INTRODUCTION — One of the strongest predictors of emotional well-being in cancer survivors, besides sexual function, appearance, and employability, is feeling healthy enough to be a good parent. Cancer survivors are often fearful that their history of cancer or its treatment will have an adverse impact on their offspring conceived after cancer treatment by placing them at risk for malignancy, congenital anomalies, or impaired growth and development. They are also concerned about the risks of cancer recurrence, infertility, miscarriage, and achieving a successful pregnancy outcome.
Despite these concerns, surveys have reported that fewer than 60 percent of respondents had received information about fertility after cancer treatment, and even fewer had received information about potential risks to offspring [1,2]. Others have reported that the rate of elective pregnancy termination among female cancer survivors was higher compared to sibling controls because of the fear that their prior cancer therapy would affect their children [3]. Patient education regarding future reproductive function is thus an important component of the care of individuals with cancer [1-4]. (See "Various rehabilitation issues in patients treated for cancer", section on 'Infertility'.)
This topic review will discuss general issues regarding fertility and pregnancy in cancer survivors. Methods of fertility preservation are discussed separately. (See "Fertility preservation in patients undergoing gonadotoxic treatment or gonadal resection".)
Pregnancies complicated by specific types of cancer are also discussed elsewhere in individual topic reviews, for example:
FERTILITY — Cytotoxic agents and radiation therapy can produce gonadal dysfunction in both men and women. (See "Pubertal development and gonadal function in survivors of childhood cancer", section on 'Infertility' and "Ovarian failure due to anticancer drugs and radiation" and "Effects of cytotoxic agents on gonadal function in adult men".)
The Childhood Cancer Survivor Study compared the pregnancy rate in female and partners of male cancer survivors to the rate in a sibling cohort [5,6]. The relative risk for female survivors of ever being pregnant was 0.81 (95% CI, 0.73-0.90) compared with female siblings; poor prognostic factors included hypothalamic/pituitary radiation dose ≥30 Gy, an ovarian/uterine radiation dose >5 Gy, a summed alkylating agent dose score of three or four, or treatment with lomustine or cyclophosphamide [6]. Males who were not surgically sterile were less likely to sire a pregnancy than siblings (HR 0.56, 95% CI, -0.49 to 0.63); poor prognostic factors included radiation therapy >7.5 Gy to the testes, higher cumulative alkylating agent dose score, or treatment with cyclophosphamide or procarbazine [5,7].
Women
Ovarian effects — Loss of ovarian function can occur after either chemotherapy or radiation therapy [8]. The frequency of amenorrhea is dependent upon the type of agent (alkylating agents are particularly deleterious to the ovarian follicle pool), the cumulative dose, and the patient's age (risk of ovarian failure increases with age). The loss of ovarian function may be permanent or temporary. Temporary amenorrhea results from destruction of maturing follicles, whereas permanent amenorrhea is caused by depletion of viable primordial follicles [9]. Importantly, ovarian reserve can be decreased even in women with regular menstrual cycles [10-12]. Ovarian failure related to radiation therapy and chemotherapy is discussed in detail elsewhere. (See "Ovarian failure due to anticancer drugs and radiation".)
Two FSH levels in the menopausal range (for the specific assay) measured at least one month apart and associated with amenorrhea/oligomenorrhea and low estradiol levels is strongly suggestive of ovarian failure.
Young women who have undergone unilateral oophorectomy generally do not have reduced fertility since young women have many primordial follicles per ovary; however, prior unilateral oophorectomy may impact fertility in older women, as they may develop diminished ovarian reserve sooner than women with two ovaries [13].
Uterine effects — There is no evidence that chemotherapy directly damages the uterus [14]. On the other hand, exposure of the uterus to radiation can cause uterine damage, including restricted growth and impaired blood flow. Obstetrical consequences include an increased risk of miscarriage, midtrimester pregnancy loss, preterm birth, low birthweight, and placental accreta (see 'Radiation therapy' below).
Methods of fertility preservation are discussed separately. (See "Fertility preservation in patients undergoing gonadotoxic treatment or gonadal resection".)
Men — Spermatogenesis is much more likely to be disrupted than testosterone production because the germinal epithelium of the testis is more sensitive to damage from cytotoxic drugs than are the Sertoli and Leydig cells. The degree of damage to the germinal epithelium is influenced by the stage of sexual maturation of the testis. In general, the postpubertal testis appears to be more susceptible to damage than the prepubertal testis. The magnitude of the effect on sperm production is both drug-specific and dose-dependent. An elevated FSH (>10 U/L) concentration after cytotoxic therapy suggests the presence of azoospermia. (See "Pubertal development and gonadal function in survivors of childhood cancer", section on 'Infertility'.)
Studies of the reproductive implications of cancer treatment do not include baseline fertility data obtained before diagnosis. Therefore, it is not known whether the quality of sperm (count and motility) harvested from cancer patients before treatment is poor because of the disease process itself.
Some types of surgery used to treat men with cancer may affect a man's future fertility [15,16]. Men with testicular cancer may undergo orchiectomy. As long as one healthy testicle remains, the sperm count should be adequate after surgery; however, some men with testicular cancer have poor fertility because the remaining testicle is not truly normal. Men who have undergone surgery for prostate cancer or retroperitoneal lymph node dissection may not be able to produce semen or ejaculate, but sperm can be taken directly from the testicles and used to fertilize the partner's egg in vitro.
PREGNANCY AND NEONATAL OUTCOME
Congenital and chromosomal anomalies — In studies including several thousand offspring, parents who have been treated for childhood cancer with chemotherapy, radiation therapy, or both are not at increased risk of having children with congenital anomalies, single gene disorders, or chromosomal syndromes [17-27]. Some representative examples are illustrated below:
In contrast to these generally reassuring studies, two studies observed an increased risk of congenital anomalies in offspring of cancer survivors. One of these involved survivors treated with radiation, the other involved offspring of male cancer survivors.
A cohort study used data on liveborn singletons from nationwide registries in Denmark and Sweden to compare the risk of congenital anomalies in offspring of male cancer survivors with the risk in offspring of fathers with no history of cancer [32]. Offspring of male cancer survivors had a slightly but statistically higher absolute rate of major congenital anomalies than offspring in the control group (RR 1.7, 95% CI 1.05-1.31; absolute rate 3.7 versus 3.2 percent).
One explanation for these findings, which are discordant with almost all previous studies, is that offspring of cancer survivors were examined more closely because of their paternal history, resulting in ascertainment bias [33]. Although data were available for only a small number of children, the rate of congenital anomalies among children of subfertile men was the same for those iatrogenically conceived with sperm cryopreserved before and after cancer treatment.
Pregnancy complications — Pregnancy outcomes that are most often evaluated are rates of miscarriage, low birth weight/preterm delivery, and stillbirth. Virtually all series are retrospective and therefore subject to recall bias. Underreporting is particularly an issue with early miscarriage [34]. A study that merged data from a hospital registry of cancer patients aged 15 to 35 years at diagnosis, and data from a birth registry, observed perinatal mortality, low birth weight, and preterm birth were higher among first births delivered to women after a cancer diagnosis [35]. There is no evidence of an increased risk of adverse outcome among female partners of male survivors of childhood cancer [36-39].
Chemotherapy — The available data do not support an adverse effect of prior chemotherapy on the risk of miscarriage, fetal growth and development, fetal demise, or uterine function [3,23,30,36,38]. The Childhood Cancer Survivor Study compared pregnancy outcome (ie, live birth, miscarriage, birth weight) in five-year female cancer survivors who were less than 21 years old at diagnosis, with outcomes in their sibling controls [3]. The most frequently used agents were cyclophosphamide, doxorubicin, vincristine, dactinomycin, and daunorubicin. Over 1900 females reported 4029 pregnancies (63 percent live births, 1 percent stillbirths, 15 percent miscarriages, 17 percent abortions, and 3 percent unknown or in gestation). There were no significant differences in pregnancy outcome between patients who had received chemotherapy and controls. However, women who had undergone pelvic irradiation were significantly more likely to have infants weighing less than 2500 g at birth (see below). Study limitations included survey design with possible recall bias.
The same investigators also compared pregnancy outcome in the partners of male survivors of childhood cancer to outcome in the partners of their male siblings [36]. Over 1200 male cancer survivors sired 2323 pregnancies (69 percent live births, 1 percent stillbirths, 13 percent miscarriages, 13 percent abortions, 5 percent unknown or in gestation). Although the percent of live births was higher than among female survivors (69 versus 63 percent, see above), it was significantly lower than in the partners of male siblings (69 versus 76 percent). This was due, in part, to a higher rate of pregnancy termination among the partners of male survivors compared to controls. The rate of miscarriage was similar in both cases and controls (13 versus 12.1 percent), as was the rate of stillbirth and birthweight distribution. Interestingly, this study showed a male:female sex ratio of 1.00:1.03, which is lower than that in female survivors and the general population. Possible explanations for this finding are chance and lower testosterone levels from exposure to chemotherapeutic agents [40,41].
Radiation therapy — Pregnancy in women who have received prior pelvic irradiation appears to be associated with complications such as miscarriage, preterm labor and delivery, low birth weight, impaired fetal growth, placenta accreta, and stillbirth [3,21,23,27,30,42-49]. However, studies reporting these findings did not consistently adjust for maternal factors that could contribute to these risks, such as maternal smoking, alcohol use, diabetes, hypertension, preeclampsia, and infection. The partners of men who received testicular radiation do not appear to be at increased risk of adverse pregnancy outcome [27,50].
Hypotheses for these findings include [51]:
The risk of these effects depends on the total radiation dose, site irradiated, and the woman's age at the time of irradiation (the prepubertal uterus is particularly vulnerable) [27,54-56]. These hypotheses have been supported by evidence that radiation therapy results in anatomic changes in the pelvic vasculature that can lead to reduced fetoplacental perfusion:
Recommendations for monitoring — Pregnant women who have received radiation therapy to the pelvis or abdomen are at higher risk for impaired fetal growth, placental attachment disorders, and possibly preterm labor and delivery. Placental attachment can be evaluated with sonography in the late second trimester and third trimesters. Ancillary techniques that may be used if the diagnosis is uncertain include color Doppler ultrasound examination and magnetic resonance imaging. (See "Diagnosis and management of placenta accreta".) Serial sonograms every four to six weeks in the late second and third trimesters are useful for assessing fetal growth. (See "Diagnosis of fetal growth restriction".) Weekly nonstress tests may be useful for women with a history of abdominal radiation before menarche, as they may have an increased risk for stillbirth or neonatal death. There are no reliable modalities for prediction and prevention of preterm labor and delivery. (See "Prevention of spontaneous preterm birth" and "Fetal fibronectin for prediction of preterm labor and delivery".)
Cancer risk in offspring — The offspring of cancer survivors are not at increased risk for cancer [57,58], unless the tumor suffered by the parent was a component of an inherited syndrome, such as retinoblastoma (see below). As an example, one series evaluated the risk of cancer among 5847 offspring of 14,652 survivors of cancer in childhood or adolescence [57]. There were 17 retinoblastomas and 27 neoplasms other than retinoblastoma, of which 22 were considered sporadic cancers. This yielded a standardized incidence ratio of 1.3 (95 percent confidence interval, 0.8 to 2.0), which was not significantly higher than the cancer risk in the general population matched for age. However, in this series there appeared to be a slightly increased risk of cancer in children of survivors who were diagnosed when less than 10 years old.
Another large series identified seven cases of cancer among 2308 offspring (0.30 percent) of 2283 case survivors and 11 cases among 4719 offspring (0.23 percent) of 3604 controls [58]. Again, no excess risk of cancer in offspring of cancer survivors was observed.
Inherited cancer susceptibility — Certain individuals may be at a higher risk of developing cancer by virtue of their family history of an inherited predisposition to develop malignancy. Retinoblastoma, and the various cancers associated with the Li-Fraumeni syndrome are among the best described of these inherited conditions. (See "Li-Fraumeni syndrome" and "Overview of retinoblastoma", section on 'Epidemiology'.)
Other malignancies that are less well recognized as having a familial component (eg, brain tumors and acute leukemia) may also aggregate in the relatives of some families. An underlying genetic predisposition for such tumors is suggested in these families, and pedigree analysis of the survivors' family should be elicited. (See "Risk factors for brain tumors", section on 'Genetic factors' and "Pathogenesis of acute myeloid leukemia", section on 'Familial leukemia'.)
Patients with heritable cancers should be counseled about risk of the disease in their offspring. A genetics counselor can discuss options for prenatal diagnosis or preimplantation genetic diagnosis, if this is desired. (See "Preimplantation genetic diagnosis".)
Cancer recurrence — With the possible exception of gestational trophoblastic disease, pregnancy does not affect the risk of recurrence of any type of cancer, although the diagnosis may be delayed because of the pregnancy. (See "Malignant gestational trophoblastic disease: Staging and treatment".) In particular, recurrence of melanoma [59,60] and breast cancer [61-63] appear to be unaffected by a subsequent pregnancy. (See "Breast cancer during pregnancy and lactation: Epidemiology and diagnosis" and "Tumor node metastasis (TNM) staging system and other prognostic factors in cutaneous melanoma".)
RISK OF COMPROMISED CARDIAC FUNCTION
Radiation induced cardiotoxity — The clinical spectrum of cardiac injury directly resulting from radiation includes acute pericarditis during therapy (rarely associated with treatment of juxta pericardial cancer), delayed pericarditis that can present abruptly or as chronic pericardial effusion or constriction; pancarditis, which includes pericardial and myocardial fibrosis with or without endocardial fibroelastosis; cardiomyopathy in the absence of significant pericardial disease; coronary artery disease, and functional valve injury and conduction defects [64]. (See "Cardiotoxicity of radiation therapy for malignancy".)
Asymptomatic valvular disease, predominantly mitral and aortic insufficiency, typically occurs 11.5 years later and becomes symptomatic 16 years post mediastinal radiation [65]. Radiation induced cardiomyopathy is more common if radiation therapy is given with simultaneous or sequential anthracycline chemotherapy. It is not clear if irradiation causes structural alterations in the valves, but myocardial fibrosis adjacent to valve rings can lead to distortion and functional impairment. Rhythm changes can occur secondary to ischemic fibrosis affecting the conduction system.
Women with a history of prior thoracic radiation therapy should undergo a baseline echocardiogram and electrocardiogram prior to pregnancy to detect subclinical radiation-induced cardiac sequelae. Consultation with a cardiologist is advised if the echocardiogram is abnormal or an arrhythmia is noted. There should be open communication between the patient's oncologist and obstetrician during the pregnancy.
Chemotherapy induced cardiotoxicity — Cancer patients receiving certain chemotherapeutic agents are at increased risk of developing cardiovascular complications even if they have normal hearts, and the risk is greater if there is a known history of heart disease. Among the serious clinical cardiac complications that have been reported are:
The anthracyclines (daunorubicin, doxorubicin, idarubicin, epirubicin, and mitoxantrone) are the class of drugs most commonly associated with long-term cardiovascular complications in oncology patients, although many other agents have the potential for cardiotoxicity (eg, trastuzumab). In general, the risk of anthracycline-related cardiomyopathy is dose dependent, and can occur many years after administration [66-68]. A series of 37 women who received doxorubicin for childhood cancer and subsequently had 72 pregnancies described generally favorable pregnancy outcomes with sustained myocardial function; however, myocardial function deteriorated in the eight women with a lower than normal left ventricular ejection fraction before pregnancy (fractional shortening <30 percent) [66]. In a multivariate analysis, female sex and a higher cumulative dose of doxorubicin were associated with depressed contractility, and there was an interaction between these two variables [69].
Both prior doxorubicin treatment and pregnancy stress cardiac reserve, thus highlighting the importance of careful follow-up of cardiac function during pregnancy and postpartum in patients previously treated with doxorubicin. In patients with a history of exposure to cardiotoxic drugs, cardiac evaluation should be performed prior to pregnancy and may include electrocardiogram, echocardiogram, radionucleotide angiocardiogram, and 24 hour Holter electrocardiogram monitoring. (See "Cardiotoxicity of anthracycline-like chemotherapy agents".)
Evaluation of women with acquired heart disease (including cardiomyopathy), pregnancy risk assessment, and treatment of heart failure in pregnant women is discussed separately. (See "Acquired heart disease and pregnancy", section on 'Assessing risk' and "Acquired heart disease and pregnancy", section on 'Cardiomyopathy' and "Management of heart failure in pregnancy".)
SUMMARY AND RECOMMENDATIONS
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.