Moderate and severe lupus nephritis, alveolar hemorrhage, vasculitis and CNS involvement are the indications for the use of immunosuppressive agents such as mycophenolate, azathioprine, or cyclophosphamide. Among these drugs, cyclophosphamide has the most devastating effects on the ovary. In order to make it clearer and easily understandable to the readers in the fields outside the reproductive medicine, we first will provide some fundamental information about normal ovarian physiology in women.
An adult human ovary harbors two different types of follicles; dormant (quiescent) primordials and the growing follicle fractions. Follicles are the functional apparatus of the ovary and consist of an oocyte surrounded by granulosa and theca cells of somatic origin.
Primordial follicles represent the earliest stage of follicular development. Ovarian reserve is determined by the number of quiescent primordial follicles in the ovary. Dormant primordial follicles constitute >90% of the follicle pool in the human ovary with only a small fraction belonging to the growing follicle pool at the primary stage and beyond [21
]. Primordial follicles do not express FSH receptor; their growth is not under the control of gonadotropins [22
]; do not produce inhibins or anti-mullerian hormone (AMH) and are not visible on ultrasound [23
]. While there is no direct marker of primordial follicle number, ovarian reserve can be estimated by evaluating the hormonally active and visible growing follicle pool using anti-mullerian hormone (AMH), antral follicle counts (AFC) and early follicular FSH levels. Any toxic insult that preferentially targets primordial follicles such as alkylating agents leads to a decrease in reproductive life span and possibly premature ovarian failure. If the loss of ovarian function develops during or shortly after the completion of therapy, it is termed acute ovarian failure (AOF). For those who retain ovarian function after the completion of gonadotoxic chemotherapy a subset will go on to experience menopause prematurely before age 40 [24
]. Chemotherapy agents, particularly those of alkylating category such as cyclophosphamidelophosphamide have the highest gonadotoxic potential. The index drug cyclophosphamidelophosphamide is metabolized to two active metabolites in the body, phosphoramide mustard and acrolein. While acrolein exerts its toxicity on the bladder causing hemorrhagic cystitis, phosphoramide mustard is the main product that is responsible for the follicular damage in the ovary [25
]. It causes ovarian damage by inducing apoptotic death of the oocytes and somatic granulosa cells (Figure 2). The clinical manifestations of ovarian damage in women at reproductive age vary from temporary menstrual irregularity to amenorrhea, infertility and premature ovarian failure depending upon the magnitude of the damage. The probability of developing permanent ovarian failure depends on the following factors: patient’s age, and the type, dose and duration of the treatment. If the patient is older and her ovarian reserve is low, they are less likely to retain or regain menstrual function than younger ones. It should be remembered that menstrual status may not be a reliable indicator of the extent of the impact of chemotherapy since many patients may experience transient menstrual irregularities and amenorrhea during chemotherapy [26
]. A portion of these patients resumed menses in the following months depending upon the age of the patients, and the type and the dose of chemotherapy regimens administered. Patients younger than age 40 are more likely to retain or regain menstrual function than those older than age 40 (22-56% vs. 11%) [26
]. Furthermore patients with critically diminished ovarian reserve and elevated FSH values may continue to menstruate [27
]. Therefore menstrual function is a crude indicator of ovarian reserve. To monitor the changes in ovarian reserve during and after chemotherapy administration non-invasive hormonal markers of ovarian reserve can be a good alternative. FSH, Anti-mullerian hormone (AMH), antral follicle count (AFC) have been utilized for this purpose. Measurement of basal FSH levels on cyclophosphamide day 2 or 3 is one of the most widely used screening tests for the assessment of ovarian reserve. It has been well-established that higher FSH levels (>10 mIU/mL) during early follicular phase are an indicator of decreased ovarian reserve and lower success rates in assisted reproduction. Although no direct correlation has been documented between the number of primordial follicles in the ovary and FSH levels, the basal FSH level will rise as the number of follicles decline in the ovary due to the regulation of FSH secretion at hypothalamus and pituitary through a feedback control of inhibin B and estradiol. Anti-mullerian hormone (AMH), a member of transforming growth factor beta family like inhibins produced by the granulosa cells of growing preantral and small antral follicles has emerged as a reliable marker of ovarian reserve [28
]. Serum AMH levels correlate well with the number of antral follicles in the ovary [29
] and the number of oocytes retrieved in IVF cyclophosphamide [30
]. AMH was more consistently correlated with the clinical degree of follicle pool depletion in young women presenting with elevated FSH levels [31
]. Currently serum AMH levels and the number of antral follicles counted on the ultrasound at early follicular phase of the menstrual cycle are the most reliable indicators of ovarian reserve.
In the recent years a number of studies documented that cyclophosphamide administration is the most significant risk factor for ovarian failure, and that AMH is a sensitive and reliable marker of ovarian reserve and damage post exposure to cyclophosphamide in female patients with SLE [18
]. For instance, Harward et al. reported a case series study of women diagnosed with SLE, vasculitis and scleroderma prior to age 35 (23 had prior cyclophosphamide exposure, 20 did not). In their series, even though women with prior cyclophosphamide exposure were 4 years younger at diganosis than those without cyclophosphamide, 30.4% of them had cessation of menses compared to 0% of those without cyclophosphamide (p<0.05). Of the women with prior cyclophosphamide exposure those with loss of menses were older at study enrollment, older at cyclophosphamide treatment and higher cumulative doses of cyclophosphamide than those who retained menstrual function [18
AMH appear to be a reliable indicator of residual ovarian reserve postexposure to cyclophosphamide based on the accumulating evidence of the aforementioned studies. But the role of this hormone in predicting the probability of subsequent pregnancy is questionable. A very recent matched cohort study consisting of 56 cyclophosphamide exposed and 56 non-exposed SLE patients found that the risk of failure to conceive was associated with cumulative cyclophosphamide dose (p=0.007) and older age (p=0.02), but not with AMH levels [35
As emphasized in the previous section, older patient and higher dose of cyclophosphamide are associated with the higher risk of menstrual dysfunction and ovarian failure. As a striking example, one study compared two different doses of cyclophosphamide (75 vs. 50 mg/body surface) in SLE patients receiving pulse cyclophosphamide therapy. More patients treated with 75 mg/m2
cyclophosphamide had sustained amenorrhea (17.5% vs. 0%, p<0.05) independently associated with treatment duration (p=0.001) and total IV cyclophosphamide dose (p=0.02) [33
]. The results of LUMINA LVIII, a multiethnic US cohort study showed that disease activity and Texan-Hispanic ethnicity and cyclophosphamide induction therapy emerged as predictors of ovarian failure in addition to well documented association of the risk of gonadal failure with the use of cyclophosphamide and older age [36