In-vitro fertilisation and retinoblastoma

The Lancet v.361, n.9354 25jan03

Commentary

In today's Lancet, Annette Moll and colleagues report retinoblastoma in five children born after in-vitro fertilisation (IVF) in the Netherlands. These investigators say that 1·0-1·5% of children in the Netherlands are conceived by IVF. With the assumption that all new retinoblastoma cases between November, 2000, and February, 2002, had been diagnosed, Moll's group calculate relative risks of 4·9-7·2 for developing this tumour in children born after IVF in the Netherlands. This is an unprecedented and alarming increased risk.

A high frequency of cytogenetic abnormalities and errors in cell-cycle regulation are detected in oocytes generated from IVF or intracytoplasmic sperm injection.1,2 Some of these aberrations, especially in regulatory genes, may lead to the development of cancer in some children born after assisted reproductive technologies. An increased risk of developing neuroblastoma3,4 and leukaemia5,6 among offspring of women treated with infertility drugs has been reported. However, the few large and comparative cohort studies in various countries found either a slight7 or no increase8 in the incidence of cancer among children born after IVF. Looking specifically at the occurrence of retinoblastoma among children born after assisted reproductive technologies, the latest findings of Moll and colleagues appear less amenable to reconciliation with the relevant and available literature.

From the many published papers about the incidence of malignancies in large cohorts of children born after assisted reproductive technologies, none has reported the incidence or prevalence of retinoblastoma. In fact, ocular manifestations among such children were largely overlooked. In 1994, after what appeared to be an unusual increased incidence of ocular malformations in children born after IVF, our group9 started routine questioning of the parents, specifically about whether the child was conceived normally or not. We soon realised that most mothers did not volunteer information about undergoing IVF. In some instances (including the one case diagnosed with retinoblastoma), this information was obtained only on subsequent follow-up visits. Our observations and the finding of one child with retinoblastoma were published in 2001. A year later another case of retinoblastoma was reported by Moll's group.10

What could explain these surprising observations? One possibility, although unlikely, is recent increased use of a new (or a combination of multiple) ovulation-inducing drug(s) having a specific influence on the retinoblastoma gene during IVF. The nationwide cohort study in the same country8 in women undergoing IVF during 1980-95 did not find any case of retinoblastoma. It is not clear from the present study by Moll and colleagues whether the IVF technique in the Netherlands has changed substantially during 1996-2000, the period covered by their study.

Could the association of ocular manifestations, especially retinoblastoma, with IVF have been missed during earlier years? After the successful delivery of a long-awaited child and the concern of parents for ocular problems in their infants, the fact that the child was conceived after IVF may not be volunteered, especially if the treating ophthalmologist has no reason to inquire. Possibly, our paper9 enhanced awareness for this potential association, which could explain the "sudden" emergence of many retinoblastoma cases in children born after IVF. However, this explanation cannot account for five new cases lacking a family history of retinoblastoma. A thorough survey of the retinas in these family members for possible retinomas is necessary. Two of the five affected children were one of a twin pair; it remains to be seen whether any of the two healthy twins will develop a retinoma or retinoblastoma.

Might the five new cases be an example of clustering? This type of clustering, detected by an "interested" observer, has been reported in retinoblastoma11 and may also occur with other "hot" diagnoses. Indeed Klip et al8 said that: "In 4 years [prior to the published study], one of the authors observed at least eight malignancies in children born after hormone stimulation for IVF and 11 malignancies in children born after insemination techniques and/or other use of fertility drugs". During the nationwide study by Klip et al, however, and a follow-up for an average 6 years, cancers were detected in "only" seven children born after IVF.

Moll and colleagues assume that 1·0-1·5 % of children are conceived after IVF in the Netherlands. Is this correct? Other researchers from the Netherlands estimated that in 1992: "almost 2·5% of all live births result from assisted reproductive technologies".8,12 Is it possible that during the period studied by Moll 3·0 or 3·5% of live births in the Netherlands were conceived after assisted reproductive technologies (ie, not 1·0-1·5%)? If so, the relative risks for retinoblastoma would be much lower than the 4·9-7·2 reported by Moll and colleagues.

Whatever the "true" incidence of retinoblastoma is after IVF, there is little doubt that a heightened awareness and a multidisciplinary approach with a closer follow-up of children conceived with assisted reproductive technologies are needed. The question recently voiced by Winston and Hardy--"Are we ignoring potential dangers of in vitro fertilization and related treatments?"--is pertinent and timely.13 An open debate on this issue is necessary to frame it in its proper context and to minimise potential harmful effects of unfounded and potentially misleading information. David BenEzra

Paediatric Ophthalmology Unit, Hadassah Hebrew University Hospital, 91120 Jerusalem, Israel benezra@md2.huji.ac.il

1 Edirisinghe W, Murch A, Junk S, Yovich J. Cytogenetic abnormalities of unfertilized oocytes generated from in-vitro fertilization and intracytoplasmic sperm injection: a double-blind study. Hum Reprod 1997; 12: 2784-91. [PubMed]

2 Voullaire L, Wilton L, McBain J, Callaghan T, Williamson R. Chromosome abnormalities identified by comparative genomic hybridization in embryos from women with repeated implantation failure. Mol Hum Reprod 2002; 8: 1035-41. [PubMed]

3 Kramer S, Ward E, Meadows AT, et al. Medical and drug risk factors associated with neuroblastoma: a case control study. J Natl Cancer Inst 1987; 78: 797-804. [PubMed]

4 Michalek AM, Buck GM, Nasca PC, et al. Gravid health status, medication use, and risk of neuroblastoma. Am J Epidemiol 1996; 143: 996-1001. [PubMed]

5 Steensel-Moll HA, Walkenburg HA, Vandenbroucke JP, et al. Are maternal fertility problems related to childhood leukemia? Int J Epidemiol 1985; 14: 555-59. [PubMed]

6 Roman E, Ansell P, Bull D. Leukemia and non-Hodgkin's lymphoma in children and young adults: are prenatal and neonatal factors important determinants of disease? Br J Cancer 1997; 76: 406-15. [PubMed]

7 Bruinsma F, Venn A, Lancaster P, et al. Incidence of cancer in children born after in-vitro fertilization. Hum Reprod 2000; 15: 604-07. [PubMed]

8 Klip H, Burger CW, de Kraker J, et al. Risk of cancer in the offspring of women who underwent ovarian stimulation for IVF. Hum Reprod 2001; 16: 2451-58. [PubMed]

9 Anteby R, Cohen E, Anteby E, BenEzra D. Ocular manifestations in children born after in vitro fertilization. Arch Ophthalmol 2001; 119: 1525-29. [PubMed]

10 Cruysberg JRM, Moll AC, Imhof SM. Bilateral sporadic retinoblastoma in a child born after in vitro fertilization. Arch Ophthalmol 2002; 120: 1773.

11 BenEzra D, Chirambo MC. Incidence of retinoblastoma in Malawi. J Pediat Ophthalmol 1976; 13: 340-43. [PubMed]

12 de Jong-van den Berg LTW, Cornel MC, van den Berg PB, et al. Ovulation inducing drugs: a drug utilization and risk study in the Dutch population. Int J Risk Safety Med 1992; 3: 99-112. [PubMed]

13 Winston RM, Hardy K. Are we ignoring potential dangers of in vitro fertilization and related treatments? Nat Cell Biol 2002; 4: (suppl) s14-s18. [PubMed]

Incidence of retinoblastoma in children born
after in-vitro fertilisation

The Lancet Volume 361, Number 9354 25jan03

Research letters

Annette C Moll, Saskia M Imhof, Johannes R M Cruysberg, Antoinette Y N Schouten-van Meeteren, Maarten Boers, Flora E van Leeuwen

Every year, 3000 women in the Netherlands are treated with in-vitro fertilisation (IVF), and results from studies suggest that the offspring of these women do not have a heightened risk of cancer. Between November, 2000, and February, 2002, we diagnosed retinoblastoma in five patients born after IVF. To calculate the relative risk of the disease, we assumed that the proportion of children conceived by IVF in the Netherlands is between 1·0 and 1·5%, and that the five patients who we diagnosed with the disease represent all new cases in the Netherlands during that period. Relative risks for retinoblastoma were significantly raised (7·2 [95% CI 2·4-17·0], and 4·9 [1·6-11·3], for 1% and 1·5% rates, respectively). This possible association of an increased risk of retinoblastoma in a population-based study needs to be established.

Lancet 2003; 361: 309-10

See Commentary

Retinoblastoma is a malignant tumour of the retina that occurs in childhood. Since 1945, the incidence of this disease in the Netherlands has been constant at around 1 per 17 000 livebirths.1 In most cases (60%), the disease is non-hereditary and affects only one eye. However, in the 40% of hereditary cases, both eyes are usually affected.

Every year, 3000 women in the Netherlands undergo in-vitro fertilisation (IVF). Although Klip and colleagues2 showed that the offspring of women who underwent IVF between 1980 and 1995 did not have an enhanced risk of cancer, the number of cancers (seven in IVF, nine in controls) of any type that was seen was very small. In the Vrije University Medical Centre, where about 95% of all patients with retinoblastoma in the Netherlands are treated, we diagnosed retinoblastoma in five patients born after IVF between November, 2000, and February, 2002. Therefore, we compared the incidence of retinoblastoma in the IVF population with that in the general population, and estimated the relative risk of retinoblastoma in children born after IVF in the Netherlands.

In this analysis, we calculated the ratio of observed to expected numbers of retinoblastoma cases in the study population using data from the Dutch retinoblastoma registry1 and the Netherlands cancer registry. We derived confidence limits of the ratio obtained by use of the Poisson distribution of the observed numbers.

Incidence of the disease was 2·6 per 100 000 children in the first year of life, 0·9 per 100 000 in those aged between 1 and 4 years, and 0·1 per 100 000 in 5-9 year olds. In the Netherlands, an estimated 1-1·5% of children are conceived after IVF.3 We calculated that 0·69 retinoblastoma cases would be expected in children conceived after IVF between 1995 and 2001 using numbers of births since 1995 and the 1-year age-specific mortality rates in the Netherlands (obtained by calendar year and sex from Statistics Netherlands), the estimate that 1% of all births are conceived by IVF, and the sex-specific and age-specific retinoblastoma incidence rates. On the assumption that the five patients we saw at our clinic represented all cases in the Netherlands born between Jan 1, 1995, and Dec 31, 2001, we calculated a significantly increased risk ratio (RR) of 7·2 (95% CI 2·4-17·0). If we used the upper estimate that 1·5% of all births are after IVF, then the RR was 4·9 (1·6-11·3). These might be conservative estimates of the true risk since some retinoblastoma cases could have been treated in other hospitals, and the oldest child with retinoblastoma was born in 1997. However, because we based the expected number of cases on all children born from 1995 onwards, we believe our results are conservative.

The table shows characteristics of the five patients with retinoblastoma. None of the patients had a family history of retinoblastoma. Two were one of a twin pair, but the siblings of these two twins have no ocular abnormalities. IVF (supplemented by intracytoplasmatic sperm injection in one case) was done in four different IVF centres.

In the three patients with unilateral retinoblastoma, the tumour was enucleated. Of the two patients with bilateral disease, one eye was enucleated; in one, the remaining eye was treated with a radioactive ruthenium plaque, and in the other, an external beam radiation therapy was used. When last seen in October, 2002, all patients were alive and free of disease.

	 Age at 		
	 diagnosis 	   Birthweight	Gestational	Number of	Cause of 	DNA 13q14	Other fertility  
	 (months)  Eye 	   (g)  	age (weeks)  	IVF cycles 	subfertility 	analysis 	treatment 
Patient 1  38 	   Both    3885 	40 		8 		Unexplained 	Exon   		6 courses of clomid  
 (F, 1997)										8 mutated
Patient 2  15	   Left    2005 	36 		2 		Maternal cause 	Normal 		None 
 (M, 1999)*
Patient 3  34 	   Right   2135 	41 		1 		Unexplained 	Normal 		None 
 (F, 1998)*
Patient 4  8·5 	   Both    2330 	39 		2 		Unexplained 	Intron  	8xAI attempts 
 (M, 2001)†										3 mutated‡
Patient 5  32 	   Left    4100 	41 		2 		Paternal cause 	Normal 		1xICSI at same time  
 (F, 1999)

M=male. F=female. AI=artificial insemination. ICSI=intracytoplasmatic sperm injection. 
*One of a pair of dizygous twins. †One of a pair of twins until co-twin died after spontaneus 
abortion at 8 weeks' gestation. ‡Variation in a non-coding part of the 13q14 gene, 
no pathogen mutation found. Characteristics of five patients with retinoblastoma born after IVF in the Netherlands 

The number of infants born after IVF is increasing steadily. Cancer incidence in this fast growing population of children has been investigated in only very few studies. In their report of ocular disorders in 47 children born after IVF in Israel, Anteby and colleagues4 describe a case of unilateral retinoblastoma in a child conceived with donor sperm; therefore, close investigation of the association between IVF and retinoblastoma was not possible.

A large retrospective cohort study from Sweden reported that 5·4% of children born after IVF had developmental problems, but that there was no increased risk of cancer in this group. This finding about cancer risk is lent support by results of a study by Bruinsma and colleagues5 that did not show a significantly increased incidence of cancer in children born after assisted conception in Australia. In a cohort of 9484 offspring of Dutch women who had ovarian stimulation before IVF, 16 cases of cancer were reported,2 which suggests that the risk of childhood cancer was not increased compared with the general population and with an internal reference group of subfertile women who were treated without IVF. Although IVF was done during 1980-95, no retinoblastoma cases were reported in children born during this time. Our five patients were all born in the Netherlands after 1995 between 1997 and 2001. That five cases arose during this time, but that none were seen during the earlier period is important, and represents a striking excess in disease frequency. Did something change in the IVF procedure itself (possibly the culture medium), or is our observation only a chance finding? Perhaps the same genetic factors are involved in infertility and retinoblastoma. However, none of the parents of our patients had had retinoblastoma, and we have not identified any published data about infertility problems of unaffected parents of retinoblastoma patients.

Whether treatment with ovulation-inducing drugs increases the risk of childhood cancer is an important matter, especially with the rising numbers of women undergoing treatment for subfertility. Future investigators should consider the number of IVF treatments, other fertility drugs given before IVF, and the possibility that serious disorders in children conceived by IVF are diagnosed earlier than those in other children who do not have such close medical surveillance. Our finding requires further research to confirm the association and to explore a possible causal mechanism.

Contributors

A C Moll, F E van Leeuwen, and J R M Cruysberg had the idea for the study. A C Moll and F E van Leeuwen designed the study. A C Moll, S M Imhof, J R M Cruysberg, A Y N Schouten-van Meeteren gathered data, A C Moll and F E van Leeuwen processed data and did statistical analysis. A C Moll, F E van Leeuwen, and M Boers drafted the manuscript. All investigators contributed to the interpretation of data and manuscript revisions.

Conflict of interest statement

None declared.

Acknowledgments

We thank W J Klokman from the Netherlands Cancer Institute for statistical advice.

1 Moll AC, Kuik DJ, Bouter LM, et al. Incidence and survival of retinoblastoma in the Netherlands: a register-based study 1862-1995. Br J Ophthalmol 1997; 81: 559-62. [PubMed]

2 Klip H, Burger CW, de Kraker J, van Leeuwen FE. Risk of cancer in the offspring of women who underwent ovarian stimulation for IVF. Hum Reprod 2001; 16: 2451-58. [PubMed]

3 Koudstaal J, van Dop PA, Hogerzeil HV, et al. Pregnancy course and outcome in 2965 pregnancies after in-vitro fertilisation in the Netherlands (in Dutch). Ned Tijdschr Geneeskd 1999; 143: 2375-80. [PubMed]

4 Anteby I, Cohen E, Anteby, BenEzra D. Ocular manifestations in children born after in vitro fertilization. Arch Ophthalmol 2001; 119: 1525-29. [PubMed]

5 Bruinsma F, Venn A, Lancaster P, Speirs A, Healy D. Incidence of cancer born after in-vitro fertilization. Hum Reprod 2000; 15: 604-07. [PubMed]

Departments of Ophthalmology (A C Moll MD, S M Imhof MD), Paediatric Oncology (A Y N Schouten-van Meeteren MD), and Clinical Epidemiology and Biostatistics (Prof M Boers MD, Prof F E van Leeuwen PhD), Vrije Universiteit Medical Centre, Amsterdam 1081 HV, Netherlands; Department of Ophthalmology, University Medical Centre, Nijmegen, Netherlands (Prof J R M Cruysberg MD); and Department of Epidemiology, Netherlands Cancer Institute, Amsterdam (Prof F E van Leeuwen)

Correspondence to: Annette C Moll a.moll@vumc.nl

Ocular manifestations in children born after in vitro fertilization.

Arch Ophthalmol 2001 Oct;119(10):1525-9

Comment in: Arch Ophthalmol. 2002 Dec;120(12):1773; author reply 1773-4.

Anteby I, Cohen E, Anteby E, BenEzra D.

Pediatric Ophthalmology Unit, Department of Ophthalmology, Hadassah University Hospital, Jerusalem, Israel. irene@hadassah.org.il

OBJECTIVE: To report the ocular abnormalities found in children born after in vitro fertilization. METHODS: Forty-seven children (25 girls and 22 boys) born after an in vitro fertilization pregnancy (mean +/- SD birth weight, 2335 +/- 817 g; range, 924-4300 g) and referred for ophthalmic evaluation were included in the study. All underwent a thorough ocular examination. Obstetric history was gathered following a detailed questionnaire with the mothers. RESULTS: Of 70 eyes among nonverbal children, visual acuity was "normal for age" in 60 (86%), "fair" in 4 (6%), and "poor" in 6 (9%). Visual acuity in 24 eyes in verbal children ranged from 6/6 to no light perception, with 4 (17%) having poor vision. Cycloplegic refraction disclosed an emmetropia in 22 (27%), hypermetropia in 47 (57%), and myopia in 13 (16%) of the eyes. Anisometropia of more than 1.0 diopters was found in 8 children. Major ocular malformations were observed in 12 (26%) of the 47 children. These malformations included Coats disease, congenital cataract, congenital glaucoma, hypoplastic optic nerve head, idiopathic optic atrophy, coloboma with microphthalmos, and retinoblastoma. CONCLUSIONS: Ocular anomalies were frequently observed in this cohort of offspring born after in vitro fertilization. A diligent and prospective prenatal search for such malformations should unveil the real prevalence of ocular malformations in children born after in vitro fertilization.