Higher clinical suspicion is needed for prompt diagnosis of esophageal adenocarcinoma in young patients

Nobel et al

Thoracic

Higher clinical suspicion is needed for prompt diagnosis of esophageal adenocarcinoma in young patients Tamar B. Nobel, MD,a,b Michael Curry, MS,c Renee Gennarelli, MS,c David R. Jones, MD,a and Daniela Molena, MDa ABSTRACT

60

Results: A total of 101,596 patients were identified with esophageal cancer. The youngest patient group (18-57 years) had the highest rate of metastatic disease (34%). No difference in tumor differentiation was observed between age groups. Younger patient groups were more likely to undergo treatment despite advanced stage at diagnosis. Overall 5-year survival was better for younger patients with local disease, but the difference was less pronounced in locoregional and metastatic cases. Conclusions: In this study, young patients were more likely to have metastatic disease at diagnosis. Advanced stage in young patients may reflect the need for more aggressive clinical evaluation in high-risk young patients. (J Thorac Cardiovasc Surg 2019;-:1-10)

58-65 years 75+ years

40 30 20 10 0

Local

Locoregional Stage at Diagnosis

Distant

Relationship between age quartile group and stage of disease at diagnosis. Central Message Perceptions of esophageal cancer as a disease of the elderly may result in the higher rate of advanced diagnostic stage observed in younger patients. Risk factors, rather than age, should guide evaluation.

THOR

Methods: A retrospective cohort study was performed using the National Cancer Database between 2004 and 2015. Patients with esophageal adenocarcinoma were divided into quartiles by age (18-57, 58-65, 66-74, 75þ years) for comparison. Clinicopathologic and treatment factors were compared between groups.

Percent of Cases

Background: Esophageal cancer is considered a disease of the elderly. Although the incidence of esophageal adenocarcinoma in young patients is increasing, current guidelines for endoscopic evaluation of gastroesophageal reflux disease and Barrett’s esophagus include age as a cutoff. There is a paucity of data on the presentation and treatment of esophageal cancer in young patients. Most studies are limited by small sample sizes, and conflicting findings are reported regarding delayed diagnosis and survival compared with older patients.

18-57 years 66-74 years

50

Perspective This population-based study showed similar clinicopathologic characteristics but more advanced disease among young patients diagnosed with esophageal adenocarcinoma. Given the poor prognosis associated with metastatic esophageal cancer, it may be time to rethink the approach for evaluation of young high-risk patients.

See Commentary on page XXX.

Esophageal cancer is the eighth most common cancer worldwide and the sixth most common cause of cancer-related death.1 Although considered a highly fatal From the aDepartment of Surgery, Memorial Sloan Kettering Cancer Center, New York, NY; bDepartment of Environmental Medicine and Public Health, Mount Sinai Hospital, New York, NY; and cDepartment of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY. This study was supported, in part, by the National Institutes of Health /National Cancer Institute Cancer Support Grant P30 CA008748. T.N. is supported, in part, by a grant from the American Cancer Society. Exempt from Institutional Review Board approval. Read at The American Association for Thoracic Surgery Thoracic Summit, New York, New York, October 12-13, 2018. Received for publication Oct 31, 2018; revisions received March 13, 2019; accepted for publication March 31, 2019. Address for reprints: Daniela Molena, MD, Department of Surgery, 1275 York Ave, New York, NY 10065 (E-mail: [email protected]). 0022-5223/$36.00 Copyright Ó 2019 by The American Association for Thoracic Surgery https://doi.org/10.1016/j.jtcvs.2019.03.095

disease, therapeutic advancements have resulted in improved 5-year overall survival from 5.1% to 19.2% over the last 4 decades.2,3 Esophageal cancer occurs primarily in the sixth and seventh decades of life; as such, it is considered a disease of the elderly. However, esophageal cancer does occur in young patients, and patients younger than 55 years currently account for approximately 13% of new diagnoses in the United States.3 Citing low risk of disease in young patients, current guidelines from the American College of Gastroenterology Scanning this QR code will take you to the article title page to access supplementary information.

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

1

Thoracic

Nobel et al

Abbreviations and Acronyms AJCC ¼ American Joint Committee on Cancer DFS ¼ disease-free survival DSS ¼ disease-specific survival EAC ¼ esophageal adenocarcinoma NCDB ¼ National Cancer Database NOS ¼ not otherwise specified OS ¼ overall survival

THOR

and American Gastroenterologic Association recommend against endoscopy for evaluation of dyspepsia in patients aged less than 60 years, even in the setting of alarm symptoms4 (Table 1). Furthermore, guidelines specify that age more than 50 years is considered a high-risk indication for endoscopic evaluation of gastroesophageal reflux symptoms or to screen for Barrett’s esophagus.5-9 Likewise, national guidelines in other western countries recommend against urgent endoscopic evaluation of dyspepsia in patients aged less than 55 years.10,11 This is concerning because studies suggest that the incidence of esophageal cancer among young patients is increasing.12,13 Furthermore, perceptions of esophageal cancer as a rare disease in young patients may result in lower clinical suspicion; as such, some studies have suggested that patients aged less than 50 years are more likely to have delayed diagnosis and consequently more advanced disease.14-18 Given that 5-year overall survival for advanced disease is 25% and less than 5% for regional and distant disease, respectively, earlier identification has a significant impact on prognosis.3 To date, esophageal cancer in young patients has received minimal attention in the literature. Most studies are single-institution experiences limited by small sample sizes

and conflicting findings relating to the presentation, treatment, and prognosis of this population; as such, the clinicopathologic presentation and outcomes of esophageal cancer in young patients are not well defined.14-25 Further complicating current understanding of this disease in young patients is the lack of consensus as to what constitutes a ‘‘young patient,’’ with many studies arbitrarily selecting age less than 50 years.14-21 Given that survival for this disease remains poor, esophageal cancer in young patients merits further investigation to better guide diagnosis and management in this population. The purpose of this study was to compare the clinicopathologic characteristics and survival of patients with esophageal cancer between age groups using a national cancer cohort in the United States. MATERIALS AND METHODS Data Source and Study Population A population-based retrospective review was performed using data from the National Cancer Database (NCDB) Participant User File. All patients diagnosed between 2004 to 2015 with adenocarcinoma of the esophagus (International Classification of Diseases-0 Code C15.0-15.9) and esophagogastric junction (C16.0, cardia not otherwise specified [NOS]) with a single primary malignancy were included. The NCDB is a cancer registry from the American College of Surgeons and American Cancer Society which collects data from more than 1500 Commission on Cancer accredited centers.21 Approximately 70% of all newly diagnosed cancer cases are captured annually. This study was exempt from Institutional Review Board review.

Variables Demographic characteristics of interest included age, gender, race, comorbidities (Charlson Comorbidity Index >0), insurance status (uninsured, private, or government), cancer center type (academic vs nonacademic), education level (percent of population within patient ZIP code without high school education 21% or <21%), year of diagnosis, region (grouped into 4 geographic areas, including Northeast, Midwest, South, and West), distance traveled to facility (miles), location (population 250,000 vs <250,000), and year of diagnosis (2004-2007, 2008-2011, 2012-2015). Histopathologic data included tumor location (lower third/ cardia NOS/abdominal, Middle Third/Thoracic, Upper Third/Cervical, Overlapping, Esophagus NOS), histology (adenocarcinoma, squamous

TABLE 1. Current guidelines for endoscopic evaluation of gastroesophageal reflux disease and Barrett’s esophagus Indications for endoscopic evaluation

High-risk features

American Society of Gastrointestinal Endoscopy (2015)5

Screening EGD for select patients with prolonged history of GERD

Age 50 y, chronic GERD symptoms (>5 y), white, male, nocturnal reflux

AGA (2008, 2011)6,7

Consider EGD to screen for BE in patients with high risk

Age 50 y, male, white, hiatal hernia, elevated BMI, intra-abdominal fat distribution

ACG (2016)8

Consider screening for BE in high risk patients

Chronic GERD symptoms (>5 y) and 2þ BE/cancer risk factors: age 50 y, white, central obesity, current/past history of smoking, family history

ACG (2013)9

Individualized EGD for GERD diagnosis in select patients for alarm symptoms and high risk

Elderly, those at risk for BE, noncardiac chest pain, patients unresponsive to PPI

AGA and ACG (2017)4

EGD should be performed in patients age >60 y with dyspepsia to rule out neoplasia

NA

EGD, Esophagogastroduodenoscopy; GERD, gastroesophageal reflux disease; AGA, American Gastroenterological Association; BE, Barrett’s esophagus; BMI, body mass index; ACG, American College of Gastroenterology; PPI, proton pump inhibitor; NA, not applicable.

2

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al

Thoracic

cell carcinoma, or carcinoma NOS), and differentiation grade (well/ moderate, poor/undifferentiated, or unknown). Treatment was defined as having received any recommended therapy for esophageal cancer, including chemotherapy, radiation, surgery, or any combination of these interventions. As previously described,26 treatments were categorized into surgery, definitive chemoradiation, induction þ surgery (surgery and neoadjuvant chemotherapy  radiation or adjuvant chemotherapy  radiation), palliative (chemotherapy or radiation alone), and no treatment (Table E1). For patients who underwent surgery, postoperative 30-day mortality and readmission rate were recorded. The American Joint Committee on Cancer (AJCC) staging criteria are recorded in concordance with diagnosis year. To account for different categorization of staging between the AJCC Sixth and Seventh editions, stage was divided into 3 categories using clinical data, including local, locoregional, and metastatic, as previously described by Rice and colleagues27 and Wong and colleagues28 (Table E2)

Statistical Analysis Patients were divided into 4 groups based on age quartiles. Baseline patient demographics and clinical factors were compared between age groups using chi-square for categoric variables and Wilcoxon ranksum test for continuous variables. Data were reported as counts (percentages) and medians (interquartile range), respectively. Overall survival was defined as time (in months) from diagnosis to death or last follow-up. Survival analysis was performed using the Kaplan– Meier method. A multivariate model using Cox regression analysis was included to determine whether age independently impacted survival after adjusting for treatment and stage. A subanalysis was performed among patients aged less than 50 years, 50 years and more, less than 60 years, and 60 years and more to reflect the guidelines discussed in Table 1. All data analyses were performed using SAS v9.4 (SAS Institute, Inc, Cary, NC) and R v3.3.3.

TABLE 2. Comparison of demographic characteristics by age group

N

18-57 y

Age quartile groups 58-65 y 66-74 y

75þ y

26,440

25,162

23,859

26,135

P overall

3567 (13.5%)

3289 (13.1%)

4101 (15.7%)

5903 (24.7%)

<.001

23,197 (87.7%)

22,855 (90.8%)

23,989 (91.8%)

21,850 (91.6%)

<.001

5215 (19.7%)

6921 (27.5%)

8403 (32.2%)

7672 (32.2%)

<.001

13.4 [1.50-143]

13.4 [1.50-148]

12.8 [1.40-146]

8.80 [1.10-104]

<.001

4117 (15.6%)

3507 (13.9%)

3513 (13.4%)

2769 (11.6%)

<.001

Insurance status Government Private None Insurance status unknown

6145 (23.2%) 17,630 (66.7%) 1774 (6.71%) 891 (3.37%)

8202 (32.6%) 15,183 (60.3%) 1027 (4.08%) 750 (2.98%)

21,526 (82.4%) 3806 (14.6%) 170 (0.65%) 633 (2.42%)

21,021 (88.1%) 2246 (9.41%) 117 (0.49%) 475 (1.99%)

Facility type Academic Nonacademic

10,964 (44.3%) 13,784 (55.7%)

10,926 (43.4%) 14,236 (56.6%)

10,656 (40.8%) 15,479 (59.2%)

8173 (34.3%) 15,686 (65.7%)

5428 (21.9%) 7211 (29.1%) 8551 (34.6%) 3558 (14.4%)

5703 (22.7%) 7233 (28.7%) 8406 (33.4%) 3820 (15.2%)

6021 (23.0%) 7306 (28.0%) 8639 (33.1%) 4169 (16.0%)

5999 (25.1%) 7000 (29.3%) 6967 (29.2%) 3893 (16.3%)

20,717 (78.4%)

19,603 (77.9%)

20,252 (77.5%)

19,085 (80.0%)

8194 (31.0%) 8867 (33.5%) 9379 (35.5%)

6844 (27.2%) 8377 (33.3%) 9941 (39.5%)

7124 (27.3%) 8404 (32.2%) 10,607 (40.6%)

7411 (31.1%) 7836 (32.8%) 8612 (36.1%)

Tumor site Abdominal/lower/stomach Overlapping Thoracic/middle Unspecified Upper/cervical

22,715 (85.9%) 740 (2.80%) 962 (3.64%) 1837 (6.95%) 186 (0.70%)

21,614 (85.9%) 751 (2.98%) 966 (3.84%) 1663 (6.61%) 168 (0.67%)

22,647 (86.7%) 667 (2.55%) 1011 (3.87%) 1654 (6.33%) 156 (0.60%)

20,089 (84.2%) 611 (2.56%) 1219 (5.11%) 1727 (7.24%) 213 (0.89%)

Differentiation grade Unknown Well/moderate Poor/undifferentiated

4801 (18.2%) 9930 (37.6%) 11,709 (44.3%)

4585 (18.2%) 9639 (38.3%) 10,938 (43.5%)

4925 (18.8%) 10,112 (38.7%) 11,098 (42.5%)

4883 (20.5%) 9140 (38.3%) 9836 (41.2%)

Female gender White race Any comorbidities Median distance to hospital, miles (IQR) Education (21% in ZIP code without high school diploma)

<.001

<.001

<.001

Location East South Midwest West Patient location (combine into 250,000) Diagnosis year 2004-2007 2008-2011 2012-2015

THOR

Characteristic, N (%)

<.001 <.001

<.001

<.001

IQR, Interquartile range.

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

3

Thoracic

Nobel et al

60

18-57 years 66-74 years

Percent of Cases

50

58-65 years 75+ years

40 30 20 10 0

Local

Locoregional Stage at Diagnosis

of local disease (37.3%), with incidence increasing with increasing age group (41.4%, 46.5%, and 55.6%, respectively). Rates of locoregional disease were similar among the youngest 3 age groups and lowest in the oldest patients. Distant disease at diagnosis decreased with increasing age; 33.9% of patients aged 18 to 57 years had metastasis, followed by 29.8%, 26.8%, and 23.1% in the respective increasing age groups. The rate of advanced disease at diagnosis, including locoregional and distant, was highest in the youngest age group (62.5%) and lowest in the patients aged 75þ years (44.4%).

Distant

FIGURE 1. Distribution of stage at diagnosis by age quartile at diagnosis.

THOR

RESULTS There were 101,596 patients included in the study cohort. The median age was 65 (57-74) years. Patients were grouped by quartiles of age distribution, which included 18 to 57 years (n ¼ 26,440) 58 to 65 years (n ¼ 25,162), 66 to 74 years (n ¼ 26,135), and 75þ years (n ¼ 23,589). The median and IQR by age group were 52 years (IQR, 47-55), 62 years (IQR, 60-64), 70 years (IQR, 67-72), and 80 years (IQR, 77-84). In this population, 31.7% of patients were aged less than 60 years. Clinicopathologic Characteristics Comparison of baseline demographics between age groups is shown in Table 2. The youngest age quartile had the highest rate of nonwhite patients compared with other age groups. Gender was similar among the younger 3 age quartiles (13.5%, 13.1%, and 15.7%, respectively); however, there were 24.7% women in the 75 years and older group. Patients in the youngest quartile were more likely to be uninsured and live in less-educated areas than other age groups. With increasing age, distance from the hospital and rate of treatment at an academic center decreased. Patients had similar tumor location and grade. Staging Distribution of tumor stage by age group is presented in Figure 1. Patients aged 18 to 57 years had the lowest rate

Treatment Disease treatment by age quartile is demonstrated in Table 3. Nontreatment was twice as likely in patients age 75þ years (32%) compared with patients in other age groups (14%, 15%, and 17%, respectively, by increasing age quartile). The rate of patients receiving definitive chemoradiation and surgery alone was similar among all groups. Patients in the youngest 2 age quartiles underwent induction þ surgery more frequently (22.3% and 21.7%, respectively) than patients in the third and fourth quartiles (17.3% and 5.2%, respectively). Among patients who underwent surgical treatment, patients aged 75þ years received local endoscopic therapy most frequently and were least likely to receive surgery. On operative pathology, nodal metastasis rates were similar among age groups; however, the incidence of advanced T stage decreased with increasing age group. Postoperative readmission rates were similar among all groups, but 30day overall mortality rates increased with age (0.7%, 0.9%, 1.4%, and 1.3%, respectively). Survival Comparison of overall survival by age quartile is presented in Figure 2. Median OS was similar in patients aged 18 to 74 years (16.5-18.3 months) and lower in patients aged 75þ years (9.6). Stratification of overall survival by age quartile and tumor stage is shown in Figure 3. In patients with local disease, median OS in patients aged 18 to 57 years and 58 to 65 years was similar (38.4 and 38.7 months); however, for age 66 to 74 years and 75þ years, OS decreased (31.2 and 13.0 months,

TABLE 3. Comparison of treatment selection between age groups Age quartile groups Treatment type

18-57 y

58-65 y

66-74 y

75þ y

Definitive chemoradiation

6211 (23.5%)

6062 (24.1%)

6641 (25.4%)

6038 (25.3%)

Induction þ surgery

5907 (22.3%)

5466 (21.7%)

4521 (17.3%)

1249 (5.23%)

Surgery

5056 (19.1%)

5108 (20.3%)

5984 (22.9%)

4459 (18.7%)

Palliative care

5698 (21.6%)

4850 (19.3%)

4649 (17.8%)

4592 (19.2%)

No treatment

3568 (13.5%)

3676 (14.6%)

4340 (16.6%)

7521 (31.5%)

4

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al

Thoracic

Overall Survival (%)

100 Age Group 18-57 58-65 66-74 75+

75

50

25

0 0

1

2 3 Years Since Diagnosis

4

5

18-57

24069

14433

8990

6120

4480

3390

58-65

22596

13418

8580

5849

4211

3103

66-74

23255

13109

8235

5492

3957

2866

75+

21666

9093

5132

3205

2163

1463

4

5

0

1

2 3 Years Since Diagnosis

THOR

Age Group

Number at risk

FIGURE 2. Comparison of overall survival by age quartile.

respectively). The differences between age groups were less apparent for patients with locoregional disease (23.5, 22.6, 18.4, and 11.9, respectively, by increasing age quartile). Among patients with metastatic disease, median OS ranged

Locoregional

Local

50 25

100 Overall Survival (%)

75

0

75 50 25 0

1 2 3 4 5 Years Since Diagnosis

9548 6969 5135 3815 2888 2183 11008 7573 5436 3899 2937 2199 12198 6038 3750 2475 1722 1197

1 2 3 4 5 Years Since Diagnosis

Age Group

9186 6777 4979 3685 2855 2240

18-57 58-65 66-74 75+

Age Group 18-57 58-65 66-74 75+

50 25

1 2 3 4 5 Years Since Diagnosis

0

Number at risk

Number at risk

0

75

0 0

985

6367 4408 2745 1723 1147

797

6079 3806 2229 1354

893

592

4520 2152 1064

373

229

0

1 2 3 4 5 Years Since Diagnosis

Number at risk

6805 4859 2986 1983 1373

605

1 2 3 4 5 Years Since Diagnosis

Age Group

0

Age Group

Metastatic

100 Overall Survival (%)

Overall Survival (%)

100

18-57 58-65 66-74 75+

from 4 to 8 months. The 5-year overall survival by age and stage is presented in Table E3. In Figure 4, median OS by age quartile was stratified by treatment selection. Among patients who underwent

18-57 58-65 66-74 75+

8078 2797 1025

452

252

165

6681 2041

700

311

176

123

6168 1730

570

239

127

75

4948 903

318

125

68

37

0

1 2 3 4 5 Years Since Diagnosis

FIGURE 3. Comparison of overall survival by age quartile, stratified by tumor stage at diagnosis.

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

5

Nobel et al

Definitive Chemoradiation

Overall Survival (%)

Overall Survival (%)

Thoracic

100 75 50 25 0 0

1

2

3

4

Induction + Surgery 100 75 50 25 0 0

5

1

Years Since Diagnosis

2950 2864 3103 2746

0

1

1447 1463 1533 1367

851 857 887 765

541 547 568 465

372 378 377 273

2

3

4

5

Strata

Strata

5611 5438 5842 5423

18-57 58-65 66-74 75+

5338 4803 3907 1076

4499 3980 3073 764

0

1

Overall Survival (%)

Overall Survival (%)

Surgery 100 75 50 25 0 1

2

3

2167 1847 1411 284

1565 1294 989 174

1187 899 694 109

2

3

4

5

4

4

5

75 50 25 0 0

5

1

2

3

Years Since Diagnosis Number at risk

3800 3839 4242 2850

0

1

2951 3068 3296 2054

2306 2422 2471 1494

1848 1885 1911 1115

1460 1472 1463 821

2

3

4

5

Strata

4701 4715 5458 4125

18-57 58-65 66-74 75+

5123 4322 4125 4151

2020 1617 1451 1264

0

1

750 628 550 474

346 296 232 217

196 181 132 123

123 125 75 71

2

3

4

5

Years Since Diagnosis

Years Since Diagnosis No Treatment 100 75

Age Group

50

18-57 58-65 66-74 75+

25 0 0

1

2

3

4

5

Years Since Diagnosis Number at risk Strata

THOR

Strata

3165 2769 2127 475

Palliative Care

Number at risk

Overall Survival (%)

5

100

Years Since Diagnosis

18-57 58-65 66-74 75+

4

Years Since Diagnosis

Years Since Diagnosis

0

3

Number at risk

Number at risk 18-57 58-65 66-74 75+

2

Years Since Diagnosis

18-57 58-65 66-74 75+

3296 3318 3923 6891

1164 1118 1240 1469

0

1

677 652 729 762

450 427 491 445

330 304 357 286

248 229 257 189

2

3

4

5

Years Since Diagnosis

FIGURE 4. Comparison of overall survival by age quartile, stratified by treatment selection.

definitive chemoradiation, no difference in median OS was observed by age group (12.8-13.6 months). Likewise, minimal difference was observed among patients who underwent palliative treatment or received no treatment. However, among patients who underwent induction þ surgery, median OS was lower in patients in the oldest quartile (24.3 months) as compared with younger groups (37.7, 37.3, and 33.4, respectively, from quartile 1-3). Likewise, among patients who underwent surgical resection alone, patients age 75þ years had lower median OS (30.1 months) compared with other groups (75.4, 81.9, and 55.3 months, respectively, from quartiles 1-3). Table 4 presents the results of multivariate Cox proportional hazards assessing survival. After adjusting 6

for stage and treatment, mortality risk increased with age quartile compared with the youngest patients. Dichotomized Analysis To account for the age cutoffs described in the guidelines in Table 1, separate analyses were conducted by dividing the population using the age cutoff of 50 and 60 years (Tables E4 and E5). Stratification by age 50 years demonstrated several differences from the quartile analysis. The racial difference between older and younger patients was greater, with the group aged less than 50 years having a 68% higher rate of nonwhite patients aged more than 50 years. Patients aged less than 50 years had a 31% higher rate of distant disease at diagnosis (36.6% vs 27.8%), and

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al

Thoracic

TABLE 4. Adjusted model for overall survival 95% CI

Age Cutoff: 50 Years 50

P value

Age quartile 18-57 58-65 66-74 75 +

Ref 1.05 1.15 1.62

Treatment Definitive chemoradiation Induction + surgery No treatment Palliative care Surgery

Ref 0.57 1.79 1.18 0.48

0.55, 0.58 1.75, 1.83 1.15, 1.20 0.47, 0.50

<.001 <.001 <.001 <.001

Stage Local Locoregional Metastatic

Ref 1.27 2.53

1.24, 1.30 2.48, 2.58

<.001 <.001

40 1.03, 1.07 1.13, 1.18 1.58, 1.65

<.001 <.001 <.001

Percent of Cases

HR

30 20 10 0

Local

Locoregional Distant Stage at Diagnosis <50 years 50+ years

A

HR, Hazard ratio; CI, confidence interval.

Age Cutoff: 60 Years 50

30

THOR

DISCUSSION The presentation and outcomes of esophageal cancer in young patients are poorly understood. Our study is the largest to date to compare characteristics of esophageal cancer by age. The majority of available studies have been small and demonstrated conflicting results.9-12,14-20 Furthermore, comparability is limited by varying definitions of ‘‘young.’’ As such, we used quartiles of age distribution (18-57, 58-65, 66-74, 75þ years) in this study to determine how young patients may differ compared with other age groups, but performed subanalyses using age 50 and 60 years as a cutoff to reflect current guideline cutoffs (Table 1). Given that the incidence of esophageal cancer, a highly morbid disease, is increasing among younger patients, there is an important need for consensus regarding the impact of age on the presentation and outcomes of esophageal cancer.7,8 Our findings support previous studies that demonstrated advanced disease presentation in younger patients.14,15,17-19 Furthermore, results of our analysis demonstrated that the incidence of metastatic disease at diagnosis decreased with increasing age quartile. Use of age 50 years and 60 years as cutoffs demonstrated that the rate of distant disease was approximately 30% higher for younger patients in both analyses. In the only other population-based study, Zeng and colleagues18 recently used Surveillance, Epidemiology, and End Results program data to evaluate 1385 patients aged 50 years or less and demonstrated that 48.5% presented with distant disease compared with 37.9% of older patients.

40 Percent of Cases

those aged less than 60 years had a 28% higher rate (33.6% vs 26.2%) (Figure 5, A and B). There was less variation in overall survival by stage and treatment between age groups compared with the quartile analysis (Figures E1 and E2).

20 10 0

Local

Locoregional Distant Stage at Diagnosis <60 years 60+ years

B FIGURE 5. Distribution of stage at diagnosis by age cutoff (A) 50 years and (B) 60 years.

The higher rate of patients with metastatic disease may reflect selection bias in the authors’ methodology, including the decision to not analyze patients with esophagogastric junction cancer and use of AJCC Sixth edition for staging. However, their results confirm our conclusion that younger age is associated with more advanced disease and that this represents an important area for intervention. This study demonstrated similar tumor grade between groups, suggesting that observed differences in stage are not a reflection of a more aggressive disease process but rather delayed diagnosis. In support of this observation, several studies noted a longer period of reported symptoms and slower referral to endoscopy among young patients.14,15 The high rate of adenocarcinoma in the youngest age quartile is consistent with the demonstrated increasing incidence of adenocarcinoma in the United States.12,13 Currently, gastroesophageal reflux disease is the only well-accepted risk factor for development of

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

7

Thoracic

Nobel et al

Esophageal Cancer – Not Only a Disease of the Elderly

Stage at Diagnosis

l na

is ta n D

18-57 years 58-65 years 66-74 years 75+ years

Lo

co r

eg

io

Lo

t

60 50 40 30 20 10 0 ca l

Percent of Cases

National Cancer Database analysis of 101,596 patients with esophageal adenocarcinoma

Current guidelines recommend against routine endoscopy for dyspepsia in patients < 60 years old

Similar clinicopathologic characteristics between age groups

Youngest patients have highest rate of distant disease at diagnosis

Risk factors, not age, should dictate evaluation for esophageal cancer FIGURE 6. Esophageal cancer diagnosed at median age of 68 years and is considered a disease of the elderly. In this large series of patients with esophageal cancer from the NCDB (2004-2015), although clinicopathologic similarities suggest similar tumor biology between age groups, the youngest age quartile had the highest rate of distant disease at diagnosis (34% vs 29.9%, 26.8%, and 23.1%, respectively, with increasing age group). Risk factors, such as obesity and prolonged history of reflux, should dictate evaluation of esophageal cancer rather than age.

THOR

esophageal adenocarcinoma. Bai and colleagues29 reported that up to 40% of people in the United States report symptoms of gastroesophageal reflux. It has previously been described that the time for progression from reflux to Barrett’s esophagus is approximately 10 years.30,31 As such, the decision to perform endoscopic evaluation in a symptomatic patient should focus on length of time he/she has had gastroesophageal reflux disease rather than age. People with known gastroesophageal reflux disease as children or adolescents should be considered at risk. As the rates of premature birth, childhood obesity, and high consumption of high fat foods in the United States continue to increase, it is possible that increasing rates of esophageal adenocarcinoma may be observed in the young.32-34 Recent data from the Barrett’s and Esophageal Adenocarcinoma Consortium assessing age-specific risk profiles demonstrated that the associations between reflux disease and obesity were stronger among those who developed cancer at a younger age.35 Our findings support the warning by Hamouda and colleagues23 against over-reliance on age cutoffs and alarm symptoms based on previous studies that have questioned the efficacy of these factors as underlying predictors of cancer.36 Instead, clinicians should focus on the patient as a whole. Age quartile analysis demonstrated minimal differences in the demographic ‘‘profile’’ of patients aged less than 75 years. However, evaluation using age 50 years as a cutoff demonstrated that patients aged less than 50 years had a significantly higher rate of nonwhite race and were more likely to be uninsured compared with older patients. This is important given that the guidelines in Table 1 mention 8

white race as a risk factor. Previous studies have demonstrated racial disparities in stage at diagnosis, treatment selection, and prognosis among nonwhite patients, which may in part reflect differences in socioeconomic status.26,37-39 Further research may investigate whether there are underlying genetic factors that may contribute to earlier development of esophageal cancer in certain patients. However, the absence of distinguishing findings support the conclusion that physicians must maintain a high clinical suspicion in patients with prolonged histories of reflux and recognize that not all patients present in the ‘‘typical’’ fashion (elderly white men). Previous studies have suggested that more aggressive treatment selection among young patients results in similar overall survival despite more advanced stage at diagnosis.14,15,17,23 Although some studies demonstrate improved 5-year overall survival among younger patients, others fail to replicate these findings.14-19,23 In this series, patients aged less than 75 years had a similar rate of receiving any surgical therapy, although patients aged less than 65 years were more likely to receive aggressive treatment with induction therapy þ surgery than older patients. However, stratification by stage demonstrated that the magnitude of survival benefits among younger patients decreased with increasing stage. This highlights the importance of early diagnosis for esophageal cancer, especially in younger patients who have less comorbidities and can better tolerate aggressive curative therapy if diagnosed at an earlier stage. Once distant disease is diagnosed, overall prognosis is poor regardless of age.

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Thoracic

Study Limitations Interpretation of our results is subject to the limitations associated with the use of NCDB data. It is not possible to determine whether young patients in this series had a delay in diagnosis because of lower clinical suspicion as it relates to patient age. However, this hypothesis is supported by current understanding of perceptions of esophageal cancer amongst providers, endoscopic screening guidelines, smaller studies that identified delayed presentation in younger patients, and the absence of markers of more aggressive tumor pathology as an underlying etiology for advanced stage at diagnosis. Additionally, the NCDB does not provide information regarding postoperative complications or toxicity because of chemotherapy or radiation. As such, it is difficult to determine why a survival difference was observed between age groups among patients undergoing surgical treatments. Furthermore, the specific chemotherapy agents given are not included, and therefore it is not possible to determine whether patients who underwent treatment at nonacademic centers received the most updated standard of treatment. Finally, it is not possible to assess the impact of birth cohort on observed differences between age groups with this dataset. However, the increasing incidence of esophageal adenocarcinoma has still been demonstrated in more recent birth cohorts.13 CONCLUSIONS The results of this study demonstrate that in a large cohort compromising the majority of cases of esophageal cancer in the United States between 2004 and 2015, young patients with esophageal adenocarcinoma had more advanced disease at diagnosis despite similar clinicopathologic tumor characteristics. Given that stage at diagnosis is the most important determinant of overall survival for this highly lethal disease, it is important that clinicians consider esophageal cancer in the differential diagnosis of symptomatic patients with high-risk histories, such as obesity and prolonged reflux. Patient symptoms and risk factors, rather than age alone, should be considered when deciding whether to perform endoscopic evaluation. Esophageal cancer is not only a disease of older patients (Figure 6). Conflict of Interest Statement Authors have nothing to disclose with regard to commercial support. The authors thank Todd Litinsky for inspiring this work and supporting Daniela Molena’s research fund.

References 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68:394-424.

2. Horner MJ, Ries LAG, Krapcho M, Neyman N, Aminou R, Howlader N, et al, eds. SEER Cancer Statistics Review, 1975-2006, National Cancer Institute. Bethesda, MD. Available at: https://seer.cancer.gov/csr/1975_2006/, based on November 2008 SEER data submission, posted to the SEER web site, 2009. Accessed October 15, 2018. 3. Noone AM, Howlader N, Krapcho M, Miller D, Brest A, Yu M, et al, eds. SEER Cancer Statistics Review, 1975-2015, National Cancer Institute. Bethesda, MD. Available at: https://seer.cancer.gov/csr/1975_2015/, based on November 2017 SEER data submission, posted to the SEER web site, April 2018. Accessed October 15, 2018. 4. Moayyedi PM, Lacy BE, Andrews CN, Enns RA, Howden CW, Vakil N. ACG and CAG clinical guideline: management of Dyspepsia. Am J Gastroenterol. 2017;112:988-1013. 5. Muthusamy VR, Lightdale JR, Acosta RD, Chandrasekhara V, Chathadi KV, Eloubeidi MA, et al. The role of endoscopy in the management of GERD. Gastrointest Endosc. 2015;81:1305-10. 6. Kahrilas PJ, Shaheen NJ, Vaezi MF. American Gastroenterological Association medical position statement on the management of gastroesophageal reflux disease. Gastroenterology. 2008;135:1383-91.e1385. 7. American Gastroenterological Association medical position statement on the management of Barrett’s esophagus. Gastroenterology. 2011;140:1084-91. 8. Shaheen NJ, Falk GW, Iyer PG, Gerson LB. ACG clinical guideline: diagnosis and management of Barrett’s esophagus. Am J Gastroenterol. 2016; 111:30-50. 9. Katz PO, Gerson LB, Vela MF. Guidelines for the diagnosis and management of gastroesophageal reflux disease. Am J Gastroenterol. 2013;108:308-28. 10. National Institute for Health and Care Excellence. Suspected cancer: recognition and referral. Available at: https://www.nice.org.uk/guidance/ng12/chapter/1Recommendations-organised-by-site-of-cancer#upper-gastrointestinal-tractcancers. updated 2017. Accessed October 15, 2018. 11. Gisbert JP, Calvet X, Ferrandiz J, Mascort J, Alonso-Coello P, Marzo M. [Clinical practice guideline on the management of patients with dyspepsia. Update 2012]. Aten Primaria. 2012;44:e721-38. 12. Guardino JMKF, Lopez R, Wachsberger DM, Richter JE, Falk GW. Barrett’s esophagus at a tertiary care center: association of age on incidence and prevalence of dysplasia and adenocarcinoma. Am J Gastroenterol. 2006;101:2187-93. 13. Thrift AP, Whiteman DC. The incidence of esophageal adenocarcinoma continues to rise: analysis of period and birth cohort effects on recent trends. Ann Oncol. 2012;23:3155-62. 14. Portale G, Peters JH, Hsieh CC, Tamhankar AP, Almogy G, Hagen JA, et al. Esophageal adenocarcinoma in patients < 50 years old: delayed diagnosis and advanced disease at presentation. Am Surg. 2004;70:954-8. 15. Hashemi N, Loren D, DiMarino AJ, Cohen S. Presentation and prognosis of esophageal adenocarcinoma in patients below age 50. Dig Dis Sci. 2008;54:1708. 16. Markar SR, Karthikesalingam A, Low DE. Outcomes assessment of the surgical management of esophageal cancer in younger and older patients. Ann Thorac Surg. 2012;94:1652-8. 17. van Nistelrooij AM, Andrinopoulou ER, van Lanschot JJ, Tilanus HW, Wijnhoven BP. Influence of young age on outcome after esophagectomy for cancer. World J Surg. 2012;36:2612-21. 18. Zeng Y, Ruan W, Liu J, Liang W, He J, Cui F, et al. Esophageal cancer in patients under 50: a SEER analysis. J Thorac Dis. 2018;10:2542-50. 19. van Nistelrooij AM, van Steenbergen LN, Spaander MC, Tilanus HW, van Lanschot JJ, Lemmens VE, et al. Treatment and outcome of young patients with esophageal cancer in the Netherlands. J Surg Oncol. 2014;109:561-6. 20. Kasagi Y, Morita M, Otsu H, Kawano H, Ando K, Hiyoshi Y, et al. Clinicopathological characteristics of esophageal squamous cell carcinoma in patients younger than 50 years. Ann Surg Oncol. 2015;22:311-5. 21. Yoon HY, Kim CB. Gastroesophageal junction adenocarcinoma of young patients who underwent curative surgery: a comparative analysis with older group. Surg Today. 2011;41:203-9. 22. Mehta SP, Bailey D, Davies N. Comparative outcome of oesophagogastric cancer in younger patients. Ann R Coll Surg Engl. 2010;92:515-8. 23. Hamouda A, Forshaw M, Rohatgi A, Mirnezami R, Botha A, Mason R. Presentation and survival of operable esophageal cancer in patients 55 years of age and below. World J Surg. 2010;34:744-9. 24. Adam DJ, Craig SR, Sang CT, Walker WS, Cameron EW. Oesophagogastrectomy for carcinoma in patients under 50 years of age. J Roy Coll Surg Edinb. 1996;41:371-3. 25. Mori M, Ohno S, Tsutsui S, Matsuura H, Kuwano H, Sugimachi K. Esophageal carcinoma in young patients. Ann Thorac Surg. 1990;49:284-6.

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

9

THOR

Nobel et al

Thoracic

26. Molena D, Stem M, Blackford AL, Lidor AO. Esophageal cancer treatment is underutilized among elderly patients in the USA. J Gastrointest Surg. 2017;21: 126-36. 27. Rice TW, Blackstone EH, Rusch VW. 7th edition of the AJCC cancer staging manual: esophagus and esophagogastric junction. Ann Surg Oncol. 2010;17: 1721-4. 28. Wong AT, Shao M, Rineer J, Osborn V, Schwartz D, Schreiber D. Treatment and survival outcomes of small cell carcinoma of the esophagus: an analysis of the national cancer data base. Dis Esophagus. 2017;30:1-5. 29. Bai Y, Li ZS, Zou DW, Wu RP, Yao YZ, Jin ZD, et al. Alarm features and age for predicting upper gastrointestinal malignancy in Chinese patients with dyspepsia with high background prevalence of Helicobacter pylori infection and upper gastrointestinal malignancy: an endoscopic database review of 102,665 patients from 1996 to 2006. Gut. 2010;59:722-8. 30. Cohen E, Bolus R, Khanna D, Hays RD, Chang L, Melmed GY, et al. GERD symptoms in the general population: prevalence and severity versus care-seeking patients. Dig Dis Sci. 2014;59:2488-96. 31. Kountourakis P, Ajani JA, Davila M, Lee JH, Bhutani MS, Izzo JG. Barrett’s esophagus: a review of biology and therapeutic approaches. Gastrointest Cancer Res. 2012;5:49-57. 32. Ong CA, Shapiro J, Nason KS, Davison JM, Liu X, Ross-Innes C, et al. Three-gene immunohistochemical panel adds to clinical staging algorithms to predict prognosis for patients with esophageal adenocarcinoma. J Clin Oncol. 2013;31:1576-82.

Nobel et al

33. Martin JA, Osterman MJK. Describing the increase in preterm births in the United States, 2014–2016. NCHS Data Brief. 2018;312:1-8. 34. Austin GL, Ogden LG, Hill JO. Trends in carbohydrate, fat, and protein intakes and association with energy intake in normal-weight, overweight, and obese individuals: 1971-2006. Am J Clin Nutr. 2011;93:836-43. 35. Hales CM, Carroll MD, Fryar CD, Ogden CL. Prevalence of obesity among adults and youth: United States, 2015-2016. NCHS Data Brief. 2017;288:1-8. 36. Drahos J, Xiao Q, Risch HA, Freedman N, Abnet CC, Andersen LA, et al. Age-specific risk factor profiles of adenocarcinomas of the esophagus: a pooled analysis from the international BEACON consortium. Int J Cancer. 2016;138: 55-64. 37. Steyerberg EW, Earle CC, Neville BA, Weeks JC. Racial differences in surgical evaluation, treatment, and outcome of locoregional esophageal cancer: a population-based analysis of elderly patients. J Clin Oncol. 2005;23:510-7. 38. Greenstein AJ, Litle VR, Swanson SJ, Divino CM, Packer S, McGinn TG, et al. Racial disparities in esophageal cancer treatment and outcomes. Ann Surg Oncol. 2008;15:881-8. 39. Nobel T, Lavery J, Barbetta A, Gennarelli R, Lidor AO, Jones DR, et al. National guidelines may reduce socioeconomic disparities in treatment selection for esophageal cancer. Dis Esophagus. November 28, 2018 [Epub ahead of print].

Key Words: esophageal cancer, young patients, esophageal adenocarcinoma, screening guidelines

THOR 10

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al

Thoracic

Local

75 50 25

100 Overall Survival (%)

Overall Survival (%)

75 50 25

1 2 3 4 5 Years Since Diagnosis

0

<50

3071 2265 1671 1248

954

757

50+

38869 25092 17629 12626 9448

7062

0

50

0

1 2 3 4 5 Years Since Diagnosis

2328 1659 1035

697

496

371

50+

21443 13566 7989

4968

3290

2232

0

1 2 3 4 5 Years Since Diagnosis

Number at risk

<50

1 2 3 4 5 Years Since Diagnosis

Age Group <50 50+

25

Number at risk Age Group

Number at risk

75

0

0 0

Age Group

Overall Survival (%)

100

0

Age Group

Metastatic

Locoregional

100

<50

3009 1096 393

174

109

80

50+

22866 6375 2220

953

514

320

1 2 3 4 5 Years Since Diagnosis

0

1 2 3 4 5 Years Since Diagnosis

THOR

FIGURE E1. Overall survival by stage and age by age cutoff (50 years).

75 50 25

100 Overall Survival (%)

Overall Survival (%)

75 50 25

0

1 2 3 4 5 Years Since Diagnosis

0

<60

11242 8273 6087 4519

3504

2742

60+

30698 19084 13213 9355

6898

5077

0

50

1 2 3 4 5 Years Since Diagnosis

Age Group <60 60+

25

1 2 3 4 5 Years Since Diagnosis

0

<60

8277 5879 3595

2374

1627

1170

60+

15494 9346 5429

3291

2159

1433

0

1 2 3 4 5 Years Since Diagnosis

Number at risk

Number at risk Age Group

Number at risk

75

0

0

1 2 3 4 5 Years Since Diagnosis

Age Group

Overall Survival (%)

100

0

Age Group

Metastatic

Locoregional

Local 100

<60

9707 3311 1202

526

292

195

60+

16168 4160 1411

601

331

205

0

1 2 3 4 5 Years Since Diagnosis

FIGURE E2. Overall survival by stage and age by age cutoff (60 years).

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

10.e1

Thoracic

Nobel et al

TABLE E1. Definition of treatment groups Surgery Surgery in first 180 d from diagnosis No radiation or chemotherapy within first 180 d Surgery in first 180 d from diagnosis Radiation and/or chemotherapy in first 180 d Number of days to surgery < days to chemotherapy/radiation Palliative care Chemotherapy only

Chemo in first 180 d from diagnosis No radiation within first 180 d No surgery within first 180 d

Radiation Group B

Radiation in first 180 d from diagnosis No chemotherapy within first 180 d No surgery within first 180 d

Definitive chemoradiation Subsequent

Chemotherapy within first 180 d from diagnosis Radiation within first 365 d Days between start of chemotherapy and start of radiation is >7

THOR

Concurrent

Chemo within first 180 d from diagnosis Radiation within first 365 d Days between start of chemotherapy and start of radiation is <7

Induction þ surgery Chemotherapy alone þ surgery

Chemotherapy in first 180 d Surgery in first 180 d Number of days to chemo < number of days to surgery Chemotherapy starts more than 7 d before surgery

Chemotherapy þ radiation concurrent þ surgery

Chemotherapy and radiation in first 180 d Surgery in first 180 d Number of days to chemotherapy/radiation < number of days to surgery Number of days between start of chemotherapy and start of radiation is < 7

Other induction þ surgery (radiation alone or chemoradiation subsequent)

Chemotherapy and/or radiation in first 180 d from diagnosis Surgery in first 180 d Number of days to chemotherapy/radiation < number of days to surgery Number of days between start of chemotherapy and start of radiation is > 7

Radiation Group A

Radiation in first 180 d from diagnosis No chemotherapy within first 180 d Radiation occurs before any surgery (surgery would be salvage)

No treatment No treatment

10.e2

No treatment within 180 d

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al

Thoracic

TABLE E2. Staging Stage

Definition

Local

Unknown or a tumor of any size, unknown or no nodes, and unknown or no metastatic disease (Tx-4, Nx-0, Mx-0)

Locoregional

Unknown or a tumor of any size, positive nodes, and unknown or no metastatic disease (Tx-4, N1-3, Mx-1A)

Distant

Unknown or a tumor of any size, unknown nodes or positive nodes, and having a metastatic diagnosis (Tx-4, Nx-3, M1-M1B)

Adapted from Wong and colleagues.28

TABLE E3. Five-year overall survival by age group and age/treatment group Survival

Lower

Upper

18-57 y

25.8

25.2

26.4

58-65 y

26.5

25.9

27.2

66-74 y

24.2

23.6

24.8

75þ y

12.6

12.1

13.1

Age

Survival by age group and stage Survival

Stage

Lower

Upper

18-57 y

local

42.7

41.6

43.9

58-65 y

local

42.2

41.1

43.3

66-74

local

37.5

36.5

38.5

75þ y

local

17.4

16.6

18.1

18-57 y

Locoregional

28.1

26.9

29.3

58-65 y

Locoregional

26.2

25

27.5

66-74 y

Locoregional

20.9

19.7

22.1

75þ y

Locoregional

11.7

10.6

12.8

18-57 y

Metastatic

4.25

3.75

4.79

58-65 y

Metastatic

3.84

3.31

4.41

66-74 y

Metastatic

3.35

2.83

3.93

75þ y

Metastatic

1.53

1.15

2

Survival by age group and treatment Survival

Age

Treatment

Lower

Upper

18-57 y

Definitive chemoradiation

58-65 y

Definitive chemoradiation

13.1

12.1

14.1

13.5

12.5

66-74 y

14.5

Definitive chemoradiation

14

12.9

15

75þ y

Definitive chemoradiation

10.1

18-57 y

Induction þ surgery

40

9.16 38.5

11.1 41.5

58-65 y

Induction þ surgery

38.7

37.1

40.2

66-74 y

Induction þ surgery

35.4

33.7

37.1

75þ y

Induction þ surgery

27.4

24.3

30.7

18-57 y

No treatment

14.8

13.4

16.2

58-65 y

No treatment

13.9

12.6

15.3

66-74 y

No treatment

12.6

11.4

13.8

75þ y

No treatment

18-57 y

Surgery

4.82 52.9

4.26 51.3

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

5.43 54.4

10.e3

THOR

Survival by age group Group

Thoracic

Nobel et al

TABLE E4. Comparison of characteristics by age cutoff (50 years) Characteristic, N (%)

<50 y

Age quartile groups 50þ y

P overall

N

9202

92,394

Female gender

1341 (14.6%)

15,519 (16.8%)

<.001

White race

7812 (84.9%)

84,079 (91.0%)

<.001

26,814 (29.0%)

Any comorbidities

1397 (15.2%)

Median distance to hospital, miles (IQR)

13.6 (1.60-147)

Education (21% in ZIP code without high school diploma)

1520 (16.5%)

11.9 (1.30-135) 12,386 (13.4%)

<.001 <.001

THOR

Insurance status Government Private None Insurance status unknown

2210 (24.0%) 5951 (64.7%) 698 (7.59%) 343 (3.73%)

54,684 (59.2%) 32,914 (35.6%) 2390 (2.59%) 2406 (2.60%)

Facility type Academic Nonacademic

3379 (45.0%) 4131 (55.0%)

37,340 (40.4%) 55,054 (59.6%)

Location East South Midwest West

1650 (22.0%) 2625 (35.0%) 2149 (28.6%) 1086 (14.5%)

21,501 (23.3%) 29,938 (32.4%) 26,601 (28.8%) 14,354 (15.5%)

Patient location (combine into 250,000)

7259 (78.9%)

72,398 (78.4%)

<.001

<.001

.247 <.001

Diagnosis year 2004-2007 2008-2011 2012-2015

3067 (33.3%) 3066 (33.3%) 3069 (33.4%)

26,506 (28.7%) 30,418 (32.9%) 35,470 (38.4%)

Tumor site Abdominal/lower/stomach Overlapping Thoracic/middle Unspecified Upper/cervical

7940 (86.3%) 267 (2.90%) 294 (3.19%) 638 (6.93%) 63 (0.68%)

79,125 (85.6%) 2502 (2.71%) 3864 (4.18%) 6243 (6.76%) 660 (0.71%)

Differentiation grade Unknown Well/moderate Poor/undifferentiated

1634 (17.8%) 3403 (37.0%) 4165 (45.3%)

17,560 (19.0%) 35,418 (38.3%) 39,416 (42.7%)

10.e4

<.001 <.001

<.001

<.001

The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al

Thoracic

TABLE E5. Comparison of characteristics by age cutoff (60 years) Characteristic, N (%) N

<60 y 32,170

Female gender White race

Age quartile groups 60þ y

P overall

69,426

4300 (13.4%)

12,560 (18.1%)

<.001

28,361 (88.2%)

63,530 (91.5%)

<.001

21,542 (31.0%)

Any comorbidities

6669 (20.7%)

Median distance to hospital, miles (IQR)

13.4 (1.50-144)

11.5 (1.30-132)

Education (21% in ZIP code without high school diploma)

4961 (15.4%)

8945 (12.9%)

<.001 <.001 <.001 <.001

Insurance status Government Private None Insurance status unknown

7456 (23.2%) 21,609 (67.2%) 2040 (6.34%) 1065 (3.31%)

49,438 (71.2%) 17,256 (24.9%) 1048 (1.51%) 1684 (2.43%)

Facility type Academic Nonacademic

13,450 (44.1%) 17,028 (55.9%)

27,269 (39.3%) 42,157 (60.7%)

Location East South Midwest West

6696 (22.0%) 10,490 (34.4%) 8871 (29.1%) 4421 (14.5%)

16,455 (23.7%) 22,073 (31.8%) 19,879 (28.6%) 11,019 (15.9%)

Patient location (combine into 250,000)

25,187 (78.3%)

54,470 (78.5%)

<.001

.560 <.001

Diagnosis year 2004-2007 2008-2011 2012-2015

9862 (30.7%) 10,755 (33.4%) 11,553 (35.9%)

19,711 (28.4%) 22,729 (32.7%) 26,986 (38.9%)

Tumor site Abdominal/lower/stomach Overlapping Thoracic/middle Unspecified Upper/cervical

27,629 (85.9%) 917 (2.85%) 1168 (3.63%) 2228 (6.93%) 228 (0.71%)

59,436 (85.6%) 1852 (2.67%) 2990 (4.31%) 4653 (6.70%) 495 (0.71%)

Differentiation grade Unknown Well/moderate Poor/undifferentiated

5816 (18.1%) 12,058 (37.5%) 14,296 (44.4%)

13,378 (19.3%) 26,763 (38.5%) 29,285 (42.2%)

<.001

<.001

IQR, Interquartile range.

The Journal of Thoracic and Cardiovascular Surgery c Volume -, Number -

10.e5

THOR

<.001

Thoracic

000

Higher clinical suspicion is needed for prompt diagnosis of esophageal adenocarcinoma in young patients Tamar B. Nobel, MD, Michael Curry, MS, Renee Gennarelli, MS, David R. Jones, MD, and Daniela Molena, MD, New York, NY Perceptions of esophageal cancer as a disease of the elderly may result in the higher rate of advanced diagnostic stage in younger patients. Risk factors, rather than age, should guide evaluation.

THOR The Journal of Thoracic and Cardiovascular Surgery c - 2019

Nobel et al