Are fertility drugs a risk factor for persistent trophoblastic tumour?

Patrick Petignat,1, Pierre Vassilakos and Aldo Campana

Department of Gynecology and Obstetrics, University Hospitals of Geneva, Geneva, Switzerland


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The introduction of ovulation-inducing drugs has raised concern that women exposed to these therapies may be at increased risk of cancer. We assessed the potential association between exposure to fertility drugs and the risk of developing persistent trophoblastic tumour (PTT). METHODS: We conducted a systematic review of the English and non-English language literature using the National Library of Medicine's Medline to identify all observations of patients with hydatidiform mole (HM) after treatment with ovulation-inducers. RESULTS: Fifty-two cases were considered including 26 singleton molar pregnancies and 26 multiple molar pregnancies consisting of an HM and one or more co-existent fetus(es) (HM-and-CF). PTT occurred in 15% of patients with singleton HM and in 42% of patients with HM-and-CF, 15% of whom had a metastasic disease. Of those patients with HM-and-CF, 16 patients delivered at <24 weeks gestation, mostly because of vaginal haemorrhage. Ten patients delivered at >=24 weeks of gestation, six of whom (25%) had a normal live child. These results are similar to spontaneously conceived pregnancies. CONCLUSIONS: Although women having an HM after therapy with ovulation-inducing drugs seem to have no added risk of PTT, multiple pregnancies are more likely to occur, and the overall risk may be increased.

Key words: hydatidiform mole/metastasis/ovulation induction/persistent trophoblastic tumour


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Hydatidiform moles (HM) are tumours originating from placental tissues and currently two main types are recognized: the complete hydatidiform mole (CHM) and the partial hydatidiform mole (PHM) (Vassilakos et al., 1977Go; Szulman and Surti, 1978aGo,bGo). In most patients, HM regresses spontaneously after uterine evacuation. Nevertheless, 18–29% of those with CHM and ~4% with PHM have persistently elevated ß-HCG values, corresponding to residual trophoblastic tissue or persistent trophoblastic tumour (PTT) in pelvic or extrapelvic sites (Berkowitz and Goldstein, 1996Go). PTT corresponds to an invasive HM with neoplasic behaviour because the molar tissue invades deeply into the myometrium, penetrates into the uterine vessels, and may be transported haematogeneously to distant sites such as the lung, brain, or liver to form metastases.

HM after fertility therapy seems to be a rare complication despite the widespread use of these treatments. The true incidence is however not known due to the lack of sufficient information regarding the number of treatments performed and the number of HMs avoided by current policies in reproductive medicine. These policies imply that only embryos derived from normally fertilized oocytes are selected for replacement. The normal fertilization process is usually considered to have occurred when two pronuclei (2PN) and two polar bodies are observed. If the number of pronuclei is different, as for monopronuclear (1PN) or tripronuclear (3PN), the embryos are generally not transferred as it is presumed that some of them could be an origin for HM. An incidence of 1/659 pregnancies has been disclosed by Merril-National Labs., Cincinnati, Ohio, USA, and reported by Schneidermann and Waxman in a group of 2 369 clomiphene-induced pregnancies (natural incidence estimated as 0.5–1.1/1000 in USA) (Schneiderman and Waxman, 1972Go). In another report Kurachi reported a rate of 1/1034 pregnancies in Japan (natural incidence estimated as 1.9–2.8/1000 in Japan) (Kurachi, 1982Go; Bracken, 1987Go).

New treatments in assisted reproductive medicine represent a major step forward but a careful evaluation is necessary to identify potential harmful effects on the mother. Molar pregnancy after exposure to fertility drugs is a rare condition that precludes an appropriate prospective study. Currently, the only available information is on the basis of case reports. These data do not permit an assessment of the real rate of PTT due to reporting bias but only allow the discussion of the potential association between the exposure to this treatment and the subsequent development of PTT.

Our study aim was to identify all published reports of molar pregnancies after ovulation induction and to evaluate if there could be an association between this therapy and the development of PTT or metastatic disease.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
All cases of HM pregnancies following the use of assisted reproductive technologies published in the literature from January 1966 through July 2001 were retrieved by a key word search of the National Library of Medicine's Medline and reviewed. Key words used were: molar pregnancy or hydatidiform mole accompanied by any of the following; ovulation induction or clomiphene citrate or gonadotrophin or HMG or HCG or FSH or IVF or ICSI. The citation lists of retrieved articles were then reviewed to source other potential publications.

All reports were reviewed with regards to (i) pre-evacuation clinical features: type of fertility therapy used, maternal age at diagnosis, estimated gestational age at diagnosis, gravidity, parity, presenting symptoms, ultrasound findings, pre-evacuation HCG; and (ii) post-evacuation clinical features: type of HM, molar karyotype, PTT, and presence of metastasis. For pregnancies combining an HM and coexistent fetus(es) (HM-and-CF), the additional following features were extracted: presence before the transfer of two pronuclei, estimated gestational age at termination, indication for termination/delivery, fetal karyotype, and fetal outcome.

Patients with HM-and-CF were divided into two groups: (i) pregnancy with a delivery <=24 weeks gestation; and (ii) pregnancy with a delivery at >=24 weeks gestation. The percentage of patients was calculated only for those with available data. Missing or incomplete clinical data of reported patients were considered as a not-available (NA) category. We have considered that patients with a mention of one negative ß-HCG or other mention attesting an absence of PTT, as having no PTT. Differences in continuous variables were evaluated using the Mann-Whitney U-test and differences in proportion by the two-tailed Fisher's exact test. Statistical significance was defined as a P-value <0.05.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Identified reports and exclusions
A total of 58 observations were retrieved. Six cases were further excluded for the following reasons: (i) no follow-up information provided concerning remission or persistence (n = 3); (ii) patients with choriocarcinoma (n = 2); (iii) patients with no histological examination (n = 1). The present study comprises 52 patients presenting an HM after ovulation induction. They were divided into two groups: (i) patients with a singleton HM (n = 26), and (ii) patients with HM-and-CF (n = 26). The first group included 11 CHM, one PHM and 14 cases of HM of unknown type (corresponding to reports mostly published pre-1980 without classification into CHM or PHM). The second group included 25 CHM with normal fetus(es), and one PHM with a normal fetus.

Comparison between singleton HM and HM-and-CF
Table IGo summarizes the clinical and laboratory parameters of the reported cases and compares patients with singleton HM and those with an HM and CF. A total of 15% of patients with singleton HM developed a PTT but none had metastasis. In patients with HM-and-CF, 42% developed a PTT with 15% developing metastasis. All patients with metastatic disease had lung metastasis and all achieved a complete remission after chemotherapy. Among the cases of HM-and-CF, 17 were twin, and there were eight triplet or quadruplet pregnancies. In these two groups, PTT was 41% (7/17) and 50% (4/8) respectively.


View this table:
[in this window]
[in a new window]
 
Table I. Hydatidiform mole after fertility therapy: comparison of clinical and biological features between singleton and multiple pregnancies combining a hydatidiform mole with coexistent fetus(es)
 
Type of fertility therapy used and risk of PTT
Table IIGo shows the type of fertility drugs used and the risk of PTT for singleton HM and HM-and-CF.


View this table:
[in this window]
[in a new window]
 
Table II. Hydatidiform mole after fertility therapy: type of ovarian stimulation and development of persistent trophoblastic tumour and metastasis
 
CHM-and-CF: Comparison between deliveries <24 and >=24 weeks gestation
Data on features of pregnancy with a gestational age of >=24 weeks are summarized in Tables III and IVGoGo. Among these women (n = 10), six women delivered live infants without any anomalies (five Caesarean section and one vaginal delivery). One subsequently developed PTT with lung metastasis. In patients who delivered <24 weeks, the information about delivery was available in 12/16 cases. This included: uterine evacuation before fetal viability due to spontaneous vaginal abortion (n = 4), vaginal bleeding (n = 4), pre-eclampsia (n = 3), and hyperemesis gravidarum with abdominal pain (n = 1). None of the patients with a triplet (n = 6) or quadruplet (n = 2) pregnancy delivered a live infant.


View this table:
[in this window]
[in a new window]
 
Table III. Hydatidiform mole after fertility therapy: clinical features of patients with complete hydatidiform mole co-existing with fetus(es) of >=24 weeks gestation.
 

View this table:
[in this window]
[in a new window]
 
Table IV. Hydatidiform mole after fertility therapy: comparison of clinical and biological features between pregnancies of <24 weeks and >=24 weeks gestation
 
Other features
Since 1985, the results of ultrasonographic examinations were available in all cases of singleton HM and in 24/25 CHM-and-CF. The diagnosis was correctly made in 7/13 (53%) and 14/24 (58%) cases respectively. In the remaining cases, it was either a misinterpretation or simply the diagnosis of HM was not recognized. The diagnosis of HM was made earlier for singleton HM than HM-and-CF (P < 0.05) where frequently it was only made at the second trimester. Chromosomal analysis or other cytogenetic examinations were reported in only one instance for patients with singleton HM, and in 16/25 of those with HM-and-CF. In other cases, the specific diagnosis was confirmed only upon the histological examination. For those cases where the genetic origin had been demonstrated, namely, androgenetic for molar tissue and biparental for the fetus, the information was available in six instances. In this group, the risk of PTT was 50%. The assessment of the early fertilization process was mentioned in six cases, and in five cases the 2PN stage was observed before the embryo transfer.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The use of ovulation-inducing drugs such as clomiphene citrate or gonadotrophin (HMG or FSH) is very popular in therapy for anovulation, but the introduction of these new technologies has raised concern that these women may be at increased risk of cancer. Several studies have investigated this topic and shown that fertility drugs were not associated with an increased maternal risk of breast, ovarian or uterine cancer (Gluds et al., 1998Go; Venn et al., 1999Go). As ovulation-inducers cause ovulation of more than one ovum, the question can be raised as to whether the increase in the production of immature or anucleated ova (an underlying mechanism of HM) by these drugs may be a contributing factor to the development of HM or invasive HM.

CHM may be considered as a precancerous condition which can transform into an invasive tumour. The aim of our study was to identify potential carcinogenic effects of ovulation-inducers and to analyse the clinical outcome of singleton HM and HM-and-CF pregnancies occurring after fertility therapy. This information could provide a basis for decision-making and the counselling of patients when an HM is diagnosed in early or late pregnancy. The present series is comprised exclusively of patients with an HM after having undergone fertility treatment and, to the best of our knowledge, no previous study has evaluated this issue, apart from isolated case reports.

Patients with singleton HM have an incidence of PTT after evacuation of the mole of 14%. This rate is similar to those with pregnancy occurring naturally. Compared with HM-and-CF, singleton HM are diagnosed earlier in the pregnancy (11.5 versus 16 weeks gestation; P < 0.05) probably because the presence of fetal heartbeat falsely reassured the clinician and is responsible for the delayed diagnosis. A statistically significant difference in age between the two groups was observed (31 versus 27.5 years; P < 0.05), but we have no immediate explanation for this finding.

Patients with an HM-and-CF have a pregnancy composed of two different conceptus; one is a normal fetus and placenta and the other is a molar pregnancy. This type of pregnancy has a more aggressive post-evacuation behaviour with a risk of PTT significantly higher than a singleton HM. Steller et al. reported 12/22 (55%) PTT in patients with HM-and-CF compared with only 10/71 (14%) in patients with single HM (Steller et al., 1994Go). Other reports have confirmed this observation with a risk of PTT which has been estimated as between 40–57% (Vejerslev, 1991Go; Miller et al., 1993Go; Bristow et al., 1996Go; Fishman et al., 1998Go; Bruchim et al., 2000Go; Matsui et al., 2000Go). We found similar results in our series with a risk of PTT occurring of 42%, suggesting that the course of CHM-and-CF after ovulation induction has a similar evolution to a natural one. It is still unclear if the greater risk of PTT is associated with a more aggressive behaviour of the molar tissue or because of delayed delivery. However, recent reports have observed that an advancement of the gestational age does not appear to increase the risk of developing a PTT (Bristow et al., 1996Go; Matsui et al., 2000Go). Our report concurs with the existing literature in that the rate of PTT in pregnancy of <24 weeks and >=24 weeks gestation is similar.

At present, it is possible to distinguish between a prenatal diagnosis of a triploid non-viable fetus and a chromosomally-normal and viable infant. The latter condition presents the patient and the physician with a critical dilemma between a therapeutic abortion or an expectant management until fetal viability. The conservative approach is supported by the fact that there have now been several reported cases of HM-and-CF after natural conception that have been carried to viability and delivery of a live infant. In a review of the literature, Bristow et al. have found 26 cases of HM-and-CF with seven cases of surviving infants (Bristow et al., 1996Go).

Once a diagnosis of gestation consisting of an HM and co-existing chromosomally-normal fetus(es) has been made, and if the clinical course is stable, the decision to allow such a pregnancy to continue should be taken with the couple. The women should be aware of the increased risk of pregnancy complications such as severe antepartum haemorrhage, pre-eclampsia or hyperemesis gravidarum that may require prompt uterine evacuation. As previously mentioned, if the gestation shows a benign clinical course, an expectant observation until infant viability must be considered. Such a management may be encouraged in those patients having undergone fertility therapy, sometimes at an advanced age and after many attempts of assisted conception, which implies a profound desire to continue the pregnancy until viability of the fetus. According to our series and that of Bristow et al., ~25% of patients will have the possibility of having a viable live birth (Bristow et al., 1996Go).

Results from the present study must be interpreted within the context and limitations of our data as they include reports from the literature only and, with meta-analyses, are subject to reporting bias. It is well known that cases with complications are documented more frequently than uneventful observations. Another limitation is that some observations have been included despite the fact that no cytogenetic analysis was carried out. However, at present, the only way to have some insight into the course of HM after ovulation-inducers, and to acquire knowledge about the optimal management of this type of pregnancy, is to retrieve from the literature all the well-documented cases.

In conclusion, our study suggests that women having a singleton or multiple pregnancy after exposure to ovulation-inducers seem to have no additional risk of PTT than those who conceive naturally. However, as this therapy is more likely to result in multiple pregnancy than spontaneously conceived singleton pregnancy, patients are at greater risk of developing PTT. In clinical practice, the overall rate of PTT after ovulation-inducing drug treatment is probably increased.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors would like to thank Rosemary Sudan for editorial assistance.


    Notes
 
1 To whom correspondence should be addressed at: Department of Gynecology and Obstetrics, Hôpital Cantonal de Fribourg, 1708 Fribourg, Switzerland. E-mail: patrick.petignat{at}hcuge.ch Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Adachi, N., Ihara, Y., Ito, H., Sasaki, M., Mandai, M., Tanada, S. and Niwa, H. (1992) Two cases of twin pregnancy with complete hydatidiform mole and coexistent fetus. Nippon Sanka Fujinka Gakkai Zasshi, 44, 1463–1466.[Medline]

Berkowitz, R.S. and Goldstein, D.P. (1996) Chorionic tumours. N. Engl. J. Med., 335, 1740–1748.[Free Full Text]

Bracken, M.B. (1987) Incidence and aetiology of hydatidiform mole: an epidemiological review. Br. J. Obstet. Gynaecol., 94, 1123–1135.[ISI][Medline]

Bristow, R.E., Shumway, J.B., Khouzami, A.N. and Witter, F.R. (1996) Complete hydatidiform mole and surviving coexistent twin. Obstet. Gynecol. Surv., 51, 705–709.[Medline]

Bruchim, I., Kidron, D., Amiel, A. Altaras, M., and Fejgin, M.D. (2000) Complete hydatidiform mole and a coexistent viable fetus: report of two cases and review of the literature. Gynecol. Oncol., 77, 197–202.[ISI][Medline]

Chao, A.S., Tsai, T.C. and Soong, Y.K. (1999). Clinical management of a quadruplet pregnancy with a classical hydatidiform mole: case report and review of literature. Prenat. Diagn., 19, 1073–1076[ISI][Medline]

Cheng, P.J., Chang, F.H., Liang C.C., Chang, T.C., Soong, Y.K. and Hsueh, C. (1995) A twin pregnancy with a hydatidiform mole and an alive, coexistent baby after in-vitro fertilization and embryo transfer. J. Assist. Reprod. Gen., 12, 389–392.[ISI][Medline]

Fishman, D.A., Padilla, L.A., Keh, P., Cohen, L., Frederiksen, M. and Lurain, J.R. (1998) Management of twin pregnancies consisting of a complete hydatidiform mole and normal fetus. Obstet. Gynecol., 91, 546–550.[Abstract/Free Full Text]

Gluds, E., Kjaer, S.K., Troisi, R. and Brinton, L.A. (1998) Fertility drugs and ovarian cancer. Epidemiol. Rev., 20, 237–257.[ISI][Medline]

Jinno, M., Ubukata, Y., Hanyu, I., Satou, M., Yoshimura, Y. and Nakamura, Y. (1994) Hydatidiform mole with a surviving coexistent fetus following in-vitro fertilization. Hum. Reprod., 9, 1770–1772.[Abstract]

Kurachi, K. (1982) Congenital malformations after clomiphene therapy. Jap. J. Fert. Ster., 27, 454.

Matsui, H., Sekiya, S., Hando, T., Wake, N. and Tomoda, Y. (2000) Hydatidiform mole coexistent with a twin live fetus: a national collaborative study in Japan. Hum. Reprod., 15, 608–611.[Abstract/Free Full Text]

Miller, D., Jackson, E., Ehlen, T. and McMurtie, E. (1993) Complete hydatidiform mole coexistent with a twin live fetus clinical course of four cases with complete cytogenetic analysis. Gynecol. Oncol., 50, 119–123.[ISI][Medline]

Montes-de-Oca-Valero, F., Macara, L. and Shaker, A. (1999). Twin pregnancy with a complete hydatidiform mole and co-existing fetus following in-vitro fertilization: case report. Hum. Reprod., 14, 2905–2907.[Abstract/Free Full Text]

Schneiderman, C.I. and Waxman, B. (1972) Clomid therapy and subsequent hydatidiform mole formation. A case report. Obstet. Gynecol., 39, 787–788.[ISI][Medline]

Shahabi, S., Naome, G., Cobin, L., Verougstraete, A., Masters, L., Zengbe, V., Noel, J.C., Avni, E.F., Donner, C., Verhest, A. et al. (1997) Complete hydatidiform mole and coexisting normal fetus. A report of two cases with contrasting outcomes. J. Reprod. Med., 42, 756–760.[ISI][Medline]

Steller, M.A., Genest, D.R. and Bernstein, M.R. (1994) Clinical features of multiple conception with partial or complete molar pregnancy and coexisting fetuses. J. Reprod. Med., 39, 147–154.[ISI][Medline]

Szulman, A.E. and Surti, U. (1978a) The syndromes of hydatidiform mole. I. Cytogenetic and morphologic correlations. Am. J. Obstet. Gynecol., 131, 665–671.[ISI][Medline]

Szulman, A.E. and Surti, U. (1978b) The syndromes of hydatidiform mole. II. Morphologic evolution of the complete and partial mole. Am. J. Obstet. Gynecol., 132, 20–27.[ISI][Medline]

van de Geijn, E.J., Yedema, C.A., Hemrika, D.J., Schutte, M.F. and ten Velden, J.J. (1992). Hydatidiform mole with coexisting twin pregnancy after gamete intra-fallopian transfer. Hum. Reprod., 7, 568–572.[Abstract]

Vassilakos, P., Riotton, G. and Kajii, T. (1977) Hydatidiform mole: two entities. A morphologic and cytogenetic study with some clinical consideration. Am. J. Obstet. Gynecol., 127, 167–170.[ISI][Medline]

Vejerslev, L.O. (1991) Clinical management and diagnostic possibilities in hydatidiform mole with coexistent fetus. Obstet. Gynecol. Surv., 46, 577–588.[Medline]

Venn, A., Watson, L., Bruinsma, F., Giles, G. and Healy, D. (1999) Risk of cancer after use of fertility drugs with in-vitro fertilisation. Lancet, 354, 1586–1590.[ISI][Medline]

Submitted on October 18, 2001; accepted on January 17, 2002.