Patient characteristics
In this cross-sectional investigation, total 591 patients were invited and 563 patients with ovarian function suppression completed surveys were collected for an estimated response rate of 95% (563/591). Twenty-eight surveys were excluded from the analysis for conflicting entries information of tumor or demographic characteristics. As shown in Fig. 1, the final group of patients with complete demographics characteristics and PHQ-9/FSFI/EORTC QLQ-BR23 scores consisting of 563 individuals. According to the type of ovarian function suppression, 174 patients with ovarian ablation for ovarian function suppression were in OA cohort and 389 patients with GnRH agonist were in GnRHa cohort. Table 1 illustrates the main characteristics between GnRHa and OA cohorts. The median ages of the female breast cancer survivors were 46.0 years in OA cohort and 43.0 years in GnRHa cohort (P = 0.013). At the time of survey response, what is striking in Table 1 is that time interval from ovarian suppression was significantly longer in OA cohort (30.3 ± 31.2 months vs. 22.8 ± 22.2 months, P = 0.009). The majority of those response to this survey were low annual income (77.0%), never or little smoking habit (98.4%), never or little alcohol drinking habit (99.1%), none co-morbidity (93.8%), mastectomy surgery (77.3%), stage I-III (73.3%), estrogen receptor (ER) or progesterone receptor (PR)-positive (97.2%), and HER2-negative (76.1%). In addition, 59.2% of the patients in GnRHa cohort were educated up to college level of above and the percentage was only 42.4% in OA cohort. Interestingly, patients received medicine GnRHa were incline to breast conserving surgery (26.0% vs 16.7%, P = 0.015).
Table 1
Baseline social demographics and clinical characteristics in breast cancer patients with ovarian function suppression
Characteristics
|
Total
|
Ovarian Ablation
OA
|
GnRH agonist
GnRHa
|
P Value
|
N
|
563
|
174
|
389
|
|
Age, years
|
|
|
|
|
Median
|
44
|
46
|
43
|
|
Range
|
22–63
|
31–61
|
22–63
|
|
Mean (SD)
|
42.7 ± 7.8
|
45.2 ± 7.0
|
41.6 ± 7.8
|
0.013
|
Duration of OFS
Mean (SD), months
|
25.1 ± 25.4
|
30.3 ± 31.2
|
22.8 ± 22.2
|
0.009
|
Educational level
|
|
|
|
0.0002
|
High school or below
|
268
|
103
|
165
|
|
College or above
|
295
|
71
|
224
|
|
Annual income (RMB)
|
|
|
|
0.360
|
≤ 50,000
|
398
|
126
|
272
|
|
> 50,000
|
165
|
48
|
117
|
|
Smoking habit
|
|
|
|
0.439
|
Never or little
|
548
|
168
|
380
|
Mostly
|
15
|
6
|
9
|
Alcohol drinking habit
|
|
|
|
0.271
|
Never or little
|
556
|
170
|
386
|
|
Mostly
|
7
|
4
|
3
|
|
Co-morbidity
|
|
|
|
0.077
|
Diabetes/Hypertension
|
51
|
22
|
29
|
|
None
|
522
|
162
|
360
|
|
Type of Surgery
|
|
|
|
0.015
|
Mastectomy
|
433
|
145
|
288
|
Breast-conserving
|
130
|
29
|
101
|
TNM staging
|
|
|
|
0.006
|
I-III
|
400
|
110
|
290
|
IV
|
163
|
64
|
99
|
Hormone Receptor Status
|
|
|
|
0.0004
|
ER or PR Positive
|
526
|
153
|
373
|
ER and PR Negative
|
37
|
21
|
16
|
HER2 Status
|
|
|
|
0.262
|
Positive
|
130
|
35
|
95
|
Negative
|
433
|
139
|
294
|
Major depression in GnRHa and OA cohorts
Closer inspection of Table 2 shows mean PHQ-9 sum score in GnRHa cohort tend to be lower than that in ovarian ablation (OA) cohort (11.4 ± 5.7 vs. 12.8 ± 5.8, P = 0.079). According to the algorithm for depression [moderate (score 8–14), major (score 15–19), and severe (score 20–27) ], 36.2% (63/174), 28.2% (49/174), and 12.6% (22/174) of patients were categorized in the moderate, major, and severe depression groups in OA cohort, and 42.9% (167/389), 22.4% (87/389), and 8.7% (34/389) of patients were categorized in the moderate, major, and severe depression groups in GnRHa cohort, respectively. Noticeably, patients with major depression, defined by PHQ-9 ≧ 15, was indicated significantly fewer in GnRHa cohort than those in OA cohort. A positive correlation was revealed between major depression and ovarian ablation, whereas the absolute difference was approximately 9.1% (31.1% vs. 40.2%, P = 0.025). Suicidal ideation was considered as a symptom of major depression. On the item level, 15 out of the 174 patients with ovarian ablation and 30 out of the 389 patients with GnRHa claimed to have suicidal ideation (8.6% vs. 7.7%, P = 0.713). What is striking about Table 3 is ovarian suppression type as the independent risk factor on major depression in uni-variate and multivariate analysis [Ovarian ablation vs GnRHa: Exp (B) = 2.483 (95% CI: 1.056 ~ 6.450), P = 0.026]. As mentioned above, patients received medicine GnRHa were incline to breast conserving surgery (26.0% vs 16.7%, P = 0.015), turning now to the investigate the relationship, however, data from Table 3, it is apparent that the type of surgery was not the independent factor in multivariate analysis on major depression. Further analysis showed that the duration of ovarian function suppression was closely correlated with major depression, and there has been a remarkable time-dependent tendency [duration of OFS > 2 year vs duration of OFS ≤ 2 year : Exp (B) = 1.651 (95% CI: 1.071 ~ 2.518), P = 0.031]
Table 2
Depression (PHQ-9) by the type of ovarian function suppression in patients with breast cancer
|
Ovarian function suppression
|
Ovarian Ablation (OA)
n, (%)
|
GnRH agonist (GnRHa)
n, (%)
|
P value
|
Total
|
174
|
389
|
|
PHQ-9 scores Median
(IQR)
|
13
(8.0–17.0)
|
11
(7.0–16.0)
|
|
PHQ-9 scores Range
|
1–24
|
0–25
|
|
PHQ-9 scores Mean (SD)
|
12.8 ± 5.8
|
11.4 ± 5.7
|
0.079
|
PHQ-9 scores subgroups
|
|
|
|
None or mild (0–7), n(%)
|
40 (23.0)
|
101 (26.0)
|
|
Moderate (8–14), n(%)
|
63 (36.2)
|
167 (42.9)
|
|
Major (15–19), n(%)
|
49 (28.2)
|
87 (22.4)
|
|
Severe (20–27), n(%)
|
22 (12.6)
|
34 (8.7)
|
|
Major Depression
(PHQ ≥ 15), n(%)
|
71
(40.2)
|
121
(31.1)
|
0.025
|
Suicidal ideation, n(%)
|
15 (8.6)
|
30 (7.7)
|
0.713
|
PHQ-9, Patient Health Questionnaire − 9 items; SE, standard errors; IQR, interquartile range (25th, 75th percentiles). |
Table 3
The associated factors on major depression (PHQ-9, scores < 15 vs ≥ 15) by logistic regression in patients with breast cancer
Characteristics
|
Univariate
|
Multivariate
|
|
OR
|
95%CI
|
P value
|
OR
|
95%CI
|
P value
|
Age, years
|
|
|
|
|
|
|
≤ 45
> 45
|
1
1.392
|
0.924–2.095
|
0.113
|
|
|
|
Educational level
|
|
|
|
|
|
|
High school or below
College or above
|
1
0.773
|
0.522–1.145
|
0.199
|
|
|
|
Annual income (RMB)
|
|
|
|
|
|
|
≤ 50,000
> 50,000
|
1
0.883
|
0.577–1.351
|
0.565
|
|
|
|
TNM staging
|
|
|
|
|
|
|
I- III
IV
|
1
1.233
|
0.641–2.373
|
0.530
|
|
|
|
Type of OFS
|
|
|
|
|
|
|
GnRH agonist
Ovarian Ablation
|
1
1.910
|
1.006–2.518
|
0.043
|
2.483
|
1.056–6.450
|
0.026
|
Type of Surgery
|
|
|
|
|
|
|
Breast-conserving
Mastectomy
|
1
2.079
|
1.338–3.230
|
0.001
|
1.459
|
0.951–3.630
|
0.058
|
OFS Time
|
|
|
|
|
|
|
≤ 2 years
> 2 years
|
1
1.555
|
1.010–1.568
|
0.037
|
1.651
|
1.071–2.518
|
0.031
|
HER2 Status
|
|
|
|
|
|
|
Negative
Positive
|
1
0.734
|
0.420–1.283
|
0.278
|
|
|
|
Endocrine Therapy
|
|
|
|
|
|
|
AI
TAM
|
1
1.030
|
0.655–1.619
|
0.898
|
|
|
|
Sexual dysfunction in GnRHa and OA cohorts
With successive diverse affect on major depression, we moved further to investigate whether varied sexual dysfunction between GnRHa and OA methods. Patients received GnRHa had lower mean and median FSFI scores compared with patients with ovarian ablation (OA), as indicated in Table 4 (mean ± SD:17.8 ± 8.7 vs 19.3 ± 8.5, P = 0.205; median: 17.8 vs 19.6), only trace difference amounts of mean FSFI sum scores were detected. If we turn to the sexual dysfunction, defined as FSFI < 23, strong evidence of GnRHa- induced sexual dysfunction was revealed, more than half of patients with ovarian ablation met the criteria for sexual dysfunction (61.5%, FSFI < 23) compared with 72.2% of patients with GnRHa (P = 0.011). From the data in Fig. 2, patients with GnRHa had a little lower most of sub-FSFI scores than patients with ovarian ablation but remained a clear significantly lower in pain section (FSFI-6 scores: mean ± SD:3.3 ± 2.2 vs 2.5 ± 2.2, P = 0.007) and significantly difference in lubrication section (FSFI-3 scores: mean ± SD:3.5 ± 2.5 vs 2.9 ± 2.6, P = 0.048). T-tests revealed no significant differences in mean scores on sexual desire (FSFI-1) and arousal (FSFI-2), in contrast to other sub-items, sexual orgasm and sexual satisfaction was more benefit in GnRHa cohort but the difference could not be identified in this analysis. Data from the uni-variate and multivariate analysis, a positive correlation between diverse ovarian function suppression type and sexual dysfunction was revealed (GnRHa vs OA: OR = 1.960, 95% CI: 1.070 ~ 3.591, P = 0.029). In uni-variate analysis, age received OFS, educational level and combined endocrine therapy were the significantly risk factors for sexual dysfunction, and further multivariate analysis were evaluated. Adjusting for other factors, age received OFS and educational level remained lower odd ratio (Age > 45 years vs Age ≤ 45 years, OR = 0.587, 95% CI: 0.356 ~ 0.967, P = 0.037; College or above vs High school or below: OR = 0.433, 95% CI: 0.258 ~ 0.726, P = 0.002). Despite of the limited samples for tamoxifen users, the associations were not found in multivariate analysis when compared combined endocrine therapy between tamoxifen users and aromatase inhibitor users (OR = 1.160, 95% CI: 0.310 ~ 4.341, P = 0.826). Further large sample research needs to be investigated the correlation between sexual dysfunction and combined endocrine therapy.
Table 4
Sexual function by the type of ovarian function suppression in patients with breast cancer
|
Ovarian function suppression
|
Ovarian Ablation (OA)
N = 174
|
GnRH agonist (GnRHa)
N = 389
|
P value
|
FSFI total scores
|
|
|
|
Mean (SD)
|
19.3± 8.5
|
17.8± 8.7
|
0.205
|
Median (IQR)
|
19.6(13.0-26.7)
|
17.8(12.2–24.3)
|
|
Range
|
2–34
|
2–34
|
|
Sexual dysfunction (FSFI < 23), n(%)
|
107 (61.5)
|
281 (72.2)
|
0.011
|
FSFI, female sexual function index; SE, standard errors; IQR, interquartile range (25th, 75th percentiles). |
Table 5
The associated factors on sexual dysfunction (FSFI, scores ≥ 23 vs < 23) by logistics regression in patients with breast cancer
Characteristics
|
Univariate
|
Multivariate
|
|
OR
|
95%CI
|
P value
|
OR
|
95%CI
|
P value
|
Age, years
|
|
|
|
|
|
|
≤ 45
> 45
|
1
0.645
|
0.403–1.030
|
0.066
|
0.587
|
0.356–0.967
|
0.037
|
Educational level
|
|
|
|
|
|
|
High school or below
College or above
|
1
0.551
|
0.341–0.891
|
0.015
|
0.433
|
0.258–0.726
|
0.002
|
Annual income (RMB)
|
|
|
|
|
|
|
≤ 50,000
> 50,000
|
1
0.926
|
0.560–1.533
|
0.766
|
|
|
|
TNM staging
|
|
|
|
|
|
|
I- III
IV
|
1
0.872
|
0.405–1.881
|
0.728
|
|
|
|
Type of OFS
|
|
|
|
|
|
|
Ovarian Ablation
GnRH agonist
|
1
1.838
|
1.041–3.244
|
0.036
|
1.960
|
1.070–3.591
|
0.029
|
Type of Surgery
|
|
|
|
|
|
|
Breast-conserving
Mastectomy
|
1
0.898
|
0.518–1.558
|
0.703
|
|
|
|
OFS Time
|
|
|
|
|
|
|
≤ 2 years
> 2 years
|
1
0.777
|
0.489–1.236
|
0.287
|
|
|
|
HER2 Status
|
|
|
|
|
|
|
Negative
Positive
|
1
1.714
|
0.764–3.844
|
0.191
|
|
|
|
Endocrine Therapy
|
|
|
|
|
|
|
AI
TAM
|
1
2.359
|
1.249–4.456
|
0.008
|
|
|
|
AI, aromatase inhibitors; FSFI, female sexual function index; OFS, ovarian function suppression; TAM, tamoxifen. |
Quality of Life in GnRHa and OA cohorts
The next section of the survey was concerned with quality of life. Mean scores of breast cancer patients with GnRHa in only one QLQ-BR23 functional dimensions, future perspective (BRFU). What stands out in Table 6 is a positive correlation between future perspective and GnRHa (GnRHa vs. OA: BRFU, 77.0 ± 31.6 verse 70.3 ± 32.0, P = 0.039). This table is quite revealing in several ways. Firstly, poor sexual functioning and enjoyment in dimensions BRSEF and BRSEE were revealed with similar tendency in FSFI scores (GnRHa vs. OA: BRSEF, 55.9 ± 37.9 verse 61.7 ± 37.8, P = 0.208; BRSEE, 35.0 ± 35.1 vs. 38.7 ± 36.6, P = 0.421). More patients suffered side effects (BRST), breast symptoms (BRBS), arm symptoms (BRAS) and upset by hair loss (BRHL) in ovarian ablation (OA) cohort, but no significant (P = 0.502; P = 0.310; P = 0.512; P = 0.785). This is an expected result, EORTC QLQ-BR23 focused on the breast cancer-related symptom and function, but not only regarding ovarian function, the differences of most sub-scales between OA and GnRHa were insignificant.
Table 6
Quality of Life Scores (EORTC-BR23) according to the type of ovarian function suppression in patients with breast cancer
EORTC-BR23 scores
(mean ± SD)
|
Ovarian function suppression
|
Ovarian Ablation
N = 174
|
GnRH agonist
N = 389
|
P value
|
Functional scales
|
|
|
|
Body image
|
BRBI
|
68.0 ± 25.8
|
72.3 ± 24.0
|
0.190
|
Sexual functioning
|
BRSEF
|
61.7 ± 37.8
|
55.9 ± 37.9
|
0.208
|
Sexual enjoyment
|
BRSEE
|
38.7 ± 36.6
|
35.0 ± 35.1
|
0.421
|
Future perspective
|
BRFU
|
70.3 ± 32.0
|
77.0 ± 31.6
|
0.039
|
Symptom scales
|
|
|
|
Side effects
|
BRST
|
54.4 ± 16.3
|
52.9 ± 15.0
|
0.502
|
Breast symptoms
|
BRBS
|
32.8 ± 26.7
|
34.5 ± 22.9
|
0.310
|
Arm symptoms
|
BRAS
|
36.6 ± 25.3
|
37.7 ± 23.0
|
0.512
|
Upset by hair loss
|
BRHL
|
90.5 ± 21.7
|
89.6 ± 23.2
|
0.785
|
BRBI, body image; BRSEF, sexual functioning; BRFU, future perspective; BRSEE, sexual enjoyment; BRST, systemic therapy side effect; BRBS, breast symptoms; BRAS, arm symptoms; BRHL, upset by hair loss. |
Replacement ovarian function suppression
In the final part of our survey, respondents were asked “Would you like to choose the other ovarian function suppression, for example, would you like to choose ovarian ablation instead of medicine GnRHa and vice versa? ” Surprisingly, the majority of both participants considered this alternation. Approximately 62.4% patients chose to exchange the type of ovarian function suppression (GnRHa: 62.7% vs OA: 60.8%). Concerns regarding costs may affect the final choice, we also asked another question “Without considering the cost and economy, would you like to choose the other ovarian function suppression, for example, would you like to choose ovarian ablation instead of medicine GnRHa and vice versa?” This is an expected outcome, without taking into account of medical-related economic cost, the exchange ratio greatly increased and was up to 70.3% in ovarian ablation cohort, whereas, no significantly increased in GnRHa cohort (67.1%). However, the observed difference between ovarian ablation and GnRHa in our study was not significant, the absolute differences of replacement were 3.2% in ovarian ablation and only 1.5% in patients with GnRHa. The majority of both cohorts may be unsatisfied the current ovarian function suppression regarding depression, sexual dysfunction and worse quality of life, and the majority of participants were reluctant to receive ovarian ablation limited to the medical-related economic costs.