Administration of cyclophosphamide at different stages of follicular maturation in mice: effects on reproductive performance and fetal malformations

D. Meirow1,5, M. Epstein2, H. Lewis2, D. Nugent3 and R.G. Gosden4

1 Department of Obstetrics and Gynecology, Rabin Medical Center, Petah Tigva, Israel, 2 Department of Obstetrics and Gynecology, Hadassah Medical Center, Jerusalem, Israel, 3 Centre for Reproduction, Growth and Development, University of Leeds, UK and 4 Women's Pavilion, Royal Victoria Hospital, Montreal, Canada


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
This study assessed reproductive performance, fetal viability and teratogenicity in female mice exposed to cyclophosphamide across a timeline corresponding to different stages of follicle maturation. Pregnancies were established in female Balb/c mice 1–4 weeks after administration of a non-sterilizing dose of cyclophosphamide (75 mg/kg). Each mating group represented a different stage of follicular growth at the time of cyclophosphamide exposure. The number of corpora lutea, pregnancies and fetal resorptions were determined. Surviving fetuses were evaluated for gross malformations. Results indicated that conceptions attributable to follicles exposed to cyclophosphamide at a mature stage had a significantly lower number of implantation sites, 4.82 ± 1.01 versus 8.27 ± 0.81 in controls (P = 0.001) and a high resorption rate, 56% ± 0.11 versus 34% ± 0.07 in controls (P = 0.05). The proportion of corpora lutea in this group which resulted in viable fetuses was extremely low, 0.2 ± 0.06 versus 0.51 ± 0.07 in controls (P = 0.001). Malformation rate was more than 10 times higher in all treated groups (P < 0.05) and a particularly high incidence of 33% (P = 0.0014) was observed in conceptions attributable to oocytes exposed to cyclophosphamide at the earliest stages of follicle growth. With an extended interval between exposure and mating the malformation rate gradually decreased towards normal values in the 12th week group. This study suggests that the effect of cyclophosphamide on female gametes and subsequently on future reproduction is influenced by the stage of oocyte maturation at the time of exposure. Early fertilization post-chemotherapy can result in a high rate of pregnancy failure and high malformation rate. This should be taken into account when considering the use of oocyte retrieval, IVF and embryo cryopreservation in patients currently undergoing chemotherapy.

Key words: cyclophosphamide/malformation/oocyte/pregnancy/resorption


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Cyclophosphamide is a commonly used chemotherapeutic and immunosuppressive agent for the treatment of a wide range of neoplastic diseases and some auto-immune diseases. With increased success of cancer treatment, due in part to the aggressive use of high-dose combination drug therapies, there has been growing concern about the long-term side-effects of this alkylating agent and other antineoplastic drugs (Sutton et al., 1990Go; Rodjer et al., 1990Go).

Many studies have demonstrated that cyclophosphamide and many other chemotherapeutic agents cause gene mutations, chromosomal breaks and rearrangements, and aneuploidy in somatic cells, as well as an increased frequency of secondary, treatment-related tumours in human cancer survivors (Sandoval et al., 1993Go; Povirk and Shuker, 1994Go; Ben-Yehuda et al., 1996Go). In the gonads cyclophosphamide treatment has been shown to cause primordial follicle destruction, and may, in high doses, result in adverse reproductive consequences, including premature menopause and sterility (Himelstein-Braw et al., 1978Go; Apperley and Reddy, 1995Go; Meirow et al., 1997Go; Chiarelli et al., 1999Go). Moreover, animal studies have shown clear evidence that cyclophosphamide causes injury to germ cells as well as induction of transmissible genetic damage (Generoso et al., 1971Go; Becker and Schoneich, 1982Go; Pdydn and Ataya, 1991Go), and this has raised serious concerns regarding the risk of abortions, birth defects, genetic or neoplastic disease in the offspring of cancer survivors who retain fertility after treatment. However, studies on pregnancy outcome in human survivors of cancer treatment have suggested that such concerns are unfounded. Despite the observed increase in germ cell mutations, offspring of women exposed to cancer treatments do not have a greater than normal risk of chromosomal or congenital abnormalities (Hawkins, 1994Go; Sanders et al., 1996Go). However, these data were collected from women who became pregnant a considerable time (often years) after cessation of therapy. Thus, the lapse of a significant period of time between exposure to the mutagenic drugs and conception may enable the oocytes within the primordial follicles to instigate DNA repair mechanisms to correct any genomic damage (Ashwood-Smith and Edwards, 1996Go), or to be eliminated.

With the advent of new reproductive technologies, centres now offer patients the option of oocyte retrieval and embryo cryopreservation before commencement of chemo/radiotherapy in order to preserve fertility. Some centres also offer oocyte retrieval to patients who enter remission following initial chemotherapy treatment cycles, prior to their exposure to more intensive sterilizing treatment. Oocytes retrieved at this point have suffered very recent exposure to chemotherapy and may have been at non-dormant growth stages rather than the primordial stage. It therefore becomes essential to determine if greater risks are associated with oocytes exposed to cytotoxic treatment during or immediately preceding growth stages. The aim of this study was to investigate whether the stage of oocyte development at the time of exposure to cyclophosphamide alters the risks in resultant conceptions. Thus, pregnancy outcome was assessed in terms of fetal viability and teratogenicity in mature female mice exposed to non-sterilizing doses of cyclophosphamide at different intervals prior to conception corresponding to different stages of oocyte maturation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Female inbred Balb/c mice (Harlan UK, Oxon, UK), 6–7 weeks of age (135 animals in total), were injected i.p. with 75 mg cyclophosphamide/kg body weight (ASTA Medica, Frankfurt, Germany). This dose was chosen based on a previous dose-response study which showed that it reduced the ovarian primordial follicle reservoir by 50% without any effect on mating or pregnancy rate (Meirow et al., 1999Go). The females were housed separately until they were introduced to proven fertile males of the same strain at a ratio of one male to a maximum of three females. Females were placed with males either 1, 2, 3 or 4 weeks post-cyclophosphamide treatment: 1 week (n = 27), 2 weeks (n = 29), 3 weeks (n = 30) and 4 weeks (n = 15), each mating group representing a different stage of follicular growth at the time of exposure to the chemotherapy. Conceptions in females mated 1 week after injection will have resulted from oocytes exposed to cyclophosphamide at late pre-antral stages of follicular development (Pedersen, 1970Go), conceptions in mice mated after a 2 week interval were from follicles exposed at growing stages and conceptions which followed a 3 or more week interval were from oocytes exposed as primordial follicles (Figure 1Go). Control animals (n = 34) were injected with sterile water in similar volumes, 0.1–0.2 ml, and mated at the same time intervals.



View larger version (28K):
[in this window]
[in a new window]
 
Figure 1. Diagrammatic representation of methods: groups of mice were mated at weekly intervals following exposure to cyclophosphamide (75 mg/kg). Oocytes, which contributed to the pregnancies, would have been at the stages indicated at the time of exposure.

 
The long term outcome following cyclophosphamide exposure was investigated in a second group of inbred Balb/c female mice, 6–7 weeks of age, which were similarly treated with cyclophosphamide (intraperitoneal injection of 75 mg/kg body weight). Those females were mated 6 weeks (n = 39), 9 weeks (n = 31) or 12 weeks (n = 32) post-cyclophosphamide treatment.

The females were examined daily for evidence of vaginal copulatory plugs, and then removed to a separate cage where they were housed for 12 days and then killed by cervical dislocation. It was decided to terminate the pregnancies after 12 days because at this stage it is still possible to count the corpora lutea, allowing for assessment of the number of ovulated oocytes, resorption sites from non-viable fetuses are visible, and the viable fetuses are developed enough to examine for malformations. The uterus and both ovaries were removed and the following measurements taken: the number of corpora lutea were counted under the light binocular microscope, the numbers of viable and non-viable fetuses were noted, each conceptus was weighed (including the placenta, membranes and amniotic fluid). Fetuses were dissected and inspected for gross anomalies under the light microscope (magnification x10–x30). Those animals which did not show vaginal plugs were separated from males after 5 days, and killed 9 days later (estimated to be approximately day 12 of pregnancy if conception occurred midway through the 5 day mating period that is the average length of an oestrous cycle). Where non-plugged mice did become pregnant, all of the above parameters were examined except for the fetal weights, since the exact gestational age was unknown.

The data were grouped into three categories, based on the following criteria (Table IGo): (i) pregnant group (had pregnancy sacs): (a) vaginal plug found (timed mating): (b) no vaginal plug (non-timed mating). (ii) non-pregnant group (no pregnancy sacs): (a) vaginal plugs and corpora lutea; (b) vaginal plugs but no corpora lutea; (c) neither vaginal plug nor corpora lutea.


View this table:
[in this window]
[in a new window]
 
Table I. Reproductive performance as indicated by the numbers of animals which achieved successful mating and pregnancies in each of the experimental groups following exposure to cyclophosphamide (Cy.) compared with control females
 
The data were analysed statistically using the non-parametric regression methods of Kruskal–Wallis and by analysis of variance.

Ethical approval for animal experimentation was received from the Ethical Committee of Hadassah Medical Center–Hebrew University.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The pregnancy rate in females mated 1 week after treatment was 44% lower than 57–62% for all other females (Table IGo, pregnant females a, b) (P = not significant). No significant difference in mating rates in all groups was observed, also the number of corpora lutea per female was similar (8–11) in all groups (Table IIGo). A high proportion of mice treated 1 week before mating did not conceive despite the presence of corpora lutea (ovulation), 18.5% compared with 2.9% of the controls and 3.3–6.9% of other treated groups (Table IGoc).


View this table:
[in this window]
[in a new window]
 
Table II. Reproductive outcome—statistical analysis of the proportion of ovulations, implantations and resorptions in each experimental group compared with controls
 
In pregnant females an analysis was made of the different parameters of reproductive outcome. The presence and number of pregnancy sacs was evaluated, and these were categorized either as viable pregnancies, or non-viable—also termed resorbed pregnancies. The data on reproductive outcome are recorded in Tables II and IIIGoGo. The results indicated that the group mated 1 week after chemotherapy exhibited poorest reproductive capability: the number of implantation sites (viable as well as resorbed fetuses) was significantly lower in this group, 4.82 ± 1.01, than in the control group, 8.27 ± 0.81 (P = 0.01). This group also had a high rate of resorptions (non-viable fetuses), 0.56 ± 0.11 compared with 0.34 ± 0.07 for controls (P = 0.05). The groups treated 2, 3, and 4 weeks before mating exhibited lower rates of resorptions, 0.19 ± 0.04, 0.31 ± 0.07 and 0.28 ± 0.06 respectively. Overall, pregnant females from the group mated 1 week after cyclophosphamide showed the lowest proportion of corpora lutea which resulted in viable fetuses, 0.20 ± 0.06 (Table IIGo, Figure 2Go), compared with 0.51 ± 0.07 in the controls (P = 0.001) and rates of between 0.53 ± 0.07 and 0.49 ± 0.07 in other treated groups.


View this table:
[in this window]
[in a new window]
 
Table III. Reproductive outcome: raw data, the total numbers of fetuses conceived to treated (Cy.) and control females including numbers of viable, non-viable and malformed fetuses
 


View larger version (31K):
[in this window]
[in a new window]
 
Figure 2. Comparison of the mean numbers (for each female) of corpora lutea, implantations (viable fetuses and resorption sites), viable fetuses and normal (not malformed) fetuses seen in the pregnant females 1, 2, 3 and 4 weeks after treatment with cyclophosphamide, as well as controls (females unsuccessfully mated were not included). P value indicates the significance of the results compared with controls. NS = not significant.

 
In order to compare fetal weights, it was important to consider the effect of litter size on any weight differences. To eliminate this bias, fetal weight was compared across three subgroups, divided according to the total number of fetuses present in the individual female as follows: 1–3, 4–6 and >7 fetuses. No significant difference between the various treated groups and control group was detected in the mean fetal weights in any of the subgroups or overall (Table IVGo). Fetal weight, which fell below 90% confidence limits, was classified as intrauterine growth retardation (IUGR), which occurred sporadically regardless of litter size. A higher but non-significant rate of IUGR was observed in the mice mated 1 (10.3%) and 4 weeks (14.0%) after treatment compared with controls or other treatment groups (2.2–5.2%).


View this table:
[in this window]
[in a new window]
 
Table IV. Fetal weight (average in mg ± SE) in treated groups according to three different categories of litter size, and the incidence of intrauterine growth retardation (IUGR) from each treated group
 
Particularly significant was the increase in the incidence of malformations of all types observed in the offspring of mice treated with cyclophosphamide The malformation rate in all treated groups was at least 10 times higher than the rate (1.2% ± 0.9) found in the control group (P < 0.05 for all groups). Several classes of malformations were observed including brain anomalies (affecting hind-, mid- and fore-brain), anteroposterior–axis aberrations, and limb abnormalities. No one particular type of malformation occurred significantly more often than any other. These results indicate that pregnancies occurring shortly after chemotherapy are at high risk for malformations. A pronounced peak in the malformation rate was observed in fetuses conceived 3 weeks after cyclophosphamide treatment (34% ± 9.4; P = 0.0014). Thereafter a significant drop in malformation rate occurred in mice which had a 4 week interval between treatment and mating (17.5% ± 5.6), possibly a sign of a recovery trend. In order to establish a critical period during which the effects of the treatment are most evident and to investigate this apparent recovery trend, embryos from females conceived after longer time intervals post-cyclophosphamide injection were evaluated. The results indicated that malformation rate decreased 6 (5/39, 12.8%) and 9 weeks (5/31, 16.1%) post-exposure, but fell to a lower value (3%, one malformed from 32 embryos) in the group mated 12 weeks after exposure (Figure 3Go). This appears to confirm a correlation between an extended time interval post-exposure and a decrease in malformation rate.



View larger version (11K):
[in this window]
[in a new window]
 
Figure 3. Incidence of malformations in the fetuses of mice mated at different time intervals (1–12 weeks) following injection with cyclophosphamide (75 mg/kg). * indicates P < 0.05.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The results presented here indicate that the effect of cyclophosphamide on female gametes and subsequently on future reproduction is time dependent and is influenced by the stage of follicle development at the time of exposure. The full span of follicle development in this species is approximately 3 weeks (Pedersen, 1970Go). Oocytes exposed to chemotherapy as late pre-antral follicles (as represented by the group mated 1 week following treatment) seem particularly vulnerable to cyclophosphamide-induced lethal damage. As the results showed (Table IGo), a larger proportion of mice treated 1 week before mating were not pregnant at autopsy. This indicates either failed fertilization, or preimplantation/implantation failure and early resorption. In this group, females achieving pregnancy had significantly fewer implantations and on day 12 of pregnancy exhibited significantly higher losses as represented by higher resorption rate than in any other treated or control group (Table IIGo). These results might imply a particular vulnerability of oocytes at the later growth stages to genetic or cytoplasmic lethal damage; however it could also partly be due to the debilitating effects of treatment on maternal physiology and health. It is less likely to be a result of poor uterine receptivity, since uterine or endometrial toxicity would have affected fetal weight and the data show no significant variation in fetal weights (Table IVGo).

The higher rate of malformations observed across the groups treated (1–9 weeks), at least 10 times greater than that seen in the control group, indicates that oocytes exposed to cyclophosphamide during growing stages suffer from increased sublethal damage as a result of the teratogenic effects of cyclophosphamide. The increase in malformation rate was not the same in all the groups. The results showed that the rate of malformations peaked in the group treated 3 weeks prior to mating (oocytes exposed as follicles just beginning the maturation process). Thereafter malformation rate dropped significantly in the group of mice treated 4, 6, and 9 weeks before mating and approached near normal levels at 12 weeks (Figure 3Go).

This would indicate that oocytes, which began the maturation process during chemotherapy treatment, were most susceptible to non-lethal damage. The subsequent drop in malformation rate following a critical period may be due to the ability of oocytes to repair induced DNA damage (Ashwood-Smith and Edwards, 1996Go). Another possible option for the reduction in malformation rate is that damaged oocytes were gradually lost during the following weeks. Increased teratogenicity cannot be due to direct effects of cyclophosphamide on the fetus since the drug and its metabolites are cleared from the body within hours (Genka et al., 1990Go). This rise in fetal malformations peaking in week 3 and then decreasing again has also been observed in another animal study which looked at the malformation rate following treatment with radiation therapy (Kirk and Lyon, 1982Go). In that study the incidence of abnormalities and loss increased with the time interval between exposure and mating, peaking in animals treated 2–3 weeks before mating, followed by a significant decrease when the interval was between 3–4 weeks. Other studies (Brewen et al., 1976Go; Caine and Lyon, 1977Go; Russell, 1977Go) also bring evidence that conceptions occurring approximately 3 weeks after exposure are most vulnerable to the mutagenic effects of radiation.

The ability of the oocyte to recover over time may explain the apparent contradiction between the increase in malformations shown in animal studies following recent exposure to chemotherapy, and clinical studies showing that offspring of female chemotherapy survivors have no greater incidence of abnormalities or malformations. Until now, children of cancer therapy survivors were by necessity conceived a considerable length of time after cessation of treatment. Those conceptions resulted from oocytes which had been exposed to therapy in a dormant stage often years earlier (Hawkins, 1994Go).

Significant numbers of cancer patients suffer infertility post-chemotherapy treatment. In order to preserve fertility, one of the options available is to freeze-bank embryos, which involves follicle aspiration and IVF prior to chemotherapy administration (Apperley and Reddy, 1995Go). Unfortunately, there are a number of associated difficulties, which prevent widespread use of this procedure for cancer patients. Time is one of the major obstacles, since several weeks of monitoring and ovarian stimulation are often required, and in most cancer patients chemotherapy cannot be delayed. For this reason, some centres tend to offer IVF and embryo cryopreservation during a suitable break in treatment (Brown et al., 1996Go) or during first remission prior to implementation of sterilizing protocols (Lipton et al., 1997Go).

The safety of using IVF and embryo cryopreservation in cancer patients who have recently undergone chemotherapy treatments is questionable. These agents may cause mutations, DNA adducts and structural breaks as well as oxidative damage. Therefore, concerns have been raised as to the quality of embryos fertilized from oocytes harvested immediately following treatment. The full span of follicle growth from the primordial to Graafian stage in humans is in the order of 6–12 months (Wasserman, 1996Go; Gougeon, 1996Go) and thus oocytes collected from patients within 6–12 months of cancer treatment could be compromised. This corresponds to approximately 3 weeks in mice.

This animal model addresses the differences in toxicity response to anti-cancer treatments at different stages of follicular maturation. Results indicate that exposure of mouse oocytes to chemotherapy during different growth stages induces a decrease in implantations and viable pregnancies, and an increase in fetal malformations, which concurs with results from other animal studies (Brewen et al., 1976Go; Caine and Lyon, 1977Go; Russell, 1977Go; Kirk and Lyon, 1982Go). Although there is not yet any evidence to suggest equivalent treatment-induced genetic abnormalities in human oocytes, it would be unwise to ignore potential risks to the offspring of women exposed to cancer treatments. These results suggest that we cannot necessarily apply the available clinical data concerning pregnancy outcome years after exposure to chemotherapy to pregnancies which result from oocyte collection, IVF and embryo cryopreservation immediately after chemotherapy treatments. Further research is needed to clarify the effects of exposure of human oocytes to cytotoxic treatment during growth stages, in particular whether they are at an increased genetic risk. If so, it will become vital to define a `safety period' between cessation of treatment and oocyte retrieval for IVF. Until definitive data are achieved, it would seem advisable to monitor the pregnancy outcome of all cancer patients who undergo oocyte retrieval and IVF, and possibly to screen fetuses and babies for chromosomal aberrations and birth defects.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We gratefully acknowledge financial support from the Israel Cancer Association (D.M.), Barclay Family Cancer Research Foundation (D.M.) and WellBeing (D.N.).


    Notes
 
5 To whom correspondence should be addressed at: Department of Obstetrics and Gynecology, Rabin Medical Center, Israel.E-mail: meirow{at}md2.huji.ac.il Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Apperley, J.F. and Reddy, N. (1995) Mechanism and management of treatment-related gonadal failure in recipients of high dose chemoradiotherapy. Blood Rev., 9, 93–116.[ISI][Medline]

Ashwood-Smith, M. and Edwards, R. (1996) DNA repair by oocytes. Mol. Hum. Reprod., 2, 46–51.[Abstract]

Becker, K. and Schoneich, J. (1982) Expression of genetic damage induced by alkylating agents in germ cells of female mice. Mutat. Res., 92, 447–464.[ISI][Medline]

Ben-Yehuda, D., Krichevsky, S., Caspi, O. et al. (1996) Microsatellite instability and p53 mutations in therapy related leukemia suggest mutator phenotype. Blood, 88, 4296–4303.[Abstract/Free Full Text]

Brewen, J., Payne, H. and Preston, R. (1976) X-ray-induced chromosome aberrations in mouse dictyate oocytes. I. Time and dose relationships. Mutat. Res., 35, 111–20.[ISI][Medline]

Brown, J.R., Modell, E., Obasaju, M. and King, Y.K. (1996) Natural cycle in-vitro fertilization with embryo cryopreservation prior to chemotherapy for carcinoma of the breast. Hum. Reprod., 11, 197–202.[Abstract]

Caine, A. and Lyon, M. (1977) The induction of chromosome aberrations in mouse dictyate oocytes by X-rays and chemical mutagens. Mutat. Res., 45, 325–331.[ISI][Medline]

Chiarelli, A.M., Marrett, L.D. and Darlington, G. (1999) Early menopause and infertility in females after treatment for childhood cancer diagnosed in 1964–1988 in Ontario, Canada. Am. J. Epidemiol., 150, 245–254.[Abstract]

Generoso, W.M., Huff, S.W. and Stout, S.K. (1971) Chemically induced dominant, lethal mutations and cell killing mouse oocytes in the advanced stages of follicular development. Mutat. Res., 11, 411–420.[ISI][Medline]

Genka, S., Deutsch, J., Stahle, P.L. et al. (1990) Brain and plasma pharmacokinetics and anticancer activities of cyclophosphamide and phosphoramide mustard in the rat. Cancer Chemother. Pharmacol., 27, 1–7.[ISI][Medline]

Gougeon, A. (1996) Regulation of ovarian follicular development in primates: facts and hypothesis. Endo. Reversus, 17, 121–155.

Hawkins, M.M. (1994) Pregnancy outcome and offspring after childhood cancer. Brit. Med. J., 309, 1034–1040.[Free Full Text]

Himelstein-Braw, R., Peters, H. and Faber, M. (1978) Morphological study of the ovaries of leukaemic children. Br. J. Cancer, 38, 82–87.[ISI][Medline]

Kirk, M. and Lyon, M.F. (1982) Induction of congenital anomalies in offspring of female mice exposed to varying doses of X rays. Mutat. Res., 106, 73–83.[ISI][Medline]

Lipton, J.H., Virro, M. and Solow, H. (1997) Successful pregnancy after allogenic bone marrow transplant with embryos isolated before transplant. J. Clin. Oncol., 15, 3347–3350.[Abstract]

Meirow, D., Lewin, A., Or, R. et al. (1997) Ovarian failure post-chemotherapy in young cancer patients – risk assessment indicate the need for intervention. Presented at the 53th annual meeting of the American Society for Reproductive Medicine, Cincinnati, USA. Fertil. Steril. (Suppl.), S218.

Meirow, D., Lewis, H., Nugent, D. and Epstein, M. (1999) Subclinical depletion of primordial follicular reserve in mice treated with cyclophosphamide: clinical importance and proposed investigative tool. Hum. Reprod., 14, 1903–1907.[Abstract/Free Full Text]

Pdydn, E. and Ataya, K. (1991) Effect of cyclophosphamide on mouse oocytes in vitro fertilisation and cleavage: recovery. Rep. Toxicol., 5, 73–78.[ISI][Medline]

Pedersen, T. (1970) Determination of follicle growth rate in the ovary of the immature mouse. J. Rep. Fertil., 21, 81–93.[ISI]

Povirk, L. and Shuker, D. (1994) DNA damage and mutagenesis induced by nitrogen mustards. Mutat. Res., 318, 205–226.[ISI][Medline]

Rodjer, S., Swolin, B., Weinfeld, A. and Westin, J. (1990) Cytogenetic abnormalities in acute leukemia complicating melphalan-treated multiple myeloma. Cancer Genet. Cytogenet., 48, 67–73.[ISI][Medline]

Russell, W. (1977) Mutation frequencies in female mice and the estimation of genetic hazards of radiation in women. Proc. Natl Acad. Sci. USA, 74, 3523–3527.[Abstract]

Sanders, J.E., Hawley, J., Levy, W. et al. (1996) Pregancies following high-dose cyclophosphamide with or without high-dose busulphan or total body irradiation and bone marrow transplant. Blood, 87, 3045–3052.[Abstract/Free Full Text]

Sandoval, C., Pui, C.H., Bowman, L.C. et al. (1993) Secondary acute myeloid leukemia in children previously treated with alkylating agents, intercalating topoisomerase II inhibitors, and irradiation. J. Clin. Oncol., 11, 1039–1045.[Abstract]

Sutton, R., Buzdar, A. and Hortobagyi, G. (1990) Pregnancy and offspring after adjuvant chemotherapy in breast cancer patients. Cancer, 65, 847–850.[ISI][Medline]

Wasserman, P.M. (1996) Oogenesis. In Adashi, E., Rock, J. and Rosenwaks, Z. (eds), Reproductive Endocrinology, Surgery and Technology. Lippincott–Raven, Philadelphia, pp. 341–357.

Submitted on September 14, 2000; accepted on January 19, 2001.