Some studies suggest that chemotherapy can improve the outcome of patients. A previous study shows that the patients of malignant non-GC SCST with stage III–IV disease who received chemotherapy had a longer median survival time than those who did not (7).
However, several studies found that adjuvant chemotherapy cannot improve prognosis. They also analyzed patients with stage I-II malignant non-GC SCST, indicating that chemotherapy had no more clear benefit of overall survival(OS) (7). Another study shows no significant difference in OS between the chemotherapy group and the observation group in patients with stage II-IV disease(10). A retrospective study showed that adjuvant chemotherapy was not associated with improved disease-free survival (DFS), analyzing 60 patients with stage IC adult granulosa cell tumor(AGCT), and the number of chemotherapy cycles was not associated with recurrence rate. (8) Another study analyzed the clinical data of 40 patients with stage IC primary GCT of the ovary treated in MITO centers reported the similar conclusion that there was no significant difference in DFS between patients who received adjuvant chemotherapy versus no treatment. (9)Another retrospective study also implied that Adjuvant chemotherapy was not associated with recurrence rate or survival in stage IA disease (11). Meisel et al. also recognized that after controlling for stage, adjuvant therapy did not improve recurrence-free interval; however, mostly stage II disease patients were included in this study, and only 11 patients received chemotherapy in this study; thus making impossible any reliable evaluation of prognosis according to adjuvant chemotherapy administration(12).
Moreover, a recent study analyzed SCST patients who underwent surgery from 2004 to 2015 were enrolled from the SEER database, developed a nomogram for prognostic prediction of OS in postoperative patients with SCST, and revealed that chemotherapy affects OS. Further analysis stratified by tumor stage found that in stages IA and IB, chemotherapy had poor OS than those who did not receive chemotherapy (P = 0.031), while OS showed no significant difference in another stage.
This retrospective study is the first to assess the impact of chemotherapy on DSS in patients with OSCSTs,and analyzed the relationship between chemotherapy and DSS stratified by stage, grade, tumor size, histopathology subtype, the extent of surgery using data from the SEER database between 1998–2016. With 323 patients, our result showed that
chemotherapy is associated with poorer DSS based on K-M survival analysis. The reverse effects were still significant in patients with Grade 1, tumor size༜10cm,GCTs ༌CSS group ༌Non- CSS group༌FSS group, Non-FSS group, and TAH-SO group. After controlling with confounding factors of tumor size, stage, grade, histopathology and surgery, there was no significant difference in DSS with and without postoperative adjuvant chemotherapy, and adjuvant chemoradiotherapy did not improve survival.
In the present study, age, whether a continuous variable or a categorical variable, did not affect the survival rate. However, some researchers believed that the age older than 40 years old, 50 years old (11, 13, 14) or 60-year-old (14) is a risk factor that can affect prognosis. For juvenile granulosa cell tumors (JGCT), age < 10 years old is an independent prognosis factor. (15) In our study, large size (> 10 cm) is an unfavorable prognostic factor. One study draws similar conclusions (13), while another study reported that large size (> 15 cm) is an unfavorable prognostic factor (16). Some studies reported a one-centimeter increase in tumor size was associated with 4% (2–6%) increased risk of death.(17) Tumor stage and degree of differentiation are consistently associated with prognosis (11, 17, 18). The histologic subtype is a risk factor in univariate analyses. However, histologic subtype showed no significant association with DSS both in multivariate analyses. A recent study revealed that SLCTs were associated with worse cancer-specific mortality than GCTs using the same database(19). Based on the results of the present study, completeness of surgical staging is not a prognostic factor for DSS. However, a previous study shows that patients who underwent FSS(CYS、USO) had a poor DFS than those who underwent RS(20). Our results revealed that lymphadenectomy was not associated with DSS. Other studies also reported no benefit of LND on DFS or OS in these patients, and LND is always associated with many complications(10, 14, 21, 22).
The poor prognosis of patients who received chemotherapy can be attributed to the side effects of conventional chemotherapy and inherent limitations associated with retrospective analyses.
Firstly, Patients were not randomly assigned to a treatment; they were grouped into chemotherapy group or non-chemotherapy group. Furthermore, we do not know whether the patients recorded as “No/Unknown” actually received chemotherapy. The regimens, duration of chemotherapy, the toxicity and other side effects of chemotherapy were not recorded in SEER database, and information about recurrence was also not available. Additionally, ,patients who received chemotherapy may have risk factors, such as elevated tumor marker, incomplete surgical staging(23), the presence of residual disease, tumor rupture, etc.; however, that information was not recorded in the SEER database. Moreover, no information about other factors that could affect prognosis is provided, including heterologous elements(24), presence of a retiform subtype, nuclear atypia(25), mitotic index (MI)༞5,TPRT mutations, KMT2D, P16 overexpression, high expression of SMAD3 and so on(26).
These results may also be explained by being insensitive to chemotherapy of patients with SCSTs. The majority of present studies show that postoperative adjuvant chemotherapy cannot improve the prognosis of patients with this disease(7–11). In a retrospective study with small sample size, among 27 patients treated with BEP, the response rate was only 22%. Moreover, the researchers performed a meta-analysis in parallel, which show that response rate of chemotherapy was 50%; however, there were 21 different chemotherapy regimens in the analysis (27). Thus, single-agent chemotherapy, or hormonal and targeted therapies need to be assessed in further prospective studies
Generally, our study has several innovative advantages. The strength of this study is that it is based on a large cohort of patients from the SEER database with strict inclusion and exclusion criteria. The follow-up time was long enough to estimate survival and it determinants. Multivariate Cox regression models and stratification were constructed to control for confounders, such as tumor size, stage, grade, histopathology, and surgical procedure type. Our study used the DSS as the endpoint to exclude the impact of other causes of survival prognosis. Despite the limitation to our study, it may help in clinical decision-making .