The Impact of Cancer Concealment on Postoperative Esophageal Cancer Patients in Stage T3: A Propensity Scores Matching Analysis

Esophageal cancer has a poor prognosis and high mortality rate, and the overall ve-year survival rate is pretty low. In addition to medical factors, some psychological and sociological factors can also contribute to it. Methods We selected postoperative T3 esophageal cancer patients hospitalized from June 2015 to December 2016 as research subjects, and divided them into a cancer-awareness group and a cancer-concealment group. Several variables are allowed for evaluating 36-month progress-free survival of the two groups. Propensity Scores Matching analysis were used to adjust selection bias. There were 72 patients enrolled in the study. Sex and cancer concealment were signicantly predicted poor survival while stage, pathological differentiation, education background and age were not (P = 0.017, 0.020, 0.302, 0.177, 0.068, 0.054, respectively). Cox multivariate regression analysis shows sex and cancer concealment were independent predictors of progress-free survival of esophageal cancer at postoperative T3 stage (harzard ratio = 0.264, 95% condence interval: 0.099–0.703, P = 0.008; harzard ratio = 2.823, 95% condence interval: 1.360–5.861, P = 0.005, respectively).


Introduction
Esophageal cancer (EsC) is one of the most fatal malignancies in the world. It is the 4th leading cause of death from cancer and the 6th most common cancer in China in 2014 (1), and is likely to hit 4th in mortality and 3rd in incidence with a total of 188044 death and 245651 new cases estimated in 2015 (2,3). The two main subtypes are esophageal squamous cell carcinoma (ESCC) and adenocarcinoma, which account for more than 95% of EsC, and ESCC is the major histology in Asian countries (4). The risk factors for ESCC include sex, smoking and achalasia (5). Important risk factors for esophageal adenocarcinoma include chronic gastroesophageal re ux disease, obesity and smoking(6-8). A prevalence study indicated that ESCC is 2-4 fold more common in males than females (9). Another study showed that the male/female ratio of Barrett's esophagus patients is about 2:1 (10), and the incidence rate of esophageal adenocarcinoma shows a 38-fold increase in males over females (10,11). It has been reported that current smokers have an increased risk of ESCC, as compared to nonsmokers (odds ratio (OR) = 2.9; 95% con dence interval (CI), 2.1-4.1) (12). Besides, some studies referred to alcohol found the average weekly alcohol intake exceeded 170g, and the OR was signi cantly increased in ESCC patients but not in esophageal adenocarcinoma patients (13).
Yoshifumi et al. (14) believed that multiple primary EsC were signi cantly associated with alcohol use and tobacco smoking, and it may be an independent predictor of poorer long-term survival. Furthermore, dietary factors can also contribute to EsC. People at a higher risk for Barrett's esophagus and esophageal adenocarcinoma may bene t from a reduction in the consumption of red meat and an increase in the consumption of fruits and vegetables (15).
EsC has a poor prognosis and high mortality rate. Despite development in diagnosis and treatment, the overall ve-year survival rate is 15-20% worldwide(16), may be because it is often diagnosed during its advanced stages, the main reason being the lack of early clinical symptoms (17). Besides, some social and psychological factors are also relevant to the prognosis of patients with EsC, such as cancer concealment. However, compared with these medical factors, there are only a few studies refer to it. Researches in other disciplines have showed that patients who gain truth telling of cancer diagnosis experience more favorable survival than those who do not. Li et al. compared 10 030 colorectal cancer patients and found that cancer concealment was an independent predictor associated signi cantly with poor survival(18). Jiang et al.
enrolled 865 non-small cell lung cancer patients and found that cancer-speci c and all-cause survival was poorer in the cancer concealment group (19). However, as far as we know, there are no study have assessed the association between cancer awareness and the prognosis of EsC patients. In according to do that, we carried out a retrospective study to evaluate it on postoperative EsC patients in stage T3.
We present the following article/case in accordance with the STROBE reporting checklist.

Patients
We selected postoperative T3 EsC patients who treated in the same medical group at Sichuan Science City 2 Trial design Cancer concealment was de ned as being concealed cancer more than one month by patients′ family members, while cancer awareness was de ned as their family members not hide or just hide the cancer to them less than one month. 36-month progress-free survival (PFS) is the major observation target. Social Page 4/12 variables, such as sex, age, education background, tumor stage and tumor differentiation, were used to evaluate it.
3 Statistical analysis SPSS 22.0 (IBM Corp., Armonk, NY, USA) software was used for the statistical analysis. Survival curves were analyzed by Kaplan-Meier method. Continuous data were expressed as the mean ± standard deviation or the median. A t-test was used to compare normally distributed data according to whether the variance was equal or not. The Mann-Whitney U test was used to compare non-normally distributed data sets. Categorical data were expressed as counts and percentages, and comparisons between groups were performed using the chisquare test or the Fisher exact probability method. Indicators with statistically signi cant results were incorporated into a Cox regression analysis for multivariate analysis of risk factors of PFS. We used Propensity Scores Matching (PSM) to adjust selection bias. The difference was statistically signi cant at P < 0.05.

Patients characteristics
There were 72 patients enrolled in the study, including 55 men and 17 women. The youngest was 45 and the oldest was 77 years old. The sex, age, education background, and location, histopathology, differentiation and stage of tumor between the two groups were not signi cantly different. The clinical characteristics of the two groups before and after PSM are shown in Table 1. Abbreviations: PSM, propensity scores matching.

PFS and risk variables of esophageal cancer progression
We rstly analyzed the PFS using Kaplan-Meier curves and discovered that sex, cancer concealment, stage and pathological differentiation were signi cantly predicted poor survival, while education background and age was not, as indicated in Fig. 1. After PSM analysis, we found that sex and cancer concealment were signi cantly predicted poor survival, and both of them were independent predictors of PFS of EsC patients at postoperative T3 stage when using cox multivariate regression analysis (Fig. 2, Table 2).  (25). However, some short-term studies showed that overall morbidity, incidences of anastomotic leakage and reoperation within 30 days, and operative time in the MIE group were signi cantly higher or longer than in the OE group, while blood loss, incidence of atelectasis and the population of patients who required more than 48h of postoperative mechanical ventilation were signi cantly lower in the MIE group than in the OE group(26, 27). Besides, many previous reports have shown the superiority of three-eld lymph node dissection (3FLD) as compared with two-eld LN dissection (2FLD). As Nishihira reported, 5-year survival was better after 3FLD (64.8%) than 2FLD (48.0%) (28), and Udagawa also showed that cervical lymph node dissection had a high e ciency index for upper and middle thoracic EsC (29).
Despite improvement in treatment, it's hard to say we won the game, because of the poor outcome. Therefore, we should transfer our attentions to other directions. Some researchers believe successful cancer treatment relies on effective communication, informed decision making, treatment adherence, and the use of effective coping strategies (30 shorter than cancer-awareness group, and cancer concealment was an independent risk factor associating with poor survival (Fig. 2 including tumor, had no increased depression, anxiety and hopelessness for them (33). Otherwise, some patients prefer to obtain information about the disease and its treatment because they want to receive psychological support from other patients or staff, learn more about other patients' experiences, learn how to manage daily problems, and provide support to other patients (34). Therefore, cancer awareness is indispensable and contributes to a longer survival for postoperative T3 stage EsC.
Our study manifested that tumor stage were not independent predictors of PFS of EsC patients at postoperative T3 stage, which was opposite to the nd of Cao et al. (35) that larger tumor size, advanced grade, depth of invasion and increased number of metastasized lymph nodes were independent prognostic factors of EsC. This is because the cases are limited and its representation is weak. Besides, even if we used PSM to adjust selection bias, this is a retrospective study and the clinical persuasiveness is feeble. Therefore, a long-term research and more cases can be involved to evaluate the relation between cancer concealment and overall survival for future prospective research.