follicles which reaches a maximum of around 7 million at 5–6 months’ gestational age, declining at an exponential rate thereafter. On average, the number of follicles remaining at birth is 2 million, falling to 300,000 by menarche. Thereafter, the rate of decline follows a bi‐exponential pattern, with acceleration after the age of 37 years when numbers reach approximately 25,000. As only 400 or so follicles are ultimately released as mature oocytes during the reproductive lifespan, the overwhelming majority of follicles are lost through the process of atresia and apoptosis. Menopause occurs when the number of remaining follicles reaches approximately 1000 [2].
Ionizing radiation can cause direct DNA damage to ovarian follicles, leading to follicular atrophy and decreased ovarian follicular reserve. This can hasten the natural decline of follicular numbers, leading to impaired ovarian hormone production and early menopause. Factors that are determinants of ovarian failure include radiation dosage, age at the time of radiation exposure and extent of radiation field treatment.
The human oocyte is generally extremely sensitive to radiation therapy. A mathematical model suggests that the dose required to destroy 50% of immature oocytes (LD50) is less than 2 Gy [4]. The effective sterilizing dose (ESD) of fractionated RT at which ovarian failure occurs immediately in 97.5% of patients was found to decrease with increasing age at treatment. The estimated ESD at birth was 20.3 Gy; at 10 years, 18.4 Gy; at 20 years, 16.5 Gy; and at 30 years, 14.3 Gy. However, there is wide individual variability in ovarian follicular reserve at time of treatment, which can explain differences in onset of premature ovarian failure between patients at similar ages [5].
A study assessing ovarian function in 100 female cancer survivors treated with chemotherapy and/or RT identified from the Childhood Cancer Registry found 17 patients had premature ovarian failure, required hormone replacement therapy and were found to have follicle depleted or undetectable ovaries [6]. A total of 70 patients with spontaneous menses had reduced ovarian volumes per ovary than controls (4.8 cm3 vs. 6.8 cm3; P <0.001) and fewer antral follicles per ovary (7.5 vs. 11; P <0.001). In addition, follicle number was inversely associated with ovarian irradiation, alkylating chemotherapy, older age at diagnosis and longer follow up. These results demonstrate that survivors with spontaneous menstrual cycles may have diminished ovarian reserve [6].
Use of higher energy photons may reduce side scatter when ovaries are outside the primary field, and external shielding can be considered to reduce external scatter from the linear accelerator. Novel radiation techniques, including IMRT and proton RT, may spare the ovaries from significant radiation and reduce potential adverse effects on fertility [2].
Investigation of ovarian function
The menstrual history and the presence or absence of menopausal symptoms can give a clinical impression regarding ovarian function. Measurement of ovarian volumes and antral follicle count using ultrasound remains a useful technique for evaluating ovarian function. Laboratory tests that are useful include cycle day‐3 follicle stimulating hormone levels, inhibin‐B and anti‐Müllerian hormone levels.
Uterine dysfunction
Pelvic irradiation increases the risk of pregnancy‐related complications, including spontaneous miscarriages, preterm labor and delivery, low birth weight infants and placental abnormalities (placenta accreta and percreta). This is probably brought about by reduced uterine volume, impaired uterine distensibility due to myometrial fibrosis, uterine vasculature damage and endometrial injury. The degree of uterine damage depends on the total radiation dose, site of irradiation and patient age at time of treatment. The prepubertal uterus is more vulnerable than the adult uterus to the effects of pelvic irradiation, with doses of 14–30 Gy likely to cause uterine dysfunction. The reduction in uterine volume is significantly higher in patients with direct uterine radiation. A study reported that following whole abdominal irradiation (20–30 Gy) in childhood, all pregnancies occurring in women with preserved ovarian function resulted in mid‐trimester miscarriage [7]. In a study of 16 pregnancies in 13 whole‐body irradiation recipients (10–15.75 Gy), none of the six patients who had received irradiation prepubertally had a live birth (five spontaneous miscarriages and one elective termination) [8].
Investigation of uterine function
Ultrasound scanning is a reliable noninvasive technique for assessing uterine size and shape, blood supply and endometrial thickness. Uterine artery blood flow may be assessed by Doppler scanning and quantified as the pulsatility index indicating degree of resistance to flow distal to the point of sampling. Endometrial biopsy may be performed with further assessment by histology and with immunohistologic techniques. Uterine distensibility is not easily assessable. A study assessed uterine characteristics in survivors of bone marrow transplant for acute lymphoblastic leukemia who had fractionated total body irradiation [7]. The patients were assessed after 3 months of a sex steroid regimen designed to mimic the normal hormonal milieu, achieving physiologic levels of serum estradiol and progesterone. There was an increase in uterine size, improvement in uterine artery blood flow and normal endometrial thickness. Whether or not pretreating patients before donor oocyte treatment in this manner improves pregnancy outcome remains unknown.
Pregnancy outcomes
Van de Loo et al. [9] assessed the impact of abdominal‐pelvic RT for childhood cancer survivors (CCS) on uterine function and pregnancy outcomes. In this nested‐cohort study, they compared RT‐exposed CCS (n = 55) with age‐ and parity‐matched nonirradiated CCSs (non–RT‐exposed CCSs; n = 110) and general population controls (n = 110). Uterine volume was assessed by three‐dimensional ultrasound, and pregnancy outcome was obtained from a self‐reported questionnaire. Among nulligravidous participants, median (interquartile range) uterine volume was 41.4 (18.6–52.8) mL for RT‐exposed CCSs, 48.1 (35.7–61.8) mL for non–RT‐exposed CCSs, and 61.3 (49.1–75.5) mL for general population controls. RT‐exposed CCSs were at increased risk of a reduced uterine volume (<44.3 mL) compared with population controls (odds ratio [OR] 5.31 [95% confidence interval 1.98–14.23]). Surprisingly, the same was true for non–RT‐exposed CCSs (OR 2.61 [1.16–5.91]). Among gravidous participants, RT‐exposed CCSs had increased risks of pregnancy complications, preterm delivery, and a low birth weight infant compared with population controls (OR 12.70 [2.55–63.40], OR 9.74 [1.49–63.60], and OR 15.66 [1.43–171.35], respectively). Compared with non–RT‐exposed CCSs, RT‐exposed CCSs were at increased risk of delivering a low birth weight infant (OR 6.86 [1.08–43.75]). They concluded that uterine exposure to RT during childhood reduces adult uterine volume and leads to an increased risk of pregnancy complications and adverse pregnancy outcomes. CCS should be counseled preconceptionally about these risks and receive appropriate obstetric monitoring in case they become pregnant. The authors did not specify if the RT exposure was pre‐ or postpuberty.
Prevention
Ovarian transposition (oophoropexy)
Repositioning the ovaries out of the irradiation field can preserve ovarian function and should be considered in women of reproductive age with pelvic malignancies or before pelvic lymph node irradiation. This can be unilateral or bilateral and carried out via laparotomy or laparoscopy. The proper location to fix the transposed ovaries depends on the planned irradiation field. For cervical cancer the ovaries are transposed high and lateral above the pelvic brim, while for pelvic lymph node irradiation (as in Hodgkin’s lymphoma) the ovaries can be medially or laterally transposed. Complications of the procedure are rare but include chronic pelvic pain, vascular injury, fallopian tube infarction and ovarian migration. For patients with external pelvic (45 Gy) irradiation with or without para‐aortic nodal irradiation (45 Gy), the amount of radiation received by the ovaries with lateral transposition (mean distance 14.4cm) was calculated to be 1.26Gy for intracavitary radiation, whereas for external pelvic with or without para‐aortic nodal irradiation the dose was 1.35–1.90 and 2.30–3.10 Gy, respectively [10]. Ovarian function preservation rates with a median follow