Chemotherapy and the Ovary

Chemotherapy induces ovarian damage in a drug- and dose-dependent manner, and is related to age at the time of treatment, with progressively smaller doses required to produce ovarian failure with increasing age [1, 2].

The stockpile of primordial follicles found in the cortex of the ovaries represents the ovarian reserve. Histological studies of human ovaries have shown chemotherapy to cause ovarian atrophy and global loss of primordial follicles [3, 4]. A reduced follicular reserve may result in premature ovarian failure (POF) and menopause many years post-treatment, even in patients undergoing chemotherapy at a very young age [5]. The odds ratio of POF is as high as 3.8 in childhood cancer survivors of Hodgkin’s disease and 3.2 for non-Hodgkin’s lymphoma [6]. In addition, women who do not undergo POF and retain ovarian functionality after chemotherapy may still have reduced fertility. Studies have shown decreased anti-Müllerian hormone (AMH) and antral follicle counts and poor response to assisted fertility procedures in cancer survivors compared to similar age controls [7–10]. However, these studies do not provide any information on the mechanism of injury. The effect of chemotherapy on the ovary is not an “all or nothing” phenomenon, and the number of surviving primordial follicles following exposure to chemotherapy correlates inversely with the dose of chemotherapy and the nature of the agent [11].

Among the most ovotoxic of the chemotherapy drug groups are the alkylating agents (cyclophosphamide, busulphan, and dacarbazine), platinum complexes (cisplatin, carboplatin), and taxanes (paclitaxel) [12] (Table 2.1). Alkylating agents and platinum complexes work in a similar fashion, creating DNA cross-links in replicating cells, which in turn cause DNA breaks, ultimately triggering apoptosis. The taxanes are microtubule-stabilizing agents, as distinct from DNA-damaging drugs, but it has been demonstrated that paclitaxel acts via Bax to induce apoptosis [13, 14]. Another drug which has been widely studied is the anthracyclin antibiotic doxorubicin (DXR), an intercalating agent that blocks DNA replication and causes double-stranded DNA breaks [15, 16]. However, while DXR has been shown to affect the follicle reserve experimentally, it is currently considered of low clinical risk for premature ovarian insufficiency (POI) [12].

Table 2.1 Risk of permanent amenorrhea in women

High risk	Stem cell transplantation External beam irradiation to fields including the ovaries Breast cancer adjuvant combination chemotherapy regimens containing cyclophosphamide, methotrexate, fluorouracil, doxorubicin, and epirubicin in women >40 years
Intermediate risk	Breast cancer adjuvant chemotherapy regimens containing cyclophosphamide in women 30–39 years, or doxorubicin/cyclophosphamide in women >40 years
Low risk (<20%)	Combination chemotherapy regimens for NHL, ALL or AML Breast cancer adjuvant chemotherapy regimens containing cyclophosphamide in women <30 years or doxorubicin/cyclophosphamide in women <40 years
Very low risk or no risk	Vincristine, methotrexate, fluorouracil
Unknown risk	Paclitaxel, taxotere, oxaliplatin, irinotecan, trastuzumab, bevacizumab, cetuximab, erlotinib, imatinib

ALL: acute lymphoblastic leukemia; AML: acute myeloid leukemia; NHL: non-Hodgkin’s lymphoma.

Risk assessment is based on amenorrhea rate. Because some therapies compromise the follicular reserve without causing amenorrhea, fertility may be compromised before the cessation of menses.

Adapted from Lee et al. J Clin Oncol. 2006; 24:2917–2931.

While apoptosis is a major mechanism of action of many chemotherapy agents, targeting preferentially dividing/proliferating cells, studies have demonstrated that the most ovotoxic chemotherapy agents do not induce apoptosis in the dormant primordial follicles [17, 18]. Granulosa cells of growing follicles are highly vulnerable to apoptotic death following chemotherapy [19, 20]. This loss of larger growing follicles causes the abrupt decline in serum AMH during chemotherapy [21, 22] and the temporary cessation of menstruation for a few months after chemotherapy.

Chemotherapy has also been shown to cause follicle loss by triggering activation and “burnout” of the dormant follicles [18, 23–26]. This accelerated follicle activation appears to be caused by dysregulation in pathways that control follicle dormancy, including upregulation of the PI3K-Akt signaling pathway and the death of growing follicles, which play a vital role in maintaining primordial follicle suppression via the excretion of suppression factors, primarily AMH [27] (Figure 2.1). The PI3K/PTEN/Akt signaling pathway is well established as a regulator of follicle quiescence, and numerous knockout mouse models [28] as well as in vitro studies on human cortical tissue [29–31] have shown that upregulation of this pathway results in activation and loss of the ovarian reserve. Anti-Müllerian hormone is a highly specific negative regulator of follicle activation produced by granulosa cells of small growing follicles [32], and studies on AMH knockout mice and in vitro culture of ovarian tissue demonstrate that loss of AMH leads to excessive activation and depletion of the primordial follicle pool [33].

Figure 2.1 Alkylating agent, cyclophosphamide (Cy), induces follicle loss by increasing follicle activation. (Adapted from Kalich-Philosoph, 2013)(A) During normal follicular development, the ovary is in a state of equilibrium. The primordial follicles (PMFs) are under balanced regulation by the PI3K/PTEN/Akt pathway, and suppressive factors produced by growing follicles (AMH) ensure that the vast majority of PMFs are maintained in a state of dormancy and very few are activated into growth. (B) Exposure to Cy disturbs the balance, upregulating the PI3K/PTEN/Akt follicle activation pathway and causing growing follicles to undergo apoptosis, thereby reducing secretion of AMH. As a result, more PMFs are recruited into growth, develop, and die, causing the reservoir to “burn out”

Injury to blood vessels, stroma, and focal fibrosis of the ovarian cortex are further patterns of ovarian damage caused by chemotherapy [34–36], which indirectly cause primordial follicle loss [35, 37–39] by making the local environment inhospitable.

The risk of ovarian failure in several commonly encountered malignancies and other disorders requiring chemo- and/or radiotherapy is presented in Table 2.2. Alkylating agent cyclophosphamide is widely used in combination chemotherapy regimens, and high-dose cyclophosphamide (200 mg/kg) is frequently utilized as conditioning therapy before bone marrow transplantation (BMT) [40]. Alkylating agent dose is related to subsequent fertility in childhood cancer survivors [41]. Treatment of Hodgkin’s lymphoma with mechlorethamine, vincristine, procarbazine, and prednisolone (MOPP) or chlorambucil, vinblastine, procarbazine, and prednisolone (ChlVPP) is associated with ovarian dysfunction in 19–63% of cases [42]. Amenorrhea is more commonly observed in older women, but long-term follow-up is necessary, as a number of young women also develop premature menopause. The BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisolone) regimen results in amenorrhea in approximately 20% of women overall, but this rises to 67% in women treated with 8 cycles of dose-escalated BEACOPP [43], with, as discussed earlier, age being an important factor: Amenorrhea was reported by 95% of women aged over 30 years compared to 51% in younger women.

Table 2.2 Ovarian failure rates (adapted and updated from ref [70]) Breast cancer

Reference	Age		Ovarian failure (%)
Lower, 1999	Premenopausal		45
	<35 years		28
Bines, 1996	Premenopausal		68
Meirow, 1999	<44 years		50
Goodwin, 1999	CMF	43.7 +/−5.2	65
Burstein, 2000	CMF	<30	19
		30–39	30–40
	CAF	<30	0
		30–39	10–25
	AC	<30	–
		30–39	13
Jonat, 2001	Premenopausal		60
Petrek, 2006	<35		15
	35–39		39–55
	>39		> 55

CMF: cyclophosphamide, methotrexate, fluorouracil

CAF: cyclophosphamide, doxorubicin, fluorouracil

AC: doxorubicin, cyclophosphamide

Hodgkin’s lymphoma

Reference	Treatment		Ovarian failure (%)
Howell & Shalet, 1998	Aggressive treatment		38–57
Meirow, 1999	Relapse post 1st treatment		32
Bokemeyer, 1994	Infradiaphragmatic Rx		50
Brusamolino, 2000	Ovarian-sparing protocol	<25 years <45 years	0
Behringer, 2005	Dose-escalated BEACOPP	>30 years <30	95 51
Decanter, 2010	ABVD		0
Kiserud, 2007	Low dose		55
	Advanced		22–27

Bone marrow transplantation

Reference	No. of patients	Age	Ovarian failure (%)
Sanders, 1996	73	38 (mean)	99
Teinturier, 1998	21	2–17	72
Thibaud, 1998	31	3.2–17	80
Meirow, 1999	63	29 (mean)	79
Grigg, 2000	19	30 (mean)	100

Systemic lupus erythematosus

Reference	Treatment		Ovarian failure (%)
Mok, 1998	70 patients		26
Boumpas, 1993	39 patients, pulse therapy (0.5–1.0 g/m2)	<7 pulses	12
		>15 pulses	39
Blumenfeld, 1996	SLE nephritis (POF)	>30	13
		20–30	15
		<20	100
Erlangen cohort	338 patients		14
Manger, 2006	63 patients	<30	39
		30–40	59

Treatment with an ABVD (adriamycin, bleomycin, vinblastine, and dacarbazine) regimen, which contains no alkylating agents or procarbazine, results in significantly less gonadotoxicity, especially in patients under 25 years of age [44]. A cohort study [45] of 518 female five-year survivors of Hodgkin’s lymphoma aged 14 to 40 years (median age: 25 years) at treatment explored the incidence of POF before age 40. Alkylating agents, especially procarbazine (HR: 8.1) and cyclophosphamide (HR: 3.5), showed the strongest associations. Ten years after treatment, the actuarial risk of premature menopause was 64% after high cumulative doses (> 8.4 g/m²) and 15% after low doses (≤4.2 g/m²) of procarbazine (29). A small study demonstrated normal fecundity in women treated with ABVD [46]. Fertility and ovarian reserve parameters related to the recent replacement of procarbazine with dacarbazine-containing protocol are not yet available. The risk of POF in Hodgkin’s lymphoma and breast cancer is summarized in Table 2.2.

In cases of germ cell tumors, fertility-sparing surgery is possible in a large proportion of patients. For patients with advanced-stage disease, maximum cytoreductive surgery appears to be beneficial. For patients who require postoperative chemotherapy, standard therapy involves a combination of bleomycin, etoposide, and cisplatin. Although POF may occur in a small proportion of patients, 80–99% of those who undergo fertility-sparing surgery and chemotherapy can expect to maintain reproductive function [47]. In a group of young women (median age: 25.5) who were treated with the VAC (vincristine, actinomycin, cyclophosphamide) protocol for germ cell tumors, 13% were found to have irregular menses, 15% oligomenorrhea or amenorrhea, and 8% persistent amenorrhea, although 11 of 16 who had attempted to become pregnant had been successful [48].

Chemotherapy is also used in patients with nonmalignant diseases such as systemic lupus erythematosus (SLE). Pulse cyclophosphamide therapy is frequently used for active lupus nephritis or neuropsychiatric lupus. The major determinants for the development of ovarian failure in patients with SLE are age at the start of therapy and the cumulative cyclophosphamide dose (number of cycles and doses) [49]. Women with SLE and related diseases provide additional challenges for fertility preservation because of the uncertain course of their disease and thus the poor predictability of the total dose of cyclophosphamide that will be required.