Tumor Differentiation as a Prognostic Marker in Clinically Staged T1bN0 Esophageal Adenocarcinoma

Abstract

Current guidelines recommend that clinically staged T1N0 esophageal cancers are to be referred to surgery or endoscopic resection. Using the National Cancer Database, we identified 733 individuals with clinically staged T1N0 esophageal carcinoma, who underwent upfront surgery and did not receive any prior treatment. We assessed upstaging, which was defined as ≥ T2 disease or positive lymph nodes. Poorly differentiated adenocarcinomas were associated with upstaging, whereas squamous cell carcinomas were not. Specifically, the percentage of upstaging among individuals with clinically staged T1b and poorly differentiated tumor was 33.8%. Therefore, clinically staged T1bN0 poorly differentiated esophageal adenocarcinomas are at high risk for upstaging following surgery.

Keywords:

• Esophageal adenocarcinoma
• stage T1b
• tumor differentiation
• prognostic marker
• misclassification

Introduction

Clinically staged T1N0 (cT1N0) esophageal cancers are referred to surgery or endoscopic resection per NCCN and ESMO guidelines. According to NCCN guidelines, neoadjuvant chemoradiation or perioperative chemotherapy (adenocarcinoma histology only) are offered in the following cases: clinically staged T2N0 tumors with poor risk features (lymphovascular invasion, tumor size ≥3 cm or poorly differentiated tumor), T3 tumors or above, or positive lymph nodes (Npos). In patients with adenocarcinoma histology, upstaging upon surgery from cT1N0 to pathologically staged T3/Npos requires offering adjuvant chemoradiation based on the INT-0116 study (1,2) or chemotherapy alone. However, in patients with squamous histology, upstaging upon surgery does not mandate any further treatment. Therefore, there is a need to define prognostic markers in cT1N0 esophageal adenocarcinoma.

Initial work-up of esophageal cancer includes chest-abdomen-pelvic CT scan. In the absence of evidence of M1 disease, endoscopic ultrasound (EUS) and fluorodeoxyglucose-positron emission tomography (PET-FDG) are recommended. The sensitivity and specificity rates of EUS for the correct evaluation of T stage are 81–92, and 94–97%, respectively (3); and for involvement of lymph nodes, the sensitivity and specificity rates are both 85% (3). Adding fine needle aspiration (FNA) to EUS improved the sensitivity and specificity rates for involvement of lymph nodes to 97 and 96%, respectively (3). The sensitivity and specificity rates of PET-FDG are 67 and 91%, respectively (4).

Previous studies have shown that upstaging is common in esophageal cancer. In a retrospective study of 355 patients with cT2N0 esophageal cancer patients, 50% had nodal disease upon surgery (5). Similarly, in the FFCD 9901 study, involving stage I–II esophageal cancer patients, the rate of node positive disease was 27.7% (54/195) based on clinical stage, and was increased to 43.8% (75/171) upon surgery (6). In a recent study, we showed that 42% of poorly differentiated cT1N0 gastric adenocarcinomas were upstaged following surgery, suggesting that these patients may benefit from preoperative chemotherapy (7).

We aimed to evaluate clinical parameters associated with upstaging in cT1N0 esophageal cancers, with either adenocarcinoma or squamous histology, in order to refine the definition of patients at high risk for upstaging upon surgery. The rationale for our study stems from the notion that those at high risk for upstaging should be considered for a more thorough work-up and/or neoadjuvant treatment.

Methods

Data source

We used data from years 2010 to 2014 in the National Cancer Database (NCDB) (8). The NCDB is a joint project of the Commission on Cancer of the American College of Surgeons and the American Cancer Society. The data used in the study are derived from a de-identified NCDB file. The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology employed, or the conclusions drawn from these data by the investigator.

Patient population

Individuals with either adenocarcinoma or squamous cell esophageal carcinoma who were clinically stage as T1N0, underwent surgery with clear margins, and did not receive either neoadjuvant chemo- or radiotherapy were included in the study. An additional analysis was performed separately using data from 2004 to 2009, due to lack of data on type of surgical procedure, as well as lack of subclassification of cT1a and cT1b.

Variables definition

Baseline characteristics included age, sex, race, patient comorbidities (Charlson–Deyo comorbidity condition, CDCC) (9,10), tumor histology and tumor grade. Race and ethnicity were categorized as White, African-American or other. Tumor grade was defined as well-, moderately-, poorly differentiated or undifferentiated. Tumor location within the esophagus was defined as found at the upper third, middle third, lower third, overlapping sites or unspecified.

Statistical analysis

Student’s t-test and Chi-squared test were used to compare patient characteristics for continuous and dichotomous variables, respectively. A two-sided p value < 0.05 was defined as significant.

All statistical analyses were performed using Stata/IC software 13.0 (StataCorp, College Station, TX, USA).

Results

We identified 733 individuals with either adenocarcinoma or squamous cell esophageal carcinoma who were clinically stage as T1N0, underwent surgery with clear margins, and did not receive either neoadjuvant chemo- or radiotherapy. The median follow-up time was 38.6 months (IQR 23.2–57.9). Of those individuals, 639 (87.2%) had adenocarcinoma and 94 (12.8%) had squamous cell carcinoma.

Out of the 733 individuals included in our cohort (Table 1), 129 (17.6%) were upstaged upon surgery, i.e., had pathologic stage ≥ T2 or positive lymph nodes (pT2+/Npos). Higher tumor grade was associated with upstaging following surgery (p < 0.001, Table 1). Specifically, 62/195 (31.8%) of individuals with a poorly differentiated tumor were upstaged following surgery. Stratified by histology, misclassified individuals with adenocarcinoma were more likely to have poorly differentiated tumor (data not shown). On the contrary, in misclassified individuals with squamous cell carcinoma, tumor differentiation was not correlated with misclassification (data not shown). Higher pathologic stage was not associated with primary tumor location (Table 2), regardless of T stage (data not shown). Higher pathologic stage was not associated with year of diagnosis (data not shown). Individuals with either pT1N0 or pT2+/Npos following surgery had similar time from diagnosis to surgery (median of 43 days (IQR, 23–70) and 45 days (IQR, 28–66), respectively).

Table 1. Patient characteristics.

	ALL N = 733	cT1N0 pT1N0 N = 604	cT1N0 pT2+/Npos N = 129	p Value
Age, median (IQR)	65 (59–70)	65 (60–71)	65 (59–70)	0.77
Sex, % male (N)	80.4 (589)	80.0 (483)	82.2 (106)	0.57
Race				0.68
White, % (N)	92.9 (681)	92.7 (560)	93.8 (121)
African-American, % (N)	3.8 (28)	3.6 (22)	^b
Other, % (N)	3.3 (24)	3.7 (22)	^b
CDCC^a				0.52
0, % (N)	63.3 (464)	63.7 (385)	61.2 (79)
1, % (N)	28.5 (209)	28.0 (169)	31.0 (40)
≥2, % (N)	8.2 (60)	8.3 (50)	7.8 (10)
Histology				0.48
Adenocarcinoma	87.2 (639)	87.6 (529)	85.3 (110)
Squamous	12.8 (94)	12.4 (75)	14.7 (19)
Grade				<0.001
Well, % (N)	16.8 (123)	18.1 (109)	10.9 (14)
Mod, % (N)	43.4 (318)	44.5 (269)	38.0 (49)
Poor, % (N)	26.6 (195)	22.0 (133)	48.1 (62)
Un, % (N)	^b	^b	^b
Other, % (N)	12.0 (88)	14.1 (85)	^b

^a Charlson–Deyo comorbidity condition.

^b Less than 10 individuals.

Table 2. Misclassification according to primary tumor location within the esophagus.

Primary site	Number of cases N = 733 (%)	T2+/Npos N (% per location)
Cervical/upper third (C150/C153)	19 (2.6)	^a
Thoracic/middle third (C151/C154)	82 (11.2)	14 (17.1)
Abdominal/lower third (C152/C155)	553 (75.4)	103 (18.6)
Overlapping sites (C158)	19 (2.6)	^a
Unspecified (C159)	60 (8.2)	^a

^a Less than 10 individuals.

Stratified by T stage, 12 individuals (7.5%) with cT1a and 42 individuals (19.4%) with cT1b were misclassified and had pT2+/Npos. Among individuals with cT1a and cT1b poorly differentiated tumors, 20.7% (six out of 29) and 33.8% (23 out of 68), respectively, were misclassified and had pT2+/Npos, compared with 8.5% (five out of 59) and 13.6% (14 out of 103) of individuals with moderately differentiated tumors, and 2.5% (one out of 40) and 17.2% (five out of 29) of individuals with well-differentiated tumors.

Misclassification association with either T or N categories was next evaluated. Misclassification due to a change in T stage, N stage or both was found in 55 (42.6%), 45 (34.9%) and 29 (22.5%) out of 129 individuals, respectively. The details of this analysis are shown in Table 3.

Table 3. Misclassification according to T and N staging.

	pN
pT	0	1	2	3	Total
1	0	^a	^a	0	^a
1A	0	^a	^a	0	^a
1B	0	31	^a	^a	36
2	43	^a	^a	^a	56
3	12	11	^a	^a	28
Total	55	57	11	^a	129

^a Less than 10 individuals.

We then analyzed data from years 2004 to 2009, for which data on type of surgical procedure were lacking, as well as lack of subclassification of cT1a and cT1b. Out of a total of 1030 individuals with cT1N0 esophageal cancer that met our inclusion criteria, 187 individuals (18.2%) were upstaged, and were characterized by higher tumor grade (p value < 0.001). Specifically, in individuals with poorly differentiated tumors, 73 out of 226 (32.3%) were upstaged.

Discussion

Among individuals with cT1N0 esophageal carcinomas, 18% were upstaged following surgery. Upstaging was found mainly in individuals with poorly differentiated tumors, and this association was statistically significant in adenocarcinoma, but not in squamous cell carcinoma. Upstaging was not associated with primary tumor location.

The percentage of upstaging among individuals with clinically staged T1b and poorly differentiated tumor was 33.8%, suggesting these parameters define a high-risk subpopulation of esophageal cancer. To the best of our knowledge, the frequency of upstaging among cT1N0 esophageal cancer was not assessed hitherto.

Our findings suggest that patients with T1b poorly differentiated esophageal adenocarcinomas should be thoroughly evaluated prior to surgery, and neoadjuvant treatment should be considered.

The FFCD 9901 phase III study randomized 195 patients with stage I (20%) and stage II (80%) esophageal cancer to either surgery alone or neoadjuvant chemoradiotherapy with cisplatin plus fluorouracil followed by surgery. There was no survival benefit for addition of neoadjuvant chemoradiotherapy. Lack of a survival benefit was shown even for patients with a node positive disease upon surgery (6). However, this study may have been underpowered to detect a significant survival benefit, if one was present (11). Moreover, current NCCN and ESMO guidelines mandate neoadjuvant treatment for patients with a node positive disease, based on the CALGB 9781 and CROSS studies (12,13).

This study had several limitations. First, the small size of the cohort of squamous cell carcinoma individuals may be responsible for the lack of statistically significant association between upstaging and poor differentiation of the tumor. Second, NCDB lacks information on known confounders in esophageal cancer, such as smoking and alcohol use. Third, NCDB does not contain information regarding the imaging modality used for staging, preventing an evaluation of a possible association between a specific imaging modality (e.g., CT scan, FDG-PET/CT and/or EUS) and the frequency of upstaging.

The main strength of this study was using a nationwide cancer registry, capturing data on 70% of cancer diagnoses in the United States.

In summary, upstaging of cT1N0 esophageal carcinoma is associated with poorly differentiated tumors and adenocarcinoma histology. Specifically, 34% of individuals with clinically staged T1b and poorly differentiated tumors were upstaged. These findings suggest that T1b poorly differentiated esophageal adenocarcinomas should be thoroughly evaluated prior to surgery. Initial work-up must include a CT scan, FDG-PET/CT and EUS, and only if there is no evidence of M1 disease, the patient should be referred to an upfront surgery. In cases where nodal disease will be detected, patients should be considered for neoadjuvant treatment.

Author contributions

Conceptualization: OM, ESS, YXY, RM and BB. Data curation: OM, RM and BB. Project administration: OM, ESS, YXY, RM and BB. Writing – original draft: OM, ESS, YXY, YRL, RM and BB. Writing – review and editing: OM, ESS, GS, YXY, YRL, ABN, IL, KAR, TG, NH, DA, BG, RM and BB. Validation: OM and BB. Software: OM, YXY, RM and BB. Investigation: OM and BB. Resources: OM, ESS, YXY, RM and BB. Formal analysis: OM, YXY, RM and BB. Methodology: OM, YXY, RM and BB. Supervision: OM and BB.

Ethical approval

N/A (all data were extracted from NCDB).

Disclosure statement

The authors declare that they have no competing interests.

Data availability statement

The data that support the findings of this study are available from the National Cancer Database, maintained by the American College of Surgeons. Restrictions apply to the availability of these data, which were used under license for the current study, and are therefore not publicly available.

Additional information

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.