Cancer is not uncommon and

Ovarian stimulation in cancer patients Hakan Cakmak, M.D. and Mitchell P. Rosen, M.D. Department of Obstetrics, Gynecology, and Reproductive Sciences, Division of Reproductive Endocrinology and Infertility, University of California, San Francisco, California The patients referred for fertility preservation owing to a malignant disease do not represent the typical population of subfertile patients treated in IVF units. Cancer may affect multiple tissues throughout the body and can result in a variety of complications during controlled ovarian stimulation. Determination of the controlled ovarian stimulation protocol and gonadotropin dose for oocyte/embryo cryopreservation requires an individualized assessment. This review highlights the new protocols that are emerging to reduce time constraints and emphasizes management considerations to decrease complications. (Fertil Steril Ò 2013;99:1476 84. Ó2013 by American Society for Reproductive Medicine.) Key Words: Cancer, fertility preservation, ovarian stimulation, random start Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/cakmakh-ovarian-stimulation-cancer-fertility-preservation/ Use your smartphone to scan this QR code and connect to the discussion forum for this article now.* * Download a free QR code scanner by searching for QR scanner in your smartphone s app store or app marketplace. Cancer is not uncommon and no longer considered to be an incurable disease among reproductive-age women. More than 790,000 new female cancer cases were estimated to be diagnosed in 2012 in the United States (1). Approximately 10% of female cancer cases occur under the age of 45 years (2). Over the past three decades, there has been a remarkable improvement in the survival rates owing to progress in diagnosing certain cancers at an earlier stage and improvements in treatment (2). From 2002 to 2012, 83% of women younger than 45 years diagnosed with cancer survived (2). As a consequence of the increase in the number of patients surviving cancer, greater attention has been focused on the delayed effects of cancer treatments on the quality of future life of the survivor (3, 4). The treatment for most of the cancer types in reproductive-age women involves either removal of the reproductive organs or cytotoxic treatment (chemotherapy and/or radiotherapy) that may partially or definitively affect reproductive function (5). The ovary is particularly sensitive to the adverse effects of cancer treatments because of the set number of follicles present in the postnatal ovary (5). Reproductive lifespan is determined by the follicle pool, and therefore, cancer treatments that cause follicular depletion accelerate the onset of menopause (6). The irreversible gonadotoxic effects of some of the chemotherapeutic agents are well documented, particularly for alkylating agents (e.g., cyclophosphamide, busulfan, and ifosfamide), which are common components of chemotherapy for breast cancer, lymphomas, leukemia, and sarcomas (7, 8). Pelvic radiation therapy is also known to cause follicular destruction, and exposure to 5 10 Gy pelvic radiation appears to be toxic to oocytes, resulting in premature ovarian insufficiency in many women (5). The risk of ovarian failure following cancer therapy Received January 29, 2013; revised March 16, 2013; accepted March 18, 2013. H.C. has nothing to disclose. M.P.R. has nothing to disclose. Reprint requests: Mitchell P. Rosen, M.D., Department of Obstetrics, Gynecology, and Reproductive Sciences, 2356 Sutter St., 7th Floor, San Francisco, California 94115 (E-mail: rosenm@obgyn. ucsf.edu). Fertility and Sterility Vol. 99, No. 6, May 2013 0015-0282/$36.00 Copyright 2013 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2013.03.029 appears to be dose related, and the effect depends on age and ovarian reserve at the time of treatment (9). Early loss of ovarian function not only puts the patients at risk for menopause-related complications at a very young age, but is also associated with loss of fertility (8). In addition, women in the United States have been delaying initiation of childbearing to later in life for social and financial reasons. The birth rate for women aged 30 34 years increased from 80.8 births per 1,000 women in 1990 to 96.5 births per 1,000 women in 2011 (10). Similarly, the rate for women aged 35 44 years rose 54% from 1990 to 2011, increasing from 37.2 to 57.5 births per 1,000 women (10). In other words, more women in their 30s to early 40s are attempting to get pregnant for the first time than ever before. Because the incidence of most cancers increases with age and many women wish to conceive using their own oocytes, delayed childbearing results in more female cancer survivors interested in fertility preservation. Multiple strategies have emerged aiming to preserve fertility in women with different types of malignancies. These include embryo and oocyte cryopreservation, cortical and whole ovary cryopreservation, ovarian transplantation, ovarian transposition, and GnRH 1476 VOL. 99 NO. 6 / MAY 2013

Fertility and Sterility agonist protection (11). Currently, embryo and mature oocyte cryopreservation following in vitro fertilization (IVF) are the only techniques endorsed by the American Society of Reproductive Medicine, and the other methods are still considered to be investigational (12, 13). Controlled ovarian stimulation (COS) for embryo or mature oocyte cryopreservation is the most preferred method for fertility preservation in cancer patients, owing to its higher success rates compared with other, more experimental, technologies (12, 13). Therefore, it should be recommended as long as the patient's medical condition does not preclude safely performing COS or oocyte retrieval and the patient has adequate time to undergo COS and oocyte retrieval (12, 13). To facilitate initiation of ovarian stimulation and avoid unnecessary delay, prompt consultation with a reproductive endocrinologist and coordination of care are necessary after the cancer diagnosis (14). The number of oocytes retrieved and their quality are imperative factors predicting the potential efficacy of the fertility preservation procedure. Consequently, information regarding the expected ovarian performance after COS is crucial when consulting with the patient. Therefore, the assessment of ovarian reserve with the use of antral follicle count (AFC) and/or antim ullerian hormone (AMH) before ovarian stimulation is necessary to provide more accurate prediction of ovarian response to COS and to determine the COS protocol and starting gonadotropin dose (15). RESPONSE TO OVARIAN STIMULATION IN CANCER PATIENTS In cancer patients, both the specific malignancy and the patient's multisystemic condition may have an impact on the response to ovarian stimulation (16). The increased catabolic state, malnutrition, and increased stress hormone levels associated with the malignancy may affect the hypothalamic-gonadal axis and decrease fertility (17). Possible adverse association between the presence of a neoplastic process and ovarian reserve or oocyte quality is also suggested (17 19). There are mixed reports about how cancer patients respond to the IVF stimulation protocols: some reporting no significant change (20 22) and others demonstrating worse ovarian response in cancer patients compared with age-matched healthy women (19, 23). In a recent meta-analysis conducted on seven retrospective studies, women with malignancies had lower numbers of total oocytes (11.7 7.5 vs. 13.5 8.4) and mature oocytes retrieved (9.0 6.5 vs. 10.8 6.8) after COS for fertility preservation compared with healthy age-matched patients (16). Moreover, the relative risk of poor response leading to cycle cancellation was higher in cancer patients than in the control group (risk ratio 1.32, 95% confidence interval 0.78 2.17) although the observed difference did not reach statistical significance, possibly due to the small size of the groups (16). BRCA genes play an essential role in double-strand DNA break repair, and their mutations are associated with an increased risk of breast and ovarian cancers (24). In patients with BRCA mutations, oocytes may be more prone to DNA damage, clinically manifesting as diminished ovarian reserve or earlier menopause (25). In BRCA mutation positive breast cancer patients, a low response to ovarian stimulation occurred more frequently than in patients without BRCA mutations (33.3% vs. 3.3%) or in breast cancer patients not tested for their BRCA status (2.9%) (18). Interestingly, all BRCA mutation positive patients with a low response to ovarian stimulation and requiring higher doses of gonadotropins for their stimulation had BRCA-1 mutations, and a low response was not encountered in women who were positive for only a BRCA-2 mutation (18). None of the studies mentioned above compared the ovarian reserve of cancer patients with healthy age-matched women. In a recent study, ovarian reserve assessed with AMH was found to be significantly lower in patients with lymphoma before chemotherapy compared with healthy control subjects (26). Moreover, we previously demonstrated that women with cancer before gonadotoxic therapy may have significantly lower AFC compared with healthy women aged 25 40 years (Table 1) (27). This lower AFC in cancer patients may be explained by either accelerated follicle loss or a defect in recruitment of antral follicles owing to disease state. It is well established that AFC correlates directly with number of follicles, number of mature oocytes retrieved, and number of embryos obtained during an IVF cycle (28). In our clinical experience, although the number of mature oocytes retrieved and embryos obtained may be lower in cancer patients compared with healthy individuals, they are appropriate for their given AFC. Moreover, their response to the gonadotropins and mature oocyte yield (i.e., number of metaphase II [MII] oocytes/afc) are similar to those of the healthy women. Therefore, if lower oocyte and embryo numbers in patients with malignancy during an IVF cycle are true, this is not due to poor response to ovarian stimulation, but likely the result of decreased number of available antral follicles to be stimulated. In conclusion, candidates for fertility preservation because of malignancy, especially BRCA-1 mutation carriers, should be informed that the expected number of oocytes retrieved after COS may be lower compared with healthy patients of similar age. However, more studies are needed to confirm these findings. GONADOTROPIN DOSE DURING OVARIAN STIMULATION Maximizing the number of embryos and oocytes cryopreserved during a fertility preservation cycle is extremely important, not only because the patient usually has a single TABLE 1 Comparison of antral follicle count (AFC) between cancer patients and healthy women in different age groups (26). Age (y) Cancer patients Healthy women n Median Range n Median Range P value 25 30 33 14 1 58 205 20 4 58 <.001 31 35 47 11 0 54 216 15 5 48.004 36 40 49 7 0 40 227 12 0 52 <.001 41 45 20 7 1 20 161 6 1 22.789 Cakmak. Ovarian stimulation in cancer patients. Fertil Steril 2013. VOL. 99 NO. 6 / MAY 2013 1477

VIEWS AND REVIEWS cycle opportunity owing to time constraints, but also to increase the chance of future pregnancies. Using higher doses of gonadotropins can be one of the strategies to increase the embryo and oocyte yield per cycle. In a study comparing a low-dose antagonist IVF protocol (150 IU FSH) and a higher-dose antagonist IVF protocol (>150 UI) in cancer patients, although the number of follicles >17 mm was greater in the higher-dose group, there was no difference in numbers of oocytes (13.3 8.7 vs. 12.3 8.0) or embryos (6.3 4.7 vs. 5.4 3.8) generated between the two groups (29). That study suggests that the use of higher doses of gonadotropins may not necessarily result in higher oocyte/embryo yield consistent with the theory that higher doses of gonadotropins may stimulate the recruitment of chromosomally abnormal or incompetent oocytes (30). However, in patients with decreased ovarian reserve as assessed with the use of AFC and/or AMH, higher doses of gonadotropins may be required. FIGURE 1 OVARIAN STIMULATION PROTOCOLS Conventional Controlled Ovarian Stimulation The choice of the specific COS protocol is generally determined based on the policy of preferences in each IVF center and influenced by the time available until the initiation of radio/chemotherapy. Although multiple different COS protocols are used, the majority of patients are treated with a GnRH antagonist based protocol, which likely allows the shortest deferral of the initiation of radio/chemotherapy. To date, there are no studies comparing agonist and antagonist protocols in women with cancer. Traditional ovarian preparation for IVF requires 9 14 days of ovarian stimulation with exogenous gonadotropins, preceded by ovarian suppression with GnRH agonists for 2 weeks to prevent premature ovulation. Because GnRH agonist is initiated in the luteal phase of the cycle, this may add up to 3 additional weeks to the process, depending on when the patient presents for treatment. The development of GnRH antagonists has significantly decreased the interval from patient presentation to embryo/oocyte cryopreservation (31). In contrast to GnRH agonists, GnRH antagonists immediately suppress pituitary release of FSH and LH and do not require the 10 14 days of administration before gonadotropin initiation. GnRH antagonists are initiated to prevent premature LH surge when the size of the lead follicle reaches 12 14 mm at approximately day 6 of gonadotropin stimulation which begins on day 2 3 of a menstrual cycle (Fig. 1A). This approach still requires awaiting menses before initiating gonadotropins, but it decreases the interval to oocyte retrieval compared to traditional IVF stimulation protocols. The use of GnRH antagonists during the preceding luteal phase was explored originally for cancer patients and then for poor IVF responders as a method to improve ovarian stimulation by inducing corpus luteum breakdown and synchronizing the development of the next wave of follicles (32, 33). For cancer patients, the idea of administering GnRH antagonists in the luteal phase was driven more by minimizing potential delays for cancer treatment (32). If a GnRH Conventional and random-start antagonist IVF protocols for cancer patients undergoing fertility preservation. COS can be started with spontaneous menses (A) or with menses following luteolysis induced by GnRH antagonist (B). COS can also be initiated in the late follicular (C) or luteal phase following spontaneous LH surge (D) or after ovulation induction with hcg or GnRH agonist (E). Cakmak. Ovarian stimulation in cancer patients. Fertil Steril 2013. 1478 VOL. 99 NO. 6 / MAY 2013

Fertility and Sterility antagonist (e.g., single dose of 3 mg cetrorelix subcutaneously) is given during the midluteal phase, menses ensues a few days later (32, 33) (Fig. 1B). As a result, ovarian stimulation would be initiated more quickly and a GnRH antagonist would be restarted in a standard fashion, when the lead follicle is at 12 14 mm to prevent premature LH surge (33). Random-Start Controlled Ovarian Hyperstimulation Conventionally, ovarian stimulation for oocyte/embryo cryopreservation is initiated at the beginning of the follicular phase with the idea that this optimizes clinical outcomes; it may require 2 6 weeks depending on the woman's menstrual cycle phase at the time of planning the treatment. Adhering to this convention may result in either significant delay of cancer treatments or forgoing of fertility preservation owing to time constraints. For cases not desirable to wait for the next menstrual period to start a stimulation protocol owing to the urgency of the cancer treatment, random-start stimulation protocols have been proposed (34 36). In a small prospective multicenter study (n ¼ 40), a novel protocol for cancer patients that initiated ovarian stimulation during the luteal phase of the menstrual cycle was described (36). Cancer patients in the luteal phase were started on GnRH antagonists to down-regulate LH and initiate luteolysis. Simultaneously, follicular stimulation was initiated with recombinant FSH only, thus avoiding exogenous LH activity which might prevent luteolysis. Compared with cancer patients stimulated during the follicular phase (n ¼ 28) with either a short flare-up protocol or an antagonist protocol, the luteal-phase group (n ¼ 12) had similar number of aspirated oocytes, number of MII oocytes, and fertilization rate (36). A report of three breast cancer patients evaluated the effectiveness of initiating ovarian stimulation at the time of patient presentation (menstrual cycle days 11, 14, and 17) rather than waiting for spontaneous menses (35). GnRH antagonist was started to prevent premature LH surge when the lead follicle measured >13 mm. The random-start ovarian stimulation resulted in a reasonable ovarian response, with 7 10 embryos cryopreserved per patient (35). The recent report presenting our clinical experience with random-start ovarian stimulation demonstrated that late follicular or luteal phase start antagonist IVF cycles were as effective as conventional (i.e., early follicular) start antagonist IVF cycles in cancer patients (34). The late follicular phase was defined as after menstrual cycle day 7 with emergence of a dominant follicle (>13 mm) and/or progesterone level <2 ng/ml. If the cancer patient presented in the late follicular phase, we proceeded with one of the following treatment plans. 1) Ovarian stimulation was started without GnRH antagonist if the follicle cohort following the lead follicle was <12 mm and continued to be <12 mm before spontaneous LH surge (Fig. 1C). After the LH surge, GnRH antagonist was started later in the cycle when the secondary follicle cohort reached 12 mm to prevent premature secondary LH surge. Or 2) ovulation was induced with hcg or GnRH agonist and ovarian stimulation was started in 2 3 days in the luteal phase (Fig. 1E) (34). If the cancer patient presented in the luteal phase or the ovulation was induced, GnRH antagonist administration was initiated similarly to conventional ovarian stimulation later in the cycle when the secondary follicle cohort reached 12 mm to prevent premature secondary LH surge (Fig. 1D and E) (34). The numbers of total and mature oocytes retrieved, oocyte yield (i.e., number of MII oocytes/afc), and fertilization rates were similar between groups (34) (Table 2). However, the length of ovarian stimulation was 2 days longer, and therefore, the total dose of gonadotropin used was significantly higher in late follicular and luteal phase start groups compared with the conventional-start group (34) (Table 2). In contrast to earlier belief, the presence of corpus luteum or luteal-phase progesterone levels did not adversely affect the follicular development, oocyte yield, or possibility of having secondary spontaneous LH surge in random-start patients (34). Overall, this approach provides a significant advantage by decreasing total time for the IVF cycle, and in urgent settings, ovarian stimulation can be started at a random cycle date for the purpose of fertility preservation without compromising oocyte yield and maturity. This is consistent with a newer concept of ovarian physiology, which indicates that there are multiple waves of follicle recruitment during TABLE 2 Comparison of characteristics and outcomes of conventional and random start antagonist IVF cycles in cancer patients. Conventional start (n [ 87; 101 cycles) Random start (n [ 24; 24 cycles) P value Age (y) 33.9 5.2 34.6 5.0 NS AFC 13 (9 19) 11.5 (6 16) NS Days of ovarian stimulation 9 (8 10) 11 (10 12) <.001 Total dose of gonadotropins (IU) 3,386 1,085 4,201 1,147.001 Follicles R13 mm 12 (6 17) 10 (8 15.5) NS Oocytes retrieved 15 (9 23) 12.5 (9 20.5) NS Mature oocytes (MII) retrieved 11 (6 16) 9 (5 14.5) NS Oocyte/AFC ratio 1.1 (0.8 1.7) 1.2 (0.9 1.7) NS Mature oocyte/afc ratio 0.8 (0.5 1.1) 0.8 (0.6 1.2) NS Fertilization rate after ICSI (2PN/MII) 0.77 0.22 0.87 0.15 NS Note: Data are presented as mean SD or median (interquartile range) (32). 2PN ¼ two pronuclei; AFC ¼ antral follicle count; ICSI ¼ intracytoplasmic sperm injection; MII ¼ metaphase II; NS ¼ not significant. Cakmak. Ovarian stimulation in cancer patients. Fertil Steril 2013. VOL. 99 NO. 6 / MAY 2013 1479

VIEWS AND REVIEWS each menstrual cycle (37). Additional clinical studies are needed to assess the efficacy of this strategy, especially regarding the rates of clinical pregnancy and of live-born infants originating from the use of cryopreserved embryos and of oocytes obtained by random start ovarian stimulation. Controlled Ovarian Stimulation in Patients with Estrogen-Sensitive Cancers During COS, there is a potential risk that the supraphysiologic E 2 levels resulting from ovarian stimulation with gonadotropins may promote the growth of estrogen-sensitive tumors, such as endometrial and estrogen receptor positive breast cancers (15). The rise in E 2 is directly proportional to the number of follicles recruited to grow; therefore, alternative and potentially safer protocols have been introduced for fertility preservation for estrogen-sensitive cancer patients, including natural-cycle IVF (without ovarian stimulation), stimulation protocols with tamoxifen alone or combined with gonadotropins, and stimulation protocols with aromatase inhibitors to reduce the estrogen production (38). Natural-cycle IVF gives only one or two oocytes or embryos per cycle and has a high rate of cycle cancellation. Therefore, this technique would likely be ineffective and is not recommended, especially when a chemotherapy treatment is imminent and the patient does not have a chance for a second cycle of IVF treatment. Tamoxifen, a nonsteroidal triphenylethylene compound related to clomiphene, has a well known antiestrogenic action on breast tissue with the inhibition of growth of breast tumors by competitive antagonism of estrogen at its receptor site, and it is accepted as the first-line drug in hormonal prevention and treatment of estrogen receptor positive breast cancer (39).Tamoxifen, besides its effect in the breast, also has an antagonist action in the estrogen receptors in the central nervous system similar to that of clomiphene. The selective antagonist action of tamoxifen interferes with the negative feedback of the estrogen on the hypothalamic-pituitary axis, leading to an increase in GnRH secretion from the hypothalamus and a subsequent release of FSH from the pituitary-stimulating follicular development. Tamoxifen can be used for COS alone starting on day 2 5 of the menstrual cycle in doses of 20 60 mg/d, or in combination with gonadotropins, similarly to the use of clomiphene (38). Even though peak E 2 levels in ovarian stimulation with tamoxifen are not altered, owing to its antiestrogenic effect on breast tissue, it is desirable to be used in estrogen receptor positive breast cancer patients. Ovarian stimulation with the use of tamoxifen for fertility preservation in cancer patients was shown to increase the mature oocyte and embryo yield compared with natural-cycle IVF (1.6 vs. 0.7 and 1.6 vs. 0.6, respectively) and reduce cycle cancellations (40). As expected, combined protocol with tamoxifen and gonadotropins further increased the number of cryopreserved oocytes and embryos (5.1 vs. 1.5 and 3.8 vs. 1.3, respectively) (41). Aromatase is a cytochrome P450 enzyme complex that catalyzes the conversion of androstenedione and testosterone to their respective estrogenic products estrone and E 2 (42). Aromatase inhibitors, such as letrozole, markedly suppress plasma estrogen levels by competitively inhibiting the activity of the aromatase enzyme (43). Aromatase inhibitors significantly reduce the risk of recurrence in postmenopausal women with hormone receptor positive breast cancer owing to profound estrogen deprivation, especially with thirdgeneration inhibitors (i.e., anastrozole and letrozole) (44). Centrally, aromatase inhibitors release the hypothalamicpituitary axis from estrogenic negative feedback, increase the secretion of FSH by the pituitary gland, stimulate follicle growth, and, thereby, can be used for ovulation induction (45). In patients with estrogen-sensitive cancers, the main advantage of adding daily letrozole to gonadotropins in ovarian stimulation protocols is to decrease serum E 2 levels to be closer to that observed in natural cycles (i.e., E 2 <500 pg/ml) without affecting oocyte or embryo yield (46, 47). Stimulation protocols using letrozole alongside with gonadotropins are currently preferred over tamoxifen protocols as treatment with letrozole results in a higher number of oocytes obtained and fertilized when compared to tamoxifen protocols (41). In a study comparing the efficacy of the letrozole plus gonadotropin protocol in breast cancer patients and the standard IVF protocol in age-matched noncancer patients with tubal-factor infertility, the breast cancer patients started to receive letrozole (5 mg/d) on menstrual cycle day 2 or 3, FSH (150 300 IU/d) was added 2 days later, all medications were discontinued on the day of hcg trigger, and letrozole was reinitiated after oocyte retrieval and continued until E 2 levels fell to <50 pg/ml (47). This letrozole plus gonadotropin protocol resulted in similar number of total oocytes retrieved and length of ovarian stimulation compared with standard IVF protocol (47). As expected, peak E 2 levels were significantly lower in the breast cancer patients receiving letrozole plus gonadotropin compared with the standard IVF group (483 278.9 pg/ml vs. 1,464.6 644.9 pg/ml) (47). The studies assessing the effect of letrozole on oocyte maturity and competence demonstrated that the addition of letrozole did not change numbers of mature oocytes retrieved and fertilization rates (47, 48). Similarly, in our practice we observed similar oocyte maturity (MII oocytes/total oocytes retrieved; 0.68 0.19 vs. 0.71 0.23) and fertilization rate (0.77 0.22 vs. 0.78 0.24) in intracytoplasmic sperm injection cycles with and without letrozole (unpublished data). The short-term follow-up of breast cancer patients, who had undergone ovarian stimulation with letrozole plus gonadotropins for fertility preservation has not been shown to raise the risk of breast cancer recurrence (49). In addition, COS with aromatase inhibitors in combination with gonadotropins has been safely used for embryo cryopreservation in endometrial cancer patients (50). Letrozole suppresses plasma E 2 levels significantly at doses of 0.1 10 mg/d (51). In our clinic, we start letrozole at 2.5 5 mg/d, depending on the ovarian reserve of the patient, with the ovarian stimulation (Fig. 1). Given the importance of keeping E 2 levels close to that observed in natural cycles in patients with estrogen-sensitive cancers, we check E 2 levels in every clinic visit and titrate letrozole dose up to 10 mg/d to keep E 2 levels <500 pg/ml (46). These letrozole doses are well tolerated by the patients during ovarian stimulation without any side effects. In addition, the mature 1480 VOL. 99 NO. 6 / MAY 2013

Fertility and Sterility oocyte/embryo yield after COS is not affected by letrozole at any dose used in our clinical practice. We also consider continuing letrozole after the oocyte retrieval if serum E 2 levels are still elevated (i.e., E 2 >500 pg/ml). In our experience, even if E 2 levels are >500 pg/ml before retrieval, only a minority of patients requires letrazole after retrieval. Discontinuation of letrazole can either be at menses or with initiation of chemotherapy. In contrast, anastrozole another thirdgeneration aromatase inhibitor failed to adequately suppress E 2 levels during COS, despite gradually increasing the dose of anastrozole to a maximum of 10 mg/d, and therefore we do not recommend its use in fertility preservation cycles (52). In summary, COS with letrozole plus gonadotropins in patients with estrogen-sensitive cancers undergoing fertility preservation is safe, well-tolerated, and yields similar number of oocytes and embryos compared with standard protocols while minimizing the risk of high estrogen exposure and not increasing the recurrence of cancer in the short term. Therefore, we highly recommend the routine use of letrozole during COS for fertility preservation in patients with estrogen-sensitive cancers. Prevention of Ovarian Hyperstimulation Syndrome in Cancer Patients Ovarian hyperstimulation syndrome (OHSS) is the most serious complication of ovarian stimulation and can be associated with intravascular depletion, ascites, liver dysfunction, pulmonary edema, electrolyte imbalance, and thromboembolic events. Although OHSS is often selflimited with spontaneous resolution within a few days, severe disease may require hospitalization and intensive care. Selecting the appropriate ovarian stimulation regimen can be challenging in embryo/oocyte cryopreservation because it is important to balance the risk of OHSS and obtaining sufficient number of oocytes or embryos to maximize the chance of a successful pregnancy in the future. The impact of OHSS can be profound in cancer patients because it may result in delaying or complicating planned life-saving cancer therapy (53). Triggering the final oocyte maturation with hcg carries the well known risk of inducing OHSS (54). GnRH agonist also induces this final oocyte maturation by promoting the release of endogenous gonadotropin stores from the hypophysis as long as the pituitary gonadotropin receptors are not down-regulated and can be used as an alternative to hcg (54). GnRH agonist trigger in GnRH antagonist based protocols dramatically reduces the risk of OHSS, owing to the short half-life of GnRH agonist induced endogenous LH surge (55). Moreover, there was a significantly lower rate of moderate/severe OHSS in the GnRH agonist group compared with the patients receiving hcg trigger (3.7% vs. 21.3%) (56). GnRH agonist trigger is particularly convenient in cancer patients pursuing oocyte or embryo banking, because luteal support is not needed to sustain a pregnancy. In a study comparing GnRH agonist and hcg as the trigger for oocyte maturation in fertility preservation cycles, GnRH agonist trigger resulted in at least similar numbers of mature oocytes and cryopreserved embryos compared with hcg (56). In addition, although hcg potentiates the endogenous production of estrogen during the luteal phase owing to its longer half-life, GnRH agonist induced endogenous LH may result in lower estrogen production, which may be an advantage for patients with estrogensensitive cancers (54). However, in our experience we have observed trigger failures with GnRH agonist trigger at both 1 mg and 4 mg dosing. The likely reason is that GnRH agonist is able to bind to only a portion of the receptors owing to competition with GnRH antagonist, yielding a limited LH surge (57). It is possible that with the increase in dose of GnRH agonist or with hcg supplementation (%1,500 IU) at the time of trigger, there will be fewer failures. Because of the possibility of failure, we do not routinely recommend GnRH agonist trigger for all patients. In our current practice, 4 mg leuprolide acetate is being used only in patients with high risk of OHSS. The number of follicles, more specifically the follicular pattern, in combination with serum E 2 levels predicts OHSS with high sensitivity and specificity (58, 59). However, one caveat is that cotreatment with aromatase inhibitors limits the use of E 2 level to help predict OHSS. In this scenario, it is important to rely on the follicular pattern and the rate of E 2 rise rather than the absolute of serum E 2 levels. If the E 2 levels are rising rapidly while administering letrozole, especially in the presence of a high number of small follicles, the patient should be considered to be be at risk for OHSS and GnRH agonist trigger should be used to lower that risk. In conclusion, we recommend GnRH agonist trigger in GnRH antagonist based fertility preservation cycles only for women who are at risk for OHSS. The trigger must be confirmed the next morning by measuring serum LH level. In the case of a GnRH agonist trigger failure determined by low post-trigger LH (in our clinic, we use a cutoff LH level of <12 miu/ml), hcg (2,500 5,000 IU) trigger can be given on the same day (60). If the patient has relatively low post-trigger LH level, but >12 miu/ml, closer attention should be given to the oocyte yield during oocyte retrieval. If no or an inappropriately low number of oocytes are retrieved after aspirating a couple of mature-size follicles, the oocyte retrieval should be stopped, oocyte maturation should be triggered again by administering hcg (2,500 5,000 IU) owing to the possibility of failing to trigger oocyte maturation with GnRH agonist, and then oocyte retrieval should be attempted again after 34 36 hours (60). Medical Considerations in Cancer Patients Undergoing COS The patients referred for fertility preservation owing to a malignant disease do not represent the typical population of subfertile patients treated in IVF units. Cancer may affect multiple tissues throughout the body and can result in variety of complications during COS. Therefore, the goals during COS in cancer patients are to prevent these serious life-threatening complications with prophylaxis, and to recognize and manage them effectively when they occur. VOL. 99 NO. 6 / MAY 2013 1481

VIEWS AND REVIEWS Cancer patients undergoing COS are at increased risk of thromboembolic events because of a hypercoagulable state induced by their malignancy and supraphysiologic serum E 2 levels (61). Therefore, these patients may require anticoagulation around the time of COS. Currently, there are no guidelines for anticoagulation during COS. However, the safety and efficacy of anticoagulation during COS and after oocyte retrieval have been reported (62). In our practice, we start prophylactic low-molecular-weight heparin with ovarian stimulation in high-risk patients and instruct the patient to take their last dose of medication 24 hours before the oocyte retrieval. Low-molecular-weight heparin is reinitiated 12 hours after the retrieval and can be continued until E 2 returns to its baseline level. The other strategy of preventing thromboembolic events is to use letrozole during COS to keep E 2 levels close to those observed in natural cycles. Letrozole at 2.5 or 5 mg/d can be started with ovarian stimulation, as in patients with estrogen-sensitive malignancies, and can be titrated up to 10 mg/d to keep E 2 levels <500 pg/ml. Letrozole can also be continued after oocyte retrieval for up to a week depending on the E 2 level at the time of ovulation induction. Malignancies with bone marrow infiltration or liver involvement may create a tendency toward bleeding during oocyte retrieval owing to thrombocytopenia, platelet dysfunction, or defective coagulation factor synthesis. Therefore, platelet count and coagulation panel should be tested before COS in patients with hematologic malignancies or with malignancies involving the liver. Platelet or fresh frozen plasma transfusion should be performed before oocyte retrieval to prevent excessive bleeding in these patients as needed. Higher risk of pelvic infection after oocyte retrieval can be a problem especially in cancer patients with neutropenia. Therefore, absolute neutrophil count should be evaluated before COS in cancer patients with possible bone marrow infiltration. In the case of neutropenia, consultation from the patient's oncologist for the use of granulocyte colonystimulating factor to increase the neutrophil count should be obtained, and prophylactic antibiotics should be given before oocyte retrieval to decrease the risk of infection. Some of the cancer-related medical conditions, including respiratory dysfunctions due to tracheal compression, mediastinal mass, or large pleural effusion, and vascular disturbances, as in superior vena cava syndrome, may preclude safe administration of conscious sedation during oocyte retrieval. Anesthesia consultation should be obtained in advance for the patients with these conditions. If safety and difficult intubation in an emergency situation are concerns, the oocyte retrieval should be performed either under general anesthesia with endotracheal intubation or only with local anesthesia. The patients with recent mastectomies for breast cancer may have special needs during COS. Owing to decreased mobility, they may need more assistance during office visits. Intravenous line placements to the upper extremity on the same side of the axillary node dissection should be avoided, owing to concerns of lymphatic system damage and inadequate lymphatic flow. In patients who have had transverse rectus abdominis myocutaneous flap for breast reconstruction after mastectomy, abdominal distention, and therefore OHSS, should be avoided to prevent wound dehiscence. The medication list for all cancer patients should be reviewed before COS. Antiepileptic medications should definitely be continued during COS in patients with brain tumor owing to increased risk of seizures. The use of Imatinib (Gleevec), a specific inhibitor of constitutively activated Bcr-Abl tyrosine kinase used in chronic myelogenous leukemia, should be temporarily stopped during COS owing to its adverse effect on ovarian hormone production and oocyte recovery (63). CONCLUSIONS Given the importance of reproduction for many young patients faced with cancer, counseling regarding fertility preservation is an essential part of comprehensive cancer care. Embryo cryopreservation is the most established method for fertility preservation, and oocyte cryopreservation has gained efficacy and is now offered at many centers. Determination of the COS protocol and gonadotropin dose for oocyte/embryo cryopreservation requires an individualized assessment. Maximizing the number of embryos and oocytes cryopreserved during a fertility preservation cycle without causing OHSS is extremely important, because most patients have only a single cycle opportunity owing to time constraints before starting their oncologic treatment. In urgent settings, random-start ovarian stimulation is emerging as a new technique for the purpose of fertility preservation without compromising oocyte yield and maturity. Letrozole plus gonadotropin protocol is an effective method for safely inducing COS in patients with estrogensensitive cancers undergoing fertility preservation. Although newly developed protocols are efficient in inducing COS and obtaining appropriate number of oocytes/embryos, only a minority of the patients have undergone thawing and embryo transfer, so there are not enough consistent data reported to evaluate the implantation and pregnancy rates of these new protocols. This issue should be appropriately addressed in the future to enable prediction of the number of oocytes/embryos needed to be preserved to offer a realistic chance for later reproduction in these patients. REFERENCES 1. American Cancer Society. Cancer facts and figures 2012. Atlanta: American Cancer Society. Available at: http://www.cancer.org/research/ cancerfactsfigures/cancerfactsfigures/cancer-facts-figures-2012; 2012. Accessed December 16, 2012. 2. Howlader N, Noone AM, Krapcho M, Neyman N, Aminou R, Waldron W, et al. (eds.). SEER cancer statistics review, 1975 2009 (vintage 2009 populations). Bethesda, MD: National Cancer Institute. Available at: http: //seer.cancer.gov/csr/1975_2009_pops09/. Accessed December 16, 2012. 3. Letourneau JM, Ebbel EE, Katz PP, Katz A, Ai WZ, Chien AJ, et al. Pretreatment fertility counseling and fertility preservation improve quality of life in reproductive age women with cancer. Cancer 2012;118:1710 7. 4. Letourneau JM, Melisko ME, Cedars MI, Rosen MP. A changing perspective: improving access to fertility preservation. Nat Rev Clin Oncol 2011;8:56 60. 5. Rodriguez-Wallberg KA, Oktay K. Options on fertility preservation in female cancer patients. Cancer Treat Rev 2012;38:354 61. 1482 VOL. 99 NO. 6 / MAY 2013

Fertility and Sterility 6. Sklar CA, Mertens AC, Mitby P, Whitton J, Stovall M, Kasper C, et al. Premature menopause in survivors of childhood cancer: a report from the childhood cancer survivor study. J Natl Cancer Inst 2006;98:890 6. 7. Maltaris T, Seufert R, Fischl F, Schaffrath M, Pollow K, Koelbl H, et al. The effect of cancer treatment on female fertility and strategies for preserving fertility. Eur J Obstet Gynecol Reprod Biol 2007;130:148 55. 8. Letourneau JM, Ebbel EE, Katz PP, Oktay KH, McCulloch CE, Ai WZ, et al. Acute ovarian failure underestimates age-specific reproductive impairment for young women undergoing chemotherapy for cancer. Cancer 2012; 118:1933 9. 9. Meirow D, Nugent D. The effects of radiotherapy and chemotherapy on female reproduction. Hum Reprod Update 2001;7:535 43. 10. Hamilton BE, Martin JA, Ventura SJ. Births: preliminary data for 2011. In: National vital statistics reports web release, vol. 61, no. 5. Hyattsville, MD: National Center for Health Statistics; 2012. 11. Anderson RA, Wallace WH. Fertility preservation in girls and young women. Clin Endocrinol (Oxf) 2011;75:409 19. 12. Practice Committee of the American Society for Reproductive Medicine. Fertility preservation and reproduction in cancer patients. Fertil Steril 2005;83:1622 8. 13. Practice Committee of the American Society for Reproductive Medicine. Mature oocyte cryopreservation: a guideline. Fertil Steril 2013; 99:37 43. 14. Lee S, Ozkavukcu S, Heytens E, Moy F, Oktay K. Value of early referral to fertility preservation in young women with breast cancer. J Clin Oncol 2010;28:4683 6. 15. Reddy J, Oktay K. Ovarian stimulation and fertility preservation with the use of aromatase inhibitors in women with breast cancer. Fertil Steril 2012;98: 1363 9. 16. Friedler S, Koc O, Gidoni Y, Raziel A, Ron-El R. Ovarian response to stimulation for fertility preservation in women with malignant disease: a systematic review and meta-analysis. Fertil Steril 2012;97:125 33. 17. Agarwal A, Said TM. Implications of systemic malignancies on human fertility. Reprod Biomed Online 2004;9:673 9. 18. Oktay K, Kim JY, Barad D, Babayev SN. Association of BRCA1 mutations with occult primary ovarian insufficiency: a possible explanation for the link between infertility and breast/ovarian cancer risks. J Clin Oncol 2010;28: 240 4. 19. Pal L, Leykin L, Schifren JL, Isaacson KB, Chang YC, Nikruil N, et al. Malignancy may adversely influence the quality and behaviour of oocytes. Hum Reprod 1998;13:1837 40. 20. Das M, Shehata F, Moria A, Holzer H, Son WY, Tulandi T. Ovarian reserve, response to gonadotropins, and oocyte maturity in women with malignancy. Fertil Steril 2011;96:122 5. 21. Knopman JM, Noyes N, Talebian S, Krey LC, Grifo JA, Licciardi F. Women with cancer undergoing ART for fertility preservation: a cohort study of their response to exogenous gonadotropins. Fertil Steril 2009;91:1476 8. 22. Robertson AD, Missmer SA, Ginsburg ES. Embryo yield after in vitro fertilization in women undergoing embryo banking for fertility preservation before chemotherapy. Fertil Steril 2011;95:588 91. 23. Klock SC, Zhang JX, Kazer RR. Fertility preservation for female cancer patients: early clinical experience. Fertil Steril 2010;94:149 55. 24. Ford D, Easton DF, Peto J. Estimates of the gene frequency of BRCA1 and its contribution to breast and ovarian cancer incidence. Am J Hum Genet 1995; 57:1457 62. 25. Lin WT, Beattie M, Chen L, Oktay K, Crawford SL, Gold EB, et al. Comparison of age at natural menopause in BRCA1/2 mutation carriers with a non clinic-based sample of women in northern California. Cancer 2013 Jan 29; http://dx.doi.org/10.1002/cncr.27952 [Epub ahead of print]. 26. Lawrenz B, Fehm T, von Wolff M, Soekler M, Huebner S, Henes J, et al. Reduced pretreatment ovarian reserve in premenopausal female patients with Hodgkin lymphoma or non-hodgkin-lymphoma evaluation by using antim ullerian hormone and retrieved oocytes. Fertil Steril 2012;98: 141 4. 27. Ebbel E, Katz A, Kao CN, Cedars MI, Rosen MP. Reproductive aged women with cancer have a lower antral follicle count than expected. Fertil Steril 2011;96:S199 200. 28. Frattarelli JL, Levi AJ, Miller BT, Segars JH. A prospective assessment of the predictive value of basal antral follicles in in vitro fertilization cycles. Fertil Steril 2003;80:350 5. 29. Lee S, Oktay K. Does higher starting dose of FSH stimulation with letrozole improve fertility preservation outcomes in women with breast cancer? Fertil Steril 2012;98:961 64.e1. 30. Baart EB, Martini E, Eijkemans MJ, Van Opstal D, Beckers NG, Verhoeff A, et al. Milder ovarian stimulation for in-vitro fertilization reduces aneuploidy in the human preimplantation embryo: a randomized controlled trial. Hum Reprod 2007;22:980 8. 31. McLaren JF, Bates GW. Fertility preservation in women of reproductive age with cancer. Am J Obstet Gynecol 2012;207:455 62. 32. Anderson RA, Kinniburgh D, Baird DT. Preliminary experience of the use of a gonadotrophin-releasing hormone antagonist in ovulation induction/invitro fertilization prior to cancer treatment. Hum Reprod 1999;14:2665 8. 33. Humaidan P, Bungum L, Bungum M, Hald F, Agerholm I, Blaabjerg J, et al. Reproductive outcome using a GnRH antagonist (cetrorelix) for luteolysis and follicular synchronization in poor responder IVF/ICSI patients treated with a flexible GnRH antagonist protocol. Reprod Biomed Online 2005; 11:679 84. 34. Cakmak H, Zamah AM, Katz A, Cedars MI, Rosen MP. Effective method for emergency fertility preservation: random-start controlled ovarian hyperstimulation. Fertil Steril 2012;98:S170. 35. Sonmezer M, Turkcuoglu I, Coskun U, Oktay K. Random-start controlled ovarian hyperstimulation for emergency fertility preservation in letrozole cycles. Fertil Steril 2011;95:2125.e9 11. 36. von Wolff M, Thaler CJ, Frambach T, Zeeb C, Lawrenz B, Popovici RM, et al. Ovarian stimulation to cryopreserve fertilized oocytes in cancer patients can be started in the luteal phase. Fertil Steril 2009;92:1360 5. 37. Baerwald AR, Adams GP, Pierson RA. Characterization of ovarian follicular wave dynamics in women. Biol Reprod 2003;69:1023 31. 38. Rodriguez-Wallberg KA, Oktay K. Fertility preservation in women with breast cancer. Clin Obstet Gynecol 2010;53:753 62. 39. Early Breast Cancer Trialists' Collaborative Group. Systemic treatment of early breast cancer by hormonal, cytotoxic, or immune therapy. 133 randomised trials involving 31,000 recurrences and 24,000 deaths among 75,000 women. Lancet 1992;339:71 85. 40. Oktay K, Buyuk E, Davis O, Yermakova I, Veeck L, Rosenwaks Z. Fertility preservation in breast cancer patients: IVF and embryo cryopreservation after ovarian stimulation with tamoxifen. Hum Reprod 2003;18:90 5. 41. Oktay K, Buyuk E, Libertella N, Akar M, Rosenwaks Z. Fertility preservation in breast cancer patients: a prospective controlled comparison of ovarian stimulation with tamoxifen and letrozole for embryo cryopreservation. J Clin Oncol 2005;23:4347 53. 42. Cole PA, Robinson CH. Mechanism and inhibition of cytochrome P-450 aromatase. J Med Chem 1990;33:2933 42. 43. Smith IE, Dowsett M. Aromatase inhibitors in breast cancer. N Engl J Med 2003;348:2431 42. 44. Winer EP, Hudis C, Burstein HJ, Chlebowski RT, Ingle JN, Edge SB, et al. American Society of Clinical Oncology technology assessment on the use of aromatase inhibitors as adjuvant therapy for women with hormone receptor-positive breast cancer: status report 2002. J Clin Oncol 2002;20:3317 27. 45. Mitwally MF, Casper RF. Use of an aromatase inhibitor for induction of ovulation in patients with an inadequate response to clomiphene citrate. Fertil Steril 2001;75:305 9. 46. Testart J, Frydman R, Nahoul K, Grenier J, Feinstein MC, Roger M, et al. Steroids and gonadotropins during the last pre-ovulatory phase of the menstrual cycle. Time relationships between plasma hormones levels and luteinizing hormone surge onset. J Steroid Biochem 1982;17:675 82. 47. Oktay K, Hourvitz A, Sahin G, Oktem O, Safro B, Cil A, et al. Letrozole reduces estrogen and gonadotropin exposure in women with breast cancer undergoing ovarian stimulation before chemotherapy. J Clin Endocrinol Metab 2006;91:3885 90. 48. Johnson LN, Dillon KE, Sammel MD, Efymow BL, Mainigi MA, Dokras A, et al. Response to ovarian stimulation inpatientsfacinggonadotoxic therapy. Reprod Biomed Online 2013 Jan 19; http://dx.doi.org/10.1016/j.rbmo.2013.01.003 [Epub ahead of print]. VOL. 99 NO. 6 / MAY 2013 1483

VIEWS AND REVIEWS 49. Azim AA, Costantini-Ferrando M, Oktay K. Safety of fertility preservation by ovarian stimulation with letrozole and gonadotropins in patients with breast cancer: a prospective controlled study. J Clin Oncol 2008;26:2630 5. 50. Azim A, Oktay K. Letrozole for ovulation induction and fertility preservation by embryo cryopreservation in young women with endometrial carcinoma. Fertil Steril 2007;88:657 64. 51. Klein KO, Demers LM, Santner SJ, Baron J, Cutler GB Jr, Santen RJ. Use of ultrasensitive recombinant cell bioassay to measure estrogen levels in women with breast cancer receiving the aromatase inhibitor, letrozole. J Clin Endocrinol Metab 1995;80:2658 60. 52. Azim AA, Costantini-Ferrando M, Lostritto K, Oktay K. Relative potencies of anastrozole and letrozole to suppress estradiol in breast cancer patients undergoing ovarian stimulation before in vitro fertilization. J Clin Endocrinol Metab 2007;92:2197 200. 53. Delvigne A, Rozenberg S. Epidemiology and prevention of ovarian hyperstimulation syndrome (OHSS): a review. Hum Reprod Update 2002;8:559 77. 54. Humaidan P, Kol S, Papanikolaou EG. GnRH agonist for triggering of final oocyte maturation: time for a change of practice? Hum Reprod Update 2011;17:510 24. 55. Engmann L, DiLuigi A, Schmidt D, Nulsen J, Maier D, Benadiva C. The use of gonadotropin-releasing hormone (GnRH) agonist to induce oocyte maturation after cotreatment with GnRH antagonist in high-risk patients undergoing in vitro fertilization prevents the risk of ovarian hyperstimulation syndrome: a prospective randomized controlled study. Fertil Steril 2008;89: 84 91. 56. Oktay K, Turkcuoglu I, Rodriguez-Wallberg KA. GnRH agonist trigger for women with breast cancer undergoing fertility preservation by aromatase inhibitor/fsh stimulation. Reprod Biomed Online 2010;20:783 8. 57. McArdle CA, Franklin J, Green L, Hislop JN. Signalling, cycling and desensitisation of gonadotrophin-releasing hormone receptors. J Endocrinol 2002;173:1 11. 58. Blankstein J, Shalev J, Saadon T, Kukia EE, Rabinovici J, Pariente C, et al. Ovarian hyperstimulation syndrome: prediction by number and size of preovulatory ovarian follicles. Fertil Steril 1987;47:597 602. 59. Papanikolaou EG, Pozzobon C, Kolibianakis EM, Camus M, Tournaye H, Fatemi HM, et al. Incidence and prediction of ovarian hyperstimulation syndrome in women undergoing gonadotropin-releasing hormone antagonist in vitro fertilization cycles. Fertil Steril 2006;85:112 20. 60. Cakmak H, Fujimoto VY, Zamah AM, Rosen MP, Tran ND, Cedars MI, et al. Metaphase II (MII) oocytes obtained at different time points in the same in vitro fertilization cycle. J Assist Reprod Genet 2012;29:1203 5. 61. Aurousseau MH, Samama MM, Belhassen A, Herve F, Hugues JN. Risk of thromboembolism in relation to an in-vitro fertilization programme: three case reports. Hum Reprod 1995;10:94 7. 62. Yinon Y, Pauzner R, Dulitzky M, Elizur SE, Dor J, Shulman A. Safety of IVF under anticoagulant therapy in patients at risk for thrombo-embolic events. Reprod Biomed Online 2006;12:354 8. 63. Zamah AM, Mauro MJ, Druker BJ, Oktay K, Egorin MJ, Cedars MI, et al. Will imatinib compromise reproductive capacity? Oncologist 2011;16: 1422 7. 1484 VOL. 99 NO. 6 / MAY 2013