diagnosis and classification, the answers were analyzed by epidemiologists specialized in birth defects and a pediatric cardiologist. To evaluate recruitment and reporting bias, the exposed and unexposed cohorts were compared with the general population in Europe [34,35].

DES exposure Exposed cohort The questionnaire section dedicated to DES exposure proposed three possible answers:

By definition, birth defects include anatomical and chromosomal abnormalities discovered prenatally or in newborn. We observed carefully those reported with discordant results in the literature, such as genital, esophageal and heart anomalies [21—28]. To improve information on congenital anomalies, details obtained on diagnoses, anatomical changes, treatments and progress allowed coding these anomalies according to ICD-10 [34] and Eurocat [35] classification. Incomplete questionnaires and minor anomalies according to Eurocat registers were excluded.

Cerebral palsy These defects are not discovered during the neonatal period like birth defects, but later on during development. They can be related to pre-term birth and they have rarely been analyzed for children of DES exposed women [29,30]. They were also detailed in open-ended questions.

Cancers Some were observed in animal studies and the results were discordant in humans [18—20,28,29]. Specific questions were asked to exposed or unexposed women on cancers in their children.

Statistical methods We compared information given by prenatally exposed and unexposed women in their children. Analysis was done with Stata for Windows (version 14, Stata Corp LP, USA). Logistic regression was used to estimate odds ratios (ORs) and Wald 95% CIs of the association of DES exposure with health effects. OR were adjusted on year of birth and computed on all children and separately for daughters and sons. Standardized incidence rate (SIR) was calculated to compare observed cases with expected incidences in the general population according to Eurocat registers.

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

+Model THERAP-12; No. of Pages 10

ARTICLE IN PRESS

Adverse health effects in children of women exposed in utero to DES

5

Results

Cancers

Population

Nine cancers reported by prenatally exposed women were compared with 16 reported by unexposed women in their respective children. There was no statistically significant difference. Exposed women questionnaires reported leukemia 3, brain 2, breast 1, eye 1, liver 1 and testicle 1. In the unexposed group, there were lymphoma 4, leukemia 3, breast 2, bones 1, brain 1, eye 1, kidney 1, melanoma 1, testicle 1 and uterus 1 (Table 4).

In the group of exposed women, the average age of the 2228 daughters of third generation was 15.2 years (range 0—43), and for the 2181 sons, it was 15.5 years (range 0—41). In the group of unexposed women, the 3054 daughters and 3149 sons were older than in the exposed group, 17.2 years (range 0—45) and 16.8 years (range 0—44), respectively.

Birth defects

Discussion

Prenatally exposed mothers reported 275/4409 children affected by birth defects (6.2%), with 101/2228 girls (4.5%) and 174/2181 boys (7.9%). In unexposed mothers, there were 163/6203 children affected (2.6%), of whom 68/3052 girls (2.2%) and 95/3149 boys (3.0%) (Fig. 1 and Table 1). Exposed and unexposed mothers reported respectively 31 and 5 children affected by more than one birth defect. The OR for exposed compared to unexposed children for any birth defect was 2.29 (95% CI 1.80—2.79) (Table 1). The six statistically significant elevated birth defects were, in decreasing order of OR: • esophagus anomalies; • other digestive defects; • all male genital defects and, in detail, hypospadias and undescended testicle; • cleft lip and palate; • musculoskeletal anomalies; • congenital heart defects.

Recruitment and reporting bias

No other significant association was observed, in particular for female organs. When compared with the general population of Eurocat registers, the SIR for all defects in exposed and unexposed children was 2.39 (95% CI 2.11—2.68) and 1.01 (95% CI 0.86—1.17), respectively. The question concerning an increased risk of congenital heart defects having been raised, we analyzed in detail these anomalies [21]. Globally, affected children were 33, 17 daughters and 16 sons in the exposed group. They were 21 children with 10 daughters and 11 sons in the unexposed group (Table 1). In Table 2, the most frequent anomaly in the exposed group, in comparison with the unexposed or general population, was tetralogy of Fallot with 6 children affected in the exposed group (3 girls and 3 boys), vs. 1 girl in the unexposed group. The difference was not significant but, compared with the general population, the SIR showed a significant increase.

Cerebral palsy Cerebral palsy was reported for 26 children in the exposed group, including 20 pre-term births, and 4 in the unexposed, including 2 pre-term births. This increase was significant (Table 3). The increase incidence of pre-term births of 24.22% in the exposed group compared with 3.35% in the unexposed group was significant.

We observed a SIR for all birth defects in the unexposed group of 1.01 (95% CI 0.86—1.17), suggestive of a good concordance between this cohort and the general population in Europe (Table 1).

Birth defects All anomalies The SIR for all birth defects in the exposed group showed a significant increase (Table 1).

Esophageal atresia and tracheoesophageal fistula A Dutch study found, for the first time, a significant association between DES mother exposure and esophageal atresia/fistula [28]. In a NCI follow-up study, this pathology was observed in 3 cases in exposed cohort and none in unexposed, but the difference was not statistically significant (P = 0.20) [21]. We found 16 cases in exposed cohort, and 1 in unexposed (Table 1). Comparison with the general population revealed a significant increase. These results, suggesting an increase of this pathology in children of DES exposed women, are in favor of a transgenerational effect of DES other than in genital tract.

Male genital anomalies Comparisons between offspring of exposed and unexposed women showed a significant increase for any male genital defect. An increase was significant for two defects: hypospadias and undescended testicle (Table 1). Compared with the general population, the SIR for hypospadias was increased significantly. Among the studies considering third generation male anomalies, the majority revealed a significant increase for hypospadias with large differences in OR, from 4.9 to 21.3 [22—27]. Our OR of 4.58 was close to those of two studies, 5.0 and 4.9 [24,25]. Increase of male genital tract defects, especially hypospadias, supports the hypothesis of an epigenetic alteration transmitted to the third generation in men similar to the transmission for males and females in animals [15—17,36].

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

Incidence of birth defects in the offspring of women prenatally exposed and unexposed to DES. Exposed n

Adjusted OR

n

1/10,000

OR

623.72 453.32 797.80

163 68 95

262.78 222.66 301.68

31.75 35.91 27.51

15 9 6

27.22 22.44 32.10

EUROCAT

SIR

P level

1/10,000

95% CI

2.29 1.90 2.61

1.8—2.79 1.3—2.6 2—3.37

< .001 < .001 < .001

261.45

259.91—263.00

24.18 29.47 19.05

1.13 1.05 1.28

0.55—2.36 0.4—2.75 0.41—3.98

.727 .906 .668

25.36

10 4 6

16.12 13.10 19.05

1.56 1.51 1.64

0.67—3.64 0.4—5.65 0.54—4.93

.296 .537 .375

13.61 8.98 18.34

5 2 3

8.06 6.55 9.53

1.88 1.22 2.62

0.56—6.3 0.17—8.71 0.56—12.2

74.85 76.30 73.36

21 10 11

33.85 32.74 34.93

1.99 2.10 1.88

11.34 17.95 4.59

2 2 0

3.22 6.55 0.00

27.22 22.44 32.10

4 1 3

72.58 53.86 91.70 36.29 26.93 45.85 8.98 334.71 100.87 183.40

Exposed

95%CI

Unexposed

95% CI

2.39

(2.11—2.68)

1.01

(0.86—1.17)

24.88—25.84

1.25

(0.68—2.1)

0.95

(0.53—1.57)

4.16

3.97—4.36

6.54

(3.38—11.43)

3.88

(1.86—7.13)

.306 .840 .220

2.04

1.91—2.18

6.67

(2.44—14.52)

3.95

(1.27—9.22)

1.14—3.44 0.96—4.62 0.87—4.06

.014 .062 .108

23.31

22.85—23.77

3.21

(2.21—4.51)

1.45

(0.9—2.22)

3.19 2.55 1.00

0.61—16.5 0.46—14 Na

.166 .281 Na

4.13

3.94—4.33

2.75

(0.88—6.41)

0.78

(0.09—2.82)

6.45 3.27 9.53

4.12 7.08 3.20

1.32—12.8 0.81—61.8 0.82—12.4

.015 .076 .093

8.55

8.28—8.84

3.18

(1.64—5.56)

0.75

(0.2—1.93)

5 2 3 1 1 0 1 24 8 12

8.06 6.55 9.53 1.61 3.27 0.00 3.27 76.21 25.40 38.11

8.90 8.11 9.48 22.20 8.87 1.00 4.47 4.31 4.07 4.58

3.45—22.9 1.8—36.5 2.8—32.1 2.93—68.7 1.04—75.3 Na 0.45—44.5 2.7—6.87 1.79—9.23 2.39—8.78

< .001 .006 < .001 .003 .045 Na .202 < .001 .001 < .001

18.65

18.24—19.07

14.23

(8.13—23.11)

0.63

(0.01—3.52)

2.55

2.40—2.71

1.95

(1.11—3.16)

0.35

(0.09—0.89)

Not in EUROCAT Not in EUROCAT Not in EUROCAT 18.23

17.83—18.64

10.06

(7.19—13.7)

2.09

(1.08—3.65)

90.72 71.81 110.04

35 17 18

56.42 55.66 57.16

1.58 1.33 1.83

0.99—2.5 0.66—2.66 0.99—3.39

.050 .420 .053

34.81

34.25—35.38

2.61

(1.86—3.55)

1.62

(1.13—2.25)

115.67 89.77 142.14

21 8 13

33.85 26.20 41.28

3.18 3.24 3.18

1.91—5.31 1.42—7.41 1.66—6.11

< .001 .005 < .001

Not in EUROCAT

20.41 22.44 18.34

18 11 7

29.02 36.02 22.23

0.61 0.54 0.71

0.27—1.36 0.18—1.57 0.2—2.44

.231 .262 .590

39.35

38.75—39.95

0.52

(0.24—0.98)

0.74

(0.44—1.17)

ICD-10: international classification of disease No. 10; DES: diethylstilbestrol; EUROCAT: European Concerted Action on Congenital Anomalies and Twins; NA: not applicable; OR: odds ratio; SIR: standardised incidence rates. Analyses ran on 4409 exposed (including 2228 girls and 2181 boys) and 6203 unexposed (including 3054 girls and 3149 boys). OR ajdusted on year of birth. a Including all children presenting at least one anomaly.

ARTICLE IN PRESS

95% CI

M. Tournaire et al.

All anomaliesa All children 275 101 Daughters 174 Sons Nervous system 14 All children 8 Daughters 6 Sons Eye 12 All children 5 Daughters 7 Sons Ear, face and neck 6 All children 2 Daughters 4 Sons Congenital heart defects 33 All children 17 Daughters 16 Sons Respiratory system 5 All children 4 Daughters 1 Sons Cleft lip and palate 12 All children 5 Daughters 7 Sons Digestive system 32 All children 12 Daughters 20 Sons 16 All children 6 Daughters 10 Sons 2 Genital female 73 Genital male 22 40 Urinary system 40 All children 16 Daughters 24 Sons Musculo skeletal system 51 All children 20 Daughters 31 Sons Chromosomal abnormalities 9 All children 5 Daughters 4 Sons

Unexposed 1/10,000

+Model

ICD-10 label

THERAP-12; No. of Pages 10

6

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

Table 1

Unexposed

Adjusted OR

n

1/10,000

n

1/10,000

OR

95% CI

1 4 2 6 0 0 1 1 8 3 2

2.27 9.07 4.54 13.61 0.00 0.00 2.27 2.27 18.14 6.80 4.54

0 7 3 1 1 1 0 0 5 0 4

0.00 11.28 4.84 1.61 1.61 1.61 0.00 0.00 8.06 0.00 6.45

1.00 0.93 0.93 7.47 1.00 1.00 1.00 1.00 1.93 1.00 0.65

NA 0.29—2.94 0.15—5.69 0.89—62.1 NA NA NA NA 0.63—5.91 NA 0.11—3.57

EUROCAT

SIR

P level

1/10,000 95% CI

Case

95% CI

Control 95% CI

NA .903 .944 .063 NA NA NA NA .250 NA .621

0.77 33.69 21.04 3.57 4.08 0.45 1.43 2.74

0.69—0.86 33.12—34.27 20.59—21.50 3.38—3.76 3.89—4.29 0.38—0.52 1.31—1.55 2.58—2.90

2.95 0.27 0.22 3.81 NA NA 1.59 0.83

0.04—16.39 0.07—0.69 0.02—0.78 1.39—8.3 NA NA 0.02—8.82 0.01—4.61

NA 0.33 0.23 0.45 0.40 3.58 NA NA

NA 0.13—0.69 0.05—0.67 0.01—2.51 0.01—2.2 0.05—19.93 NA NA

3.75 3.78

3.56—3.95 3.59—3.98

NA 1.71

NA 0.46—4.37

1.81 0.36—5.3 1.20 0.13—4.33

ICD-10: international classification of disease No. 10; DES: diethylstilbestrol; EUROCAT: European Concerted Action on Congenital Anomalies and Twins; NA: not applicable; OR: odds ratio; SIR: standardised incidence rates. Analyses ran on 4409 exposed (including 2228 girls and 2181 boys) and 6203 unexposed (including 3054 girls and 3149 boys). OR adjusted on year of birth.

ARTICLE IN PRESS

Common arterial troncus Ventricular septal defect Atrial septal defect Tetralogy of fallot Pulmonary valve stenosis Ebstein anomaly Aortic valve atresia/stenosis Hypoplastic left heart Unspecified Persistent ductus arteriosus Coarctation of aorta

Exposed

+Model

ICD-10 Label

THERAP-12; No. of Pages 10

Incidence of severe heart defects in the offspring of women prenatally exposed and unexposed to DES.

Adverse health effects in children of women exposed in utero to DES

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

Table 2

7

+Model

ARTICLE IN PRESS

THERAP-12; No. of Pages 10

8

M. Tournaire et al. Table 3 to DES.

Incidence of pre-term births and cerebral palsy in the offspring of women prenatally exposed and unexposed

ICD-10 Label

Exposed

Unexposed

Adjusted OR

n

1/100

n

1/100

OR

95% CI

P level

Pre-term births Children Daughters Sons

1068 515 553

24.22 23.11 25.36

208 97 111

3.35 3.18 3.52

8.71 8.73 8.71

7.4—10.1 6.9—10.9 7—10.7

< .001 < .001 < .001

Cerebral palsy

n

1/10,000

n

1/10,000

26 13 13

58.97 58.35 59.61

4 2 2

6.45 6.55 6.35

10.50 10.20 10.80

3.6—30.6 2.26—46.2 2.41—48.8

< .001 .003 .002

20 9 11

187.27 174.76 198.92

2 1 1

96.15 103.09 90.09

2.17 1.97 2.43

0.50—9.46 0.249—15.7 0.301—19.7

< .001 .519 .403

6 4 2

17.96 23.35 12.29

2 1 1

3.34 3.38 3.29

6.01 7.61 5.31

1.1—30.5 0.732—79.1 0.545—51.7

.031 .089 .150

All births Children Daughters Sons Pre-term births Children Daughters Sons In-term births Children Daughters Sons

ICD-10: international classification of disease No. 10; DES: diethylstilbestrol; OR: odds ratio. Analyses ran on 4409 exposed (including 2228 girls and 2181 boys) and 6203 unexposed (including 3054 girls and 3149 boys). OR adjusted on year of birth.

Table 4

Incidence of cancers in the offspring of women prenatally exposed and unexposed to DES.

ICD-10 label

Cancers Children Daughters Sons

Exposed

Unexposed

Adjusted OR

n

1/10,000

n

1/10,000

OR

95% CI

P level

9 5 4

20.41 22.44 18.34

16 7 9

25.79 22.92 28.58

0.96 1.46 0.68

0.41—2.2 0.44—4.85 0.2—2.24

.918 .534 .527

ICD-10: international classification of disease No. 10; DES: diethylstilbestrol; OR: odds ratio. Analyses ran on 4409 exposed (including 2228 girls and 2181 boys) and 6203 unexposed (including 3054 girls and 3149 boys). OR adjusted on year of birth.

Circulatory system An increase of cardiovascular defects was first reported by Titus-Ernstoff et al. in 2010 [21], with an OR of 1.84 (95% CI 0.91—3.72). However, comments of investigators concluded that this result might be due to an artefact of under-reporting of these conditions by the unexposed mothers. In all, we found 33 defects in 4483 subjects in the exposed group and 21 in 6329 unexposed ones. OR and SIR were significantly increased (Table 2). Of the 11 types of defects analyzed, the main increase concerned the tetralogy of Fallot. These results suggest an increase of cardiovascular defects that requires careful interpretation. A selection bias due to our retrospective method and voluntarily based recruitment cannot be excluded.

Other increased defects As shown Table 1, a global rise of incidence was observed for three defects:

• there was a significant global increase of digestive system defects other than esophageal anomalies in the exposed group, including 6 pyloric stenosis, 1 duodenal stenosis and 3 imperforate anus, compared with 2 pyloric stenosis in the unexposed group, but none of these increases reached a significant level; • the significant increase of cleft lip or palate could represent another example of extra-genital transgenerational transmission; • for all musculoskeletal anomalies, there was a significant increase. However, for each anomaly,such as clubfoot or omphalocele, there was no significant rise.

Female genital tract anomalies Attention is focused on these defects for two reasons. First, DES families and their physicians express a comprehensive concern for girls who may replicate genital anomalies of their mothers exposed in utero. Second, studies in exposed

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

+Model THERAP-12; No. of Pages 10

ARTICLE IN PRESS

Adverse health effects in children of women exposed in utero to DES mice found that alterations in expression of genes by prenatal DES exposure may be transmitted to the next generation in particular for genitalia [15—17]. Two studies evaluated female genital tract in women of third generation. Kaufman et al. [37] found no anomalies in 28 third generation girls. In Titus-Ernstoff et al. [21], none of the mothers reported in their 3808 daughters any reproductive tract anomalies. In our study, prenatally exposed women reported two genital defects in their 2268 daughters, a Rokitansky syndrome and a bicornuate uterus. These defects, with a well-known origin, are different from anomalies, such as T-shaped uterus observed after DES exposure in utero. This study provides a new reassuring message for the daughters of women exposed in utero. However, this evaluation requires to be continued. In 2013, date of the questionnaires, in this third generation of girls aged 43 years or less, half of them were less than 18 years old.

Cerebral palsy It is established that cerebral palsy is related to pre-term birth [29,30]. Our prematurity incidence of 24% in the DES exposed group (Table 3) was comparable with 26% in an American cohort study [7]. However, comparing exposed and unexposed group, our OR of 8.71 (95% CI 7.4—10.1) was higher than the OR of 4.68 (95% CI 3.74—5.86) reported in the American study. This difference may be explained by an over-report of pre-term births in our exposed cohort or, more likely, by an under-report in the unexposed one. Incidences of cerebral palsy in our exposed group were 187/10,000 after pre-term birth and 18/10,000 in the interm group (Table 3). They can be compared with the figures reported in a review study where incidence of cerebral palsy varied according to the gestational age at birth. It decreased, for 10,000 pre-term births, from 822 before 28 weeks, 431 between 28 and 31 weeks, to 67 between 32 and 36 weeks. It was 13 for 10,000 in-term births [30]. In brief, our study suggests the occurrence of a significant increase of cerebral palsy associated with a higher incidence of preterm births in the exposed cohort, in agreement with the literature [29,30].

Cancers The attention was attracted to genital cancers by experiments in mice revealing an elevated incidence of cancers in female, (uterus and ovary) and male reproductive tract in the offspring of animals prenatally exposed to DES [18—20]. The NCI follow-up study found no global increase of cancer, but the incidence of ovarian cancers, 3 cases, was higher than expected and another study had reported a case of ovarian cancer [31,32]. Our comparison between exposed and unexposed revealed no difference for all cancers and, in particular, no breast, ovarian or uterus cancer in the exposed group (Table 4). However, the conclusions of this study are limited by the young age of the population, aged 43 years or less, whereas the incidence of most cancers increases with age.

9

Conclusion Our results confirmed a transgenerational transmission to the third generation of birth defects in male genital tract. With caution attached to possible bias associated with this retrospective method, our data suggested an increase for esophageal anomalies, cleft lip or palate, musculoskeletal and heart defects. An encouraging point is the absence of significant increase, for these third generation women, of defect or cancer in genital tract, but the young age of our cohort does not allow to definitely conclude. A significant increase of cerebral palsy was observed, associated with a higher incidence of premature births.

Ethical standards This study was submitted to the Ethics Review Committee ‘‘Comité de protection des personnes Île-de-France III’’ which testified that ‘‘the study appears to be in accordance with the scientific principles generally accepted, and to the ethical standards of research; the study was performed in the respect of the French law and regulation’’.

Funding The study was funded by ANSM (Agence nationale de sécurité du médicament et des produits de santé) and managed by Réseau DES France, a non-profit patients association.

Acknowledgements The authors would like to acknowledge Nathalie Lafaye and Liliane Lamolère (Réseau DES France), Marylène Jandon (Mutualité franc¸aise) and Arnaud Lacaze-Masmontel (Infotronique, Paris) for the study management; Gérard Breart (Inserm unit, Hôpital Cochin, Paris), Pamela Solere (Réseau DES France) and Edwin L. Cooper (University of California, Los Angeles) for reviewing the manuscript; Nathalie Lelong (INSERM unit 1153) for coding the data; Robert Hoover and Rebecca Troisi (National Cancer Institute) for their advice and encouragement; Agence nationale de sécurité du médicament et des produits de santé for funding; Mutualité franc¸aise for supporting the study.

Disclosure of interest The authors declare that they have no competing interest.

References [1] Kaufman RH, Adam E. Genital tract anomalies associated with in utero exposure to diethylstilbestrol. Isr J Med Sci 1978;14:353—62. [2] Barnes AB, Colyon T, Gundersen J, Noller KL, Tilley BC, Strama T, et al. Fertility and outcome of pregnancy in women exposed in utero to diethylstilbestrol. N Engl J Med 1980;302:609—13.

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006

+Model THERAP-12; No. of Pages 10

ARTICLE IN PRESS

10 [3] Melnick S, Cole P, Anderson D, Herbst A. Rates and risks of diethylstilbestrol-related clear-cell adenocarcinoma of the vagina and cervix. An update. N Engl J Med 1987;316:514—6. [4] Palmer JR, Wise LA, Hatch EE, Troisi R, Titus-Ernstoff L, Strohsnitter W, et al. Prenatal diethylstilbestrol exposure and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 2006;15:1509—14. [5] Troisi R, Hatch EE, Titus-Ernstoff L, Hyer M, Palmer JR, Robboy SJ, et al. Cancer risk in women prenatally exposed to diethylstilbestrol. Int J Cancer 2007;121:356—60. [6] Verloop J, van Jeeuven FE, Helmerhorst TJM, van Boven HH, Rookus MA. Cancer risk in DES daughters. Cancer Causes Control 2010;21:999—1007. [7] Hoover RN, Hyer M, Pfeiffer RM, Adam E, Bond B, Cheville AL, et al. Adverse health outcomes in women exposed in utero to diethylstilbestrol. N Engl J Med 2011;365:1304—14. [8] Tournaire M, Epelboin S, Devouche E. Diethylstilbestrol story. Therapie 2014;69:101—14. [9] Tournaire M, Devouche E, Espié M, Asselain B, Levadou A, Cabau A, et al. Cancer risk in women exposed to diethylstilbestrol in utero. Therapie 2015;70:433—41. [10] Palmer JR, Herbst AL, Noller KL, Boggs DA, Troisi R, TitusErnstoff L, et al. Urogenital abnormalities in men exposed to diethylstilbestrol in utero. Environ Health 2009;8:37. [11] Perez KM, Titus-Ernstoff L, Hatch EE, Troisi R, WactawskiWende J, Palmer JR, et al. Reproductive outcomes in men with prenatal exposure to diethylstilbestrol. Fertil Steril 2005;84:1649—56. [12] Strohsnitter WC, Noller KL, Hoover RN, Robboy SJ, Palmer JR, Titus-Ernstoff L, et al. Cancer risk in men exposed in utero to diethylstilbestrol. J Natl Cancer Inst 2001;93:545—51. [13] Spira A, Goujard J, Henrion R, Lemerle J, Robel P, Tchobroutsky C. L’administration de diéthylstilbestrol (DES) pendant la grossesse, un problème de santé publique. Rev Epidem Sante Publ 1983;31:249—72. [14] Epelboin S, Bulwa S. Pharmacovigilance et reproduction : le mauvais exemple du distilbène. Contracept Fertil Sex 1993;21:658—72. [15] Nelson KG, Sakai Y, Eitzman B, Steed T, McLachlan J. Exposure to diethylstilbestrol during a critical developmental period of the mouse reproductive tract leads to persistent induction of two estrogen-regulated genes. Cell Growth Differ 1994;5:595—606. [16] McLachlan JA, Burow M, Chiang TC, Li SF. Gene imprinting in developmental toxicology: a possible interface between physiology and pathology. Toxicol Lett 2001;120:161—4. [17] Mahawong P, Sinclair A, Li Y, Schlomer B, Rodriguez Jr E, Ferretti MM, et al. Prenatal diethylstilbestrol induces malformation of the external genitalia of male and female mice and persistent second-generation developmental abnormalities of the external genitalia in two mouse strains. Differentiation 2014;88:51—69. [18] Turusov VS, Trukhanova LS, Parfenov YuD, Tomatis L. Occurrence of tumours in the descendants of CBA male mice prenatally treated with diethylstilbestrol. Int J Cancer 1992;50:131—5. [19] Newbold RR, Hanson RB, Jefferson WN, Bullock BC, Haseman J, McLachlan JA, et al. Increased tumors but uncompromised fertility in the female descendants of mice exposed developmentally to diethylstilbestrol. Carcinogenesis 1998;19:1655—63. [20] Newbold RR, Hanson RB, Jefferson WN, Bullock BC, Haseman J, McLachlan JA. Proliferative lesions and reproductive tract tumors in male descendants of mice exposed developmentally to diethylstilbestrol. Carcinogenesis 2000;21:1355—63.

M. Tournaire et al. [21] Titus-Ernstoff L, Troisi R, Hatch EE, Palmer JR, Hyer M, Kaufman R, et al. Birth defects in the sons and daughters of women who were exposed in utero to diethylstilbestrol (DES). Int J Androl 2010;33:377—84. [22] Klip H, Verloop J, van Gool JD, Koster ME, Burger CW, van Leeuwen FE, et al. Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet 2002;359:1102—7. [23] Palmer JR, Wise LA, Robboy SJ, Titus-Ernstoff L, Noller KL, Herbst AL, et al. Hypospadias in sons of women exposed to diethylstilbestrol in utero. Epidemiology 2005;16:583—6. [24] Pons JC, Papiernik E, Billon A, Hessabi M, Duyme M. Hypospadias in sons of women exposed to diethylstilbestrol in utero. Prenat Diagn 2005;25:418—9. [25] Brouwers MM, Feitz WF, Roelofs LA, Kiemeney LA, de Gier RP, Roeleveld N. Hypospadias: a transgenerational effect of diethylstilbestrol? Hum Reprod 2006;21:666—9. [26] Epelboin S, Giorgis Allemand L, Tournaire M, et al. Influence de l’exposition intra-utérine au DES (Distilbène® ) sur la survenue d’hypospadias et autres anomalies congénitales dans la descendance. Agence nationale de la sécurité des médicaments et des produits de santé (ANSM); 2011. [27] Kalfa N, Paris F, Soyer-Gobillard MO, Daures JP, Sultan C. Prevalence of hypospadias in grandsons of women exposed to diethylstilbestrol during pregnancy: a multigenerational national cohort study. Fertil Steril 2011;95(8):2574—7. [28] Felix JF, Steegers-Theunissen RP, de Walle HE, de Klein A, Torfs CP, Tibboel D. Esophageal atresia and tracheoesophageal fistula in children of women exposed to diethylstilbestrol in utero. Am J Obstet Gynecol 2007;197:38e1—5. [29] Himpens E, Van den Broeck C, Oostra A, Calders P, Vanhaesebrouck P. Prevalence, type, distribution, and severity of cerebral palsy in relation to gestational age: a meta-analytic review. Dev Med Child Neurol 2008;50:334—40. [30] Oskoui M, Coutinho F, Dykeman J, Jetté N, Pringsheim T. An update on the prevalence of cerebral palsy: a systematic review and meta-analysis. Dev Med Child Neurol 2013;55:509—19. [31] Titus-Ernstoff L, Troisi R, Hatch EE, Hyer M, Wise LA, Palmer JR, et al. Offspring of women exposed in utero to diethylstilbestrol (DES): a preliminary report of benign and malignant pathology in the third generation. Epidemiology 2008;19:251—7. [32] Blatt J, Van Le L, Weiner T, Sailer S. Ovarian carcinoma in an adolescent with transgenerational exposure to diethylstilbestrol. J Pediatr Hematol Oncol 2003;25:635—6. [33] Association Réseau DES France. Distilbène étude 3 générations: questionnaires filles DES et groupe témoin; 2013 http://www.des-france.org/documents/questionnaires-2013fillesdes-et-femmestemoins-etude-des-3-generations.pdf [Accessed January 18th, 2016 (8 pp.)]. [34] European Surveillance of Congenital Anomalies Twins (Eurocat). Eurocat Register; 2016 http://www.eurocat-network.eu/ [Accessed January 18th, 2016]. [35] CIM 10. Classification Internationale des Maladies. Chapitre XVII Malformations congénitales et anomalies chromosomiques; 2016 http://taurus.unine.ch/icd10?term=&select=7114#select [Accessed January 18th, 2016]. [36] Newbold RR, Padilla-Banks E, Jefferson WN. Adverse effects of the model environmental estrogen diethylstilbestrol are transmitted to subsequent generations. Endocrinology 2006;147(6 Suppl.):S11—7. [37] Kaufman RH, Adam E. Findings in female offspring of women exposed in utero to diethylstilbestrol. Obstet Gynecol 2002;99(2):197—200.

Please cite this article in press as: Tournaire M, et al. Adverse health effects in children of women exposed in utero to diethylstilbestrol (DES). Therapie (2016), http://dx.doi.org/10.1016/j.therap.2016.01.006