Quality of Life for Patients with Esophageal / Gastric Cardia Precursor Lesions or Cancer : A One-year Prospective Study

Esophageal cancer is a common malignancy. Its incidence and mortality, respectively, rank at eighth and sixth among all cancers, with 456,000 new cases and 400,000 deaths in 2012 worldwide (Ferlay et al., 2013). About half of these cases and of deaths occur in China, giving it the highest incidence and mortality rates in the world. Although the mortality of esophageal cancer has decreased over the last three decades, it remains the fourth leading cause of cancer deaths in China (Wei et al., 2011). There is also wide disparity in the burden of esophageal cancer. In some rural areas with limited health resources and under-developed economy, the mortality of esophageal cancer is three times higher than the average of the country (Chen, 2008). Esophageal cancer is considered as a major public health challenge in these high risk areas (Wei et al., 2011b; Wang et al., 2013).


Introduction
Esophageal cancer is a common malignancy.Its incidence and mortality, respectively, rank at eighth and sixth among all cancers, with 456,000 new cases and 400,000 deaths in 2012 worldwide (Ferlay et al., 2013).About half of these cases and of deaths occur in China, giving it the highest incidence and mortality rates in the world.Although the mortality of esophageal cancer has decreased over the last three decades, it remains the fourth leading cause of cancer deaths in China (Wei et al., 2011).There is also wide disparity in the burden of esophageal cancer.In some rural areas with limited health resources and under-developed economy, the mortality of esophageal cancer is three times higher than the average of the country (Chen, 2008).Esophageal cancer is considered as a major public health challenge in these high risk areas (Wei et al., 2011b;Wang et al., 2013).

Quality of Life for Patients with Esophageal/Gastric Cardia Precursor Lesions or Cancer: A One-year Prospective Study
Ying Wen 1 , Xiong-Fei Pan 1 , Wen-Zhi Huang 2 , Zhi-Mei Zhao 1 , Wen-Qiang Wei 3 , Feng Chen 1 , Hui Lan 1 , He Huang 1 , Chun-Xia Yang 1 *, You-Lin Qiao 3 To date, major risk factors for esophageal cancer are not well understood and specific primary prevention strategies are lacking; screening, early diagnosis and treatment are regarded as important methods for esophageal cancer prevention and control.Screening for esophageal cancer aims to detect the lesion at an early stage and alter its natural course before symptoms are observed (Tomizawa and Wang, 2009).The prognosis of patients with esophageal cancer is dismal, with overall 5-year survival rates of 10-30% even in developed countries (Rouvelas et al., 2005;Jemal et al., 2006).Most esophageal cancer cases are asymptomatic at an early stage, and once diagnosed, the condition may be locally advanced and unresectable.Since the survival of esophageal cancer is strongly correlated with the stage at diagnosis, the detection of early neoplasia and subsequent treatment can lead to an improved outcome, with the 5-year survival rate increasing over 85% (Wang et al., 2004).In such context, the Early Detection and Early Treatment of Esophageal Cancer (EDETEC) Program, funded by Chinese Central Financial Transfer Payment Program, has been conducted in several high-risk areas in China since 2005 (Wei et al., 2011a).
In clinical practice, survival and mortality rates are the main outcome measurements in most cancer research and practice, because of their obvious relevance and ease of calculation and interpretation (Scarpa et al., 2013).However, with the development of self-administered structured and validated instruments for the assessment of quality of life (QoL) for cancer patients, QoL research appears increasingly important (Darling, 2013).The QoL outcome not only reflects the influence of disease and treatment on patients' physical, psychological and social function and informs clinical decision, but also provides information for health economic evaluation to select suitable prevention strategies of esophageal cancer, such as early detection and treatment (Avery et al., 2007;Lagergren et al., 2007;Zeng and Liu, 2012).Although QoL assessments for patients with esophageal cancer have attracted attention in academia, most studies focus on QoL for patients receiving different treatment modalities (Reynolds et al., 2006;Zeng and Liu, 2012;Zapletal et al., 2014), and very few have explored QoL for patients with different clinical stages of esophageal cancer during treatment and follow-up.
There are two types of rating scales used to measure QoL among cancer patients: one is for general measures of QoL for patients with various cancers (such as EORTC QLQ-C30 and FACT-G), and the other for measuring disease-specific issues for patients with specific cancers (such as QLQ-OES18 and FACT-E for esophageal cancer patients and QLQ-LC13, FACT-L for lung cancer patients).For esophageal cancer patients, QLQ-C30 and QLQ-OES18 have shown good psychometric and clinical validity, and the combined use of the two rating scales can reflect patients' QoL accurately and objectively (Aaronson et al., 1993;Blazeby et al., 2003;Parameswaran et al., 2010).
This study was conducted using QLQ-C30 v3.0 and the esophageal cancer-specific module QLQ-OES18 to evaluate the QoL for patients with esophageal/gastric cardia cancer at different clinical stages and its change along the timeline of treatment in Linzhou, a high risk area in China.Gastric cardia cancer was not distinguished from esophageal cancer regarding diagnosis and treatment in clinical practice in many hospitals in China when the study was conducted.Thus our study did not deliberately focus only on esophageal precursor lesion and cancer, and gastric cardia cancer and precursor cases were also enrolled.We aimed to explore the relationship between early detection and QoL for esophageal/gastric cardia cancer patients and provide information to facilitate improved QoL in clinical practice.

Research subjects
The study enrolled patients with severe dysplasia or more of esophageal/gastric cardia mucosa at Linzhou Cancer Hospital through the EDETEC program or who were referred to the hospital after diagnosis between 2009 and 2010 and followed to 2011.All patients were aware of their disease condition and voluntarily participated in the survey.None of them were unable to understand the content of the questionnaires or had other previous or concurrent malignancies.Informed consent was obtained from all participants prior to survey.The study was approved by the Institutional Review Board of the Cancer Institute of Chinese Academy of Medical Sciences.
For the purpose of this study, participants were divided into three groups based on pathology diagnosis: the precursor lesion group, including severe dysplasia and carcinoma in situ; the early stage cancer group, including intramucosal carcinoma and submucosal carcinoma; and the advanced cancer group, i.e. invasive carcinoma.Patients were treated according to recommended clinical guidelines.For patients with precursor lesion, endoscopic mucosal resection (EMR) or argon plasma coagulation (APC) was selected; for those with early stage cancer, EMR or surgery was suggested; and for those with advanced cancer, common treatment modalities, including surgery, radiotherapy, chemotherapy, or a combination, were chosen depending on physical condition, disease severity and patients' requests for the treatment.

Questionnaire survey
All participants were surveyed with the Chinese versions of QLQ-C30 v3.0 and QLQ-OES18 before primary treatment, and at 1, 6 and 12 months after treatment (±2 weeks).Both questionnaires were developed by the European Organization for Research and Treatment of Cancer (EORTC) and well validated in Chinese population (Wan et al., 2005;Xu et al., 2007).The QLQ-C30 v3.0 has been used to assess QoL for patients with different cancers in international clinical trials (Aaronson et al., 1993;Farooqui et al., 2013).It includes 30 items in one global QoL scale, five function scales (physical, role, cognitive, emotional, and social), three symptom scales (fatigue, nausea/vomiting, and pain), and six single items (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial difficulties).The QLQ-OES 18 is a QLQ-C30 supplement that is designed to collect information on esophageal/gastric cardia cancer-specific symptoms and side effects.Its 18 questions are categorized into four symptom scales (dysphagia, eating difficulties, reflux, and esophageal pain) and six single items (trouble swallowing saliva, choking when swallowing, dry mouth, trouble with taste, trouble with coughing, and speech difficulties).There are four responses in each question of both questionnaires: "not at all", "a litter", "quite a bit", and "very much", except for the global QoL scale, which has seven responses ranging from "very poor" to "excellent".All responses are converted linearly into scores on a 0 to 100 scale according to the EORTC scoring manual (Fayers et al., 2001).High scores for global QoL and function scales represent better levels of overall QoL and function, while high scores for symptom scales/items indicate more or worse symptoms.A 5 to 10-point difference in mean scores between comparison groups or time points is considered DOI:http://dx.doi.org/10.7314/APJCP.2015.16.1.45Quality of Life for Patients with Esophageal/Gastric Cardia Precursor Lesions or Cancer: A One-year Prospective Study clinically significant (Osoba et al., 1998).
The patients were interviewed face to face in the first survey before primary treatment, and by phone at different time points after treatment.All interviewers were welltrained postgraduate students or health care workers.When the patients had difficulty in comprehending the meaning of items in questionnaires, neutral explanations were made by the interviewers to avoid biased responses.Demographic and clinical information was collected from their medical records and verified in the surveys.

Statistical analysis
IBM SPSS Statistics version 20.0 (Armonk, New York, United States) was used for statistical analysis.Demographic and clinical data were compared between different stage groups using Chi-square test or Student's t-test.QoL scores for each measure were expressed as mean±standard deviation (SD).Comparisons between groups were made using multivariate analysis of variance (MANOVA) at each time point (for each of the measures), adjusting for age and sex, to identify QoL differences between patients with different clinical stages.Comparisons between specific assessment points (withinclinical-stage group) were conducted using analysis of variance for repeated measures to assess QoL changes over time.p<0.05 was considered statistically significant for Chi-square test or Student's t-test, whereas for comparison of QoL scores, p<0.01 was a level of significance due to multiple testing.

Characteristics of patients
A total of 166 patients participated in the survey before primary treatment, including 60 with precursor lesion, 58 with early stage cancer, and 48 with advanced cancer.During the 12 months after treatment, 59 with precursor lesion, 57 with early stage cancer, and 43 with advanced cancer (5 were died before the last survey) were successfully followed up.The patients who did not complete the survey at all four time points were excluded from subsequent data analysis.

Comparisons of QoL scores for patients with different stages of esophageal/gastric cardia lesions
Before primary treatment, there were no significant differences in all measures of QoL between the precursor lesion and early stage cancer groups (p>0.01), and function scales were generally good in both groups (mean scores>90).Patients in the advanced cancer group reported worse cognitive and social functions, and more problems with insomnia, appetite loss, financial difficulties, eating, reflux, and choking compared with those in the other two groups (difference between means>10, p<0.01) (Table 2).
At 1 month after treatment, patients with early stage or advanced cancer had worse QoL in most function and symptom measures than those with precursor lesion (Table 2).QoL declined most in the early stage cancer group, which showed lower global QoL, physical, role, and social function scores, and higher fatigue, pain, dyspnea, and appetite loss scores compared with corresponding baseline values (difference between means>5, p<0.01).There was significant deterioration in fatigue, pain, and insomnia scores in the precursor lesion group, and in global QoL, social function, pain, and financial difficulties in the advanced cancer group (difference between means>5, p<0.01) (Table 3).
At 6 months after treatment, scores for fatigue, pain, appetite loss, and dry month domains in the precursor lesion group returned to baseline values (p>0.01)(Table 3).Compared with 1 month after treatment, physical, role, social, fatigue, pain, and dyspnea measures in the early stage cancer group showed significantly superior results (difference between means>5, p<0.01), while most domains in the advanced cancer group did not demonstrate At 12 months after treatment, patients with early stage and advanced cancer still had lower global QoL scores than those with precursor lesion (difference between means>5, p<0.01) (Table 2).There were no significant differences in all measures compared to baseline levels in the precursor lesion group (p>0.01).However, appetite loss, reflux, and dry month in the early stage cancer group, role function, financial difficulties, and dysphagia in the advanced cancer group, and global QoL, social function, pain, and insomnia in both groups showed significantly worse results than the baseline (difference between means>5, p<0.01) (Table 3).

Comparisons of QoL scores for patients at different time points
In the precursor lesion group, function and symptom scores did not change over time, except fatigue, pain, insomnia, appetite loss, and dry month (p<0.01)(Table 3).In the early stage cancer group, some measures were worse at 1 month after treatment, and then gradually recovered, but some constantly deteriorated, such as appetite loss and reflux.Patients with advanced cancer consistently reported problems with insomnia, appetite loss, financial difficulties, dysphagia, eating, reflux, and choking, for which mean scores exceeded 15 points throughout the study.The early stage cancer group scored better than the advanced cancer group for most function and symptom measures (p<0.01),although the global QoL scores in both groups were similar between 1 month to 1 year after treatment, which were lower than those in the precursor lesion group (difference between means>5, p<0.01) (Table 2).Scores for constipation, diarrhea, saliva problems, taste, cough, and speech difficulties in three groups were around zero and did not change much over time, and thus were not presented in tables.Nausea/vomiting and esophageal pain results showed similar trends as those

Discussion
In the context of bio-psychosocial medical model, patients' QoL is evolving into a key issue in modern cancer research and clinical practice (Farooqui et al., 2013).QoL outcomes provide information about patients' subjective feeling of physical and psychosocial health as well as indications of prognosis (Gockel et al., 2010;Djarv and Lagergren, 2011;Quinten et al., 2014).QoL measures with survival and mortality data reflect curative effects on malignancies and rehabilitation, inform treatment choices, and improve doctor-patient communication (Velikova et al., 2004;Lin et al., 2012).In the current study, although the pretreatment QoL scores for all measures did not significantly differ between the precursor lesion and early stage cancer groups, patients with precursor lesion scored better than those with early stage cancer in most function and symptom measures at time points following treatment.The advanced cancer group always had the worst QoL scores among the three.In addition, the measures which deteriorated at 1 month after treatment in the precursor lesion group finally recovered to baseline at last assessment, while some measures in the early cancer group did not return to baseline values.In the advanced cancer group, some measures did not demonstrate much change during the follow-up period, while others continually deteriorated.
In most measures, the precursor lesion and early cancer groups reported superior QoL scores to the advanced cancer group throughout the study.This discrepancy was probably due to significant variability among patients' baseline condition, treatment approaches, demographic characteristics, and mental states.Before primary treatment, precursor lesion and early esophageal/gastric cardia cancer are usually asymptomatic and do not affect daily activities of patients, while advanced cancer is always accompanied by a series of symptoms, like eating difficulties, reflux, and pain in the chest and back.Besides, different treatment approaches may have differing effects on patients' QoL.Our patients diagnosed with precursor lesion usually received endoscopic treatments such as EMR and APC, which are effective, minimally invasive, and cost-effective (Peters et al., 2005), while those with esophageal/gastric cardia cancer were treated with surgery alone or combined with neoadjuvant radiotherapy and/or chemotherapy, and advanced cancer often needed a longer period of treatment.As the patients with advanced cancer were older than those with early stage lesions, tumor biology and therapeutic methods may have a larger impact on them.Moreover, advanced cancer patients usually suffer greater shock and concern, which is reflected by the fact that emotional function scores in this group were lower than those in the other two groups throughout the current study.All these led to inferior QoL in advanced cancer patients.Since most precursor lesion and early cancer patients were identified through the screening program, while most advanced cancer patients were referred to the hospital with typical symptoms, it is important to note that better QoL will be achieved if esophageal/gastric cardia lesions are detected at an early stage.In addition, our study showed that the advanced cancer group reported more problems with financial difficulties than the other two groups during the follow-up.The mean scores for financial difficulties in this group gradually increased (got worse) over time.These may resonate with an early observation that screening, early diagnosis and treatment for esophageal/gastric cardia cancer could provide great cost savings (Wei et al., 2011a;Yang et al., 2011).
QoL scores showed distinct patterns of change along the timeline of treatment for different clinical stages of esophageal/gastric cardia lesions.For patients with precursor lesion or early esophageal/gastric cardia cancer, some QoL domains, like fatigue and pain, deteriorated in the first month after primary treatment and gradually improved during the remaining follow-up period.We presume that prescribed treatment modalities for esophageal/gastric cardia cancer are related to marked injuries of body and psychology during the months following treatment; for example, dyspnea in patients treated with surgery may be associated with reduced lung volume due to the presence of an intrathoracic stomach (Lagergren et al., 2007).As a result, early posttreatment QoL scores are impaired in comparison to baseline levels, which is consistent with other studies (Reynolds et al., 2006;Lagergren et al., 2007;Parameswaran et al., 2010).With fading immediate side effects of treatments and an increased tolerance towards residual symptoms with time, QoL for patients with early stage lesions would improve within 6 to 12 months after treatment.But for advanced cancer patients, some QoL measures did not change much and some gradually got worse over time.So it is important to inform patients with different clinical stages what problems may occur and what to except after treatment.Most patients with early stage lesions can expect a generally good QoL.For advanced cancer patients who consistently report problems in some QoL domains, physicians should take corresponding measures and offer extra counseling to relieve their concern and improve their QoL (Azmawati et al., 2014).Timely supportive interventions may help patients to cope with their posttreatment problems more easily.Besides, since QoL has prognostic value and may be associated with mortality and survival (Djarv and Lagergren, 2011;Quinten et al., 2014), health care providers should focus on not only clinical parameters but also patients' perception of their QoL.Clinicians must realize that the patient-reported QoL is one of the most important considerations, and QoL assessments should be incorporated into routine clinical practice to guide decisions.
It seems unusual that although most measures in the early cancer group were better than those in the cancer group after treatment, patients in both groups had similar global QoL scores.Besides, changes of many QoL measures between baseline and 1 month after treatment in early stage cancer patients were larger than those in advanced cancer patients.These may be explained by response shift.Adjustments to impairments of physical function and esophageal/gastric cardia cancer symptoms experienced before primary treatment may lead to reevaluation of internal standards in advanced cancer patients.As a result, an adaptation of the perception of QoL (i.e.response shift) may occur, and many measures, including global QoL scale, in the advanced cancer group would be affected after treatment (Sprangers and Schwartz, 1999;Blazeby et al., 2005).There may be much larger score change associated with patients who reported worsening in this group.Response shift is an important confounding factor in current QoL research and difficult to measure.Even using a combination of generic and disease-specific tools, it may be still too crude to detect discrepancies between patients with different clinical stages (Blazeby et al., 2005).
The study is subject to certain limitations.Firstly, because of a small sample size and incomplete information, we were not able to quantify the effects of treatments and other clinical characteristics as contributing factors to the variation of QoL scores in the follow-up and between different stages of esophageal/gastric cardia lesions.Based on this study, large-scale and long-period follow-up investigations should be conducted to evaluate the QoL and its influential factors for patients with different clinical stages of esophageal/gastric cardia cancer.Secondly, at baseline, patients had already been diagnosed with esophageal/gastric cardia lesions, and their QoL may be affected by tumor biology and psychosocial factors in an unpredictable and individual way, so using pretreatment data as comparator values to assess changes in QoL may lead to inherent flaws (Derogar and Lagergren, 2012).However, a true baseline (i.e.before the disease has occurred) is not feasible to obtain.
In conclusions, the QoL was better for patients at an early stage of esophageal/gastric cardia lesions.At 12 months after treatment, all measures in the precursor lesion group showed no significant differences compared with pretreatment values, while some measures in advanced esophageal/gastric cardia cancer patients gradually deteriorated.Our results indicate that screening, early diagnosis and effective treatment among high risk populations enhance the QoL for esophageal/gastric cardia cancer patients.As QoL plays an important role in clinical practice, QoL assessments should be incorporated into routine health care and guide clinical decisions together with mortality and survival.Target intervention and counseling should be given by health care providers during treatment and follow-up to facilitate the QoL improvement.
Figure1shows changes in mean scores for representative QoL domains (physical function, fatigue, and dysphagia) for 3 groups along the timeline of treatment.

Figure 1 .
Figure 1.Changes in Mean Scores for Representative QoL Domains for Different Clinical Stage Groups Along the Timeline of Treatment.A: Physical function; B: Fatigue; C: Dysphagia.(T1, before primary treatment; T2, 1 month after treatment; T3, 6 months after treatment; and T4, 12 months after treatment) 0

Table 1 . Demographic and Clinical Characteristics of Patients in Different Stage Groups
Scores for insomnia symptom increased in both the early stage and advanced cancer groups (Table2).

Table 2 . Mean Scores (SD) for Selected QLQ-C30 and QLQ-OES18 Domains at Different Time Points for Different Stage Groups
T3, 6 months after treatment; and T4, 12 months after treatment.For global QoL and function scales, higher scores represent better levels of overall QoL and function; for symptom domains, higher scores represent more or worse symptoms.Compared with the precursor lesion group.*p<0.01 alone; **p<0.01 and difference between means>5; ***p<0.01 and difference between means>10