Synuclein-γ (SNCG) expression in ovarian cancer is associated with high-risk clinicopathologic disease

Background Synuclein gamma (SNCG) expression is associated with advanced disease and chemoresistance in multiple solid tumors. Our goal was to determine if SNCG protein expression in ovarian cancer was correlated with clinicopathologic variables and patient outcomes. Methods Tissue microarrays from primary tumors of 357 ovarian, fallopian tube, and primary peritoneal cancer patients, who underwent primary surgery at Roswell Park Cancer Institute between 1995 and 2007, were immunohistochemically stained for SNCG. A pathologist blinded to patient data scored tumors as positive if ≥10 % of the sample stained for SNCG. Medical records were reviewed for clinicopathologic and demographic variables. Between the positive and negative groups, Wilcoxon rank-sum test was used to compare the median ages and Fisher’s exact test was used to compare groups in categorical variables. Cox proportional hazard models examined associations between SNCG and overall and progression-free survival. Results The median follow-up was 36 months, median overall survival was 39 months, and median progression-free survival was 18 months. SNCG presence was associated with clinical variables of serous histology, grade 3 disease, suboptimal debulking, ascites at surgery, FIGO stage III-IV cancer, or initial CA-125 level >485. There was no significant difference in overall survival (HR 1.06 95 % CI 0.81–1.39 P 0.69) or progression-free survival (HR 1.16 95 % CI 0.89–1.50 P 0.28) for patients with or without SNCG expression. Conclusions SNCG expression in ovarian cancer is frequent in patients with high-risk features, but it does not correlate with chemotherapy response, overall survival, or progression-free survival. Electronic supplementary material The online version of this article (doi:10.1186/s13048-016-0281-4) contains supplementary material, which is available to authorized users.


Background
Ovarian cancer is a leading cause of cancer mortality in women, accounting for more deaths than any other gynecologic malignancy in the United States [1]. While overall cancer incidence and mortality rates for gynecologic malignancies have declined in the past decade, progress in ovarian cancer outcomes has been slow. Despite developments in cytotoxic chemotherapy, five-year survival rates for women diagnosed with ovarian cancer remain less than 50 % [2].
High rates of recurrence and associated mortality, coupled with advancements in the characterization of intracellular signaling pathways in carcinogenesis, have prompted investigation into other potential targets that can be used in the treatment of ovarian cancer, such as intracellular signaling pathways [3]. Among such targets, synuclein gamma (SNCG) was proposed as a potential target in ovarian cancer therapy [4].
Synucleins are a family of neuronal proteins expressed primarily in the peripheral nervous system. To date, three synuclein proteins have been identified: synuclein-α (SNCA), synuclein-β (SNCB), synuclein-γ (SNCG) [5,6]. The former two, SNCA and SNCB, have been implicated in neurodegenerative disorders, such as Parkinson's and Alzheimer's disease [7,8], while SNCG has been primarily linked with cancer. SNCG was first discovered in breast cancer tissue [9] and has since been detected in multiple solid tumors, including breast, lung, liver, esophagus, colon, bladder, pancreatic, and prostate cancers [10]. SNCG has also been identified in gynecologic processes, including benign pathology (endometriosis), as well as endometrial and ovarian cancers [10,11]. SNCG expression occurs with advanced disease and chemoresistance in many cancers, and in breast cancer, SNCG has been causatively linked to increased proliferation, metastasis, and drug resistance [12].
Recent studies have demonstrated the role of synucleins as potential biomarkers in several cancer types, including ovarian cancer [13]. Several studies by immunohistochemistry (IHC) analysis demonstrated high levels of SNCG expression in up to 73 % of epithelial ovarian cancers [14][15][16] and one study showed that SNCG overexpression may promote the metastatic potential of ovarian cancer cells [17]. These findings suggest that SNCG may be a potential prognostic marker as well as a target for therapeutic drug development. Studies examining the correlation of SNCG expression with clinical outcomes are lacking, however, limited only to a single meta-analysis of gene expression profiles in ovarian cancer [18].
In this study, SNCG expression levels were examined using immunohistochemistry in primary and metastatic tumors of 357 patients with ovarian, fallopian tube, and primary peritoneal cancer. In particular, we examined the association of SNCG expression in ovarian cancer with worse outcomes, such as decreased progression free-and overall-survival and/or increased chemotherapy resistance.

Patient population
After obtaining IRB approval, the Pathology archive at Roswell Park Cancer Institute, Buffalo, New York was searched for ovarian, fallopian tube, and primary peritoneal cancer cases from 1995 to 2007. A chart review was conducted with extraction of clinical information, including patient's age at the time of diagnosis, the surgical stage, postoperative treatment. All patients underwent a primary surgical staging surgery, including total abdominal hysterectomy with bilateral salpingo-oophorectomy, with or without pelvic and para-aortic lymph nodes dissection and pelvic washings, depending on tumor stage. Patients were treated according to National Comprehensive Cancer Network guidelines (https://www.nccn.org). Patient's general information and tumor features are summarized in Table 1.
Histologic evaluation and high-density tissue microarray (TMA) preparation Tumor grade was assessed using the International Federation of Gynecology and Obstetrics (FIGO) system and tumor stage was assigned based on the 2014 FIGO surgical staging guidelines. All slides were examined by a gynecologic pathologist for confirmation of tumor morphology and tumor grade. The viable tumor tissues and control tissues (fallopian tube) from each case were circled by pathologists. 0.6 mm tissue cores were punched and arrayed into high-density TMA receipting blocks.

Immunohistochemistry
Tissues were prepared for analysis, as previously described by Mhawech-Fauceglia et al. [19]. Briefly, high-density TMA blocks were sectioned in 4 μm thickness followed by deparaffinization with xylene, then washed with ethanol. Sections were cooled for 20 min and incubated for 10 min with 3 % H2O2 to quench endogenous peroxidase activity. Blocking was performed using serum-free protein block, Dakocytomation (Carpenteria, CA) for 30 min. Antigen retrieval was done using a citrate-based buffer (pH 6; Bond Epitope Retrieval Solution, Leica Biosystems) and sections were incubated with the SNCG antibody (Abcam) for 1 h at room temperature on the Dako Autostainer Plus. The diaminobenzidine complex was used as a chromogen. Immunostained slides were blindly reviewed and scored by two gynecologic pathologists. Immunostain for SNCG was present in both the nucleus and cytoplasm. Immunoreactivity was semiquantitatively scored in immunointensity of 0 (negative), 1 + (weak), 2+ (moderate) and 3+ (strong) and immunopercentage of <10 %, 10-50 % and >50 %. For the sake of statistical analysis, tumors were grouped as positive (SNCG+) or negative (SNCG-). Tumors were considered SNCG+ if ≥10 % of the tumor epithelial cells were immunoreactive with immunointensity of ≥2. Examples of positive and negative cases are illustrated in Fig. 1a-d.

Real time PCR
RNA from benign fallopian tube epithelium and ovarian cancer tissues was reverse transcribed. First-strand cDNA synthesis was performed using 700 ng of RNA and M-MLV reverse transcriptase (Life Technologies). Real time PCR was performed using Taqman reagent in a QuantStudio 5 system and primers to SNCG and the housekeeping gene TBP (Applied Biosystems/Life Technologies). Fold-change values were calculated using TBP as the housekeeping gene.

Statistical analyses
To test the association between the biomarker and the clinical parameters, Wilcoxon rank-sum test was used to compare the median ages and Fisher's exact test was used to compare frequencies and percentages for categorical variables. Progression-free survival (PFS), was defined as observed length of time from date of diagnosis to recurrence or censoring at date of last contact. Overall survival (OS) was defined as observed length of life from date of diagnosis to death or censoring at date of last contact. For survival analysis, Kaplan-Meier curves were used to estimate PFS, and OS curves. Curves were compared for tumors with and without SNCG expression using the logrank test. Univariate cox proportional hazards models were used to estimate hazard ratios for SNCG status for both the OS and PFS outcomes. Multivariate Cox regression was then conducted with SNCG expression as the main predictor and adjustment for clinical factors that were associated with SNCG expression (CA125 level, ascites, debulking status, FIGO grade, tumor stage and histologic subtype). Since several of the clinical covariates were highly associated with each other, sensitivity of the hazard ratio estimates for SNCG was examined after adjustment for each clinical variable in separate models. All reported p values are two sided with P < 0.05 significance. Statistical analyses  was performed using SAS v 9.4. Unpaired t-test was used to compare SNCG mRNA levels in benign fallopian tube and ovarian cancer tissues using Graphpad Prism version 6.0 (Graphpad Software, La Jolla, CA, USA).

Patient characteristics
The clinical and pathologic features of 357 patients with primary ovarian, fallopian tube, and primary peritoneal cancer are summarized in Patients with ascites also showed significantly higher SNCG expression than those without ascites (p = 0.01). When dividing CA-125 scores of below and above 485, SNCG expression was marginally associated with high CA-125 expression as well (p = 0.02). There was no association between age at diagnosis or site of primary disease and SNCG expression.
To provide a quantitative measure of SNCG levels using real time PCR, RNA from tissue samples outside of the cases used for the tissue microarray, was analyzed. SNCG expression from benign human fallopian tube epithelium and ovarian cancer tissues was compared. Benign fallopian tube epithelium expressed less SNCG than ovarian cancer tissues (Additional file 1: Figure S1), which supports the immunohistochemical analysis of SNCG (Fig. 2).

Survival analyses
The  (Fig. 3). The lack of association between SNCG expression and OS and PFS persisted after adjustment for all clinical variables associated with SNCG (Additional file 2: Table S1).

Metastatic disease
We further explored SNCG expression in metastases.

SNCG expression and chemotherapy response/platinum resistance
We further explored the impact of SNCG on chemotherapy response. Table 4 demonstrates the distribution of SNCG expression by platinum-refractory (defined as disease progression with first-line chemotherapy) or platinum-sensitive (defined as disease recurrence within 6 months of first-line chemotherapy) disease status compared to those patients with no evidence of disease (NED). Results show a marginally greater prevalence of platinum-refractory cases in cases with positive SNCG expression (p = 0.08). There was no significant difference in SNCG status in patients with platinum-sensitive disease (p = 0.79).

Discussion
This study is the first to formally evaluate the proposed association between SNCG protein expression and clinical outcomes in patients with ovarian, fallopian tube, and primary peritoneal cancers. SNCG is a new potential biomarker and demonstrates to be specific and reliable for immunohistochemistry in formalin-fixed and paraffinembedded tissue sections. SNCG has been examined in several different tumor types [10,[13][14][15][16]. Although the cut-off value of immunoreactivity for positive and negative SNCG has not been established for each tumor type, we found it was meaningful and reproducible in ovarian cancer when 10 % of tumor cells have moderate to strong immunoreactivity for SNCG. In our series of 357 patients, expression of SNCG was identified in 72 % of primary tumors, a percentage similar to the 73 % expression that has previously been reported in ovarian carcinomas [14]. The hypothesis, namely that SNCG expression was associated with worse clinical outcomes, was refuted, as the results of our study found no association between SNCG expression and OS or PFS in primary tumors. While SNCG expression did not correlate with clinical outcomes such as PFS, OS, or chemoresistance, the results of our study did find that there was a significant association between SNCG expression and high-risk clinicopathological factors, such as serous histology, high grade disease, advanced stage, and suboptimal debulking surgery. These associations suggest that while SNCG expression might not be a single prognostic marker, it A B p=0.69 p=0.27   [20][21][22][23]. The next direction would be to test serum samples from patients that were collected at the same time as tumor tissue in order to correlate serum and tissue SNCG levels.
In regards to SNCG expression in metastatic ovarian cancer, our data did not demonstrate a significant difference in OS among patients in whom primary tumors gained or lost SNCG expression; however, our data did demonstrate that the majority of metastatic tumors (83 %) demonstrated SNCG expression. This data supports prior studies that implicate SNCG overexpression may promote the metastatic potential of ovarian cancer cells [17] and further supports the association of SNCG expression with high-risk clinicopathologic disease. Our findings of high levels of SNCG expression in ovarian cancer were consistent with SNCG expression in multiple cancer types, including breast, liver, prostate, and colon cancer [14,21,24,25]. However, unlike breast [12] and endometrial cancers [19] in which high SNCG expression has been correlated with adverse clinical outcomes, our study found no correlation between high SNCG expression and clinical outcomes.
There are several possibilities for the lack of association with OS or PFS in this study in ovarian cancer. First, the study may have been underpowered to detect a significant difference in clinical outcomes. The association of SNCG expression with high-risk disease would suggest that there is a difference in clinical outcomes such as PFS or OS, however this was not seen in our cohort. Next, ovarian cancer is a heterogeneous cancer with multiple histologic subtypes and underlying molecular aberrations [26] and thus SNCG expression may influence tumor phenotype differently depending on co-existing histologic and molecular factors. Furthermore, as SNCG expression is not specific to ovarian cancer, its function in ovarian cancer may be a feature of malignancy in general. Finally, we evaluated the presence of SNCG, not its function, nor its level of expression. Tumors may change such genomic outcomes and our study is insufficient to determine either the underlying somatic or germline polymorphisms or the epigenetic actions on SNCG genes. Future studies might focus on the correlation of SNCG protein and gene expression in combination with other markers of disease, such as CA125 levels.
In conclusion, this study found that while SNCG expression is often present in ovarian carcinoma, the positive or negative expression of SNCG protein alone is not independently associated with clinical prognosis. Our results did indicate that SNCG expression is associated with clinicopathologic features of high-risk disease, suggesting that SNCG expression may play a role in severity of disease and be a marker for aggressive disease. Furthermore, as targeted drug therapy develops, recognition of SNGC expression in ovarian cancer will remain important as we continue to discover ways to improve outcomes in patients diagnosed with this deadly disease.

Conclusions
Expression of SNCG is associated with clinicopathologic variables of aggressive and advanced disease but not with overall survival or progression free survival. SNCG may serve as a novel biomarker for aggressive or advanced ovarian carcinoma and warrants further investigation to determine its role in this disease.

Additional files
Additional file 1: Figure S1. Expression of SNCG mRNA in benign fallopian tube epithelium and ovarian cancer tissues. RNA from human benign fallopian tube epithelium and ovarian cancer tissues was subjected to RT-real time PCR for SNCG. SNCG as a ratio of the housekeeping gene TBP is presented using delta delta CT calculations. * = p < 0.05. FTfallopian tube; OvCaovarian cancer. (PPTX 64 kb) Additional file 2: Table S1.