Does Esophagectomy Provide a Survival Advantage to Patients Aged 80 Years or Older? Analyzing 5066 Patients in the National Database of Hospital-based Cancer Registries in Japan

With the recent increases in the elderly population in developed countries, cancer treatments, especially invasive surgical treatments, for the elderly are receiving attention.¹ When selecting cancer treatments for elderly patients, one must evaluate whether the expected outcome merits the risk, given the patients’ shorter life expectancy and lower stress tolerance. Although there is no worldwide consensus on the age limit after which one is considered elderly, for highly invasive surgical treatments such as esophagectomy, patients aged 75 years or older are generally recognized as elderly.^2–4 An important statement from the National Comprehensive Cancer Network in the United States is that advanced age alone should not preclude the use of effective treatments that could improve quality of life or extend meaningful survival.¹ In clinical practice, treatment strategies are determined based on the patient's general conditions and underlying disease, in addition to the cancer stage, while respecting the patient's will.³ The underlying problem is that most elderly patients have multiple comorbidities, are in comparatively poor physical condition, and a calculated healthy longevity that is not long. For such elderly patients, there are limitations to using the full standard cancer treatment as specified in clinical practice guidelines.^3,5–6 This is in part because the clinical trials on which the guidelines are based do not include data on the elderly. As a result, little clear standard treatment for the elderly has been established.

Esophageal cancer is a relatively common malignancy, with 572,034 new cases in 2018 around the world (international Agency for Research on Cancer, Cancer Today, https://gco.iarc.fr/today/home). In Japan, there were 25,845 new cases in 2016, and the peak population was 65 to 75 years old. Among these patients, 4319 (16.7%) were 75 to 79 years old and 5013 (19.4%) were 80 years old or older (Cancer Incidence in Japan 2016. https://www.mhlw.go.jp/content/10900000/000553552.pdf). As in Japan, esophageal cancer patients in the United Kingdom have a median age at diagnosis of 68 years, with around 41% of new cases each year occurring in people aged 75 or older. (http://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/oesophageal-cancer/incidence). Esophagectomy is definitely a curative and powerful treatment for resectable thoracic esophageal cancer (TEC). However, it remains a highly invasive treatment modality that brings a high mortality rate and potentially fatal postoperative complications, despite recent advances in the surgical techniques and perioperative management.^7–8 In addition, a clear advantage is provided by neoadjuvant and adjuvant treatments using chemotherapy and che-moradiotherapy (CRT).^9–11 As a result, elderly patients do not always receive the full array of cancer treatments. One definitive treatment, alternative to esophagectomy, is CRT consisting of cisplatin based chemotherapy and more than 50 Gy radiotherapy.^12–13 Its therapeutic effect of definitive CRT has already been demonstrated in Stage I TEC. However, definitive CRT is not sufficient for Stage II-III TEC, and it is associated with severe adverse events and treatment-related death.^12,14 Therefore, a key question is whether esophagectomy provide a long-term survival advantage in elderly patients with resectable Stage II-III TEC. To make effective use of limited medical resources in the future, we should now begin by analyzing in detail currently available real-world big data on treatment strategies for elderly TEC patients. To address this important issue, in the present study we analyzed data from 5066 elderly patients aged 75 years or older, who were treated with esophagec-tomy, chemotherapy and/or radiation for TEC using the National Database of Hospital-based Cancer Registries, a large nationwide Japanese database on cancer.

PATIENTS AND METHODS

This study was approved by the Ethics Committee of Akita University Graduate School of Medicine (No. 2113).

We retrieved the 2008 to 2011 data from the National Database of Hospital-based Cancer Registries from the National Cancer Center, Tokyo, Japan.¹⁵ The registry data included the following information on individual cancer patients: (i) clinical profiles, including birth date, sex, tumor topology, and histology code defined by the International Classification of Disease for Oncology, third edition (ICD-O–3); (ii) clinical and pathological tumor-node-metastasis (6th Edition) stage, based on the Union for International Cancer Control guidelines; (iii) diagnosis year and month; (iv) first-line treatment provided at the registering facility; and (v) survival information.

We extracted the data for patients aged 75 years or older who were diagnosed with thoracic esophageal cancer (ICD-O–3 topography code: C151, C153–155) and treated with some type of anticancer therapy: esophagectomy, chemotherapy and/or radiation. Both the clinical and pathological Union for International Cancer Control tumor-node-metastasis classifications were registered in the National Database of Hospital-based Cancer Registries; however, we adopted a pretreatment clinical stage (cStage), tumor depth (cT), node status (cN), and metastasis status (cM) in this study to compare survival curves between groups with and without surgical resection.³ The clinical staging was defined by physicians affiliated with each institution enrolled, based mainly on findings from endoscopy, esophagography, and contrast-enhanced computed tomography. Systematic [¹⁸F] fluo-rodeoxyglucose-positron emission tomography/computed tomography (FDG-PET/CT), ultrasonography of the neck and abdomen, and endo-scopic ultrasonography were also performed when judged to be indicated. Survival time was defined as the follow–up time after diagnosis of TEC. To ensure the quality of the survival data, the survival data analyzed were limited to those from facilities able to provide 3-year survival data for all cancers for more than 90% of their patients.

Statistical Analysis

We divided the patients into 2 groups, a >75 group (75–79 years) and a >80 group (80 years or older), and compared the patient backgrounds and survival curves. Further analyses were performed after dividing the patients based on first-line treatment: esophagec-tomy with and without preoperative and/or postoperative adjuvant therapy, CRT, chemotherapy alone, and radiotherapy alone. Statistical comparisons between 2 groups were made using the chi-squared test or Cuzick Wilcoxon-type test for trend, depending on the type and distribution of the variables. Overall survival was characterized using Kaplan-Meier curves. Survival curves were compared between the 2 groups using the log-rank test. A multivariable Cox proportional hazards regression model was developed to evaluate the effect of esophagectomy on survival compared to that of CRT in the two groups after adjusting the analysis stratified based on cStage and sex in cStage II-III patients (N = 2112). Two-sided P-values of <0.05 were used to define statistical significance. We performed all statistical operations using STATA 14-MP (Stata Corp LP, College Station, TX).

RESULTS

The Hospital-based Cancer Registry for 2008 to 2011 listed 2935 patients who received cancer treatment, excluding endoscopic treatment, for TEC in the >75 group and 2131 patients in the >80 group. Table 1 shows the characteristics of the patients in the 2 groups. There were slight but statistically significant differences in sex and cN, but no statistically significant difference in cT, cM or cStage. On the other hand, 873 patients in the ≥75 group and 791 patients in the ≥80 group did not receive cancer treatment or received endoscopic treatment for TEC (Supplement Table, https://links.lww.com/SLA/C750). In these patients, more than 70% in the ≥75 group and 45% in the ≥80 group were classified as cStage 0-I.

More than 90% of the patients in the 2 groups were diagnosed with squamous cell carcinoma, whereas 3% were diagnosed with adenocarcinoma. Overall, treatment was predominantly nonsurgi-cal in both groups (Table 1). Nonetheless, a larger proportion of patients were treated with esophagectomy in the ≥75 group (34.6%) than the ≥80 group (18.4%) (P < 0.001). Among cStage I patients, 41.8% in the ≥75 group and 19.5% in the ≥80 group (P < 0.001) were treated with esophagectomy. The same trend was seen with cStage II patients, but not with cStage III-IV patients, among whom a smaller number (33.8% with cStage III and 11.1% with cStage IV) in the ≥75 group were treated with esophagectomy. Open thoracot-omy was the predominant approach in both groups, though thoraco-scopic esophagectomy was performed in 21.3% and 17.9% of patients in the ≥75 and ≥80 groups, respectively. Adjuvant therapies before and/or after surgery were performed in 39.9% of patients in the ≥75 group and 19.4% of patients in the ≥80 group (P < 0.001). Adjuvant therapy was less frequently performed in the >80 group. As a nonsurgical treatment, CRT was administered more frequently in the ≥75 group than the ≥80 group (64.7% vs 40.3%) (P < 0.001). Radiation alone was more common in the ≥80 group (53.4%).

Analysis of the treatment method at each stage revealed that for cStage I-II patients in the ≥75 group, esophagectomy was the most frequently used treatment, followed by CRT. For cStage III-IV in the ≥75 group, CRT was the predominant treatment (Table 1). On the other hand, in the ≥80 group, radiation alone was administered most frequently at all stages, and next was CRT.

Comparison of the Kaplan-Meier survival curves using logrank tests showed a statistically significant difference between the ≥ 75 and ≥ 80 groups at cStages I-IV (Table 2). Among the patients who received esophagectomy, the 3-year survival rate was 51.1 [95% confidence interval (CI): 48.0–54.1] % in the ≥75 group and 39.0 (95% CI: 34.2–43.9) % in the ≥80 group (P < 0.001). There was also a significant difference in the 3-year survival rate between those who received open and thoracoscopic esophagectomy in the 2 groups (47.8% vs 37.4% and 63.7% vs 46.7%). Among the patients who received esophagectomy, the 6-month survival rate was 80.0% (95% confidence interval [CI]: 88.0–91.7) % in the ≥75 group and 84.1 (95% CI: 80.1–87.4) % in the ≥80 group. The 6-month survival rate after thoracoscopic esophagectomy tended to be higher than after open thoracotomy in both groups (96.7% vs 88.2% in the >75 group and 90.0% vs 82.8% in the ≥80 group) (Table 2). However, among the patients who received CRT, radiation alone or chemotherapy alone, there was no statistically significant difference (P = 0.17, 0.93, and 0.83) in in 3-year survival.

We then compared survival curves between patients receiving esophagectomy or CRT at each cancer stage in the 2 groups (Fig. 1 ). For patients in the ≥75 group, the survival rates at cStages II-IV were significantly better among those receiving esophagec-tomy than CRT (P < 0.05) (Table 3). This was not the case for cStage I patients, however (P = 0.268) (Table 3). Among patients in the ≥80 group, the survival curves for those receiving esoph-agectomy and those receiving CRT did not significantly differ at any stage (Fig. 1 ). This means that for patients aged 80 years or older, survival rates among patients treated with CRT was not significantly lower than that among patients treated with esoph-agectomy (Table 3). Esophagectomy improved survival rates for patients with cStage II-IV TEC only in the ≥75 group, not in the 80°C group.

Multivariable Cox proportional hazard analysis of the ≥75 group stratified based on cStage and sex (cStage II-III, N = 1403) revealed that esophagectomy was significantly associated with better survival (adjusted HR: 0.731) (95%CI: 0.645–0.829, P < 0.001) when compared with CRT. On the other hand, in the ≥80 group (cStage II-III, N = 709), esophagectomy was not significantly associated with better survival (adjusted HR: 1.035) (95%CI: 0.875– 1.225, P = 0.685).

We further compared survival curves between patients receiving open or thoracoscopic esophagectomy in the ≥75 and ≥80 groups (Table 4). For cStage I, II, IV patients in the ≥75 group, there was no significant difference in 3-year survival between patients treated with open or thoracoscopic esophagectomy, and no difference at any disease stage in the ≥80 group. The minimally invasive thoracoscopic approach significantly increased the survival rate only among cStage III patients in the ≥75 group (57.5% vs 33.4%). Likewise, adjuvant therapy with surgery improved the 3-year survival rate only among cStage III patients in the ≥75 group. However, the proportion of patients receiving adjuvant therapy differed between the ≥75 and ≥80 groups.

DISCUSSION

This study yielded several noteworthy results. First, there was a significant difference in the percentage of patients who received esophagectomy between the ≥75 group (34.6%) and ≥80 group (18.4%). Moreover, among patients with early stage tumors (cStage I-II), more than twice as many in the ≥75 group received esoph-agectomy. Nonetheless, patients in both groups predominately received nonsurgical treatment. In the ≥80 group, radiation alone was the most frequent therapy in all stages, followed by CRT. In the ≥ 75 group, CRT was the most frequent therapy in more advanced stages (cStage III-IV). Second, the percentage of patients receiving standard full treatment (ie, surgery with adjuvant therapy or radiotherapy with chemotherapy [CRT]) was significantly smaller in the ≥80 group than the ≥75 group. This means that most patients in the ≥80 group received surgery alone or radiation alone. Third, and most importantly, for cStage II-III patients in the ≥75 group, esophagectomy significantly improved the survival rate (adjusted HR: 0.731) as compared to CRT, but that was not the case in the ≥80 group (adjusted HR:1.035). Fourth, minimally invasive thoracoscopic esophagectomy and adjuvant therapy improved the 3-year survival rate only for cStage III patients in the ≥75 group but not the ≥ 80 group.

The survival benefit of esophagectomy for patients aged 80 years or older is unclear. A large number of older cancer patients are frail or prefrail.¹⁶ This leads to treatment (especially esophagec-tomy)-associated negative outcomes, including postoperative morbidity and mortality along with long-term mortality.¹⁷ Therefore, frailty should be routinely assessed in elderly patients being considered for esophageal cancer treatment.^18,19 However, the assessment of actual frailty is not easy given the lack of universally-accepted standard tools for evaluation. Whether age is an independent risk factor for esophagectomy and whether elderly patients gain a survival benefit from surgical treatment remain unclear, and the definition of “elderly” has yet been determined. Consequently, there have been several comparisons between elderly and not-elderly patients -that is, 70, 75, or 80 years old or older versus younger. A large retrospective study assessing esophagectomy outcomes of patients ≥ 65 years (N = 27,957) reported that operative mortality significantly increased with age, from 8.8% in the 65 to 69 years group to 13.4% in the 70 to 79 years group and 9.9% in the >80 years group.²⁰ Those investigators concluded that esophagectomy carries significant risks for older patients. Clinical outcomes after esophagectomy were also compared in a meta-analysis covering 9 studies that included a total of 522 patients aged 80 years or older and 4,424 younger (70–79 years old) patients.²¹ It observed higher in-hospital mortality (odds ratio [OR] 2.00) and higher incidence of cardiac (OR 1.55) and pulmonary (OR 1.57) complications in patients aged 80 years or older than younger patients. In addition, patients aged 80 years or older had a poorer overall survival rate (HR 2.66) than the younger patients. In this analysis, however, patients aged 70 to 79 years were merged with even younger patients. Nonetheless, it is evident that younger patients experience fewer postoperative complications and better long-time survival.

On the other hand, there have been several reports that suggest the non-inferiority of outcomes after esophagectomy in elderly patients. Kanda et al compared postoperative outcomes between elderly (age >75 years, N = 50) and non-elderly (age <75 years, N = 100) patients after conducting a propensity score analysis.²² They observed no significant differences in the overall incidences of postoperative complications, rates of repeat surgery and death caused by surgery, and length of postoperative hospital stay between the 2 groups, although the incidences of postoperative pneumonia, arrhythmia and delirium were slightly higher in the elderly group.

They also reported that there was no significant difference in disease-specific or overall survival between the 2 groups. In addition, Miyata et al. divided 722 patients who underwent esophagectomy into 4 age groups (<70 [N = 509], 70–74 [N = 117], 75–79 [N = 73], and ≥80 years old [N = 23]) and analyzed short- and long-term outcomes after esophagectomy.²³ They reported that as compared to younger patients, advanced age tended to be associated with higher frequencies of postoperative pulmonary and cardiovascular complications, but not with a change in mortality rate. They also reported that patients aged 75 years or older, especially octogenarians, had poorer prognoses than younger patients. They suggested that the reason was that they less often received neoadjuvant therapy. A major problem with these studies is patient selection bias. The physical condition of patients receiving surgery is generally better than that of patients receiving nonsurgical treatment.

Does the minimally invasive approach have the potential to improve survival after esophagectomy for TEC in elderly patients? In a multicenter, randomized controlled trial, Biere et al found that minimally invasive esophagectomy (MIE) reduces morbidity as compared to open esophagectomy (59 MIE vs 56 open esophagec-tomies).²⁴ They concluded that the incidences of pulmonary infection in the first 2 weeks (9% vs 29%) and in-hospital (12% vs 34%) were significantly lower in the MIE group than the open esopha-gectomy group. Unfortunately, this study enrolled a relatively small number of patients, and the follow-up failed to show evidence of a survival benefit of MIE, as there were no differences in disease-free or overall 3-year survival between MIE and open esophagectomy.²⁵ In the present study, thoracoscopic esophagectomy improved the 3-year survival rate only for cStage III patients in the ≥75 group; we detected no survival advantage at the other cStages in the ≥75 group and at no cStage in the ≥80 group. However, a new advanced MIE, robot-assisted thoracoscopic esophagectomy (RATE), has been rapidly coming into use in recent years. In a single-center, randomized controlled trial, van der Sluis et al showed that use of RATE resulted in a lower incidence of overall surgery-related and cardiopulmonary complications with lower postoperative pain, better short-term quality of life, and a better short-term postoperative functional recovery than open transthoracic surgery.²⁶ In addition, we reported that the extent of mediastinal lymph node dissection was more thorough with RATE, which may have survival benefit.²⁷ We anticipate that for frail elderly patients, a new advanced MIE, RATE will improve survival in the near future, though that will need to be evaluated over the next 10 years or so.

Does adjuvant therapy before and/or after esophagectomy improve the survival rate among elderly patients? Unfortunately, our data do not enable us to answer that question, as elderly patient, especially octogenarians, usually did not receive adjuvant therapy due to their fragility. In addition, we do not know the specific treatment set used in each individual case (neoadjuvant and/or adjuvant). However, we anticipate that future studies will enable resolution of this difficult question.

Using the Surveillance Epidemiology and End Results (SEER)-Medicare Linked Database (2001 –2009) and identifying locally advanced esophageal cancer patients >65 years old (median age 75), Molena et al analyzed overall treatment-dependent sur-vival.⁴ The majority of these patients were treated with definitive CRT (48.49%). One third (34.74%) received no treatment of any kind, although esophagectomy was performed in 26.8% of patients. They concluded that elderly patients with locally advanced esoph-ageal cancer who received treatment had better 5-year survival than those without treatment. In our study, although we cannot know the percentage of patients who received no cancer treatment, the percentage will be smaller than in the United States. Surgical treatment was performed in 27.6% of 75-year-old patients, which is comparable to the data summarized above.

The strengths of our study include the following. The use of a large database enabled us to analyze patients with relatively rare characteristics while maintaining a higher degree of power. Second, this study revealed real world outcomes using the National Database of Hospital-based Cancer Registries. Another important merit of the study is the accuracy of the survival data achieved through the use of big data. On the other hand, our study design has several limitations. First, the cancer data were collected only from Designated Cancer Care Hospitals and only for first-course treatments provided by the registering facility. For example, patients receiving esophagectomy for a recurrent tumor within a year after endoscopic mucosal resection were omitted. Second, there was clearly patient selection bias between the esophagectomy and nonsurgery groups. Patients receiving esophagectomy may have been in better condition before treatment than those who did not undergo surgery and our data did not include the information on comorbidity and performance status. Third, cStage tends to be underestimated, and staging accuracy varies among institutions. This is an inherent weakness of all studies using cancer registration databases. Fourth, our study took only survival as the outcome and did not evaluate the quality of life of the patients, which is an important indicator of postoperative outcome, especially in elderly patients.

In conclusion, our real-world data indicate that for cStage II-III TEC patients aged 75 to 79 years, esophagectomy improves survival rates. However, many octogenarians do not necessarily get a survival benefit from esophagectomy. Patients should therefore be evaluated based on their overall health and by assessing comorbid-ities and surgical risk before ruling out surgery based on age alone. Advanced MIE may be a means to gain a survival benefit from this surgery for the elderly.

Supplemental Digital Content

• SLA_2020_10_19_MOTOYAMA_ANNSURG-D-20-01354_SDC1.xlsx; [Excel] (11 KB)