Investigating sexual health after breast cancer by longitudinal assessment of patient-reported outcomes

Abstract

Background

Sexual concerns are a major unaddressed need among survivors of breast cancer (BC) with significant negative effects on quality of life. We longitudinally analyzed sexual health over time, using patient-reported outcomes.

Methods

Patients with stage I-III BC prospectively included from the CANcer TOxicity cohort (CANTO) provided data at diagnosis, then 1, 2, and 4 years afterward. Sexual concerns outcomes included poor body image (score ≤91/100), poor sexual functioning (≤16/100), poor sexual enjoyment (≤66/100), and sexual inactivity (EORTC QLQ-B23). Multivariate generalized estimating equation models assessed associations with sexual concerns after diagnosis, adjusting for age, sociodemographic, tumor, treatment, and clinical characteristics.

Results

Nearly 78.1% among 7895 patients reported at least one sexual concern between diagnosis and 4 years’ follow-up. Over time, the proportion of patients reporting sexual concerns either increased or remained constant with diagnosis. Less than half (46%, range 11.4-57) of the patients with sexual concerns reported the use of supportive care strategies, including gynecological or psychological consultations (range 11.4-57.4). Factors consistently associated with sexual concerns up to 4 years after diagnosis included already reporting the same concern at diagnosis [odds ratio (OR)_{poor body image} 3.48 [95% confidence interval (CI) 3.11-3.89]; OR_{sexual inactivity} 9.94 (95% CI 8.84-11.18), OR_{poor sexual function} 9.75 (95% CI 8.67-10.95), OR_{poorsexual enjoyment} 3.96 (95% CI 3.34-4.69)], endocrine therapy use [OR_{poor body image} 1.15 (95% CI 1.01-1.31); OR_{sexual inactivity} 1.19 (95% CI 1.02-1.39), OR_{poor sexual function} 1.17 (95% CI 1.01-1.37), OR_{poor sexual enjoyment} 1.23 (95% CI 1.00-1.53)], and depression [OR_{poor body image} 2.00 (95% CI 1.72-2.34); OR_{sexual inactivity} 1.66 (95% CI 1.40-1.97), OR_{poor sexual function} 1.69 (95% CI 1.43-2.00), OR_{poor sexual enjoyment} 1.94 (95% CI 1.50-2.51)]. Outcome-specific associations were also identified.

Conclusions

Sexual concerns seem frequent, persistent, and insufficiently addressed. Pretreatment concerns, endocrine therapy, and emotional distress are commonly associated factors. A proactive evaluation of sexual health across the care continuum is needed, to promptly identify patients suitable for multidisciplinary counseling, referral, and supportive interventions.

Highlights

• •Sexual concerns are frequent at BC diagnosis and seem to increase or persist in the after-treatment period.
• •Less than 50% of patients with reported sexual concerns consulted with proper supportive care specialists.
• •This large longitudinal cohort study helps identify patients that are more likely to report persistent sexual concerns.
• •A proactive evaluation and management of sexual health must exist across the BC care continuum.

Introduction

In Europe and North America, up to 5 million people live with and after breast cancer (BC).^1,2 A substantial proportion of these patients endure troublesome symptom burdens and deteriorated overall quality of life (QoL).^3,4 Among the several domains of QoL, sexual health is often impacted by cancer and cancer treatments. Data also show that sexual health is often insufficiently approached by health care professionals during routine follow-up care of survivors of BC, remaining largely unaddressed.⁵ Sexuality is multidimensional, encompassing physical, psychological, interpersonal, social, and cultural elements. As a result, survivors of BC may experience impaired sexual health through a variety of symptoms.⁶ Most commonly these symptoms include decreased sexual desire, arousal disorders, lubrication problems, dyspareunia, poor nipple sensation, and poor body image.⁵ Among women with a history of BC, sexual health may be influenced by a multitude of factors that add to the impact of a cancer diagnosis and treatment. These include the patient’s own health, partner relationships and communication, previous sexual history, view of self-identify, self-esteem, concomitant medications, menopausal status, but also concomitant symptoms such as fatigue, pain, emotional distress, and urinary incontinence.⁵

Prior studies evaluating sexual health and sexual concerns in survivors of BC lacked a comprehensive longitudinal evaluation. Most of these studies were cross-sectional surveys7, 8, 9, 10 or longitudinal cohorts with small sample size, short follow-up, or limited clinical and treatment-related data.¹¹ In addition, several studies focused only on young patients.12, 13, 14

The prospective multicenter CANcer TOxicity (CANTO) cohort (ClinicalTrials.gov identifier: NCT01993498) aims at characterizing toxicities of BC treatment, building on an extensive longitudinal collection of clinical, behavioral, tumor, treatment, and patient-reported outcome data.¹⁵ In this study, we used CANTO to comprehensively study sexual health over time after a diagnosis of early BC and investigate clinical, psychosocial, and behavioral factors associated with sexual concerns.

Methods

Data source

We used data from the prospective CANTO cohort, which includes patients diagnosed with stage I-III primary BC and no prior history of cancer within the past 5 years. Data were collected in 26 French cancer care centers. Treatment and tumor classification were extracted from medical files. Patients’ medical history, prediagnosis comorbidities, and a set of physical treatment-related toxicities and use of supportive care strategies were collected during face-to-face health examinations by trained clinical research nurses. Patient-reported QoL, including sexual health, and psychological outcomes were collected by validated self-reported paper questionnaires assessing physical and psychological outcomes. Socioeconomic data were gathered through an ad hoc self-reported paper questionnaire gathering items from diverse French population-based surveys.^16,17 For the current study, these data were collected prospectively at four time points: at diagnosis of BC (T0) and during follow-up visits at year 1 (T1), 2 (T2), and 4 (T3) after diagnosis. Surgery, chemotherapy, and/or radiation therapy are completed 3-6 months before T1. The study was approved by the French Regulatory Authorities. All patients enrolled in the study were aged ≥18 years and provided written informed consent.¹⁵ Data were reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement.¹⁸

Study cohort

This study included the data of 9597 patients enrolled in the CANTO cohort between 2012 and 2018. Women that did not provide information about sexual health (body image, sexual function, and sexual enjoyment) at any of the time points of interest were excluded from the analysis (n = 1612). The final study sample included a total of 7985 patients. For the outcomes of sexual enjoyment, analyses were restricted to women who declared themselves to be sexually active at diagnosis (n = 3956); see Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2024.102236.

Definition of variables of interest

Outcome variables

Sexual health was evaluated by the assessment of specific sexual concerns outcomes that included (i) poor body image, (ii) sexual inactivity, (iii) poor sexual functioning, and (iv) poor sexual enjoyment. These concerns were measured using the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire (QLQ)-BR23 (see Supplementary Material, Description of Sexual Health Outcomes, available at https://doi.org/10.1016/j.esmoop.2024.102236).

Exposure variables

Socioeconomic, behavioral, clinical, tumor, and treatment characteristics (assessed at diagnosis) as well as treatment-related toxicities and symptoms (assessed at T1, e.g. closest time point after primary treatment completion) were categorized according to Table 1. Anxiety and depression were assessed with the Hospital Anxiety and Depression Scale, and each was categorized as not at risk (score = 0-7), doubtful risk (score = 8-10), or at risk (score = 11-21) of developing clinical disorders. Additional domains of QoL were assessed using the EORTC QLQ-C30 and categorized as poor or severe using thresholds of <60/100 for functions and <40/100 for symptoms.¹⁹

Table 1.

Overall cohort description

Characteristics	Total (n = 7985)
Demographic and socioeconomic information
Age (years), mean (range)	56.3 (22.2-89.1)
<40	580 (7.3)
40-49	1867 (23.4)
50-64	3520 (44.1)
>65	2018 (25.3)
Marital status, n (%)
Not partnered	1679 (21.5)
Partnered	6119 (78.5)
Missing, n	187
Number of children, n (%)
0	273 (3.8)
1	1548 (21.3)
≥2	5444 (74.9)
Missing, n	720
Education, n (%)
Primary	1090 (14.1)
High school	3596 (46.4)
College	3070 (39.6)
Missing, n	229
Monthly household income (euros), n (%)
<1500	1091 (14.3)
≥1500-<3000	3307 (43.3)
≥3000	3246 (42.5)
Missing, n	341
Medical history and health behavior information
Charlson Comorbidity score, n (%)
0	5967 (81.2)
≥1	1386 (18.8)
Missing, n	632
Smoking behavior, n (%)
Current smoker	1360 (17.3)
Former smoker	1727 (21.9)
Never smoker	4782 (60.8)
Missing, n	116
Daily alcohol consumption, n (%)
Less than daily	6685 (85.9)
Daily	1095 (14.1)
Missing, n	205
Body mass index (kg/m2), WHO definition, n (%)
Lean	4081 (51.3)
Overweight	2319 (29.1)
Obese	1557 (19.6)
Missing, n	28
Level of physical activity, WHO definition, n (%)
Insufficiently active (<10 MET-h/week)	3401 (43.1)
Sufficiently active (≥10 MET-h/week)	4499 (56.9)
Missing, n	85
Breast cancer treatment information
Chemotherapy, n (%)
No	3748 (46.9)
Yes	4237 (53.1)
Endocrine therapy, n (%)
No	1450 (18.2)
Yes	6531 (81.8)
Missing, n	4
Use of ovarian function suppression, n (%)
No	7785 (97.70)
Yes	183 (2.30)
Missing, n	17
Anti-HER2 therapy, n (%)
No	7001 (87.7)
Yes	981 (12.3)
Missing, n	3
Adjuvant radiotherapy, n (%)
No	656 (8.2)
Yes	7324 (91.8)
Missing, n	5
Axillary surgery, n (%)
Dissection sentinel	2988 (37.4)
Lymph node excision	4997 (62.6)
Breast surgery, n (%)
Mastectomy	2115 (26.5)
Conservative	5870 (73.5)
Breast reconstruction at T1, n (%)
No	7621 (95.5)
Yes	359 (4.5)
Missing, n	5
Symptoms and quality of life measured at T1
Anxiety, n (%)
Noncase	4260 (56.2)
Doubtful + case	3321 (43.8)
Missing, n	404
Depression, n (%)
Noncase	6166 (81.4)
Doubtful + case	1413 (18.6)
Missing, n	406
Vaginal dryness, n (%)
No	5109 (65.7)
Yes	2670 (34.3)
Missing, n	206
Hot flashes, n (%)
No	3013 (38.7)
Yes	4767 (61.3)
Missing, n	205
Poor role functioning, n (%)
No	6361 (84.0)
Yes	1213 (16.0)
Missing, n	411
Poor social functioning, n (%)
No	6497 (86.0)
Yes	1059 (14.0)
Missing, n	429
Severe fatigue, n (%)
No	4801 (63.5)
Yes	2765 (36.5)
Missing, n	419
Severe pain, n (%)
No	37 945 (49.9)
Yes	3802 (50.1)
Missing, n	388
Severe insomnia, n (%)
No	4763 (63.1)
Yes	2789 (36.9)
Missing, n	433
Supportive care consultation during T0-T1
Gynecologist/sexologist, n (%)
Yes	2413 (30.2)
No	5572 (69.8)
Psychiatrist/psychologist, n (%)
Yes	1315 (16.5)
No	6670 (83.5)

HER2, human epidermal growth factor receptor 2; MET, metabolic equivalent; T0, at the moment of diagnosis; T1, year 1 after diagnosis; WHO, World Health Organization.

Statistical analysis

First, descriptive statistics were used to summarize cohort characteristics and prevalence of sexual concerns at diagnosis T1, T2, and T3. We then fitted multivariate generalized estimating equations models to assess the association between exposure variables and sexual concerns outcomes over time. Separate models were constructed for each outcome. An exchangeable working correlation matrix structure was specified to account for within-subject correlations. The variance function for the binomial distribution and the logit-link function were used (binary response data). To assess relative differences between levels of exposure variables, model-based multivariable-adjusted odds ratios (ORs) and respective 95% confidence intervals (CIs) were obtained. Statistical analyses were carried out using SAS statistical software version 9.4 (SAS Institute Inc., College Station, TX). Statistical significance was defined with a 2-sided P value <0.05.

Results

Cohort characteristics

Baseline characteristics of the 7985 patients included in the study cohort are summarized in Table 1. Their mean age was 56.3 (range 22.2-89.1) years; most patients were in a couple (78.5%), had at least a high school educational level (86%), and were actively working (54.6%). The majority of patients were postmenopausal (61.9%), had no comorbidities (81.2%), never smoked (60.8%), and had less than daily alcohol consumption frequency (85.9%). A body mass index >25 kg/m² was encountered in 48.7% of patients and 43.1% presented with insufficient physical activity levels (<10 metabolic equivalents-h/week).

Up to half of the patients were treated with chemotherapy (53.1%) and the majority received adjuvant endocrine therapy (ET; 81.8%, of which 2.3% received ovarian-function suppression) and radiotherapy (91.8%). Breast conservative surgery was carried out in 73.5% of patients and 62.6% underwent axillary dissection. At T1, ∼43.8% of patients had a doubtful or clinically suggestive diagnosis of anxiety and 18.6% of depression. Vaginal dryness was reported in 34.3% and 61.3% of patients reported hot flashes. In addition, severe fatigue, pain, and insomnia were reported by 36.5%, 24.4%, and 36.9% of the patients at T1, respectively. In the entire cohort, 30% of the patients reported having consulted with a gynecologist between diagnosis and the post-treatment consultation at T1. Specifically, 0.2% consulted a sexologist and 16.5% a psychologist/psychiatrist (Table 1). A description of clinical and sociodemographic data of women that did not provide information about sexual health is presented in Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2024.102236.

Sexual health over time and use of supportive care strategies

Overall, 78.2% of patients reported at least one sexual concern among poor body image, sexual inactivity, poor sexual functioning, and poor sexual enjoyment between diagnosis and 4 years after. Over time, the proportion of patients reporting sexual concerns either increased or remained as elevated as at diagnosis (Figure 1). An underutilization of supportive care strategies was observed among patients with reported sexual concerns (Figure 2). At T3, the percentage of patients reporting a consultation with a gynecologist/and psychologist/psychiatrist was 49.4% and 18.1% for patients reporting poor body image, 45.1% and 11.2% for patients with sexual inactivity, 45.9% and 11.4% for patients with poor sexual functioning, and 57.3% and 13.4% for patients with poor sexual enjoyment at T3.

Figure 1.

Prevalence of sexual concerns (poor body image, sexual inactivity, poor sexual functioning, and poor sexual enjoyment) over time.T0, at diagnosis; T1, year 1 after diagnosis; T2, year 2 after diagnosis; T3, year 4 after diagnosis.

Figure 2.

Use of supportive care strategies (psychosocial counseling and gynecologist consultation) in patients with reported sexual concerns at T1 and T2.T0, at diagnosis; T1, year 1 after diagnosis; T2, year 2 after diagnosis; T3, year 4 after diagnosis.

Factors associated with sexual health outcomes

Body image

A poor body image was reported by 32.2% of the patients at T0, then by 57.8% at T1, 52.6% at T2, and 48.3% at T3. A higher likelihood of reporting poor body image was observed in patients with a lower educational level (OR 0.73, 95% CI 0.62-0.86 for college versus primary school, respectively], smoking behavior (OR 1.19, 95% CI 1.05-1.34 for current smoker and OR 1.33, 95% CI 1.16-1.53 for prior smoker), with a body mass index ≥25 kg/m² (OR 1.24, 95% CI 1.12-1.37), those treated with mastectomy (OR 1.99, 95% CI 1.74-2.28 versus partial), adjuvant chemotherapy (OR 1.88, 95% CI 1.68-2.10 versus not), and adjuvant ET (OR 1.15, 95% CI 1.01-1.31 versus not). In addition, patients with a doubtful or clinically suggestive diagnosis of anxiety (OR 1.80, 95% CI 1.62-1.99 versus not) or depression (OR 2.00, 95% CI 1.72-2.34 versus not), those with vaginal dryness (OR 1.16, 95% CI 1.05-1.29 for any grade versus not), hot flashes (OR 1.22, 95% CI 1.10-1.35 for any grade versus not), severe fatigue (OR 1.48, 95% CI 1.31-1.67 versus nonsevere), severe pain (OR 1.31, 05% CI 1.14-1.50 versus nonsevere), severe insomnia (OR 1.16, 05% CI 1.04-1.29 versus nonsevere), and poor social functioning (OR 1.65, 95% CI 1.39-1.97 versus nonpoor) were more likely to report poor body image. Reporting poor body image already at diagnosis was strongly associated with persistent poor body image between diagnosis and year 4 (OR 3.48, 95% CI 3.11-3.89; Table 2).

Table 2.

Multivariate analysis of factors associated with poor body image, sexual inactivity, poor sexual functioning, and poor sexual enjoyment

Exposable variable	Body image			Sexual functioning			Sexually inactive			Sexual enjoyment
	OR	95% CI	P value	OR	95% CI	P value	OR	95% CI	P value	OR	95% CI	P value
Age, <40 years versus >65 years	1.08	0.86-1.36	0.460	0.17	0.12-0.23	<0.0001	0.18	0.13-0.24	<0.0001	0.49	0.34-0.71	0.001
Age, 40-49 years versus >65 years	1.12	0.95-1.31	0.152	0.22	0.18-0.27	<0.0001	0.22	0.18-0.27	<0.0001	0.81	0.60-1.08	0.160
Age, 50-64 years versus >65 years	1.02	0.90-1.16	0.658	0.49	0.43-0.57	<0.0001	0.49	0.43-0.57	<0.0001	0.85	0.66-1.11	0.260
Marital status, partnered versus not partnered	0.95	0.83-1.09	0.519	2.22	1.89-2.60	<0.0001	2.29	1.95-2.68	<0.0001	0.74	0.51-1.07	0.112
Education, high versus primary school	1.13	1.01-1.26	0.027	1.02	0.89-1.16	0.715	1.02	0.90-1.16	0.673	0.95	0.80-1.13	0.578
Education, college versus primary school	0.73	0.62-0.86	0.001	1.10	0.91-1.33	0.300	1.14	0.95-1.38	0.149	1.58	1.09-2.30	0.015
Monthly income, ≥1500-<3000 euros versus <1500 euros	1.10	0.98-1.23	0.076	0.86	0.75-0.98	0.031	0.84	0.73-0.96	0.011	0.85	0.72-1.01	0.079
Monthly income, ≥3000 euros versus <1500 euros	1.04	0.88-1.22	0.626	1.20	0.99-1.45	0.056	1.22	1.01-1.48	0.031	1.09	0.76-1.54	0.627
Comorbidity score, ≥1 versus 0	1.00	0.88-1.14	0.914	1.04	0.90-1.21	0.525	1.03	0.89-1.19	0.649	0.95	0.76-1.18	0.668
Smoke behavior, current versus never	1.19	1.05-1.34	0.003	0.97	0.84-1.13	0.779	0.96	0.83-1.10	0.601	0.87	0.72-1.06	0.180
Smoke behavior, former versus never	1.33	1.16-1.53	<0.0001	1.08	0.92-1.27	0.322	1.09	0.93-1.29	0.256	0.81	0.65-1.02	0.075
Daily alcohol consumption, yes versus no	1.00	0.87-1.15	0.923	0.99	0.84-1.17	0.940	0.99	0.84-1.17	0.941	0.92	0.73-1.16	0.506
BMI, ≥25 kg/m2 versus <25 kg/m2	1.24	1.12-1.37	<0.0001	1.13	1.01-1.28	0.032	1.12	0.99-1.26	0.054	0.96	0.81-1.13	0.627
Level of physical activity, sufficiently versus insufficiently active	1.04	0.94-1.14	0.405	0.99	0.88-1.11	0.869	1.01	0.90-1.14	0.740	1.04	0.89-1.22	0.585
Chemotherapy, yes versus no	1.88	1.68-2.10	<0.0001	1.19	1.03-1.36	0.013	1.16	1.01-1.33	0.029	0.99	0.81-1.20	0.928
Endocrine therapy, yes versus no	1.15	1.01-1.31	0.026	1.19	1.02-1.39	0.023	1.17	1.01-1.37	0.032	1.23	1.00-1.53	0.046
Use of ovarian function suppression, yes versus no	0.98	0.70-1.37	0.912	1.33	0.91-1.93	0.128	1.37	0.95-1.98	0.091	1.03	0.63-1.67	0.885
Anti-HER2 therapy, yes versus no	0.99	0.84-1.15	0.913	1.10	0.91-1.34	0.299	1.14	0.94-1.38	0.181	1.28	1.02-1.62	0.031
Radiotherapy, yes versus no	0.79	0.64-0.97	0.029	0.92	0.73-1.17	0.533	0.93	0.74-1.17	0.546	1.01	0.72-1.42	0.944
Axillary surgery, dissection versus sentinel/no	1.08	0.96-1.21	0.171	1.00	0.87-1.15	0.977	1.02	0.89-1.17	0.751	1.09	0.90-1.32	0.332
Breast surgery, mastectomy versus conservative	1.99	1.74-2.28	<0.0001	1.05	0.89-1.24	0.502	1.04	0.88-1.22	0.629	1.09	0.88-1.35	0.406
Anxiety (at T1), doubtful/case versus noncase	1.80	1.62-1.99	<0.0001	0.91	0.80-1.03	0.152	0.91	0.80-1.04	0.181	0.99	0.84-1.17	0.959
Depression (at T1), doubtful/case versus noncase	2.00	1.72-2.34	<0.0001	1.66	1.40-1.97	<0.0001	1.69	1.43-2.00	<0.0001	1.94	1.50-2.51	<0.0001
Vaginal dryness (at T1), yes versus no	1.16	1.05-1.29	0.002	0.73	0.64-0.82	<0.0001	0.73	0.64-0.83	<0.0001	1.82	1.55-2.14	<0.0001
Hot flashes (at T1), yes versus no	1.22	1.10-1.35	<0.0001	1.02	0.90-1.15	0.694	1.01	0.89-1.14	0.819	1.20	1.01-1.43	0.037
Poor role functioning (at T1), yes versus no	0.89	0.75-1.06	0.204	1.04	0.85-1.27	0.700	1.05	0.86-1.29	0.578	0.97	0.73-1.29	0.857
Poor social functioning (at T1), yes versus no	1.65	1.39-1.97	<0.0001	1.22	0.98-1.50	0.063	1.17	0.95-1.45	0.117	1.24	0.92-1.66	0.147
Severe fatigue (at T1), yes versus no	1.48	1.31-1.67	<0.0001	1.07	0.92-1.25	0.318	1.11	0.95-1.28	0.163	1.21	0.99-1.48	0.053
Severe pain (at T1), yes versus no	1.31	1.14-1.50	<0.0001	1.13	0.96-1.34	0.123	1.13	0.96-1.33	0.138	1.12	0.90-1.40	0.302
Severe insomnia (at T1), yes versus no	1.16	1.04-1.29	0.005	1.13	1.00-1.29	0.050	1.13	1.00-1.29	0.048	1.17	0.98-1.40	0.081
Consultation with a gynecologist (at T1), yes versus no	0.99	0.89-1.10	0.919	0.74	0.65-0.84	<0.0001	0.75	0.66-0.86	<0.0001	0.81	0.69-0.95	0.011
Consultation with a psychiatrist/psychologist (at T1), yes versus no	1.11	0.97-1.27	0.099	1.11	0.94-1.31	0.205	1.10	0.94-1.30	0.214	0.96	0.78-1.17	0.698
Sexual concern (at T0), yes versus no	3.48	3.11-3.89	<0.0001	9.94	8.84-11.1	<0.0001	9.75	8.67-10.9	<0.0001	3.96	3.34-4.69	<0.0001

BMI, body mass index; CI, confidence interval; HER2, human epidermal growth factor receptor 2; OR, odds ratio; T0, at the moment of diagnosis; T1, year 1 after diagnosis.

Sexual inactivity

Around 40% of the patients declared themselves as sexually inactive at T0, 37.6% at T1, 38.2% at T2, and 46.3% at T3. Older patients had higher chances of self-reporting themselves as sexually inactive (OR 5.55, 95% CI 4.16-7.69). Being in a couple (OR 2.29, 95% CI 1.95-2.68) and lower income (OR 1.19, 95% CI 1.36-1.04) were associated with higher chances of declaring sexual inactivity as well as the receipt of adjuvant chemotherapy (OR 1.16, 95% CI 1.01-1.33), ET (OR 1.18, 95% CI 1.01-1.37), a doubtful or clinically suggestive diagnosis of depression at T1 (OR 1.69, 95% CI 1.43-2.00), and severe insomnia at T1 (OR 1.13, 95% CI 1.00-1.29). Self-reported sexual activity was associated with vaginal dryness (OR 0.73, 95% CI 0.64-0.83) and having a consultation with a gynecologist in the past year (OR 0.76, 95% CI 0.67-0.86). Sexual inactivity at baseline was strongly associated with persistent sexual inactivity between diagnosis and year 4 (OR 9.75, 95% CI 8.68-10.96; Table 2).

Sexual functioning

Poor sexual functioning was reported by 39.3% of the patients at T0, 37.6% at T1, 38.2% at T2, and 40.3% at T3. Sociodemographic factors associated with higher chances of reporting poor sexual functioning were older age (>65 years versus <40 years; OR 5.88, 95% CI 4.34-8.33), being in a couple versus single (OR 2.20, 95% CI 1.90-2.60), and lower financial income (<1500 euros versus ≥3000 euros; OR 1.16, 95% CI 1.33-1.02). Patients with a body mass index >25 kg/m² had higher chances of reporting poor sexual functioning (OR 1.13, 95% CI 1.01-1.28). The use of adjuvant chemotherapy (OR 1.19, 95% CI 1.03-1.37) and ET was also associated with poor sexual functioning (OR 1.19, 95% CI 1.02-1.39). A doubtful or clinically suggestive diagnosis of depression (OR 1.66, 95% CI 1.40-1.97) at T1 was significantly associated with poor sexual functioning at T2. Consulting with a gynecologist in the past year was a protective factor for poor sexual functioning reporting (OR 0.74, 95% CI 0.65-0.85). Poor sexual functioning at diagnosis (T0) was strongly associated with persistent poor sexual functioning between diagnosis and year 4 (OR 9.94, 95% CI 8.84-11.18; Table 2).

Sexual enjoyment

Poor sexual enjoyment was reported by 29.2% of the patients at T0, 39.7% at T1, 38.1% at T2, and 37.4% at T3. Older age (>65 years versus <40 years; OR 2.04, 95% CI 1.40-2.94) and high educational level (OR 1.58, 95% CI 1.09-2.30) were sociodemographic factors associated with higher chances of reporting poor sexual enjoyment. Use of adjuvant ET (OR 1.23, 95% CI 1.00-1.53), the presence of vaginal dryness at T1 (OR 1.82, 95% CI 1.55-2.14), hot flashes (OR 1.20, 95% CI 1.01-1.43), and a possible diagnosis of depression (OR 1.94, 95% CI 1.50-2.51) at T1 were significantly associated with poor sexual enjoyment. Having a consultation with a gynecologist in the past year (OR 0.81, 95% CI 0.69-0.95) was associated with a lower chance of reporting poor sexual enjoyment. Poor sexual enjoyment at diagnosis (T0) was strongly associated with persistent poor sexual enjoyment between diagnosis and year 4 (OR 3.96, 95% CI 3.34-4.69; Table 2).

Discussion

This study included 7985 patients with early-stage BC from a prospective longitudinal cohort, followed for over 4 years since diagnosis. We found that 78% of patients reported at least one sexual concern over the study period and identified several common as well as outcome-specific associations, spanning several sociodemographic, treatment, and toxicity-related characteristics. Among other findings, we observed that poor body image was the most common sexual concern, being reported by over 50% of patients 2 years after diagnosis, particularly by patients that had lower education, were current smokers, or had received more intense treatment including more extensive surgery, chemo, or ET. In addition, this study highlights a gap between frequent patient complaints and the relatively low use of key services to address these sexual concerns (e.g. only 11.4% of patients with poor sexual functioning received psychosocial counseling and 45.9% had seen a gynecologist, with a specific sexologist consultation reported only by 0.2%).

Sexual health is defined by the World Health Organization (WHO) as ‘a state of physical, emotional, mental and social well-being in relation to sexuality [and] not merely the absence of disease, dysfunction or infirmity’²⁰ and it is considered a human right. Our findings stress the notion that sexual health and related concerns are a major need among BC survivors, which seems to be largely unaddressed.

Several important points should be highlighted:

Sexual concerns are frequent from diagnosis throughout the whole after-cancer trajectory. Our data are consistent with prior literature in the field. Concerns related to sexual health may occur in up to 30%-100% of female cancer survivors,²¹ with a meta-analysis suggesting a rate reaching 70% among survivors of BC.²² Specific retrospective or cross-sectional studies were carried out among BC survivors with rates of sexual concerns varying from 50% to 75% in both high- and low- and middle-income countries.^10,23, 24, 25, 26 This is comparable with what was observed in our longitudinal cohort, with up to 78% of patients reporting at least one type of sexual concern at any time point studied.

Sexual concerns were prevalent before receiving any treatment for BC. Up to 40% of patients in our cohort were declared to be sexually inactive at diagnosis which is not surprising because sexual concerns are prevalent in the general female population and therefore at diagnosis.^10,27 Importantly, in our study, the presence of these preexisting sexual concerns was strongly associated with persistent sexual concerns at 4 years’ follow-up for all studied domains, illustrating the fact that untreated sexual concerns typically persist or may also worsen over time, as validated by our findings.^5,28,29

Our study stresses the multitude of factors associated with sexual concerns driven by a confluence of biological (age, treatment-related side-effects, comorbidities), psychological (emotional distress), socioeconomic (educational level and income), physical (mastectomy, high body mass index), and interpersonal (partnered versus single) factors.

Regarding treatment-related side-effects, it is well known that estrogen deprivation due to menopause is associated with sexual concerns in the overall female population.³⁰ In BC survivors, menopause-related symptoms are often induced or augmented with the use of adjuvant ET. Specific adverse events of ET may include hot flashes, insomnia, vulvovaginal problems, reduced arousal, and libido.³¹ Our data demonstrate an association between adjuvant ET and self-reported sexual concerns as well as specific ET side-effects with some sexual health concerns. Importantly, due to the low number of patients using ovarian function suppression in our cohort, it is likely that our analysis underestimates the effect of ET in younger patients, for whom such regimen is known to add considerable impairment in several QoL domains including sexual health.^32,33 When prescribing this highly effective therapy, it is important that physicians keep in mind its impact on sexual health, offering patients validated strategies to mitigate its side-effects.³¹

Another consistently associated factor with sexual concerns is emotional distress, expressed as either anxiety or depression; proper screening and management of distress and psychosocial needs are recommended per international guidelines, although not yet systematically implemented.³⁴ Additional domain-specific risk factors were identified including some that are actionable and modifiable.

Remarkably, our data document the gap between reported sexual concerns and supportive care use, suggesting that this is a largely unaddressed concern.^5,35, 36, 37 Our results likely reflect known barriers to optimal management of sexual health, mainly lack of patient and physician empowerment and knowledge on sexual health and unavailability or uncoordinated referral processes for multidisciplinary supportive care strategies.

Several studies have shown that health care professionals face difficulties talking about sexual concerns with patients, while patients expect health care providers to be the ones to bring up the topic.^5,27,38 Clinicians have often reported feeling unprepared, having limited time, or facing gender and cultural concerns about ‘offending’ the patient when speaking about sexual health.35, 36, 37 In addition, a lack of available resources, funding, and reimbursement for sexual health professionals in cancer centers has been reported.³⁹ As a result, it is estimated that fewer than half of women with BC receive any communication on sexual health over their cancer care journey.35, 36, 37

Our finds call for the implementation of a proactive approach to ensure the evaluation and management of sexual health at BC diagnosis and continuously across the whole cancer care continuum regardles of age, educational or income level.⁴⁰ This approach should include (i) continuous screening and normalization, empowering clinicians and patients with proper education. Several initiatives have been proposed to facilitate communication around sexual health, including the use of leaflets,⁴¹ checklists,⁴² and short questionnaires⁴³ during consultation, although very few studies have evaluated the effectiveness of such approaches. Both in-person⁴⁴ and technology-delivered⁴⁵ specific training in sexual health communication seem feasible and acceptable for clinicians with promising results in improving communication skills. (ii) Coordinated referral processes to offer patients evidence-based strategies targeting specific sexual concerns⁴⁶ (Table 3), allowing a multidisciplinary approach with a dedicated team including psychologists, sexologists, gynecologists, and supportive care specialists.47, 48, 49, 50, 51, 52 It is important that oncologists and program directors actively work to foster team medicine within their centers to build this network.⁵³ Promoting periodic institutional educational activities with practical case discussions, identification of engaged health care professionals, development of symptom-management decision aids, and referral processes may help initiate this multidisciplinary dynamic.⁵⁴ (iii) Addressing modifiable risk factors for sexual concerns as appropriate, including the screening and management of emotional distress and weight balance,⁵⁵ and supporting patients with nonmodifiable risk factors such as the use of adjuvant therapies by offering them strategies to mitigate its treatment-related side-effects. (iv) Assess and reassess sexual health across the whole cancer care continuum at varied points of the patient journey to address any changes.⁴⁶

Table 3.

Effective interventions for managing sexual health in breast cancer survivors

Body image distress	Sexual response (desire, arousal, or orgasm)	Genital symptoms	Intimacy/relationships
Psychosocial and/or psychosexual counseling -Couples-based interventions if partnered -Self-compassionate writing	Psychosocial and/or psychosexual counseling, including cognitive behavioural therapy Regular stimulation (including masturbation)	Vaginal and/or vulvar atrophy such as dryness: -Lubricants for sexual activity or touch in addition to vaginal moisturizers -Temporary, low-dose vaginal estrogens (gynecologist evaluation + discussion of risks and benefits) -Lidocaine for persistent introital pain and dyspareunia (gynecologist evaluation)	Psychosocial and/or psychosexual counseling, including cognitive behavioural therapy Couples-based interventions
		Vaginismus/vaginal stenosis: -Vaginal dilators
		Urinary tract symptoms -Pelvic floor (Kegel) exercises -Cognitive behavioral therapy -Referral to a urologist/urogynecologist
		Pelvic floor dysfunction/ pelvic pain -Pelvic floor physiotherapy
		Vaginal discharge/leakage -Skin protectants/sealants
Patient education and symptom management advice should be offered to all patients based on the patient’s diagnosis and potential sexual health determinants: -Concomitant symptoms: pain, fatigue, and depression/anxiety -Social aspects: finances -Treatment-induced imaged changes: alopecia, nail changes, and breast surgery -Weight management

Adapted from Carter et al.⁴⁶

Digital health may facilitate the delivery of this proactive approach. The use of electronic patient-reported outcomes has been highlighted as a useful and simple tool for implementing a continuous screening and evaluation of sexual health in clinical practice.^56,57 In addition, digitally delivered interventions have proven to be effective in managing sexual concerns (e.g. cognitive behavioral therapy⁵⁸) and improving patients' ability to communicate sexual problems to the care team⁵⁹ and may complement in-person resources.

Our study has several strengths, including its prospective design, with longitudinal data collection of a large representative cohort. We controlled for several patients, context, and tumor-related factors to reduce confounding bias, and we used a large sample with adequate power. In addition, the CANTO cohort provided a unique opportunity to study the impact of other aspects beyond clinical and treatment data such as social, behavioral, and concomitant symptoms on sexual health. Limitations are the absence of a noncancer control group, the absence of detailed information regarding the reason for a patient not consulting a supportive care specialist, and some attrition over the course of follow-up. In addition, because no formal threshold for the definition of sexual concerns with the use of EORTC-QLQ-B23 exists, the possibility that our results may under- or overestimate sexual concerns should be considered, even though our threshold selection was based on a recent publication of real-world reference scores of EORTC-QLQ-B23 of considerable sample size.⁶⁰ Multicultural factors such as ethnicity and religion were not assessed according to the French legislation. Partner-related issues and sexual orientation may also play an important role in sexual health and were not specifically addressed in this study. Finally, it is possible that the investigated sexual health items could be influenced by the cultural and social environment of France, and results could potentially vary if the study was elsewhere conducted.

In conclusion, a considerable proportion of women experience sexual concerns before and after a BC diagnosis with the rate of reported sexual problems increasing or persisting over time, being associated with an underutilization of key supportive care services. There is a major need to improve care delivery in this domain empowering and educating patients as well as habilitating health care professionals to proactively evaluate and manage such concerns, preferably referring patients to a multimodal approach that includes pharmacological, psychological, behavioral, and social care strategies.⁴⁶