Article Text
PDF
Abstract
Objective To determine oncological outcomes and associated prognostic factors in women younger than 45 years diagnosed with non-epithelial ovarian cancer.
Methods A retrospective, multicenter Spanish study was performed including women with non-epithelial ovarian cancer younger than 45 years between January 2010 and December 2019. All types of treatments and stages at diagnosis with at least 12 months of follow-up were collected. Women with missing data, epithelial cancers, borderline or Krukenberg tumors, and benign histology, as well as patients with previous or concomitant cancer, were excluded.
Results A total of 150 patients were included in this study. The mean±SD age was 31.45±7.45 years. Histology subtypes were divided into germ cell (n=104, 69.3%), sex-cord (n=41, 27.3%), and other stromal tumors (n=5, 3.3%). Median follow-up time was 58.6 (range: 31.10–81.91) months. Nineteen (12.6%) patients presented with recurrent disease with a median time to recurrence of 19 (range: 6–76) months. Progression-free survival and overall survival did not significantly differ among histology subtypes (p=0.09 and 0.26, respectively) and International Federation of Gynecology and Obstetrics (FIGO) stage (I-II vs III-IV) with p=0.08 and p=0.67, respectively. Univariate analysis identified sex-cord histology with the lowest progression-free survival. Multivariate analysis showed that body mass index (BMI) (HR=1.01; 95% CI 1.00 to 1.01) and sex-cord histology (HR=3.6; 95% CI 1.17 to 10.9) remained important independent prognostic factors for progression-free survival. Independent prognostic factors for overall survival were BMI (HR=1.01; 95% CI 1.00 to 1.01) and residual disease (HR=7.16; 95% CI 1.39 to 36.97).
Conclusions Our study showed that BMI, residual disease, and sex-cord histology were prognostic factors associated with worse oncological outcomes in women younger than 45 years diagnosed with non-epithelial ovarian cancers. Even though the identification of prognostic factors is relevant to identify high-risk patients and guide adjuvant treatment, larger studies with international collaboration are essential to clarify oncological risk factors in this rare disease.
- Granulosa Cell Tumor
- Leydig Cell Tumor
- Sertoli-Leydig Cell Tumor
- Sex Cord-Gonadal Stromal Tumors
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Non-epithelial ovarian cancer is a rare malignancy and includes a heterogeneous group of tumors. Evidence regarding risk factors associated with oncological outcomes is scarce.
WHAT THIS STUDY ADDS
Similar 5-year progression-free survival and overall survival were found among different histological subtypes of non-epithelial ovarian cancer. The 5-year progression-free survival and overall survival were not different when early was compared with advanced International Federation of Gynecology and Obstetrics (FIGO) stage. Body mass index (BMI) and sex-cord histology are the most important independent prognostic factors for progression-free survival. Independent prognostic factors for overall survival were BMI and residual disease.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The prognostic factors identified in this study might help physicians decide whether women younger than 45 years with non-epithelial ovarian cancer might benefit from adjuvant treatment.
Introduction
Non-epithelial ovarian cancer is a rare malignancy1 arising from the primordial germ cells,2 from the sex-cord, and from the ovarian stroma. It includes a heterogeneous group of tumors, where granulosa cell tumors are the most frequent among non-epithelial ovarian cancer accounting for almost 50% of them.3 Even though these tumors are commonly diagnosed at an early stage4 with favorable prognosis compared with the majority of their epithelial counterparts, some concerns regarding the risk factors for relapse and survival remain.
As non-epithelial ovarian cancer involves a vast heterogeneity of tumors, survival varies among the different histologic subtypes. In recent decades, only a few studies have analyzed oncological outcomes in this group of rare ovarian tumors.5–7 In this regard, a Danish study recently showed that overall mortality was associated with year of surgery, age, and stage at diagnosis.3 However, the study included information on patients from a national database over a period of 40 years. Other studies, moreover, have analyzed different risk factors for relapse and death, but have focused on sex-cord5 or germ cell8 histologic subtypes. In addition, there is no description in the literature regarding other potential associated factors for survival such as body mass index (BMI), residual disease after surgery, size of primary tumor, and tumor histology. Therefore, a multicenter Spanish study was designed that aimed to determine the oncological outcomes and associated prognostic factors in women younger than 45 years diagnosed with non-epithelial ovarian cancer.
Methods
The institutional review board of each center approved this retrospective, multicenter, observational study. Patients younger than 45 years treated for non-epithelial ovarian cancer between January 2010 and December 2019 at 39 Spanish hospitals were included in the study. Patients who received any type of treatment and were diagnosed at all International Federation of Gynecology and Obstetrics (FIGO) stages with at least 12 months of follow-up were selected. Women with missing data, epithelial cancers, borderline or Krukenberg tumors, and benign histology, as well as patients with previous or concomitant cancer, were excluded. Missing data included absence of the following information: age, BMI, surgical residual disease, size of lesion, and lost to follow-up.
Individual records of all patients were reviewed by each investigator and all possible variables influencing oncological prognosis collected. Data registered included patients’ demographical characteristics, disease management such as type of surgical procedure and surgical approach (minimally invasive vs open surgery), histological findings, and newly diagnosed cancer after risk reduction salpingo-oophorectomy, tumor markers (CA125, alpha-fetoprotein (AFP), human gonadotropin hormone, B-inhibin, HE4, carcinoembryonic antigen (CEA), and CA 19.9), diagnostic imaging (pelvic ultrasound, pelvic magnetic resonance imaging (MRI), thorax-abdomino-pelvic computed tomography (CT) scan), and type of adjuvant chemotherapy.
Malignant non-epithelial ovarian cancers were classified according to the World Health Organization (WHO) in 2014.9 10 Tumors were staged following FIGO classification in 2014.11 Although a protocol of follow-up was not homogeneous across all institutions, patients were generally followed with pelvic examinations, vaginal/cervical Pap smear, CT scan, and MRI. The follow-up schedule was generally as follows: every 3 months for the first 2 years, every 6 months for the following 3 years, and yearly thereafter. Disease status was recorded at the most recent follow-up visit. Analyzed oncological outcomes were progression-free survival and overall survival. Risk factors associated with oncological outcomes included: age, BMI, type of surgery (fertility-sparing, surgical staging, cytoreduction), residual disease (0 mm vs >1 mm), size of largest lesion at diagnosis, FIGO stage (I-II vs III-IV), intra-operative ovarian tumor rupture (yes vs no), and histology subtype (germ cell, sex-cord, stromal tumor).
Statistical Analysis
Kolmogorov–Smirnov with Lilliefors correction was used to evaluate the normal distribution of the data of the collected variables. Whereas frequencies and proportions were used as summary statistics for categorical variables, mean (SD) or medians and interquartile (IQ) range were used for continuous variables.
Progression-free survival time was calculated from the date of diagnosis to the date of any recurrence, death, or the date of last visit, whichever occurred first. Overall survival time was calculated from the date of diagnosis to the date of death from any cause, or date of last follow-up. In the absence of events, survival was determined by the date of the last visit. Progression-free survival and overall survival were plotted using the Kaplan–Meier method. The log-rank test was used to assess survival differences between groups by univariate analysis. Multivariate Cox proportional hazards regression modeling was used to identify the prognostic clinical–pathological features independently associated with overall survival and progression-free survival. All the tests were two-sided and a p value ≤0.05 was considered statistically significant. The analyses were performed using STATA 15.
Results
After excluding 21 patients due to missing data, 150 patients were included in this study. The mean±SD age was 31.45±7.45 years. Median (IQR) BMI was 23.39 (range: 21.15–26.66) kg/m2. Histological subtype was divided in germ cells in 104 (69.3%) patients, sex-cord tumors in 41 (27.3%) patients, and other stromal tumors in 5 (3.3%) cases. Mean tumor size at surgery was 114.1±62.12 mm. Adjuvant chemotherapy was received by 74 (49.3%) patients. Detailed information on baseline characteristics of the population is shown in Table 1. Distribution of FIGO stages at diagnosis among different histological subtypes is detailed in Online supplemental material 1. Of note, 121 (80.6%) patients were diagnosed at early stages (FIGO I-II).
Supplemental material
Patients’ baseline characteristics (n=150)
A total of 82 (54.6%) patients underwent fertility-sparing surgery at diagnosis, 62% by minimally invasive surgery. Surgical staging by pre-operative apparent early-stage disease was performed in 49 (32.6%) patients; 41% by minimally invasive surgery. Surgical cytoreduction for advanced stages (FIGO III-IV) was performed in 20 (13.2%) women; 11 (55%) of these had a surgical tumor resection (Online supplemental material 2).
Supplemental material
Median (IQR) follow-up time was 58.58 (range: 31.10–81.91) months. Nineteen (12.6%) patients had recurrent disease with a median time to recurrence of 19 (range: 6–76) months. Five (3.3%) women died of the disease, while one case (0.6%) died due to another reason. The nine women with relapse who underwent fertility-sparing surgery received unilateral adnexectomy with or without peritoneal staging (omentectomy and biopsies). A detailed description of the women with relapsed disease is shown in Table 2.
Characteristics of patients with relapsed disease (n=19)
No significant differences (p=0.09) were found in the 5-year progression-free survival according to histological subtype, with rates of 92%, 82%, and 76% for germ cell, sex-cord, and stromal tumors, respectively (Online supplemental material 3). The same differences (p=0.26) were observed for the 5-year overall survival according to histological subtype, which were 94%, 96%, and 77% for germ cell, sex-cord, and stromal tumors, respectively (Figure 1).
Supplemental material
Overall survival according to histology subtype.
The 5-year progression-free survival rates related to FIGO stage did not show significant differences (p=0.08) and were 92% for stages I–II versus 78% for stages III–IV (Figure 2). The 5-year overall survival analysis also showed no significant differences between stages I–II versus III–IV (95% vs 91%, respectively; p=0.67) (Figure 3). In the univariate analysis (Online supplemental material 4) we found that sex-cord histology was significantly associated with progression-free survival. Using Cox proportional hazards regression, a multivariate model identified BMI (HR=1.007; 95% CI 1.00 to 1.01; p=0.004) and sex-cord histology (HR=3.6; 95% CI 1.17 to 10.9; p=0.025) as the most important independent prognostic factors for progression-free survival. Conversely, independent prognostic factors for overall survival were BMI (HR=1.01; 95% CI 1.00 to 1.01; p=0.001) and residual disease (HR=7.16; 95% CI 1.39 to 36.97; p=0.019) (Online supplemental material 5).
Supplemental material
Supplemental material
Progression-free survival according to International Federation of Gynecology and Obstetrics (FIGO) stage.
Overall survival according to International Federation of Gynecology and Obstetrics (FIGO) stage.
Discussion
Summary of Main Results
In this study we observed that residual disease and sex-cord histology were independent prognostic factors associated with worse oncological outcomes in women younger than 45 years diagnosed with non-epithelial ovarian cancers.
Results in the Context of Published Literature
Non-epithelial cancers arising from the ovary are rare tumors and represent only 3–8% of all ovarian cancers.1 7 This rarity makes it difficult to build robust evidence based on prospective studies. In the present study, 70% of patients had germ cell tumors, 27.3% sex-cord tumors, and 3.3% stromal tumors. A Danish retrospective study, using a nationwide population-based cancer registry data, was published recently. The authors reported 720 cases of non-epithelial ovarian cancer over a period of 38 years from 1978 to 2016. Of note, the distribution of histology subtypes differed in comparison with the present study. The investigators described 49% of germ cell tumors, 38% of sex-cord tumors, and 11% of stromal tumors. Cox regression showed that overall mortality was associated with year of surgery, patient’s age, and FIGO stage.3
The identification of risk factors not only provides useful prognostic information but also helps guide the selection of initial treatment in each patient. In this regard, the International Germ Cell Cancer Collaboration Group (IGCCCG) classification stratifies male granulosa cell tumor patients requiring chemotherapy into good-, intermediate-, and poor-risk groups, with estimated 5-year overall survival rates of 91%, 79%, and 48%, respectively.12 The model is based on histology, primary tumor site, tumor marker levels, and sites of metastasis. In addition, international guidelines in non-epithelial ovarian cancer use prognostic risk factors to recommend adjuvant chemotherapy to improve patients’ survival.13 14
Some factors associated with worse prognosis have been suggested for women with germ cell tumors, including age >40 years at diagnosis,8 presence of metastatic disease,8 residual tumor,15 stage of disease,16 17 incomplete peritoneal staging,16 17 as well as slow rate of post-operative alpha-fetoprotein decline among patients with yolk sac tumors.18 The recurrence risk in germ cell tumors was also evaluated in the MITO-9 trial.16 The investigators observed that disease recurrence was associated with incomplete peritoneal surgical staging (p=0.04; OR 2.37; 95% CI 1.04 to 5.44). In addition, the presence of the yolk sac component was predictive of better survival, with a 5-year overall survival rate of 83.7% in patients with yolk sac or mixed tumors compared with 98.6% in those with other germ cell histology subtypes (p<0.05).16 In a retrospective chart review at Memorial Sloan Kettering Cancer Center, progression-free survival and overall survival were also significantly improved in early stage in comparison with stage III-IV disease (p=0.029 and p=0.018, respectively).19
Sex-cord ovarian tumors are more infrequent than germ cell tumors.3 However, two studies observed a correlation between prognostic factors responsible for survival in this group of patients.20 21 The first study published in 2005 analyzed 85 women (median age: 49 years) with a median follow-up of 58 (range: 1–310) months. The median and 5- year disease-specific survival of women with stage I–II versus III–IV was 180 months and 85% compared with 58 months and 48%, respectively (p=0.012). Furthermore, age <50 years (p=0.003), pre-menopausal status (p=0.013), tumor size <10 cm (p=0.003), lack of lymph node involvement (p<0.0005), and absence of residual disease (p=0.002) were all significant predictors of improved survival. Of the patients who received adjuvant treatment, chemotherapy did not impact survival (p=0.11). In multivariate analysis, age <50 years, smaller tumor size, and absence of residual disease remained as independent prognostic factors.20
The other study published in 2007 reported 376 women with a median age of 51 years. Women with stage I–II disease had a 5-year disease-specific survival of 95% compared with 59% in those women with stage III–IV cancers (p<0.001). Patients ≤50 years had a survival advantage over those >50 years (93% vs 84%; p<0.001). This age-associated survival advantage was observed for early-stage (97% vs 92%; p=0.003) but not for advanced-stage (68% vs 53%; p=0.09) patients. On multivariate analysis, age ≤50 years (p=0.001) and early-stage disease (p<0.001) remained significant prognostic factors for improved survival.21
As the present study analyzed women younger than 45 years, we were not able to find patient age as an independent prognostic factor of oncological outcomes as did previous studies.3 8 20 21 Even though the present study observed a trend to worse progression-free survival and overall survival in women with advanced FIGO stage, this clinical factor has been shown to be the most important prognostic factor in most studies.3 16 17 19–21 The low number of cases analyzed in the present study with low number of recurrences and deaths might be the main reasons why advanced FIGO stage at diagnosis was not identified as prognostic factor in this group of patients.
Tumor size as a prognostic factor in sex-cord tumors is a matter of controversy. While some authors have not been able to validate the prognostic significance of tumor size in this histologic subtype,21 22 other studies observed that women with larger tumors lived for a shorter period of time.20 23 In the study of Miller et al that included 70 patients, larger tumor size (13.5 cm vs 10 cm) was associated with decreased progression-free survival, but this difference was no longer significant in multivariate analysis after accounting for stage of disease.23 In a multi-institution series of 83 patients, Chan et al showed that tumor size ≥10 cm contributed to decreased survival rates in both univariate and multivariate analyses.20
The present study observed that higher BMI was an independent risk factor for poor progression-free survival and overall survival. This is the first study analyzing this association in women with non-epithelial ovarian cancer, and the mechanism to explain this finding is still unclear in the literature.24 A recent meta-analysis analyzed 929 women with ovarian cancer in 19 cohort studies. The study noted that higher BMI was associated with poor patient survival (HR=1.35; 95% CI 1.03 to 1.76). However, the results for obesity at diagnosis depended on whether BMI was analyzed as a categorical or continuous variable. Analysis of obesity with BMI as a categorical variable did not affect ovarian cancer prognosis (pooled HR 1.07; 95% CI 0.95 to 1.21), whereas obesity with BMI as a continuous variable showed poorer survival with each incremental increase in BMI (pooled HR 1.02; 95% CI 1.01 to 1.04).25
Strengths and Weaknesses
The strength of this study includes the fact that we specifically designed it to describe oncological outcomes in women younger than 45 years diagnosed with a non-epithelial ovarian cancer. However, there are a number of limitations. First, the retrospective collection of data might be associated with misinterpretation and missed information, as well as with unidentified patients. In this regard, the retrospective nature of this study renders it susceptible to the inherent biases of a clinical review. Since non-epithelial ovarian cancers are uncommon tumors with a tendency to late recurrence, it is difficult to accrue a sufficient number of patients to complete either prospective or randomized studies in a reasonable period of time. Second, while the study included centers in Spain with specialized pathologists, a central pathology review was not performed to confirm non-epithelial ovarian cancer. Third, even though the patients’ characteristics are homogeneous, the sample size is still too small to generate robust results and the conclusions should therefore be interpreted with caution.
Implications for Practice and Future Research
Even though the identification of prognostic factors is crucial to identify high-risk patients and guide adjuvant treatment, larger studies with international collaboration are essential to clarify oncological risk factors in this rare disease.
Conclusions
Body mass index, residual disease, and sex-cord histology were prognostic factors associated with worse oncological outcomes in women younger than 45 years diagnosed with non-epithelial ovarian cancers.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
The study was previously approved by the Internal Review Board in accordance with the ethical standards of the Helsinki Declaration of 1975, revised in 1983. For this type of study formal consent was not required. Ethics approval was done at Hospital Clinico San Carlos, Madrid, Spain.
Acknowledgments
The authors want to thank the whole YOC-Care Group Collaborative for their full support and contribution to the manuscript: Alcalá MM (Hospital de Poniente, Almería), Alkourdi A (Hospital Virgen de las Nieves, Granada), Alonso M (Hospital Universitario de Guadalajara, Guadalajara), Alonso T (Hospital Universitario de Burgos, Burgos), Álvarez R (Hospital Universitario Santa Cristina, Madrid), Amengual J (Hospital Universitario Son Espases, Palma de Mallorca), Aparicio I (Complejo Hospitalario Universitario, Pontevedra), Arencibia O (Hospital Universitario Materno Infantil, Canarias), Azcona L (Hospital Universitario Virgen de la Macarena, Sevilla), Baciu A (La Paz University Hospital, Madrid), Bayón E (Hospital Clínico de Valladolid, Valladolid), Bellete C (Hospital General de Segovia, Segovia), Bellón M (Hospital Clínico San Carlos, Madrid), Boldó A (Hospital La Plana, Castellón), Brunel I (Hospital Quirón Málaga, Málaga), Bustillo S (Hospital Universitario de Salamanca, Salamanca), Cabezas MN (Hospital Universitario Virgen de la Macarena, Sevilla), Cano A (Hospital Universitario Fundación Alcorcón, Madrid), Cárdenas L (Hospital Josep Trueta, Gerona), Chacón E (Clínica Universitaria Navarra, Navarra), Coronado P (Hospital Clínico San Carlos, Madrid), Corraliza V (Hospital Universitario Ramón y Cajal, Madrid), Corredera FJ (Hospital Universitario de Salamanca, Salamanca), Couso A (Hospital Príncipe de Asturias, Madrid), Díaz B (Hospital Clinic Barcelona, Barcelona), Duch S (Hospital Universitario Infanta Sofía, Madrid), Fernández M (MD Anderson Cancer Center, Madrid), Fernández MJ (Complejo Asistencial de Zamora, Zamora), Fidalgo S (Hospital Universitario Central de Asturias, Asturias), García J (Fundación Jiménez Díaz, Madrid), Gil B (Hospital Universitario 12 de Octubre, Madrid), Gilabert J (Hospital Universitari Valencia, Valencia), Gómez AI (Hospital General de Segovia, Segovia), Gómez B (Hospital Clínico Universitario Virgen de la Arrixaca, Murcia), González L (Hospital Universitario de Torrejón, Madrid), González MH (Hospital Príncipe de Asturias, Madrid), Gorostidi M (Hospital Universitario de Donostia, San Sebastián), Herrero S (Hospital Universitario Puerta de Hierro, Madrid), Hidalgo E (Hospital Universitario San Cecilio, Granada), Iacoponi S (Hospital Quirón Madrid, Madrid), Izquierdo R (Hospital de Igualada, Barcelona), Lamarca M (Hospital Universitario Miguel Servet, Zaragoza), Lara AM (Hospital Virgen de las Nieves, Granada), Lete I (Hospital Universitario Araba, Vitoria), Llueca A (Hospital de Castellón, Castellón), López I (Hospital HM Montepríncipe y San Chinarro, Madrid), López CA (Hospital General Universitario de Ciudad Real, Ciudad Real), Mancebo G (Hospital del Mar, Barcelona), Marcos J (Hospital General Universitario de Alicante, Alicante), Marino M (Hospital Universitario de Salamanca, Salamanca), Martí MD (Hospital Universitario Bellvitge, Barcelona), Martínez M (Hospital Virgen de la Luz, Almería), Martínez MC (Hospital Son Llatzer, Palma de Mallorca), Melero LM (Hospital Virgen del Rocío, Sevilla), Minig L (Instituto Valenciano Oncología, Valencia), Montesinos M (Hospital Universitario y Politécnico La Fe, Valencia), Moral R (Hospital Poniente, Almería), Morales S (Hospital Universitario Infanta Leonor, Madrid), Negredo I (Hospital Miguel Servet, Zaragoza), Nieto A (Hospital Virgen de la Arrixaca, Murcia), Níguez I (Hospital Virgen de la Arrixaca, Murcia), Ribot L (Hospital Universitario Parc Tauli, Barcelona), Rosado C (Hospital de Mataró, Barcelona), Sancho V (Hospital Universitario de Salamanca, Salamanca), Tetilla V (Hospital General de Segovia, Segovia), Torres MC (Hospital del Perpetuo Socorro, Badajoz), Valencia I (Hospital Universitario Puerto Real, Cádiz), Veiga A (Hospital General Universitario Gregorio Marañón, Madrid), Vidal R (Complejo Hospitalario de Pontevedra, Pontevedra), Vilches JC (Hospital Quirón Málaga, Málaga).
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @Quique_ChC
Contributors Conception or design of the work: IZ, MG, LM. Data collection: all authors. Data analysis and interpretation: IZ, MG, LM. Drafting the article: IZ, MG, LM. Critical revision of the article: all authors. Final approval of the version to be published: all authors. All authors contributed to the concept, design and writing of the manuscript, and are responsible for the overall content as guarantors.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.