Marital Status is an Independent Prognostic Factor in 7 Sites of Squamous Cell Carcinoma: A Propensity-Adjusted SEER Database Analysis

Background: Marital status has been proved as an independent prognostic factor in many cancer types. However, no detailed investigation of marital status on squamous cell carcinoma (SCC) has been evaluated. The aim of this essay is to explore the relationship between marital status and SCC in 7 tumor sites. Methods: All patients diagnosed with SCC were collected from the SEER database (1975-2016). We analyzed the survival of all included SCC patients in four marital status. We utilized propensity-score matching analysis to balance baseline characteristics between married and unmarried SCC patients in 7 tumor sites. The inuence of marital status on overall survival (OS) in each site was performed by Cox regression analysis. Results: A total of 180009 SCC patients were involved in this study. After propensity-score matching, patients in the married group were 1:1 matched with patients in the unmarried group for each sites. Married group exhibited higher 5- year OS rate than unmarried group (27.3% vs 19.8%). More precisely, being divorced and widowed were observed to be related to have worse survival than single patients in most sites. Furthermore, patients with clinical stage IV were more common in the unmarried group which having a lower proportion of receiving treatment. Conclusions: This study indicated that marital status was a signicant factor for OS of SCC in 7 tumor sites. Married patients always behaved more favorable than unmarried including single, divorced, and widowed patients.


Introduction
Squamous cell carcinoma is a form of cancer that originates from squamous cells of the skin and mucous membranes. SCC occurs in tissues that provide a barrier between an organism and environment, such as the skin, oral cavity, esophagus and lung [I1]. Statistically, SCC accounted for more than 90% of the head and neck cancers and esophageal cancers [2][3]. SCC tended to have different mechanisms of tumorigenesis and progression and are frequently associated with smoking and alcohol consumption [4][5][6]. Besides, oral SCC have also been demonstrated to be related to human papilloma virus (HPV) infection and a lack of vitamins [7]. High-risk HPVs (HPV16, HPV18, and HPV31) were observed to be relevant to at least 99.7% of cervical SCC [8].
The correlation between the prognosis of cancer and clinical features including age, gender, TNM stage, tumor size was frequently discussed. However, other confounding factors, such as dietary habits, socioeconomic status, educational attainment, and marital status were rarely evaluated. Several previous studies had reported the relationship between marital status and SCC of the head and neck and penis [9][10][11]. They all indicated that marital status was an independent prognostic factor among these cancers and married patients showed a signi cantly better prognosis. However, other sites of SCC were less investigated.
The surveillance, epidemiology, and end result (SEER) database consisted of 18 cancer registries and covers 30% of US population [12]. In our study, we analyzed the data from the SEER database to nd out the in uence of marital status on SCC in 7 common sites.

Patient selection
All patients were identi ed in the SEER database (version 8.3.8). Patients with a pathological diagnosis of SCC (ICD-O-3 histology code: 8050-8089) in the site of the cervix, esophagus, larynx, lung, nasopharynx, oropharynx, or tongue were enrolled in our study. The TNM staging was based on the American Joint Committee on Cancer (AJCC) 6th edition. Exclusion criteria were listed as the following: (a) Patients with the metastatic tumor or more than one primary tumor; (b) Patients with incomplete clinicopathological characteristics.
Propensity-score matching Marital status was divided into the married group and unmarried group. To evaluate the importance of marital status of SCC, propensity-score matching was performed to regulate the difference of clinical features between two patient groups in 7 sites. After matching, clinicopathological variables and the in uence of marital status were re-analyzed within the matched pairs.

Statistical analysis
The relationship between marital status and a series of clinical information, including age, gender, race, tumor site, stage, surgery, radiation, and chemotherapy was analyzed by Pearson's chi-square test. The primary endpoint of this  Table 1. Results of Chi-square tests showed that signi cant differences were observed in the age, gender, race, stage, surgery, radiotherapy, and chemotherapy between married and unmarried groups. We also calculated the proportion of stage IV patients at the rst diagnosis between two groups. The results showed that the ratio of distant metastases in the unmarried group was all higher than the married group in 7 sites. This phenomenon was particularly noticeable in the site of the larynx. Besides, rates of surgery, radiotherapy, and chemotherapy were computed in both two groups ( Table 2). We observed that the married group had more therapeutic advantages than the unmarried group.
Propensity-score matching for married and unmarried groups Baseline characteristics of 7 tumor sites between married and unmarried groups were different. So propensity-score matching analysis was performed to balance potential confounding factors between two groups. After matching, a total of 158880 patients matched pairs were enrolled, which including cervix (18620), esophagus (8142), larynx (22555), lung (73101), nasopharynx (2478), oropharynx (15024), and tongue (18960), respectively. Table 3 showed the median survival time (MST) of 4 marital status in 7 sites after propensity-score matching. The MST in the married group was always signi cantly better than that in the unmarried group (P<.001). Among 7 sites of SCC, head and neck and cervical SCC had more favorable survival. But the SCC of the esophagus and lung was associated with poor survival. The 5-year survival rate of the married and unmarried group was 27.3% and 19.8%, respectively. Figure 1 displayed the survival differences between the two groups in 7 SCC sites by survival curves (P<.001). The curves showed the survival advantage of the married group intuitively. A cox regression analysis was used to compare 4 marital status related to OS ( Table 4). The relative hazard of the unmarried group including single, divorced, and widowed was all higher than the married group. This was especially obvious for the site of the oropharynx. Single (HR 2.008), divorced (HR 1.856), and widowed (HR 3.078) group showed much higher risks than the married group.

Discussion
Several studies have indicated that there is a close relationship between marital status and survival of cancers [13][14][15][16]. But the correlation between marital status and SCC remained elusive. Seven high-prevalence sites of SCC were selected from the SEER database to address this issue. After propensity-score matching, marital status was con rmed to be signi cantly associated with the prognosis of SCC patients in all 7 sites. Moreover, the survival period of patients who were married was signi cantly longer than those who were single, divorced or widowed.
Unmarried patients were more likely to have distant metastases and were less likely to undergo surgery, radiotherapy or chemotherapy. Besides, it is the rst study to indicate that marriage has an independent bene cial effect on survival of SCC in seven common sites.
Previous studies had demonstrated that marital status was signi cantly related to superior prognosis of penis, oral cavity and tongue SCC [17][18][19]. Studies had shown that marriage was correlated with a reduction in cancer death ranging from 12-33% [20]. Some literature suggested that married people were more able to receive early screening and treatment due to the support of spouse [20][21]. Other potential advantages of being married included the increasing likelihood of having insurance coverage, good living habits, mental state or even social support [22]. The premise must be a benign marriage without adverse factors such as domestic violence. Also, the detailed classi cation of unmarried patients in our study exhibited that divorced and widowed patients had worse survival than those who were single. This may be attributed to psychological distress suffered by patients who were divorced or widowed. Furthermore, the poor prognosis of widowed patients may be also related to older age [23]. It was widely known that the association between tobacco smoking and the incidence of esophageal, lung, nasopharyngeal, head and neck SCC was well established [24][25][26][27]. Meanwhile, SCC often occurs in male smokers. The ratio of males to females in our study also supported this opinion. Previous researches had found that marriage reduced smoking and alcohol intake [28]. Males and black had a higher smoking-related mortality rate [29], so this type of patients may

Conclusion
In conclusion, we have demonstrated that marital status is an independent prognostic factor of SCC patients.
Married patients behaved more superior OS than unmarried patients of SCC in 7 sites. We tried to help unmarried patients minimize the survival gap by analyzing the reasons behind the longer survival time of married patients.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and material
The datasets generated during and/or analysed during the current study are available in the SEER database [https://seer.cancer.gov/].

Competing interests
The authors declare that they have no competing interests.