Impact of marital status on renal cancer patient survival

Marital status is an independent prognostic factor for various cancer types. The present study used the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute (NCI) to analyze the impact of marital status on renal cancer patient survival outcomes. We identified a total of 62,405 eligible patients (23,800 women and 38,605 men). Overall 5-year renal cancer cause-specific survival (CSS) was 80.3% in the married group, 69.2% in the widowed group, 78.9% in the single group, and 76.5% in the divorced/separated group. The widowed patient group had the highest female/male ratio, more distant metastases, and fewer high-grade (III/IV) tumors. Most widowed patients (90.4%) were elderly (>60 years old). In our study, male renal cancer patients benefited more from marriage than females. We also found that white married patients had better survival outcomes than other white patient groups, but black unmarried and married patients exhibited similar survival outcomes. Our results show that, in general, unmarried patients have higher rates of cancer-specific mortality and highlight the importance of psychological intervention for cancer patients during treatment.


INTRODUCTION
Renal cancer causes 140,000 deaths per year, and is the seventh most common cancer in the world [1]. In 2013, more than 350,000 people were diagnosed with renal cancer [1]. While most renal cancers are localized, low-grade tumors, nearly 17% of patients had distant metastases at the time of diagnosis [2]. Several factors, such as smoking tobacco [3], hypertension [4], obesity [5,6], and red meat consumption [7], are associated with renal cancer progression. However, little is known about the roles of socioeconomic status and psychological supports, such as marital status, in renal cancer prognosis. Marital status is an independent parameter to predict survival outcome in various cancers [8][9][10], and married patients exhibit better survival outcomes than unmarried patients. Married status may be associated with improved social support, higher income, and healthier behaviors, which might improve cancer patient rehabilitation results [11][12][13][14]. To our knowledge, the effects of marital status on renal cancer patient survival have not yet been studied. Here, we collected data from the Surveillance, Epidemiology, and End Results (SEER) cancer-registry program, including individuals diagnosed with renal carcinoma between 2004 and 2013, and explored the impact of marital status on renal cancer patient cause-specific survival (CSS).
Multivariate analysis showed that marital status was also a prognostic factor. Widowed (hazard ratio [HR], 1.162; confidential interval [CI], 1.095-1.234) and  The other covariates were also validated as independent factors in predicting renal cancer patient outcome. Female patients had better CSS than male patients (HR 0.912, 95% CI 0.876-0.950), while black patients had a higher mortality risk than white patients (HR 1.105, 95% CI 1.038-1.177). Patients with larger, advanced stage, or high-grade tumors, and elderly patients (>60 years of age) had higher mortality rates (Table 2).

Subgroup analysis of the effects of marital status
We evaluated the impact of marital status on survival at each stage of renal cancer. Widowed and divorced/separated patients had worse survival rates compared with married patients in each tumor stage (Table 3, Figure 1). However, although single patients had a lower 5-year CSS than married patients with SEER distant stage tumors (12.5% vs 15.1%, P<0.001), multivariate analysis showed no difference between the two groups in localized (P=0.174), regional (P=0.410), and all tumor stages combined (P=0.057) ( Table 3). Widowed patients had the lowest 5-year CSS. Compared to married patients, widowed patient 5-year CSS was 7.8% lower in localized stage (87.0% vs 94.8%, P<0.001), 9.8% lower in regional stage (64.2% vs 74.0%, P<0.001) and 4.3% lower in distant stage tumors (10.8% vs 15.1%, P<0.001). Compared with the married group, divorced/separated patient 5-year CSS was 1.4% lower in localized stage (93.4% vs 94.8%, P<0.001), 5% lower in regional stage (69% vs 74%, P<0.001) and 2% lower in distant stage tumors (13.1% vs 15.1%, P<0.001). Multivariate analysis also indicated that widowed and divorced/separated patients had worse survival outcomes ( Table 3).

Effect of marital status on renal cancer CSS by patient sex
We analyzed the influence of marital status on male and female patient survival separately. Log rank χ 2 test results indicated that marital status affected renal cancer CSS (P<0.001) in both men and women (Figure 2

Effect of marital status on renal cancer CSS by patient race
Multivariate analysis results showed that black patients had worse survival outcomes than white patients, while other races had no survival differences compared with whites. Log rank χ2 test results showed that marital status affected renal cancer CSS (P<0.001) in both black and white patients (Figure 3). Similar to previous findings, white married patients had better survival outcomes than other white patient groups. However, black unmarried and married patients exhibited similar survival outcomes (Table 5).

DISCUSSION
The present study found that marital status is an independent prognostic factor for renal cancer, even after eliminating the influence of patient sex, age, race, tumor size, grade, laterality, and SEER stage. Widowed and separated/divorced patients had poorer survival rates compared with married patients. While a previous study suggested that male patients had better survival rates [15], female patients in our study had better 5-year CSS compared with male patients. As indicated by multivariate analysis, black patients had poorer outcomes than white patients. Patients with larger, more advanced stage, and higher-grade tumors had shorter survival times. Patients with bilateral tumors had the lowest 5-year CSS, possibly due to the presence of metastasis at diagnosis. Finally, patients with right-side renal tumors had slight better outcomes than those with left-side tumors. The reasons for this result need to be further investigated.
Our results also showed that male renal cancer patients benefited more from marriage than females. This trend was also observed in lung, colorectal, breast, pancreatic, prostate, liver/intrahepatic bile duct, non-Hodgkin lymphoma, head/neck, ovarian, and esophageal cancer patients in another SEER-based study [16]. A potential reason for this disparity is that male patients might receive more social supports from their relatives and friends [16]. However, a systematic review and metaanalysis found no differences with respect to the effects of marriage on male and female patient outcomes [12]. Race also influenced survival in patients with different marital statuses. Our results showed that marriage improved survival in white, but not black patients. Potential reasons for these findings must be clarified in future studies.
Although the association between marital status and cancer outcomes has been extensively investigated [8-10, 14, 15], the intrinsic mechanisms behind this association remain poorly understood. On explanation might be that better financial resources and health care are available to married patients, enabling earlier disease detection and more favorable treatment options [14]. However, a recent assessment of patient insurance and neighborhood socioeconomic statuses indicated that marriage-associated survival benefits may not mediated by better access to material resources [17]. Another possible explanation is that married patients have better access to social and psychological support [14]. Cancer patients usually experience distress and anxiety from the time of diagnosis [18,19]. Without company and encouragement from a spouse, unmarried patients are more vulnerable to negative emotions. Evidence also suggests that psychological stress could promote tumor progression by disturbing normal immune and endocrine system functions [12,14,20]. Chronic psychological stress promotes cortisol secretion [21], which downregulates cortisol in white blood cells. This downregulation impairs cell responses to antiinflammatory signals and induces excessive cytokinemediated inflammatory processes [22], which are associated with cancer progression [23,24]. Moreover, long term stress dysregulates immune function by altering the Type 1-Type 2 cytokine balance, inducing low-grade chronic inflammation, and inhibiting immune protective cell functions [25], all of which are associated with cancer metastasis [26][27][28]. Additionally, diurnal cortisol rhythms were associated with cancer patient survival [29,30]. High quality psychological support from a spouse and perceived social support were associated increased natural killer cell activity [31]. Depression was also correlated with patient noncompliance with medical treatment recommendations [32].
This study had certain limitations. First, the SEER database only documented patient marital status at diagnosis. Patient marital statuses might have changed over the course of treatment, which could have influenced survival outcomes. Additionally, marriage quality might also impact spousal support levels, and this information  was not included in the database. Second, the SEER database did not provide other important information, such as patient access to socioeconomic resources, which could influence the association between prognosis and marital status.
Despite these limitations, our study demonstrated a direct association between cancer patient marital status and prognosis based on a large and representative population. In summary, married patients had better survival rates than unmarried patients. We speculated that psychosocial and socioeconomic statuses might contribute to the better outcomes of married patients. This study highlights the importance of psychological intervention for cancer patients during treatment, especially for those who are unmarried.

Data sources
All primary data were obtained from the SEER Program, including cancer incidence, stage, grade, therapy type(s), and population data, such as patient age, sex, race, and geographic region. The current dataset used for this study was based on Incidence-SEER 18

Patient selection and data extraction
We assessed renal cancer patients from the SEER database using SEER*Stat 8.3.2. Patients were selected according to the following criteria: (1) aged ≥18 years at diagnosis; (2) diagnosed between 2004 and 2013; (3) histological types were limited to clear cell adenocarcinoma and renal cell carcinoma (code 8310, 8312); (4) marital status was limited to married (including common law), divorced, separated, widowed, or single (never married). Patients with unknown cause of death, unknown survival time (months), or incomplete date information were excluded. Gender, age, race, marital status, grade, tumor size, laterality, SEER stage, cause of death, survival time, and survival months were extracted from the SEER database for each patient. This study was approved by the ethics committee of the Fourth Affiliated Hospital of Harbin Medical University (Harbin, China).

Statistical analysis
Data were analyzed based on patient age, gender, race, marital status, tumor size, tumor grade, laterality, and Patient baseline characteristics were analyzed using the χ 2 test. Patient survival rates were calculated using the Kaplan-Meier method. A multivariate Cox regression model was built to analyze survival outcome risk factors. The primary endpoint of this study was cancer causespecific death. Death resulting from renal cancer was assessed via events, and deaths due to other causes was considered censored events. All statistical analyses were performed using SPSS for Windows, v22 (SPSS Inc, Chicago, IL, USA). P<0.05 (two-sided) was considered statistically significant.