Endoscopic Treatment and Surveillance of Esophageal Cancer: GI Perspective

Charles J. Lightdale, MD

Columbia University

New York, NY

Barrett's

metaplasia

Chronic

inflammation

Squamous

esophagus Injury

Acid & bile reflux

Genetics

Gender, race,

Evolution of

Barrett’’’’s esophagus

Adenocarcinoma

High-grade

dysplasia

Low-grade

dysplasia

Accumulate

Genetic

Changes

1.0%/yr

<0.5%/yr

Hvid-Jensen, et al. N Engl J Med 2011;365:1375-83

Barrett’s Prevalence Estimates

• 1.6% of general adult population (3.3 M)– Ronkainen J, et al. Prevalence of BE; Gastroenterology

2005;129:1825-31.

• 6.8% of persons over age 40 (8.7 M)– Rex DK, et al. Screening for Barrett’s... Gastroenterology 2003;

125:1670-77.

• Majority without GERD – Gerson LB, et al. Prevalence of Barrett’s;Gastroenterology

2002;123:461-7.

3

Accuracy and Acceptability of

Cytology Sponge with IHC for TFF3

• 501/504 patients swallowed sponge

capsule

• Compared to endoscopy and biopsy:• Compared to endoscopy and biopsy:

>1 cm BE >2 cm BE

Sensitivity 73.3% 90.0%

Specificity 93.8% 93.5%

Kadri, et al. BMJ 2010;341:4372

Corley, et al. Gastroenterology 2002;123:633-640

Technique of Biopsy in Endoscopic SurveillanceTechnique of Biopsy in Endoscopic Surveillance

BE With and Without Dysplasia: Confounding Factors

• Endoscopic Biopsy Sampling Error

– Persists even with “Seattle Protocol”

• Pathology Interpretation • Pathology Interpretation

– LGD: Κ = 0.32 (fair)

– HGD/IMC: Κ = 0.65 (substantial)

– HGD vs IMC: Κ = 0.42 (fair)Ormsby AH, et al. Gut 2002;51:671-76

Montgomery, Canto. CG & H 2006;4:1434-39

Adherence to Surveillance Biopsy Guidelines by Year

56.3%

44.7%49.9% 48.3%

54.8% 55.5%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

Pe

rce

nta

ge

0.0%

10.0%

20.0%

30.0%

40.0%

2002 2003 2004 2005 2006 2007

Year

Pe

rce

nta

ge

Abrams JA, et al. Clin Gastroenterol Hepatol 2009;736-42

Optical Contrast EndoscopyOptical Contrast Endoscopy

WLE NBI

Mucosal Patterns:

Regular/Irregular

3708

Confocal Probe BE

IM

HGD

Wide Area Transepithelial Sample (WATS)

15

WATS method is

designed to decrease

biopsy sampling

Flat HGD and

IMCA can occur

in a mosaic

pattern missed

by 4-Q biopsy

biopsy sampling

error

17

Early Esophageal Adenocarcinoma

Risk of lymph node metastases:

• Tis (intraepithelial) 0

• T1a (intramucosal) 2%• T1a (intramucosal) 2%

• T1b (submucosal) 25% Nigro, et al. J Thorac Cardiovasc Surg 1999;117:16-25

Stein, et al. Ann Surg 2000;232:733-742

Rice, et al. J Thorac Cardiovasc Surg 2001;122:1077-90

Hulscher, et al. N Engl J Med 2002;347:1662-1669

EMR Techniques

• INJECTION ASSISTED:– Inject and snare (saline-assisted polypectomy)

– Inject, lift, and cut with snare (2-channel scope)

• CAP ASSISTED:• CAP ASSISTED:– inject, endoscopic suction, and snare

• LIGATION ASSISTED:– Band and snare

• ENDOSCOPIC SUBMUCOSAL DISSECTION) (ESD)– Inject and cut with free-hand knife for en-bloc resection

EMR for HGD/Early Carcinomain Barrett’s Esophagus

• Outpatient procedure

• Major risk is bleeding, usually mild, < 5%

• Perforation is rare, < 1%• Perforation is rare, < 1%

• Pathology specimen for evaluation of tumor depth and margins.

Ell. Gastroenterology 2000;118:670-7.

Alvarez Herrero. Endoscopy 2011;43:177-83.

Mino-Kenudson. Gastrointest Endosc 2007;66:660-6.

Peters. Gastrointest Endosc 2008;67:604-9.

Moss. Am J Gastroenterol 2010;105:1276-83.

Good Risk Lesions For EMR in Early Esophageal Adenocarcinoma

• Non-ulcerated lesions, < 2.0 cm in diameter

• Invasion limited to mucosa• Invasion limited to mucosa

• No lympho-vascular invasion

• No poorly differentiated histology

Ell et al. Gastrointest Endosc 2007;65:3-10.

EMR in HGD/T1a complicating

Barrett’s esophagus

• 100 patients, 144 resections

• No major complications; 11 mild bleeding

• Complete local remission in 99% after a • Complete local remission in 99% after a maximum of 3 resections

• 11% recurred in 36.7 months, all successfully retreated with EMR

Ell. Gastrointest Endosc 2007;65:3-10

Complete Removal of BE with EMR:Radical Sequential EMR

• Length of Barrett’s segment is a major factor

• SSBE < 3 cm has best results• SSBE < 3 cm has best results

• Strictures >50% if >3.0 cm length or > ¾ circumference of lumen

• Longer segments: focal EMR + ablationPeters et al. Am J Gastroenterol 2006;101:1449-57.

Larghi et al. Endoscopy 2007; 39:1086-91.

Van Vilsteren, et al. Gut 2011;60:765-73

Currently Available Endoscopic Ablation Techniques for BE

• Argon Plasma Coagulation (APC)

• Multipolar Electrocoagulation (MPEC)

• Heat Probe• Heat Probe

• Lasers (Nd:YAG, KTP, Argon)

• Photodynamic Therapy (PDT)

• Cryotherapy

• Radiofrequency Ablation (RFA)

Prasad, et al. Gastroenterology 2007;132:1226-1233

Radiofrequency Ablation Balloon

Focal Radiofrequency Ablation Device

RFA versus Surveillance at 1-YearShaheen, et al. N Engl J Med 2009;360:2277-88

RFA ComplicationsShaheen, et al. N Engl J Med 2009;360:2277-88

Serious Adverse Events:

3/84 (3.6%). 1 UGI bleed, 2 chest pain

Strictures:

5/84 patients (6%) resolved:mean 2.6 dilations5/84 patients (6%) resolved:mean 2.6 dilations

Subsquamous intestinal metaplasia:

RFA Surveillance

Baseline 25% 26%

1-year 5% 40%

Longterm Durability of Ablation

38Shaheen, et al. Gastroenterology 2011:141:460-468.

EMR+RFA vs. Esophagectomy

in BE with HGD/IMC

• All patients treated at USC 2001-2010

EMR+RFA(40)

Esophagectomy(61)Esophagectomy(61)

Survival (3 yr) 94% 94% ns

Morbidity 0% 39% p <

0.0001

Zehetner, et al. J Thorac Cardiovasc Surg 2011;141:39-47.

Conclusions

• Low-cost screening for BE may become feasible.

• New methods may improve surveillance.

• Need for better risk stratification in BE.• Need for better risk stratification in BE.

• EMR for staging of all focal dysplastic lesions.*

• HGD/IMCA: Endoscopic Therapy (EMR, Ablation)

preferred in most patients to esophagectomy.**AGA Medical Position Statement on the Management of BE

Gastroenterology 2011;140:1084-1091.