Analysis of Mexican young women with primary ovarian insufficiency induced by gynaecological and haematological cancer management

Abstract

This was a retrospective study that included 114 women younger than 40 years with induced primary ovarian insufficiency. Patients who presented vasomotor symptoms had a higher proportion (26 [63.41%] versus 58 [79.45%], OR 2.23, 95% CI 0.95–5.23, p = .065) to initiate hormone replacement therapy. Vasomotor symptoms were present in patients with ovarian cancer (OR 0.27, 95% CI 0.09–0.8, p = .18), haematologic cancer (OR 0.11, 95% CI 0.2–0.65, p = .014), radiotherapy (OR 2.62, 95% CI 1.04–6.54, p = .039) and chemotherapy with radiotherapy (OR 2.72, 95% CI 1.01–7.35, p = .049). Having ovarian or haematological cancer, being managed with radiotherapy and/or chemotherapy, and having follicle-stimulating hormone parameters higher than 35 mUI/mL are factors that significantly increase the risk of presenting vasomotor symptoms.

Impact Statement

• What is already known on this subject? In young women with cancer, induced primary ovarian insufficiency can result as an ovarian surgery or as an adverse effect of chemotherapy or radiotherapy. Regardless of aetiology, patients are going to manifest early climacteric symptoms with an increased risk for cardiovascular disease, metabolic syndrome and osteoporosis.

• What do the results of this study add? Patients who presented vasomotor symptoms had initially a higher proportion of hormone replacement therapy. Patients that were treated exclusively with radiotherapy or with chemotherapy and concomitant radiotherapy have a significantly increased risk to manifest vasomotor symptoms.

• What are the implications of these findings for clinical practice and/or future research? Having ovarian or haematological cancer, being managed with radiotherapy and/or chemotherapy and having follicle-stimulating hormone parameters higher than 35 mUI/mL are factors that significantly increase the risk of presenting vasomotor symptoms.

Keywords:

• Gynaecological oncology
• adult oncology
• adult radiotherapy
• adult surgery
• adverse events and chemotherapy

Introduction

The incidence of cancer among women younger than 50 years old has increased in the past 20 years, being an increasingly more common disease in premenopausal patients (Islami et al. 2021). Evidence has shown that gynaecological and haematological cancers are leading causes of cancer deaths in young women (Islami et al. 2021). According to the latest Histopathological Registry of Malignancies, these types of cancer are very prevalent in Mexican women, being cervical cancer the most frequent gynaecological cancer in the Mexican population with an incidence that peaks among women aged 30–44 years (Isla-Ortiz et al. 2020). This group of patients has special needs that must be addressed under multidisciplinary guidelines to avoid the risk of developing a primary ovarian insufficiency and ultimately affect their quality of life in the short and long term (Luisi et al. 2015; Vermeulen et al. 2017; Li et al. 2020).

Primary ovarian insufficiency is defined as the early cessation of menstruation accompanied by levels of follicle-stimulating hormone (FSH) 20–40 mIU/mL presented before the age of 40 (Ghahremani-Nasab et al. 2020). Secondary or induced primary ovarian insufficiency can result from distinct mechanisms: as a direct consequence of ovarian surgery (therapeutic or prophylactic bilateral salpingo-oophorectomy), or as an adverse effect of chemotherapy or radiotherapy (Donnez et al. 2010; Chapman et al. 2015). Regardless of aetiology, patients are going to manifest early climacteric symptoms with an increased risk for cardiovascular disease, metabolic syndrome and osteoporosis (Van Dijk et al. 2015; Faubion et al. 2015). In fact, the most frequent symptom reported by patients are vasomotor symptoms, followed by insomnia, vaginal dryness and dyspareunia (Van Dijk et al. 2015). These considerations should be properly addressed during follow-up to identify patients that can benefit from hormone replacement therapy to improve their quality of life. Studies have shown that a history of gynaecological cancer is not an absolute contraindication to start hormone replacement therapy in young patients (Fenton 2015; Hamoda 2017; Angioli et al. 2018; Edey et al. 2018; Deli et al. 2020; Lee et al. 2020). Unfortunately, data are scarce regarding the clinical and therapeutic characteristics of Mexican young women with cancer and induced primary ovarian insufficiency. However, the Instituto Nacional de Cancerología (INCan) in Mexico City has incorporated a primary ovarian insufficiency clinic for young women to provide supportive care to follow-up patients with gynaecological and haematological cancer. The aim of this study is to explore and analyse the clinical and therapeutic characteristics of primary ovarian insufficiency patients with gynaecological and haematological cancer.

Materials and methods

This retrospective and descriptive study included women aged 40 and younger with induced primary ovarian insufficiency due to gynaecological cancer or haematological cancer treatment in follow up at the primary ovarian insufficiency clinic for young women of the INCan in Mexico City during 2016–2018. Demographic, clinical and therapeutic variables were collected. The exclusion criteria are patients older than 40 years, breast cancer, incomplete files and those with disease recurrence at the time of inclusion.

The INCan’s electronic patient registry was used for data extraction. Cardiovascular risk was assessed using the Framingham electronic calculator (ASCVD n.d.). Oncological treatment was administered depending on the neoplasm and treating oncologist. In general, for radiotherapy, 45–50 Gray were administered to the pelvis (uterus, adnexa, and depending on the clinical stage, pelvic and para-aortic nodes); and in the case of chemotherapy, 4–6 cycles of platinum-based schemes were administered for gynecological neoplasms, and schemes based on alkylating agents in hematological neoplasms.

Descriptive statistics were used for analysis, using measures of central tendency and dispersion according to data distribution. Normality was assessed using the Shapiro–Wilk test. To assess differences, Student’s t test was used for parametric distributions and Mann–Whitney U test for non-parametric distributions; Fisher's exact test and chi-square test were used when appropriate. Univariate and multivariate binomial regression models were performed to assess the factors associated with the appearance of climacteric symptoms and the initiation of hormone replacement therapy. All analyses were two-tailed, and a value of p ≤.05 was set for significance. The statistical software STATA 14.1 (Company, College Station, Texas, USA) was used for calculations.

Results

A total of 114 patients with induced primary ovarian insufficiency were included. Median age was 32.5 years (interquartile range [IQR] 27 − 36 years). Only two patients (1.8%) had hypertension and none were diabetic (Table 1). Eighty (70%) patients had never smoked, 16 (14%) were active smokers, and 18 (16%) were former smokers. Regarding cancer type, seven (6.1%) had haematologic cancer and 107 (93.9%) had gynaecologic cancer (39% cervical cancer, 39% ovarian cancer, and 12.3% endometrial cancer). As regards to the cancer treatment employed, 71 (62.3%) had surgery, 49 (43%) received pelvic radiation therapy, 40 (35%) received concomitant chemoradiotherapy, and 33 (29%) had only chemotherapy (Table 2).

Table 1. Clinical and medical characteristics of patients with primary ovarian insufficiency. (n = 114).

Age^a	32.5 (27–36)
Hypertension^b	2 (1.8)
Smoking^b
Negative	80 (70)
Positive	16 (14)
Former smoker	18 (16)
Type of cancer^b
Cervical cancer	45 (39)
Ovarian cancer	45 (39)
Endometrial cancer	14 (12.3)
Haematological cancer	7 (6.1)
Other	3 (2.6)
Received treatment^b
Surgery	71 (62.3)
Radiotherapy	49 (43)
Chemotherapy and radiotherapy	40 (35)
Chemotherapy	33 (29)
Systolic blood pressure (mmHg)^a	112 (101–123)
Diastolic blood pressure (mmHg)^a	70 (58– 82)
Total cholesterol (mg/dL)^c	175.9 ± 32.8
High density lipoprotein cholesterol (mg/dL)^a	45 (29–61)
Triglycerides (mg/dL)^a	132 (32– 232)
Glucose (mg/dL)^a	93 (83–103)
Follicular-stimulating hormone (mUI/mL)^c	68.4 ± 37.5
Bone densitometry (T-score)^a
Column	–1 (–1.6 – −0.2)
Right hip	0 (–1.1 − 0.7)
Left hip	–0.1 (–1 − 0.8)
Cardiovascular risk^a (16)	0.10 (0–0.225)
Vasomotor symptoms	84% (74)
Insomnia	40% (35)
Vaginal dryness	45% (39)
Diminished libido	18% (16)
Restarted a sexually active lifestyle	47% (60)
Initiated an exercise program	19% (17)
Takes supplementary calcium	92% (72)
HTR
Yes	73% (64)
No	41% (36)
Type of HRT
Combined	39% (53.4)
Only oestrogen	34% (46.6)

HRT: hormone replacement therapy

^a Median (interquartile range).

^b Absolute frequency (percentage).

^c Mean (standard deviation). ∞ Six patients with no clinical evidence of primary ovarian insufficiency FOI and 2 that were on hormone implants.

Table 2. Risks associated with the initiation of HRT (n = 114).

Variable	No HRT^b 41 (36%)	HRT^b 73 (64%)	Odds ratio	Confidence interval 95%	p Value
Age^a	31 (26–36)	34 (28–36)	1.02	0.95–1.10	.44
Smoking^b
Negative	28 (68.29)	52 (71.23)	1	Reference	Reference
Positive	6 (14.63)	10 (13.70)	0.89	0.29–2.7	.84
Former smoker	7 (17.07)	11 (15.07)	0.84	0.29–2.42	.75
Type of cancer^b
Cervical cancer	15 (36.59)	30 (41.1)	1	Reference	Reference
Ovarian cancer	14 (34.15)	31 (42.47)	1.1	0.45–2.68	2.82
Endometrial cancer	6 (14.63)	8 (10.96)	0.66	0.19–2.27	.51
Haematological cancer	5 (12.2)	2 (2.74)	0.2	0.03–1.15	.072
Other	1 (2.44)	2 (2.74)	1.0	0.08 − 11.93	.99
Received treatment^b
Surgery	24 (58.54)	47 (64.38)	1	Reference	Reference
Radiotherapy	17 (41.46)	32 (43.84)	1.10	0.51–2.39	.81
Chemotherapy and radiotherapy	12 (29.27)	28 (38.36)	1.51	0.66–3.42	.33
Chemotherapy	15 (36.59)	18 (24.66)	0.56	0.25–1.29	.18
FSH^c	57.57 ± 39.23	73.75 ± 35.80	1.01	0.99–1.02	.077
Bone densitometry^a
Lumbar	–0.8 (–1.6 – −0.5)	–1.1 (–1.6 – −1.1)	1.07	0.75–1.53	.68
Right hip	–0.25 (–1– 0.5)	0 (–1.1 − 0.7)	1.15	0.76–1.73	.50
Left hip	–0.25 (–1.1– 0.7)	0 (–0.9 − 0.9)	1.22	0.82–1.80	.32
Vasomotor symptoms^b	26 (63.41)	58 (79.45)	2.23	0.95–5.23	.065
Insomnia^b	17 (41.46)	3 (31.51)	0.64	0.29–1.43	.28
Vaginal dryness^b	19 (46.34)	26 (35.62)	0.64	0.29–1.39	.26
Diminished libido^b	9 (22.5)	9 (12.33)	0.48	0.17–1.34	.16
Exercise^b	4 (9.76)	15 (20.57)	2.39	0.73–7.76	.15

HRT: hormone replacement therapy; FSH: follicular-stimulating hormone.

^aMedian (interquartile range).

^bAbsolute frequency (percentage).

^c Mean (standard deviation).

FSH concentration levels were assessed at baseline during patients’ first medical appointment at the clinic, having a mean of 68.4 mUI/mL (standard deviation [SD] 37.5). Other baseline reported data were median systolic and diastolic blood pressure measures of 112 mmHg (IQR 101–123) and 70 mmHg (IQR 58–82), respectively. As for laboratory parameters, median high-density lipoprotein cholesterol was 45 mg/dL (IQR 29–61), triglycerides 132 mg/dL (IQR 32–232), glucose 93 mg/dL (IQR 83–103), and total cholesterol 175.9 mg/dL (±32.8). The cardiovascular risk assessment was performed in 66 patients that met the criteria and a median of 0.1 (IQR 0–0.225) was reported. The median results of the bone mineral density test for spine, right hip, and left hip were −1 (IQR −1.6 to −0.2), 0 (IQR −1.1 to 0.7), and −0.1 (IQR −1 to −0.8), respectively (Table 1).

Regarding the question of symptoms, 84 (74%) patients had vasomotor symptoms, 40 (35%) insomnia, 45 (39%) vaginal dryness and 18 (16%) decreased libido. Most patients (60%) stated having restarted a sexually active lifestyle, 19 (17%) having initiated an exercise program, and 92 (72%) being on supplementary calcium intake. Seventy-three (64%) patients reported having initiated hormone replacement therapy, 39 (53.4%) were on combined therapy (oestrogen/progestin) and 34 (46.6%) on oestrogens. On the contrary, 41 (36%) were not on hormone replacement therapy because they did not wish to initiate therapy (24.4%), had absolute contraindications (29.3%) cause history of drug allergy or deep vein thrombosis, did not attend the scheduled medical appointment (47.4%), no clinical indication (42.1%), not completed cancer treatment (10.5%), and other reasons (46.3%).

Initially, a univariate analysis was done to establish the factors associated with the onset of hormone replacement therapy finding that patients with haematologic cancer received in a lower proportion (5 [12.2%] versus 2 [2.74%]; odd ratio [OR] 0.2, confidence interval [CI] 95% from 0.03 to 1.15, p = .072). Likewise, patients that received hormone replacement therapy had higher serum FSH levels 73.75 [SD 35.80] IU/dL; OR 1.01, 95% CI 0.99–1.02, p = .077) versus (57.57 [SD 39.23] IU/mL and presented vasomotor symptoms in a higher proportion 58 (79.45%) patients versus (26 (63.41%) patients; OR 2.23, 95% CI 0.95–5.23, p = .065) (Table 2). A multivariate analysis was performed but no independent associations were found with the initiation of hormone replacement therapy.

Given that a high association was found between the presence of vasomotor symptoms with the onset of hormone replacement therapy, a univariate analysis was performed to determine the factors that associate with the presence of these symptoms. According to this analysis, vasomotor symptoms were affected patients with ovarian cancer (OR 0.27, 95% CI 0.09–0.8, p = .18) and haematologic cancer (OR 0.11, 95% CI 0.2–0.65, p = .014) in a lower proportion compared to patients with cervical cancer. According to the same analysis, the type of treatment also contributed to the appearance of vasomotor symptoms. Patients that received radiotherapy (OR 2.62, 95% CI 1.04–6.54, p = .039) and concomitant chemoradiotherapy (OR 2.72, 95% CI 1.01–7.35, p = .049) had a high association with the presence of vasomotor symptoms. Finally, elevated serum FSH levels were also associated with the presence of vasomotor symptoms (OR 1.02, 95% CI 1.00–1.03, p = .04), specifically when FSH levels were greater than 35 mIU/dL (OR 4.00, 95% CI 1.19 − 13.42, p = .025) (Table 3). A multivariate analysis of possible risk factors associated with having vasomotor symptoms was performed, but no independent factors were found.

Table 3. Risks associated with the presence of vasomotor symptoms (n = 114).

Variable	No vasomotor symptoms^b n = 30 (26)	With vasomotor symptoms^b n = 84 (74)	Odds ratio	Confidence interval 95%	p Value
Age^a	30.5 (26–36)	33.5 (27.5–36)	1.03	0.96–1.12	.35
Smoking^b
Negative	22 (77.33)	58 (69.05)	Reference	Reference	–
Positive	5 (16.67)	11 (13.10)	0.83	0.26–2.67	.76
Former smoker	3 (10)	15 (17.86)	1.89	0.49–7.19	.35
Type of cancer^b
Cervical cancer	6 (20)	39 (46.43)	Reference	Reference	–
Ovarian cancer	16 (53.3)	29 (34.52)	0.27	0.09–0.8	.018
Endometrial cancer	4 (13.33)	10 (11.9)	0.38	0.09–1.62	.19
Haematological cancer	4 (13.33)	3 (3.57)	0.11	0.2–0.65	.014
Other	0	3 (3.57)	NA	NA	NA
Received treatment^b
Surgery	20 (66.67)	51 (60.71)	0.77	0.32–1.86	.56
Radiotherapy	8 (26.67)	41 (48.81)	2.62	1.04–6.54	.039
Chemotherapy and radiotherapy	6 (20)	34 (40.48)	2.72	1.01–7.35	.049
Chemotherapy	13 (43.33)	20 (23.81)	0.41	0.16–0.98	.046
FSH (mUI/dL)^a	53.32 (9.69–74.99)	70.7 (50.5–95.9)	1.02	1.00–1.03	.04
>35 mUI/mL^b	13 (65)	52 (88.14)	4	1.19–13.42	.025
>60 mUI/mL^b	9 (45)	36 (61.02)	1.91	0.68–5.32	.21

FSH: follicular-stimulating hormone.

^aMedian (interquartile range).

^bAbsolute frequency (percentage).

^c Mean (standard deviation).

Discussion

Women with primary ovarian insufficiency have an increased risk of cardiovascular disease and premature death as a result of endothelial dysfunction, abnormal lipid profile and metabolic syndrome along with other illnesses like osteoporosis, dementia and parkinsonism (O’Donnell et al. 2016). Therefore, it is relevant to give them periodical evaluations and reinforce them to change or improve their habits. Additionally, attention should be given to the psychosocial sphere of these patients because they revealed infertility issues, sexual dysfunction and psychological distress at an early age. A way to improve their quality of life and increase their life expectancy is by considering hormone replacement therapy as an early intervention (Fenton 2015; O’Donnell et al. 2016; Hamoda 2017; Angioli et al. 2018; Edey et al. 2018; Deli et al. 2020; Lee et al. 2020).

According to our study, patients with vasomotor symptoms tend to initiate hormone ovarian therapy (OR 2.23, CI 0.95–5.23, p = .065). However, it is striking that this result was not significant given the fact that the main indication for hormone replacement therapy in women with primary ovarian insufficiency is to lessen climacteric symptoms, especially the vasomotor ones, which affected most of our patients (74%) (Fenton 2015; Hamoda 2017). This discordance could be explained by the fact that a relatively high proportion of patients (36%) did not start hormone replacement therapy. Even though a great emphasis is made on the benefits of hormone replacement therapy over its risks during the medical appointments in our clinic, 24% of symptomatic patients did not desire to receive treatment. Diverse studies have not found if initiating hormone replacement therapy increases recurrence risks or affects survival of patients with gynaecological cancer; but have shown their benefits over symptom control. As a matter of fact, there are several studies support initiating hormone replacement therapy in patients with past medical history of gynaecological cancer (except breast cancer) to alleviate symptoms secondary to primary ovarian insufficiency (Edey et al. 2018).

In addition, we found that the appearance of vasomotor symptoms associates with the type of treatment received. Patients that were treated exclusively with radiotherapy or with chemotherapy and concomitant radiotherapy have a significantly increased risk to present vasomotor symptoms. Radiation seems to have more toxic effects on the ovaries compared to chemotherapy. Studies have shown that a 2 Gray dose of radiation to the ovaries damages a high proportion of oocytes, and a higher dose of 6 Greys can cause irreversible hypogonadism. These effects increase with increasing age (Wallace et al. 2003; Bradley and Petereit 2006).

Furthermore, this report shows that the appearance of vasomotor symptoms associates with the type of cancer the patient has. According to our results, patients with a history of haematological or ovarian cancer are significantly less likely to present vasomotor symptoms (OR 0.11, 95% CI 0.2–0.65, and OR 0.27, CI 0.09–0.8, respectively). This could be partially explained by the fact that haematological cancer treatment consists mainly of chemotherapy (Meirow 2000). Not knowing the specific type of haematological cancer in these patients limits our analysis greatly, but we could argue that most malignancies of this type are treated with chemotherapy. This premise does not apply to patients with ovarian cancer, whose type of treatment depends on clinical stage, which is not reported in this study. On the contrary, according to the results of this univariate analysis, receiving only chemotherapy as treatment decreases the risk of presenting vasomotor symptoms in a significant manner (OR 0.41, 95% CI 0.16–0.98, p = .046).

Also, the presence of vasomotor symptoms significantly increases with increasing serum FSH levels (OR 1.02 for each mUI/dL of FSH, 1.00–1.03 p = .04). This supports and highlights the definition of primary ovarian insufficiency, which states that it affects patients younger than 40 years with serum FSH levels greater than 20–40mIU/mL (NICE 2021)

Patients of this cohort had good metabolic and cardiovascular health since they did not have evidence of dyslipidaemia, hypertension, hyperglycaemia or osteoporosis, but 30% had past medical smoking history (Table 1). As previously mentioned, it is important to define strategies to prevent cardiovascular disease and osteoporosis in this group of patients. According to the NICE clinical management guidelines, patients with primary ovarian insufficiency, regardless of the aetiology, should receive hormone replacement therapy until the age when "natural" menopause occurs in the general population (NICE 2021) (Langrish et al. 2009). The intention of this is to stop the development of osteopenia and osteoporosis, atherosclerotic plaques, a prothrombotic state, chronic articular pain and climacteric symptoms (Ibeanu et al. 2011; Honigberg et al. 2019). For every delayed year of menopause appearance, the cardiovascular mortality risk is decreased by 2% (Reslan and Khalil 2012). The benefits of exogen oestrogen act by activating the nitric oxide synthetase through phosphatidylinositol 3-kinase and AKT signalling pathways and ultimately promote coronary vasodilation, and, in the long term, cardiac tissue tolerance to ischaemia (Murphy 2011; Knowlton and Lee 2017). If the patient does not accept taking hormone replacement therapy, other preventive interventions should be offered such as complementing calcium intake with supplements, quit smoking, exercise regularly, or take serotonin recapture inhibitors (Langrish et al. 2009; O’Donnell et al. 2016). In our study, there weren’t thromboembolic events associated with hormone replacement therapy and this could be associated with the lowest dose of oestrogen used, but we should take this cautiously because of the number of patients included and besides, it is not clear the risk of events due to low incidence of arterial disease among young women (Gialeraki et al. 2018).

Among the strengths of this study are that the analysis was done in a specific oncologic population with primary ovarian insufficiency and that the impact of both neoplasia and oncological management was evaluated. With this evidence, we will be able to identify patients at risk for presenting primary ovarian insufficiency and advise them on the symptoms they may present, in order to offer hormone replacement therapy in a timely manner and positively impact their quality of life, sexual, bone and cardiovascular health. In the future, as in INCan, other cancer centres will pay special attention to this group of patients with adequate monitoring and management of conditions associated with primary ovarian insufficiency. As for study limitations, its retrospective nature and the small size of the cohort restricted statistical analysis. Additionally, variables such as clinical stage and treatment for each type of cancer are needed to deepen the discussion of certain associations of the analysis. We concluded that it is important to individualise hormone replacement therapy and inform patients on other available options to improve their quality of life.

Conclusion

This is the first study that characterises Mexican patients with primary ovarian insufficiency secondary to gynaecological or haematological cancer. In summary, most patients had good metabolic and cardiovascular health, a significant proportion of patients were on hormone replacement therapy, and the most frequently reported climacteric symptom were of vasomotor nature. Having ovarian or haematological cancer, being managed with radiotherapy and/or chemotherapy, and having FSH parameters higher than 35 mUI/mL are factors that significantly increase the risk of presenting vasomotor symptoms. Results from this study favour the identification of primary ovarian insufficiency patients who are prone to present vasomotor symptoms and hence address them in a more individual and targeted manner.

Ethical approval

Ethics approval for this study was waived by Research Ethics Committee with approval number REF/INCAN/CI/0059/2019.

Author contributions

Clinical team: SABM RSH, DIO and DCDL. Conceptualisation: SABM and MGG. Data curation: SABM, DCDL, AGE and VCG. Formal analysis: SABM, RSH and DIO. Investigation: SABM RSH, DIO and DCDL. Literature review: SABM, MGG, CMSR and JCGR. Methodology: SABM. Writing – original draft: SABM and MGG. Writing – review & editing: SABM, MGG, CMSR and JCGR.

Acknowledgement

The authors thank “Programa Integral para La Atención, Investigación, Difusión y Capacitación del Cáncer Cervicouterino en sus diferentes etapas: MICAELA”. Presupuesto de Egresos de la Federación 2021, INCan for their support.

Disclosure statement

No potential conflict of interest was reported by the authors.