The aim of the present study was to investigate clinical features and outcomes among patients with cervical SmCC to ensure that doctors and patients clearly understand the different clinical and survival features for cervical SmCC. As far as we know, this was first population-data study to compared SmCC and SCC according to specified inclusion criteria.
In our study, we found the average morbidity age of the two malignances was similar, but cervical SmCC patients were somewhat a little younger. Both pathological types are more likely to occur in single women and in white patients. This could be explained by the fact that socioeconomic psychological factors and genetic diversity might play an important role in tumorigenesis and tumor progression(20–22). Several previous studies did demonstrate that marital status and ethnicities were important etiological and prognostic factors in several solid tumors, including cervical cancer (23–26). Similarly, in our study better survival was seen in married patients and in white women.
Significant differences between patients with cervical SmCC versus SCC were also observed in our study. Patients with SmCC of cervix were much more likely than patients with SCC of cervix to present with larger tumor size, local lymph node involvement, distant metastasis and advanced disease stage, all of which might contribute to poorer survival in SmCC patients. These observations were in accordance with previous studies showing that FIGO stage, tumor size, distant metastases were important clinicopathological characteristics that dictated the survival of patients with cervical SmCC (27–29). Moreover, cox regression models in our study indicated that pathology SmCC and higher stage primarily led to poor prognosis of patients with cervical cancer. In addition, stratified analyses by stages showed that SmCC was an independent prognosis factor for poor survival in stage I-II, III and IV patients.
More importantly, to balance the confounding factors, we conducted a PSM study to match confounding factors between patients with cervical SmCC and SCC. Multivariate regression analysis in matched cohorts suggested that SmCC histology and FIGO stage were crucial indictors for poor survival. Kaplan-Meier curves clearly displayed that OS was lower in stages I-II, III, and IV in patients with cervical SmCC compared to patients with the same stage of cervical SCC. Furthermore, stratified analyses by stage in matched cohorts yielded similar results, indicating that SmCC pathology was the main influencing factor for poor survival. These findings suggested that cervical SmCC itself was highly malignant, undifferentiated, and aggressive, associated with poorer prognosis. Consistently, precious studies had proved that SmCC of cervix was associated with a dismal survival(9, 30). However, in our study we offered a relatively large sample size of patients with SmCC and PSM and stratified analyses were performed to more convincingly illustrate this fact.
However, the mechanism was not yet clear due to its rarity. Some studies showed that most patients with SmCC of cervix were etiologically associated with high-risk HPV, especially HPV 18, and that HPV infection was involved at an early stage of oncogenesis in SmCC of cervix(31–33). Moreover, SmCC of cervix are characterized by high mitosis, massive necrosis and common lymph-vascular involvement(34). Besides, existing studies have demonstrated that additional driven events also involved in the progression of SmCC of cervix. For example, loss of heterozygosity has been reported to be a driven event in SmCC of cervix(35). Besides, there were studies showing that recurrent mutations involving PIK3CA, KRAS, and TP53 genes in SmCC of cervix. And in SmCC of cervix genetic alterations involving the MAPK, PI3K/AKT/mTOR, and TP53/BRCA pathways were found in a recent study(36). These findings may have implications to targeted therapy for the rare disease, offering the potential individualized management for this aggressive tumor. There was study suggesting that cervical SmCC showed a higher expression of the synaptophysin than that in lung SmCC, while a higher expression TTF-1 in tumor cell nuclei was observed in lung SmCC. Both pulmonary and cervical SmCC had similar immunoreactive staining for CD56 and chromogranin A(37). These results were useful for early diagnosis. But we did not know much about the mechanism of cervical SmCC and more efforts were needed to explore the mechanism for this aggressive tumor.
The optimal therapeutic strategies for cervical SmCC were not clear due to lack of standardized guidelines for treatment. Treatment options for cervical SmCC were according to the data for treating small carcinoma of lung because SmCC of the cervix share similar histological and clinical characteristics with small cell lung cancer.(38) Accepted treatment modalities included surgery, radiotherapy and chemotherapy. Surgery is mainly used to diagnosis or a resection to early stage disease(39). For limited stage disease, radiotherapy is an important approach and it is usually given with concurrent etoposide/cisplatinum(40). Platinum-based combination chemotherapy was the standard primary approach for advanced stage disease, which could achieve a high response rates of 50–79% but also developed recurrent or progressive chemoresistant disease.(41) Future prospective studies designed to explore the treatment for cervical SmCC were necessary.
As a retrospective study based on SEER databases, there were limitations existing. Firstly, information of detailed treatment regimens was insufficient in SEER databases, which may lead to different survival. Secondly, Selection bias may exist, which is inevitable for clinical observational studies even when PSM was used. But our analysis also had some strengths. Firstly, we collected 278 cases with cervical SmCC from 2004 to 2015, which offered a relatively large sample size to perform accurate and multiple forms of analysis. Secondly, PSM analysis and stratified analyses were used to analyze potential confounding factors.