Long-term quality of life in non-epithelial ovarian cancer survivors: TMRG-GINECO-Vivrovaire rare tumors case–control study Authors: Coraline Dubot, Isabelle Ray-Coquard, Justine Lequesne, Patricia Pautier, Tiphaine Renaud, Frédéric Selle, Dominique Berton-Rigaud, Sophie Frank, Thibault De La Motte Rouge, Elsa Kalbacher, Magali Provansal, Cyriac Blonz, Hubert Orfeuvre, Jérôme Alexandre, Paule Augereau, Cédric Nadeau, Jean-Emmanuel Kurtz, Brunhilde Hanvic, Brice Fresneau, Djihane Ahmed-Lecheheb, François Christy, Jean-Michel Grellard, Bénédicte Clarisse, François Gernier, Florence Joly Published: 2025 ABSTRACT AbstractBackgroundPatients treated for rare ovarian germ cell tumors (GCT) and sex cord stromal tumors (SCST) often experience long-term survival. Treatments include surgery with or without chemotherapy (CT), which can cause late side effects that negatively impact quality of life (QoL). This study assessed long-term QoL among GCT and SCST survivors compared to age-matched healthy women (HW), focusing on patients treated with CT.MethodsCancer-free GCT and SCST survivors (≥ 2 years post-treatment) were recruited from the INCa French Network TMRG. HW were recruited via the “Seintinelles” platform. Participants completed validated questionnaires assessing QoL (FACT-G/FACT-O), fatigue (MFI), anxiety/depression (HADS), insomnia (ISI), neurotoxicity (FACT/GOG-NTX), cognition (FACT-COG), and sexuality (FACT-O OCS). Clinically relevant differences were defined as ≥ 5% between groups.ResultsA total of 144 survivors (99 GCT and 45 SCST) were included and compared with 284 matched HW (median age: 35.2 years). The median time from the end of treatment was 6 years [Q1 3.5–Q3 9.3]. At inclusion, 42% of patients were menopausal compared to 17% of HW (p < 0.001), with reduced sexual QoL. Survivors reported more cognitive impairment (32% vs. 21%, p = 0.025) and severe neurotoxicity (23% vs. 8%, p < 0.001). QoL, fatigue, anxiety/depression, and insomnia scores were similar for patients and HW.ConclusionsWith a median of 6 years after CT, most GCT and SCST survivors reported similar overall QoL compared to HW but faced premature menopause, reduced sexual QoL, and persistent cognitive and neurotoxic effects.Trial registration numberNCT03418844.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12916-025-04540-x.Keywords: Germ cell ovarian neoplasms, Sex cord stromal tumors, Long-term effects, Survivorship, Chemotherapy, Long-term fatigue, Quality of life, Menopause, Sexuality Background Germ cell tumors (GCT) and sex cord stromal tumors (SCST) are rare ovarian tumors that mainly affect young women, with early stage and good prognosis [1, 2]. Treatment includes surgery (with fertility-sparing for young women) combined with chemotherapy [bleomycin, etoposide, and platinum (BEP)], depending on the stage and prognostic factors [3]. Approximately 2/3 of these tumors are limited to stage I at diagnosis. Therefore, unilateral salpingo-oophorectomy with conservation of the contralateral adnexa and uterus is recommended in women who desire future fertility [4]. Patients with advanced stage are recommended to be managed with debulking surgery, sometimes associated with neoadjuvant chemotherapy. Recommended adjuvant chemotherapy consists of 3 to 4 courses of BEP. A small percentage of patients have persistent disease, progress during treatment, or have a recurrence after chemotherapy [5]. Some recurrences that are amenable to surgery may offer prolonged survival, so any resectable residual disease should be removed. Treatment failures are rare and categorized as platinum-resistant (progression during or within 4–6 weeks) or platinum-sensitive (recurrence beyond 6 weeks of initial therapy). Very few data exist regarding treatment of resistant patients, and most information is extrapolated from testicular cancer reports. Long-term follow-up of testicular cancer survivors treated with the same chemotherapy found that chemotherapy side effects and factors associated with cancer such as stress, anxiety, and depression had a negative impact on testicular cancer survivors’ quality of life (QoL) across the physical, psychological, sexual, and social domains [6, 7]. Higher chronic cancer-related fatigue among long-term testicular cancer survivors (17%) compared with general population (10%) (p < 0.001) has been described [8]. In a longitudinal study exploring fatigue in 812 testicular cancer survivors, the prevalence of long-term fatigue increased significantly over time. After 19 years of follow-up, 27% of patients reported fatigue and the risk of long-term fatigue was increased in patients with high levels of peripheral neuropathy, anxiety, and depression [9]. Very few studies have focused on the impact of chemotherapy on the general health and different domains of QoL of ovarian GCT and SCST survivors [10–13]. Most of them were performed without a comparative control group [11–13], not with a large comprehensive assessment of long-term QoL and symptoms. No previous study focused on long-term fatigue and specific aspects of QoL such as cognition. This study aimed to describe exhaustively the long-term side effects and the QoL of ovarian GCT and SCST survivors after treatment with surgery (with or without fertility-sparing) and chemotherapy, treated homogeneously in a large national dedicated cancer network for rare gynecological cancers, in comparison with healthy women (HW). Methods This is a multicenter case–control study using the French network for rare malignant ovarian tumors TMRG (Tumeurs Malignes Rares Gynécologiques) (NCT03418844). Network organization was already published and include a systematic histological review, dedicated multidisciplinary tumor board to propose initial standard of care management, and options at relapse [14]. Study design has been previously published [15]. Eligible patients have been recruited from the INCa French network TMRG. They have been selected in 19 of the 22 centers of the French cooperative GINECO group (Groupe d’Investigateurs Nationaux pour l’Étude des Cancers Ovariens et du sein). HW have been recruited from the Seintinelles® network (https://www.seintinelles.com/home), which is a collaborative research platform linking researchers and citizens to accelerate cancer research. Female cancer-free volunteers have been contacted to complete online self-questionnaires. This study has included ovarian GCT or SCST patients treated with surgery ± chemotherapy who were cancer-free at ≥ 2 years post-treatment (most recent treatment may have been conducted at primary diagnosis or at relapse). This study has been conducted in two phases to assess the following: i)“Part 1”: long-term fatigue and several domains of QoL “Part 1”: long-term fatigue and several domains of QoL ii)“Part 2”: the long-term medical side effects of chemotherapy with a focus on cardiovascular and pulmonary disorders “Part 2”: the long-term medical side effects of chemotherapy with a focus on cardiovascular and pulmonary disorders Here we present the results of study “Part 1” assessing long-term fatigue and QoL among patients treated with chemotherapy. Primary objective was to assess long-term fatigue in survivors treated for ovarian GCT or SCST with surgery and chemotherapy, as compared with matched healthy women (HW). Survivors treated for ovarian GCT or SCST with surgery and chemotherapy were matched with cancer-free HW issued from the Seintinelles network. Matching (2:1 nearest neighbor propensity score matching without replacement) was adjusted on age (± 2 years) and level of education (primary/middle school, secondary school, and higher education). All participants completed self-reported questionnaires including standardized and validated instruments previously detailed in study protocol [15]. Fatigue Questionnaire (Multidimensional Fatigue Inventory: MFI-20) [16, 17]; Functional assessment of cancer therapy scale general (FACT-G) and module for ovarian cancer (FACT-O) [18]; Hospital Anxiety and Depression Scale (HADS) [19]; Peripheral Neurotoxicity Questionnaire (FACT/GOG-NTX) [20]; FACT-Cog Questionnaire [21]; Insomnia Severity Index (ISI) [22]; Physical Activity Questionnaire (IPAQ) [23]; and a specific questionnaire adapted from the “living conditions” questionnaire used in previous studies [24]. A minimal 5% difference of scores between groups was considered as clinically relevant [25]. HADS score ≥ 11 indicated elevated anxiety or depression. ISI score 22–28 indicated severe clinical insomnia. Severe neurotoxicity was defined by Ntx score < 33 [26]. Perceived cognitive impairment (PCI) categories were issued from the FACT-Cog normative data from a healthy population. Clinically significant symptoms of cognitive complaints were defined as ratings lower than the 10th percentile of a normative sample [27–29]. Patients were asked to return completed questionnaires anonymously. For the HW control group, the Seintinelles network published the study information, and the questionnaires were completed online. Items in the questionnaires administered to patients aimed to compare their QoL prior to cancer diagnosis with their QoL at the time of the survey. These questionnaires were accordingly adapted for the HW by assessing their QoL 3 years before the survey in comparison to their QoL at the time of the survey. Missing questionnaire data were imputed using the mean of the available items when more than 50% of the items were completed by the patient at the corresponding time point. Patient’s medical data were collected from patient records. Menopausal status was obtained from self-reported questionnaires and corrected when discrepancies with medical records were identified (e.g., after radical surgery). Categorical variables are presented as frequencies and percentages and were compared using the chi-square test (or Fisher exact test if frequencies < 5), whereas continuous variables are displayed as mean ± SD and were compared using the Student t-test (or Wilcoxon-Mann–Whitney test for non-gaussian distributions). All tests were 2-tailed, and a p value < 0.05 was considered to indicate statistical significance. A sensitivity analysis was performed on the patient population, excluding those who experienced a disease recurrence. The association between fatigue and group (patients vs. HW) was assessed using multivariable linear models adjusted on confounding factors identified in univariate analysis. The association between cognitive impairment and group, and between severe neurotoxicity and group were assessed using multivariable logistic regression models adjusted on confounding factors identified in univariate analysis. Subgroup analyses depending on histological type and treatment options (type of surgery and duration of chemotherapy) have been performed. The results are briefly presented here and will be discussed in more detail in a subsequent manuscript. Results Among 231 ovarian GCT or SCST survivors who were included from 19 centers, 144 (99 GCT, 45 SCST) were treated with surgery and chemotherapy (≥ 1 cycle) and available for matching with 284 HW (Fig. 1). Patients’ and HW’s characteristics are summarized in Table 1. Patients were young (median age at diagnosis 26.1 years), had mostly FIGO stage I/II tumors (n = 76, 52%), and treated with fertility-sparing surgery for half of them (n = 73, 51%). Most SCST patients were treated with radical surgery (n = 31 (69%)), whereas GCT patients mostly had fertility-sparing surgery (n = 60 (61%)). Median BMI was higher for patients than HW (24.2 [20.5–28.3] vs. 22.7 [20.3–26.1], respectively; p = 0.01, data not shown). Menopause was more frequent in patients than HW (42% vs. 17%, p < 0.001), with younger age at menopause (median age at menopause of 42 years [IQR: 31.0–49.5] vs. 50 years [IQR: 48.0–53.0], respectively, p < 0.001). Patients had more premature ovarian insufficiency (POI: menopause age ≤ 40 years) (24 patients (44%) vs. 2 HW (4%), respectively, p < 0.001, data not shown). Among patients with conservative surgery, 8 (12%) were post-menopausal with a median age of menopause of 42.0 years [IQR: 31.0–45.0] (data not shown). Differences in menopausal status were observed according to histological subtypes: 27% of GCT survivors were post-menopausal compared to 6% of HW (p < 0.001), with a median age at menopause of 36 vs. 50 years (p < 0.001), whereas menopause was reported in 73% of SCST survivors compared to 44% of HW (p = 0.004), with a younger age at menopause (45 vs. 50 years, p = 0.001). Flow-chart of patient selection *Matching (2:1 by nearest neighbor propensity score matching) on age and level of education. QoL: quality of life Table 1Patients and healthy women (HW) characteristicsPatientsHW(N = 144)(N = 284)Age (years) median (min–max)35.2 (18.7–71.3)35.0 (18.0–71.0)Age at diagnosis (years) median (min–max)26.1 (4.75–64.7)–Body mass index (BMI), n (%) BMI ≥ 3024 (17%)26 (9%) 25 ≤ BMI < 3039 (27%)61 (21%) BMI < 2580 (56%)197 (69%)Education, n (%) University degree80 (56%)178 (63%) Highschool34 (24%)80 (28%) Primary/middle school30 (21%)26 (9%)Menopausal statusa Post-menopausal, n (%)57 (42%)49 (17%) Age at menopause (years) median (min–max)42 (12.0–57.0)50 (33.0–58.0)Time since the end of treatment (years) median [IQR]6.0 [3.5–9.3]FIGO stage, n (%) I64 (44%) II12 (8%) III/IV39 (27%) Unknown29 (20%)Histology, n (%) Germ cell99 (69%) Sex cord stromal Granulosa38 (26%) Sertoli-Leydig6 (4%) Otherb1 (1%)Surgeryc, n (%) Conservative surgery73 (51%) At primary diagnosis72 (50%) At relapse1 (1%) Non-conservative surgery70 (48%) At primary diagnosis3 (2%) At relapse67 (46%)Chemotherapy (CT), n (%) CT at primary diagnosis119 (83%) Relapse treated with CT34 (24%) Previous CT at primary diagnosis9 (6%) No previous CT25 (17%) Type of CT BEP (or EP)138 (96%) Other CTd6 (4%) Lines of CT 1 line132 (92%) 2 lines12 (8%)aMenopausal status: 8 missing data and 3 status unknown for patient’s groupbHistological type: 1 patient with non-epithelial rare ovarian tumor which was not germ cell and either sex cord stromal tumorcConservative surgery includes ovarian and uterine conservative surgery; non-conservative surgery includes at least hysterectomy and bilateral adnexectomy ± supplemental resections (when needed)dSome patients did not receive BEP (or EP) at any line of treatment. For these patients, other CT type was as follows: carboplatin + paclitaxel (n = 3), carboplatin monotherapy (n = 1), VeIP (vinblastine, ifosfamide, and cisplatin) (n = 2) Patients and healthy women (HW) characteristics aMenopausal status: 8 missing data and 3 status unknown for patient’s group bHistological type: 1 patient with non-epithelial rare ovarian tumor which was not germ cell and either sex cord stromal tumor cConservative surgery includes ovarian and uterine conservative surgery; non-conservative surgery includes at least hysterectomy and bilateral adnexectomy ± supplemental resections (when needed) dSome patients did not receive BEP (or EP) at any line of treatment. For these patients, other CT type was as follows: carboplatin + paclitaxel (n = 3), carboplatin monotherapy (n = 1), VeIP (vinblastine, ifosfamide, and cisplatin) (n = 2) Most patients received one line of CT (n = 132, 92%). CT was administrated at primary diagnosis for 119 (83%) of them. BEP CT was the most common regimen (n = 138, 96%). Patient characteristics based on disease relapse and treatment are described in Additional file 1: Table S1. Detailed CT regimen for the 12 patients treated with 2 lines of CT are synthetized in Additional file 1: Table S2. Patient-reported outcomes were compared between GCT or SCST survivors and matched HW in Table 2. Total score of fatigue was similar between patients and HW, with 48.3 vs. 46.4, respectively (p = 0.2), as well as for all fatigue subscores (general fatigue p = 0.06; mental fatigue p = 0.57; reduced activities p = 0.67; motivation p = 0.59). Levels of anxiety, depression, and insomnia were similar in both groups (p = 0.19, p = 0.53, and p = 015, respectively). FACT-G total score was also similar between patients and HW (p = 0.45). Statistical differences were observed in physical well-being in the FACT-G subcategories (p = 0.03) and QoL assessed by ovarian cancer subscale (p < 0.001), but this was not clinically relevant (difference in QoL scores < 5%). Statistically higher social well-being assessed in the FACT-G subcategories was observed in patients compared with HW (p < 0.001), but this was not clinically relevant either (difference in QoL scores < 5%). A sensitivity analysis was performed on the 110 patients who did not experience disease relapse and was compared with matched HW. The results were similar to the QoL observed in the overall patient population (data not shown). Table 2Patients and healthy women (HW) issues in quality of lifePatients (N = 144)HW (N = 284)∆QoL scorese (%)P valueFatigue (MFI French version)aTotal, mean (SD) (min–max)48.3 (14.8) (24–83)46.4 (14.6) (24–86)1.90.2 General/physical24.9 (7.25) (13–42)23.6 (7.25) (12–42)1.30.06 Mental13.4 (5.63) (6–28)13.1 (5.43) (6–30)0.30.57 Reduced activities6.3 (3.08) (3–15)6.2 (2.82) (3–14)0.10.67 Motivation3.7 (1.94) (2–10)3.6 (1.89) (2–10)0.10.59Insomnia (ISI), N (%) Moderate and severe43 (30%)65 (23%)–0.15Emotional status (HADS) N (%) Anxiety (HADS-A score ≥ 11)53 (37%)86 (30%)–0.19 Depression (HADS-D score ≥ 11)13 (9%)33 (12%)–0.53Quality of life (FACT-G score)bTotal, mean (SD) (min–max)81.2 (15.4) (30–107)80.1 (14.8) (31.8–104)1.00.48 Physical well-being23.2 (4.2) (10–28)24.2 (4.2 (5–28.0) − 0.90.032 Social well-being20.8 (5.5) (4.67–28)18.7 (5.4) (0–28)1.9 < 0.001 Emotional well-being17.3 (4.5) (2–24)17.5 (4.9) (0–24) − 0.20.68 Functional well-being19.9 (5.2) (6.00–28.0)19.7 (4.9) (0–28.0)0.20.8Ovarian cancer subscalebMean (SD) (min–max)33.0 (5.3) (18.7–44)34.8 (4.8) (14–44) − 0.8 < 0.001Sexual concernsc Less interest in sex45 (32%)34 (12%)– < 0.001 Feel less like a woman46 (32%)58 (21%)–0.013 Afraid not having childrend54 (40%)37 (19%)– < 0.001aHigher MFI score indicates higher fatiguebHigher quality of life scores indicate better functioning for the global health scale or functioning scalescQuality of life (FACT-O) itemsdMissing data: 8 for patient’s group, 97 for HW’s groupePercentage of differences observed in QoL scores between patients and HW Patients and healthy women (HW) issues in quality of life Fatigue (MFI French version)a Total, mean (SD) (min–max) Insomnia (ISI), N (%) Moderate and severe Quality of life (FACT-G score)b Total, mean (SD) (min–max) Ovarian cancer subscaleb Mean (SD) (min–max) aHigher MFI score indicates higher fatigue bHigher quality of life scores indicate better functioning for the global health scale or functioning scales cQuality of life (FACT-O) items dMissing data: 8 for patient’s group, 97 for HW’s group ePercentage of differences observed in QoL scores between patients and HW Patient’s sexual QoL, assessed by FACT-O items, was deteriorated as compared with HW, with patients having less interest in sex (32% vs. 12%, p < 0.001), feeling less like a woman (32% vs. 21%, p = 0.013), and being afraid not having children (40% vs. 19%, p < 0.001). Patients perceived more cognitive impairment than HW (median CogPCI score 59.5 vs. 61.8, p = 0.05; percentage of patients with clinically significant cognitive complaint 32% vs. 21%, p = 0.025, respectively) (Fig. 2A). Of note, the impact of cognitive impairment on QoL was greater in patients treated with multiple lines of chemotherapy (median cog QoL score of 13.0 (0–16.0) for patients treated with 1 CT line vs. 6.50 (3.00–16.0) for patients treated with 2 CT lines, p = 0.012, ∆QoL scores = 40%; data not shown). Analysis of cognition according to histological subtype found that GCT survivors experienced more cognitive impairment than HW (37% vs. 20%, respectively; p = 0.002), whereas SCST survivors had approximately similar cognitive impairment than HW (20% vs. 15%, respectively; p = 0.643). Cognition Impairment and severe peripheral neuropathy among patients and matched-HW. A Description of CogPCI score (box plot/left y-axis) and rate of cognitive complaint (histogram/right y-axis) in patients (red) and matched HW (blue). The boxplot shows data distribution: central lines are the median, the boxes represent the interquartile range, the dots are the mean, and whiskers extend up to 1.5 times the IQR. PCI category was issued from the French version of the Functional Assessment of Cancer Therapy—Cognitive Function (FACT-Cog), normative data from a healthy population [25]. B Description of FACT/GOG-NTX subscale score (box plot/left y-axis) and rate of severe peripheral neuropathy (FACT/GOG-NTX subscale score < 33 [24]) (histogram/right y-axis) in patients (red) and matched HW (blue). The boxplot shows data distribution: central lines are the median, the boxes represent the interquartile range, the dots are the mean, and whiskers extend up to 1.5 times the IQR Peripheral neuropathy was more frequent in patients than HW (mean FACT/GOG-NTX score 36.3 vs. 39.6, respectively, p < 0.001; percentage of patients with clinically significant severe peripheral neuropathy, defined by Ntx score < 33 of 23% vs. 8%, for patients and HW, respectively; p < 0.001) (Fig. 2B). Severe peripheral neuropathy was higher in SCST survivors than HW (36% vs. 7%, respectively, p < 0.001), whereas it was similar between GCT survivors and HW (16% vs. 9%; p = 0.09). Factors associated with long-term fatigue in both univariate and multivariable analyses are described in Table 3. After adjustment on age, education level, and factors associated with fatigue in univariate analysis, independent risk factors of fatigue were cognitive impairment (p < 0.001), insomnia (p < 0.001), anxiety (p < 0.001), low physical activity (p < 0.001), and high BMI (p = 0.002) (Table 3). Table 3Factors associated with fatigue and group effect after adjustment: results from univariate and multivariate modelsUnivariate analysisMultivariate analysisNBeta (95% CI)p valueBeta (95% CI)p valueAge4280.01 (− 0.04, 0.06)0.670.02 (− 0.03, 0.07)0.38Education428–––– Primary/middle school––––– Secondary school– − 1.1 (− 3.4, 1.2)0.34 − 0.1 (− 2.2, 2)0.92 Higher education– − 2.6 (− 4.6, − 0.46)0.017 − 0.96 (− 2.8, 0.91)0.31Patients or HW group428––––HW–––––Group–1.4 (− 0.08, 2.8)0.0641.1 (− 0.52, 2.7)0.18BMI4270.25 (0.11, 0.40) < 0.0010.20 (0.08, 0.33)0.002PCI cognitive impairmenta428––3.4 (1.9, 4.9) < 0.001Insomnia (ISI)4280.71 (0.61, 0.80) < 0.0014.8 (3.3, 6.2) < 0.001Anxiety (HADS)4270.82 (0.62, 1.0) < 0.0010.52 (0.34–0.71) < 0.001Depression (HADS)4270.39 (0.17, 0.61) < 0.0010.14 (− 0.11, 0.4)0.27Physical activity (IPAQ)413–––– Low––––– Moderate– − 0.95 (− 2.6, 0.74)0.27 − 0.91 (− 2.3, 0.48)0.20 High– − 3.8 (− 5.8, − 1.9) < 0.001 − 3.4 (− 5, − 1.7) < 0.001Menopausal status420–– Premenopausal––– Postmenopausal–0.43 (− 1.2, 2.1)0.60 Unknown– − 0.44 (− 3.9, 3.0)0.80aPCI category issued from the French version of the Functional Assessment of Cancer Therapy—Cognitive Function (FACT-Cog), normative data from a healthy population [25]bold are statistically significant (p<0.05) Factors associated with fatigue and group effect after adjustment: results from univariate and multivariate models aPCI category issued from the French version of the Functional Assessment of Cancer Therapy—Cognitive Function (FACT-Cog), normative data from a healthy population [25] bold are statistically significant (p<0.05) Factors associated with cognitive impairment in both univariate and multivariable analyses are described in Additional file 1: Table S1. After adjustment on age, education level, and factors associated with cognitive impairment in univariate analysis, independent risk factors for cognitive impairment were age (p = 0.032), being a survivor treated with surgery and CT (p = 0.007), insomnia (p = 0.021), and fatigue (p < 0.001) (Additional file 1: Table S3). Factors associated with severe peripheral neuropathy in both univariate and multivariable analyses are described in Additional file 1: Table S2. Especially, independent risk factors for severe neuropathy were anxiety (p = 0.07) and fatigue (p < 0.001) (Additional file 1: Table S4). Discussion This is one of the largest comparative studies exploring exhaustively long-term QoL in ovarian GCT and SCST survivors in comparison with HW. Although fatigue was similar in both groups, we reported that, after a median of 6 years from chemotherapy, patients had more cognition impairment, more severe peripheral neuropathy, and more premature menopause than HW. Scores of QoL described in our study are consistent with previous reports in non-epithelial ovarian cancer survivors [10, 11]. GCT and SCST survivors had similar long-term fatigue, as compared with HW. Long-term fatigue was associated with high BMI, low physical activity, and insomnia. No correlation was observed between fatigue and patient’s group, suggesting that most of patients have recovered from cancer-induced fatigue after a median of 6 years from chemotherapy. Patients reported anxiety and depression for 37% and 10% of them, respectively, levels that were similar to that of the HW. Our results shed new light on those previously reported in a non-comparative study of 150 GCT and SCST survivors where 34% of patients had severe mental distress (HADS total score ≥ 13), mostly those without fertility-sparing treatment [12]. Our study underlines the importance of a comparison with a control group. Of note, a trend was observed for higher social well-being in patients than HW. Although this observation could not be considered as clinically relevant in our study, this is consistent with previous reports [30]. These findings on enhanced social well-being are consistent across several studies involving cancer survivors (including testicular [31] and breast cancer [32]) and reflect underlying coping mechanisms. This suggests that GCT and SCST survivors, as described in other young cancer survivors [33, 34], were able to develop coping strategies and re-prioritize their new life. Taken together, these positive and negative variations across QoL subdomains may have influenced the overall scores, potentially masking global differences. This is the first study evaluating and describing cancer-related cognitive impairment (CRCI) in this population of GCT and SCST survivors. CRCI has been described in a variety of cancer types [35, 36], and notably in breast cancer [37] and epithelial ovarian cancer survivors [38] and testicular GCT survivors [39] however never evaluated among ovarian GCT and SCST, well known to be enriched in long-term survivors compared to other ovarian cancer [40]. CRCI was associated with age, fatigue, and insomnia. Interestingly, GCT survivors appeared to be more susceptible to CRCI than SCST survivors. To prevent the consequences of CRCI on GCT and SCST survivors QoL, special attention could be given to systematically assess cognition impairment, particularly in the young population of GCT survivors. Counseling patients to prevent sleep disorders and how to manage CRCI would be interesting to be integrated in survivorship care plans. Peripheral neuropathy was high with 23% of severe symptoms, consistent with those reported in male patients treated with BEP for testicular cancer: in a cohort of 1409 men with 10.7 years (range = 4–21 years) of follow-up, 29% (95% CI = 25% to 33%) reported symptoms of peripheral neuropathy [41]. In a cohort of 93 ovarian GCT survivors, at 15 years from diagnosis, severe neurotoxicity was observed in 10% of patients who received ≥ 4 cycles of platinum-based chemotherapy [42]. Peripheral neuropathy is also reported in epithelial ovarian cancer survivors treated with platinum and taxane, notably, in a study of 130 patients, 26.9% had severe neuropathy [43]. Evidence on peripheral neuropathy in SCST survivors is scarce. Our findings suggest they may be more susceptible to long-term severe peripheral neuropathy than HWs and GCT survivors. We found that risk factors for severe neuropathy were fatigue and anxiety. Correlation between fatigue and neuropathy has previously been described in a study of 812 testicular cancer survivors where the risk of chronic fatigue was increased 3-to-fourfold in patients with high levels of neuropathy compared with no neuropathy [9]. In a cross-sectional study of 623 adult cancer survivors, an association was found between perceived stress and chemotherapy-induced neuropathy [44]. Moreover, in a study of 111 women receiving neoadjuvant chemotherapy for breast cancer, pre-treatment anxiety was associated with chemotherapy-induced neuropathy (OR = 5.02; 95% CI, 1.84–13.70; p = 0.002) [45]. Evaluation and management of emotional distress, lifestyle recommendations to fight against fatigue in GCT and SCST survivors may provide potential strategies to prevent long-term neurologic sequalae. Although almost half of patients had fertility-sparing surgery, premature menopause was higher in patients as compared with HW. Premature menopause has previously been reported in GCT and SCST survivors from 3 to 5% [46, 47], suggesting the effect of treatments on long-term ovarian function. Rates of premature menopause in our study were high because we focused on patients treated with CT (in first line or subsequent lines), including patients who experimented disease recurrence requiring radical surgery. Furthermore, patients had worse sexual QoL and more reproductive concerns than HW, which is consistent with previous reports [10]. Given the young age of the population in our study (median age 35 years), these results rise concerns on the potential long-term impact on the cardiovascular, bone, and gynecological health of GCT and SCST survivors. Follow-up of testicular cancer survivors treated with BEP chemotherapy found persistent long-term side effects of chemotherapy such as cardiovascular disease, neurotoxicity, hypogonadism, and a higher risk of secondary cancer [6, 8, 41, 48, 49]. The few studies focusing on ovarian GCT and SCST survivors found hypertension, hypercholesterolemia, and reproductive concerns (50, 51, 52).. A way to prevent some of these long-term side effects is the use of hormonal replacement therapy (HRT) which is recommended for ovarian GCT and acceptable in rare case of SCST [4]. Unfortunately, the use of HRT is disparate among these population of survivors. Furthermore, chemotherapy did not demonstrate survival benefit in early-stage SCST (53). The use of chemotherapy in this population should be balanced with long-term side effects. For all these reasons, we intend to explore, in our study part 2, the long-term side effects of chemotherapy on ovarian GCT and SCST survivors. This multicentric study, focusing on rare tumors, has a large population and a long follow-up period. Comprehensive assessment of the different categories of QoL has been exhaustively performed. Another strength of this study is that it relies on 19 of the 22 centers of the French national TMRG network, covering the whole territory and linked together. The comparison with age-matched healthy women (HW) enhances the validity of the findings. Notably, the HW group was sourced from the “Seintinelles” network, which consists of volunteers who are informed about cancer and its risk factors, thereby introducing a potential selection bias in this control group. Limitations of the matching are evident with respect to educational level: the patient group included some very young women who were diagnosed with cancer while still studying, whereas the control group was generally more knowledgeable about cancer and had a higher educational level. Another limitation of the study is the lack of statistical power due to the small sample size of subgroups, which is inevitable when studying rare tumors. Assessment of QoL relied on self-reported outcomes, using multiple questionnaires to explore all dimensions of QoL. This approach is subject to the risk of missing data due to incomplete questionnaire responses from patients or HWs, as well as potential recall bias, particularly among HWs. Response shift bias may also affect this analysis based on self-reported outcomes, as the interpretation of certain constructs and items can change over time. Patients may understand and respond to these items differently as they undergo new life experiences. Although this study is not a longitudinal assessment of QoL over time, the evaluation of the different QoL domains is comprehensive and reported major information to prepare future long-term survivorship care plans for these patients. Conclusions For the first time, compared with healthy women, even if GCT and SCST survivors treated with chemotherapy have similar long-term global QoL, we observed persistent side effects. Fertility concerns and premature menopause as well as neurotoxicity and CRCI have a negative impact on their QoL and need to be detected and managed as early as possible. Supplementary Information Additional file 1. Tables S1–S4. Table S1 Characteristics of patients according to chemotherapylines. Table S2 Detailed chemotherapyregimen for relapsed patients treated with 2 lines. Legend: BEP: bleomycin etoposide cisplatin; VeIP: vinblastine, ifosfamide, and cisplatin; Intensification: intensified CT with autologous stem cell rescue. Table S3 Factors associated with cognition impairment and group effect after adjustment: results from univariate and multivariate models. Table S4 Factors associated with severe peripheral neuropathyand group effect after adjustment: results from univariate and multivariate models. Acknowledgements We are grateful to all the patients who have consented to participate. We acknowledge the ARCAGY-GINECO Intergroup (academic clinical research group specializing in gynecological oncology) for its scientific support, and the Data Processing Center (DPC) of the North West Canceropole (Centre de Traitement des Données du Cancéropôle Nord-Ouest) in charge of data management. We thank the Seintinelles network (Guillemette Jacobs) for the constitution of the HW group, and the TMRG (Centres Experts Tumeurs Malignes Rares Gynécologiques, Louise Mbengue, Amandine Charenton, Patricia Lambert) supported by INCa (Institut National du Cancer). We are grateful to all the study coordinators involved in the study (Marie Fernette for Baclesse Cancer Center, Caen, France) and Gilles Giraud (librarian, Baclesse Cancer Center, Caen, France) for bibliographic research. Abbreviations Bleomycin etoposide cisplatin Body mass index Confidence interval Chemotherapy-induced cognition impairment Chemotherapy Functional assessment of cancer therapy scale, module for cognition Functional assessment of cancer therapy scale, general Functional assessment of cancer therapy scale, Peripheral Neurotoxicity Questionnaire Functional assessment of cancer therapy scale, module for ovarian cancer International Federation of Gynecology and Obstetrics Germ cell tumors Groupe d’Investigateurs Nationaux pour l’Étude des Cancers Ovariens et du sein Hospital Anxiety and Depression Scale Hormone replacement therapy Healthy women Physical Activity Questionnaire Interquartile range Insomnia Severity Index Multidimensional Fatigue Inventory Perceived cognitive impairment Premature ovarian insufficiency Quality of life Sex chord and stromal tumors Tumeurs Malignes Rares Gynécologiques Authors’ contributions Conceptualization: FJ Methodology: JL Funding search: DAL, BC, FG Investigation: CD, IRC, PP, TR, FS, DBR, SF, TDLMR, EK, MP, CB, HO, JA, PA, CN, JEK, BH, BF, FG, FJ Data management : JMG Data analysis: FC, JL, FG, CD, FJ Writing: CD Revisions: CD, IRC, PP, TR, FS, DBR, SF, TDLMR, EK, MP, CB, HO, JA, PA, CN, JEK, BH, BF, FG, FJ, DAL, FC, JMG, BC. All authors read and approved the final manuscript. Funding This work is supported by the ARC Foundation for Cancer Research. In the context of this major external funding, the study protocol has undergone peer review by the funding body. The funding body played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript. Data availability The data that support the findings of this study are available on request from the corresponding author, Coraline DUBOT, upon reasonable request. Restrictions apply to the availability of these data due to ethical and privacy concerns. Declarations Ethics approval and consent to participate The study was approved by Ile de France V ethics committee on 10th January 2018, ID RCB: 2017-A03028-45. Informed consent form was signed by every patient who has been included in the study (inform consent form available on request). Consent for publication Not applicable. Competing interests Coraline Dubot: Payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events: GSK, AMGEN, MSD; Support for attending meetings and/or travel: Astra Zeneca, MSD, GSK, Eisai, Novartis, Lilly. Cyriac Blonz: receipt consulting/advisory fees (DAIICHI SANKYO, Lilly, MSD, Pfizer, Astra Zeneca). Jérôme Alexandre: Payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events: GSK, Astra Zeneca, MSD, Eisaï, Novartis; Support for attending meetings and/or travel: Astra Zeneca, MSD, Eisai. Frédéric Selle: Payment or honoraria for speakers’ bureaus, consulting or advisory role: Astra Zeneca, MSD, GSK-Tesaro, Eisai, Seagen, Abbvie. Sophie Frank: honoraria for lectures: Astra Zeneca; Support for attending meetings : Pharma And, Lilly, Pfizer. Florence Joly: Payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events: GSK, Astra Zeneca, MSD, Pharmand, Esai; Support for attending meetings and/or travel: GSK, ESAI, MSD. Patricia Pautier: Consulting fees: MSD, GSK; Payment or honoraria for lectures, presentations, speakers’ bureaus, manuscript writing or educational events: MSD, GSK, PharmaMar; Support for attending meetings and/or travel: PharmaMar, MSD, GSK, Roche, Astra Zeneca. Elsa Kalbacher: Grants or contracts from any entity: GSK, MSD, Astrazeneca, Seagen, Tesaro; Consulting fees: GSK, MSD, Astrazeneca, Seagen, Tesaro; Payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events: GSK, MSD, Astrazeneca, Seagen, Tesaro; Support for attending meetings and/or travel: GSK, MSD, Astrazeneca, Seagen, Tesaro, pharmamar, Esai, pharmaand. Thibaut De La Motte Rouge : Grants or contracts from any entity: MSD, Novartis, Pfizer; Consulting fees: Astra-Zeneca, pharmaand, Eisai, MSD, Natera, Pfizer, Roche, Sanofi, Seagen, GSK; Support for attending meetings and/or travel: Astra-Zeneca, MSD, Pfizer, Roche, GSK. Jean-Emmanuel Kurtz : Payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events: GSK, Johnson&Johnson; Support for attending meetings and/or travel: GSK, PharmaMar, MSD, AstraZeneca; Participation on a Data Safety Monitoring Board or Advisory Board: GSK Dominique Berton Rigaud, Tiphaine Renaud, Magali Provansal, Paule Augereau, Cédric Nadeau, Hubert Orfeuvre, Brunhilde Hanvic, Isabelle Ray-Coquard, Brice Fresneau, François Gernier, Djihanne Ahmed-Lecheheb, Benedicte Clarisse, Jean-Michel Grellard, Justine Lequesne, François Christy : none Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1. Tables S1–S4. Table S1 Characteristics of patients according to chemotherapylines. Table S2 Detailed chemotherapyregimen for relapsed patients treated with 2 lines. Legend: BEP: bleomycin etoposide cisplatin; VeIP: vinblastine, ifosfamide, and cisplatin; Intensification: intensified CT with autologous stem cell rescue. Table S3 Factors associated with cognition impairment and group effect after adjustment: results from univariate and multivariate models. Table S4 Factors associated with severe peripheral neuropathyand group effect after adjustment: results from univariate and multivariate models. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author, Coraline DUBOT, upon reasonable request. Restrictions apply to the availability of these data due to ethical and privacy concerns.