Amenorrhea after Lung Cancer Treatment

Abstract

Objective

More than 5,000 premenopausal women are diagnosed with lung cancer annually in the United States. Limited data exist regarding the risk of treatment-related amenorrhea, a surrogate for infertility and early menopause, after systemic therapies for lung cancer.

Methods

Premenopausal women diagnosed with lung cancer under age 50 were surveyed at diagnosis and annually thereafter about their menstrual status as a part of the Mayo Clinic Epidemiology and Genetics of Lung Cancer Research Program. Types of lung cancer-directed treatments were recorded, and frequencies of self-reported menopause at each survey were calculated.

Results

A cohort of 182 premenopausal women were included in this study, with average age at lung cancer diagnosis 43 years (SD 6). Among the 85 patients who received chemotherapy, 64% self-reported that they had become menopausal within a year of diagnosis. Platinum salts were universally included in these chemotherapy regimens, and the majority of these women also received taxanes within one year of diagnosis. Only 15% of the 94 patients who did not receive systemic therapy within one year of diagnosis experienced self-reported menopause. Three patients received targeted therapy alone, two of whom remained premenopausal at the final qualifying survey, completed a median of 3 years after diagnosis.

Conclusions

Chemotherapy for lung cancer patients appears to increase risk of early loss of menses in survivors.

Introduction:

Approximately 415,000 Americans are lung cancer survivors, with more than 220,000 new cases diagnosed annually in the United States (1, 2). While the rate of new lung cancer diagnoses in men has decreased by 32% since 1975, it has risen 94% for women, and lung cancer has now surpassed breast cancer as the leading cause of cancer death among women in this country (1, 2). While lung cancer is more common in older adults, women are diagnosed at younger ages compared to men, and approximately 5,000 premenopausal women are diagnosed with lung cancer annually in the United States (1, 2). Thus, it is important to study the challenges facing the young female lung cancer patient population.

Unique to the premenopausal survivor population is the concern that chemotherapy may diminish or eradicate the ovarian reserve. Chemotherapy may cause acute amenorrhea by killing growing follicles, and also may cause ovarian failure, i.e. menopause, by destroying remaining quiescent follicles (3–5). Alkylating agents are especially notorious for their gonadotoxic effects. They are thought to destroy quiescent ovarian follicles by triggering follicle activation and apoptosis concurrently, and result in rates of premature menopause between 40–80% among women with breast cancer (5–11). Women who remain amenorrheic two years after chemotherapy do not frequently regain menstrual function (12).

Permanent loss of ovarian function results in infertility, estrogen deprivation symptoms (e.g., hot flashes and vaginal dryness), and bone loss (13). These symptoms may cause significant distress and impaired quality of life in lung cancer survivors. Furthermore, it may influence treatment decisions: in one study of 620 female breast cancer survivors under age 40, over half were concerned about infertility after treatment, and 26% reported that fertility concerns had affected their treatment decisions (4). Thus, understanding the risk for premature menopause after lung cancer-directed therapy is important to inform pre-therapy counseling. Amenorrhea, an absence of normal menses, is frequency measured, as it may help to identify patients at risk for infertility and menopausal symptoms. However, rates of amenorrhea after lung cancer treatment regimens are understudied; a literature search of amenorrhea and lung cancer yielded no results. Extrapolations of risks for amenorrhea must be based on studies of the individual components of the chemotherapy regimens utilized to treat cancers in other populations.

Platinum salts are the frequent backbone of treatment for both non-small-cell and small cell lung cancer. They cross-link DNA with a mechanism of action similar to that of alkylating agents, which suggests they may cause similar, detrimental ovarian function effects (6). This is supported by a recent rat study that reports they are as gonadotoxic as cyclophosphamide (14). However, when given to very young women with germ cell tumors (median age 24.3 years), were not associated with high rates of long-term amenorrhea (15). The effect of platinums on older premenopausal human ovaries has not been well studied. The other components of the “platinum doublet” include topoisomerase inhibitors and taxanes. Etoposide is a frequent cause of amenorrhea in children with Hodgkin’s lymphoma and in young adults with AML (8, 16, 17). It is difficult to ascertain if taxanes contribute to amenorrhea as they are frequently administered in combination with alkylating agents and anthracyclines (18), but one study of paclitaxel and trastuzumab (a nongonadotoxic monoclonal antibody) in women diagnosed with breast cancer at a median age of 44 found a relatively low rate of amenorrhea (28% at 4 years), suggesting that paclitaxel is not highly gonadotoxic (19). In addition, another study that assessed amenorrhea after anthracycline-based chemotherapy for breast cancer did not find that adding a taxane to doxorubicin-cyclophosphamide substantially increased the rate of amenorrhea (20).

While the rates, implications, and risk factors for chemotherapy-induced amenorrhea have been widely studied in survivors of a variety of other cancers (particularly breast cancer and lymphoma), there is a paucity of information regarding amenorrhea in patients with lung cancer. As it has been shown that older premenopausal cancer survivors with a history of tobacco use are more likely to experience premature menopause; as lung cancer is traditionally a disease affecting older tobacco users, it could be posited that risks for amenorrhea may be higher in lung cancer survivors than in the aforementioned populations (20–22). The objective of this study was to evaluate rates of chemotherapy-induced amenorrhea among lung cancer patients.

Methods:

The Mayo Clinic Epidemiology and Genetics of Lung Cancer Research Program is an ongoing clinic-based registry of lung cancer patients (23–25). Since 1997, patients identified with newly diagnosed primary lung cancer and treated at Mayo Clinic in Rochester, Minnesota have been invited to participate in a longitudinal cohort study. Participants are actively followed by mailed questionnaires, beginning 6 months after lung cancer diagnoses, and then annually. Participants included in the study analysis were under the age of 50 and premenopausal at the time of their initial lung cancer diagnosis. Data collection was limited to surveys completed between 1999–2016, with at least one survey completed 1–20 years after the cancer diagnosis. The study was reviewed and approved by the Mayo Clinic institutional review board.

Participants were asked about the types of cancer treatment they underwent, including names of chemotherapies and/or targeted therapies and the dates of treatment. In the survey, patients were asked to identify whether they considered themselves premenopausal, peri-menopausal, or postmenopausal at the time of survey completion. Women who self-reported they were postmenopausal in surveys given after the time of diagnosis were considered to have amenorrhea. “Immediate” treatment-related menopause was defined as reporting age of menopause as the same as the age at diagnosis of cancer or 1 year older. For instance, a woman would be considered to have “immediate” treatment-related menopause if she were 40 years old and premenopausal at the time of diagnosis, and then completed a survey 12 months after diagnosis at age 41 in which she reported she was postmenopausal. Statistical analyses were completed with percentages and averages.

Results:

Out of the 2735 female patients who completed at least one survey between 1999–2016, 182 were under age 50 and reported that they were premenopausal at diagnosis. All surveys that were completed in the cohort were utilized in analysis: some women completed only one survey, and some completed multiple surveys over time. Median time from diagnosis to first survey was 20.4 months (SD 23 months, range 12–202 months). At the time of data analysis in January 2017, there were 101 participants that had provided more than one qualifying questionnaire; the first and last questionnaires were an average of 3.7 years apart (range: 0.4 to 14 years). The average time between diagnosis and last questionnaire was 4.0 years (range 1–17 years), with the last questionnaire having been received between November 1999 and November 2016. There were 81 participants who provided only one qualifying survey, which was obtained an average of 2.6 years after first diagnosis (range 1–17 years).

Table 1 provides participant’s characteristics. Of these 182 participants, 85 (median age at diagnosis 44 years, SD 5, range 34–48) received chemotherapy during the year after diagnosis, 26 of whom also received targeted therapy during that year. Ninety-four women did not receive systemic therapy within a year of diagnosis. Table 2 describes the types and frequencies of chemotherapy agents received, among the 85 women who received chemotherapy. As patients self-reported cancer-directed therapies, occasionally they were not clear on the therapies received, so such treatments were reported as “unknown.” There were a number of different targeted therapy agents utilized by participants, including EGFR, VEGF, CD20, ALK, and HER2 inhibitors.

Table 1.

Participant Characteristics.

Characteristic	All participants, N=182
Age	Mean: 42.7 ± 5.7
	Median: 44.0
	Range: 20–49
Years from diagnosis to first survey	Mean: 2.2 ± 1.9
	Median: 1.7
	Range: 1.0–16.8
Cigarette Smoking
Never	48.4%
Former	24.1%
Current	27.5%
Chemotherapy within 1 year of diagnosis
Yes	46.7%
No	16.5%
Unknown	36.8%
Targeted therapy within 1 year of diagnosis
Yes	15.9%
No	9.9%
Unknown	74.2%

Table 2.

Chemotherapy agents administered within one year of diagnosis.

Chemotherapy Agents	Number of Chemotherapy Recipients (N=85)
Platinum (cisplatin, carboplatin)	85 (100%)
Taxane (paclitaxel, docetaxel)	54 (64%)
Antimetabolite (pemetrexed, gemcitabine)	27 (32%)
Topoisomerase inhibitor (etoposide, topotecan)	25 (29%)
Other (epothilone, vinorelbine)	10 (12%)

Among women who received chemotherapy, those who were older at diagnosis reported becoming postmenopausal after fewer years than women who women who were younger at diagnosis. The menopausal status of women who received chemotherapy is reported in Table 3. Of the women who reached menopause within one year of diagnosis, 44% were never-smokers; of the women who remained pre- or peri-menopausal at their final survey, 85% were never-smokers.

Table 3.

Menopausal status of patients who received chemotherapy.

	Percentage who reached menopause		Percentage who did not reach menopause
Years after diagnosis	Within 1 year	2+ years	Time of final survey Mean: 3 years± 2 [1–10]
Percentage who reached menopause	46%	9%	45%
Mean age at diagnosis	47±2 [41–49]	43±5 [35–48]	40±5 [25–48]

Three patients in the study cohort received targeted therapy alone, two of whom remained premenopausal at the final qualifying survey at a mean age at diagnosis of 40. The third, diagnosed at age 43, became post-menopausal more than two years after her cancer diagnosis. All of these patients had a current or former history of tobacco use.

The remaining 94 women (mean age at diagnosis of 42 years, SD 6 years, range 20–49) received no systemic therapy within a year of diagnosis. The menopausal status of women who did not receive chemotherapy or targeted therapy is shown in Table 4. Of the women who reached menopause within one year of diagnosis, 71% were never smokers; of the women who remained pre- or peri-menopausal at their final survey, 51% were never-smokers. Due to the differing ages at diagnosis, times to survey completion, variable smoking histories among patients, and sample size, further analyses were unable to be completed.

Table 4.

Menopausal status of patients who did not receive chemotherapy and/or targeted therapy.

	Percentage who reached menopause		Percentage who did not reach menopause
Years after diagnosis	Within 1 year	2+ years	Time of final survey Mean: 4 years± 3 [1–10]
Percentage who reached menopause	15%	16%	69%
Mean age at diagnosis	45±3 [41–49]	43±4 [36–49]	41±7 [20–49]

In order to describe changes in reported menstrual status longitudinally, menstrual status was followed over time for the group of participants who returned multiple qualifying surveys. Of the 54 participants who returned multiple surveys, 31 participants (mean age at diagnosis of 45 years, SD 3, range 39–49) reported becoming postmenopausal at a mean age of 46 years (SD 2, range 41–50). Twelve participants (five who received chemotherapy and seven who had no systemic therapy) reported that they were pre- or peri-menopausal at least once after reporting that they became postmenopausal. Four of these 12 continued to report premenopausal status until their final qualifying survey; five reported peri-menopausal status; and three temporarily reported pre- or peri-menopausal status, but then by the time of the final qualifying survey, again reported postmenopausal status.

Discussion:

While young women comprise an important cohort of lung cancer survivors, their risk of premature amenorrhea has not been previously investigated. Our results suggest that self-reported menopause occurs soon after a lung cancer diagnosis for more than half of all premenopausal women with lung cancer. While our conclusions are limited by the sample size and heterogeneity of the study population, this rate is only slightly lower than the rates of amenorrhea seen in similarly aged breast cancer survivors who received cyclophosphamide-containing regimens (8–11). While prior studies suggest that alkylating agents are responsible for much of the ovarian toxicity experienced by young women with certain cancers, it is clear from these findings and others’ that non-alkylating agents can also damage ovarian function (14, 16, 17, 26). It is notable that etoposide and platinum agents have been linked to amenorrhea in prior studies (6, 8, 14, 16, 17, 27).

It will be important to build upon our findings to more thoroughly study the unique consequences of gonadal toxicity in survivors of lung cancer. Unlike in breast cancer, premature menopause is unlikely to improve prognosis in lung cancer, but preservation of fertility and subsequent pregnancies may be complicated by other sequelae of lung cancer treatment, such as chronic dyspnea.

While this study is the first to comment on amenorrhea rates among a premenopausal lung cancer population, it is limited by heterogeneity of reported cancer-directed therapies, number of surveys completed by each participant, and the timing of completion of initial qualifying surveys. This was largely due to the survey from which data was extracted; this survey was initially created to define menopausal status among participants in a lung cancer tumor registry, rather than to evaluate the frequency of chemotherapy-induced amenorrhea. Unfortunately, the resulting heterogeneity precluded further statistical analyses, including correlative studies. Small sample size was also a limitation of this study; there were too few women under the age of 45 to analyze younger women separately. Thus, conversions from pre- to postmenopausal status may concordantly be explained by variables such as time to follow-up, age, and smoking status.

Further studies are needed to carefully assess the gonadotoxic effects of each different systemic agent and regimen utilized to treat lung cancer. Targeted and immunotherapy agents are increasingly important in lung cancer treatment strategies, but even though many of the signaling pathways that are targeted by these new therapies are also present in ovaries (for instance, the epidermal growth factor receptor, EGFR, pathway), the ovarian effects of these therapies are understudied (28). While this study utilized patient self-report of amenorrhea as the primary outcome measure, further studies should correlate self-reported menstrual status with laboratory evaluations. Laboratory measures of ovarian function such as Anti-Mullerian hormone (AMH) are under investigation in survivors of other cancers, and may reflect the degree of chemotherapy-induced gonadal damage in patients with lung cancer as well (29–38).

Conclusions:

Premenopausal lung cancer patients should be educated about the risk for chemotherapy-related amenorrhea and the aforementioned lung cancer-specific amenorrhea considerations prior to therapy initiation. If future fertility is desired, reproductive endocrinology should be consulted to discuss options for embryo and oocyte cryopreservation, the gold standard techniques for fertility preservation in this setting (39, 40). Gonadotropin-releasing hormone agonist administration could also be considered in light of preliminary data for possible efficacy in patients with other cancers (6, 41–46).