Long‐term changes in bone mineral density in postoperative patients with esophageal cancer

Abstract Aim The aim of this study was to investigate long‐term changes in bone mineral density (BMD) after esophagectomy, identify the risk factors for postoperative osteoporosis in patients with esophageal cancer and survival outcomes related to osteoporosis. Methods We retrospectively evaluated BMD changes for 197 consecutive patients with thoracic esophageal cancer who were disease‐free for 5 years after radical esophagectomy. Osteoporosis was diagnosed using computed tomography with an L1 attenuation threshold of ≤110 HU. Survival analysis was performed on 381 consecutive patients with 5‐year follow‐up after radical esophagectomy. Results BMD decreased annually after esophagectomy. The median attenuation (HU) was 134.2 before surgery and 135.2, 127.4, 123.3, 115.2, 105.6, and 102.4 at 6 months and 1, 2, 3, 4, and 5 years after surgery, respectively. Osteoporosis was diagnosed in 25.9% patients before surgery and 23.3%, 29.4%, 40.1%, 46.7%, 54.8%, and 60.4% patients with osteoporosis were observed at 6 months and 1, 2, 3, 4, and 5 years after surgery, respectively. Postoperative BMD did not decrease in patients aged ≤54 years, those who had never been smokers, and those with no weight loss after esophagectomy. Multivariate analysis identified that age (≥65 years) at surgery and smoking history were independent risk factors for osteoporosis at 5 years after esophagectomy. Patients with preoperative osteoporosis tended to have worse prognosis in disease‐free survival and overall survival than those without osteoporosis, who were more likely to die due to non‐esophageal cancer. Conclusion Esophageal cancer survivors are more likely to develop osteoporosis after esophagectomy, and preoperative osteoporosis might be associated with prognosis.


| INTRODUC TI ON
Osteoporosis is a systemic skeletal disease characterized by low bone mass and microarchitectural deterioration in bone tissue. It is initiated by an imbalance between bone resorption and formation due to modifiable risk factors (inadequate nutritional absorption, weight loss, cigarette smoking, alcohol consumption, sarcopenia), nonmodifiable risk factors (older age, sex, white ethnic background, prior fracture), or secondary disease (corticosteroid use, hyperparathyroidism, vitamin D deficiency, diabetes). 1,2 In adults of all ages, bone mineral density (BMD) loss is associated with increased fracture risk, decreased bone strength, diminished quality of life, and increased mortality. 3 Nguyen et al. demonstrated that a high rate of BMD loss was an independent predictor of all-cause mortality in elderly men and women, independent of incident fractures and concomitant diseases. 4 Patients with cancer also exhibit significant bone loss and fracture risks due to the disease, therapy for malignancy, and age-related osteoporosis. 5 Shapiro et al. 6 showed that cancer therapy induced more rapid and severe bone loss than postmenopausal bone loss in women or that in normal age-related osteoporosis in men. Moreover, cancer survivors often have additional risk factors for osteoporosis, such as advanced age, smoking, excessive alcohol consumption, impaired mobility, and reduced physical activity. 7 Therefore, the risk of osteoporosis in patients with cancer should be carefully evaluated in order to provide appropriate treatment and ensure satisfactory quality of life in the long term.
Esophageal cancer is the eighth most common and the sixth most deadly cancer worldwide. 8 Increased early detection and advances in multimodal treatments have improved survivorship among patients with esophageal cancer. 9,10 Moreover, the number of elderly patients with esophageal cancer undergoing radical surgery has increased with the increasing age of patients and technological innovations. 11 Therefore, there is an increasing focus on qualitative outcomes and various issues impacting the long-term quality of life in terms of survivorship. 12,13 Among them, osteoporosis has become an important concern for postoperative patients with esophageal cancer. However, few studies have investigated the association between esophageal cancer and osteoporosis. Moreover, there are no studies on the long-term risk for osteoporosis and survival outcomes for patients with osteoporosis.
In the present study, we aimed to investigate long-term BMD changes after esophagectomy and identify the risk factors for postoperative osteoporosis in patients with esophageal cancer. Furthermore, we investigated survival outcomes related to osteoporosis after radical esophagectomy. Finally, 197 patients who maintained a disease-free status during the 5-year follow-up period were included in this retrospective study ( Figure S1). Data on patient characteristics, surgical outcomes, clinicopathological features, and postoperative findings were reviewed from the medical reports. The patients were evaluated using esophagoscopy, computed tomography (CT), or positron emission tomography. The histopathological findings were classified according to the Union for International Cancer Control (UICC) Tumor, Nodes, Metastasis (TNM) classification system. 14 Heavy drinking was defined as consumption of three units or more of alcohol per day; one unit equals 20 mg of pure alcohol. Postoperative complications were assessed by the Clavien-Dindo classification. 15 Postoperative dysphagia was evaluated by a videofluoroscopic swallowing study as previously described. 16 The rate of weight change was assessed relative to the preoperative weight. Since the median weight change at 1-year after surgery was 88% in the present study, we defined severe weight loss as ≤85% weight change between the time of surgery and post-surgery. Previous studies show that 10%-15% of unintentional weight loss is predictive of poorer clinical outcomes for patients with cancer and thus reduces the possibility of successful postoperative recovery. 17

| Preoperative treatment
In our institute, preoperative chemotherapy has been the standard treatment for advanced esophageal cancer, and it involves a triplet regimen with docetaxel (70 mg/m 2 ), cisplatin (70 mg/m 2 ), and 5-fluorouracil (FU) (700 mg/m 2 /day). A doublet regimen with cisplatin (80 mg/m 2 ) and 5-FU (800 mg/m 2 /day) was selected depending on the patient's condition, age, and other comorbidities.
Chemoradiation therapy (CRT) was selected as a preoperative treatment for patients with thoracic esophageal cancer with suspected cervical esophageal infiltration, disease refractory to initial chemotherapy, suspected cT4 disease, simultaneous double cancer, and for those who requested it. The CRT regimen comprised simultaneous radiation with cisplatin (70 mg/m 2 ) and 5-FU (700 mg/m 2 ).
External beam radiation therapy was administered at 1.8-2.0 Gy per fraction per day, five fractions per week, for a total dose of 40-60 Gy. The radiation field was designed to encompass the primary tumor and metastatic lymph nodes with an ample margin.
The field of preventative irradiation included the bilateral supraclavicular fossae and superior mediastinal lymph nodes in patients with upper esophageal cancer, and the mediastinal lymph nodes in patients with middle or lower esophageal cancer. Surgery was scheduled within 6 weeks after completion of the last cycle of CRT or chemotherapy.

| Bone mineral density (BMD)
BMD was assessed by CT attenuation values for the L1 vertebral body, as previously described by Pickhardt et al. 18,19 CT attenuation in a region of interest (ROI) was recorded in Hounsfield Units (HU), similar to the method used to generate BMD measurements using dual X-ray absorptiometry (DXA), and it represents a linear transformation of the attenuation coefficient. A single oval click-and-drag ROI was placed over an area of trabecular bone in the vertebral body, and CT attenuation was measured between the T12 and L5 vertebral levels, with emphasis on the L1 measures. ROI placement near areas that would distort the BMD measurements (posterior venous plexus; focal heterogeneity or lesion, including compression fracture; and imaging-related artifacts) was avoided. Osteoporosis was diagnosed using an L1 attenuation threshold of ≤110 HU, as previously reported. [18][19][20] In the study by Pickhardt, these criteria specifically distinguished osteoporosis from osteopenia and normal BMD. 18 L1 attenuation was measured before and after 6 months, and 1, 2, 3, 4, and 5 years after radical esophagectomy.

| Statistical analysis
Results are expressed as median (range) for continuous variables or percentage for categorical variables. We retrospectively analyzed the associations between groups using χ 2 tests, the Mann-Whitney U test, and one-way repeated measures analysis of variance (anova).
Cox proportional hazards regression analysis was used for the univariate and multivariate analyses. Disease-free survival (DFS) was calculated from the date of surgery to the date of the first evidence of recurrence, death owing to any cause, or last follow-up in patients without recurrence. Overall survival (OS) was calculated from the date of surgery to the time of death owing to any cause or to the last follow-up. Cancer-specific survival and non-cancer-specific survival were defined as the time from the date of diagnosis to the date of death due to esophageal cancer and diseases other than esophageal cancer, respectively. Survival analyses were performed using the Kaplan-Meier method, and survival curves were compared using log-rank test. All statistical tests were two-sided, and the threshold for statistical significance was set at p = 0.05. Statistical analyses were performed using JMP® Pro 15.1.0 (SAS Institute Inc.). Table 1 shows the clinicopathological characteristics and surgical outcomes in 5-year disease-free patients. The median age was 65.0 (range 37-79) years. Risk factors for osteoporosis, such as female gender, heavy alcohol consumption, smoking history (current smoker/ex-smoker), and respiratory dysfunction, were noted for 20% (n = 39), 43% (n = 84), 87% (n = 171), and 12% (n = 23) of patients, respectively. Retrosternal reconstruction was selected in most patients (81%, n = 160).

| Body weight change after esophagectomy
We investigated body weight changes in patients with esophageal cancer. Body weight significantly decreased until 6 months after surgery, with no changes thereafter (Figure 2A). Body weight decreased in more than 90% of patients after esophagectomy; however, the proportion of patients with body weight change remained almost constant from 6 months after esophagectomy ( Figure 2B).    Figure 3C). There were no significant differences in postoperative BMD changes according to the sex of the patient ( Figure 3D).

| Evaluation of risk factors for osteoporosis at 5 years after esophagectomy
Potential risk factors for osteoporosis at 5 years after esophagectomy included preoperative characteristics, surgical factors, pathological characteristics, and postoperative outcomes. Univariate analysis showed that ages ≥65 at surgery, heavy drinking at diagnosis, smoking history (current smoker/ex-smoker), percentage forced expiratory volume in the first second (FEV1%) <70% at diagnosis, and pT3 disease were significant risk factors for osteoporosis at 5 years after esophagectomy. Multivariate analysis revealed that age ≥65 years at surgery and smoking history were independent risk factors for osteoporosis at 5 years after esophagectomy (p < 0.001 and p = 0.024, respectively) ( Table 2).

| Survival outcomes
Survival analysis was performed in 381 patients with 5-year followups ( Figure S1). Among them, 121 patients (32%) were diagnosed with osteoporosis before surgery. We compared preoperative patient characteristics between the osteoporosis and non-osteoporosis groups in Table S1. Patients with preoperative osteoporosis were significantly older than those without osteoporosis (median 68.0 vs. 64.0, p < 0.001), while there were no significant differences in sex, BMI, preoperative biochemical micronutrients (calcium and phosphate), preoperative treatment, and pathological tumor-nodemetastasis (pTNM) stage.
Patients with preoperative osteoporosis tended to have worse prognosis in DFS and OS than those without osteoporosis (log rank p = 0.068 and 0.082, respectively) ( Figure 5A,B). The 5-year OS was 51.2% in osteoporosis and 63.5% in non-osteoporosis patients ( Figure 5B). There was no significant difference in cancer-specific OS between the preoperative osteoporosis and non-osteoporosis groups (log rank p = 0.304) ( Figure S2A). However, patients with preoperative osteoporosis tended to have worse prognosis in non-cancer-specific OS than those without osteoporosis (log rank p = 0.084) ( Figure 5C).

| DISCUSS ION
In the present study, we found that postoperative BMD decreased annually over a period of 5 years after esophagectomy, and more than half of the patients developed osteoporosis by 5 years after esophagectomy. Multivariate analyses demonstrated that independent risk factors for osteoporosis at 5 years after esophagectomy were ages ≥65 years at surgery and smoking history.
Moreover, patients with preoperative osteoporosis tended to have worse prognosis than those without osteoporosis and were more likely to die due to non-esophageal cancer. This is the first study to report long-term BMD changes after esophagectomy and survival outcomes related to osteoporosis in patients who undergo radical esophagectomy.
Habitual alcohol consumption and smoking are the most common risk factors for esophageal cancer. 21   However, no significant differences in OS were observed in 238 patients ( Figure S2B) 24 Third, although we revealed the incidence of osteoporosis after esophagectomy, we could not investigate bone fracture and the long-term quality of life in patients with osteoporosis after esophagectomy. In general, bone loss enhances bone fragility and increases fracture risk, which results in impaired quality of life. 3 Moreover, osteoporosis can be a predictor of all-cause mortality in elderly men and women. 3,4 Further studies are required to clar- interventional treatments are required in patients with a high risk of developing osteoporosis, such as the elderly or those with a smoking history.

ACK N OWLED G EM ENTS
We would like to thank Editage (www.edita ge.com) for English language editing.

D I SCLOS U R E
Conflict of Interest: The authors declare no conflict of interests for this article.