Barrett's Oesophagus

Section: UGI Sub-section: Oesophagus Curriculum: Curriculum, page 94

Definition

• Metaplasia of the distal oesophageal mucosa (> 1cm above GOJ) in response to chronic exposure to gastric juice
• Replacement of the squamous epithelium by columnar epithelium
• Seen endoscopically and confirmed histologically
• Australian definition requires
  • Columnar lined epithelium in tubular oesophagus and Goblet Cells (intestinal metaplasia)
• British definition
  • Doesn’t require goblet cells
• Pre-malignant condition

Incidence

• Found in up to 2% of general population
• Potentially increasing (or ? More endoscopy)
• In ≈ 10% of pts with symptomatic / chronic, severe GORD
• Risk is related to duration & severity of reflux.
• Increases with age
• More frequent in Men > 50yo
• Caucasians most common
• In Barrett’s patients
  • Incidence of low-grade dysplasia = 15-25%
  • High grade dysplasia = 5-10%
  • Risk does NOT regress if reflux controlled
• Seen in 15% of oesophagitis, 45% of peptic strictures
• Hiatus hernia more common in Barrett’s

Risk Factors

• Age
• GORD
• Male, Caucasian
• Central Obesity
• Smoking
• EtOH equivocal
• Protective factors
  • H. pylori (Atrophic gastritis = Less acid)
  • Height

Pathogenesis

• 60% related to chronic GORD

  • Adaptive response to increased cell loss as a result of chronic inflammation due to GORD
  • Sensitive to acidic/alkaline and biliary reflux
    • Leads to cell loss, necrosis and ulceration
• Also other causes chronic oesophagitis:
  • NSAID
  • Chemotherapy
  • Viral infection
• Exacerbated by obesity/smoking unclear
• Reflux healed by pluripotent stem cells
  • Get gastric differentiation, true metaplasia
• Believed to induce metaplasia, dysplasia, adenocarcinoma sequence

Risk of Cancer Progression

• 10-50x increased risk for oesophageal adenoca
  • 140 x with high grade dysplasia
  • ≈ 0.2-0.5% / year risk = 10% at 20 years
• Non-dysplastic Barrett’s risk of progression
  • 0.12-0.33%
  • 1-3 cancers per 1000 patient years
• Low Grade Dysplasia
  • 0.7-1%
• High Grade Dysplasia
  • 6-8%
• BUT < 10% of Oesophageal Adenocarcinoma due to Barrett’s

Pathophysiology of Cancer Progression

• ? Progression Stepwise
  • Barrett’s → Low grade dysplasia (LGD) → High grade dysplasia (HGD) → Cancer
  • Likely due to accumulation of genetic mutations (APC, p53, DCC, MTS1/p16)
• Initiation
  • Reflux induced ulceration/inflammation induces tumour suppressor gene mutations (p53/p16) in stem cell populations in oesophageal gland squamous ducts
• Clonal proliferation
  • Following this ‘initiation’ phase multiple distinct clones of metaplastic tissue compete to colonise the oesophagus creating a mosaic pattern
• Dominant clone
  • Clonal proliferation of population with greater selective advantage leads to dominant clones
  • Once initiated the propagation is dependent on surrounding environment with emphasis on pro-inflammatory cytokines (IL-1, TNF and transforming growth factor b TGF-β) and reduction in E-cadherin with loss of cell adhesion
• Leads to metaplasia-dysplasia-adenocarcinoma sequence

Risk Factors for Cancer Progression

• Length of Barrett’s segment
• Extent of dysplasia
• HGD > LGD
• Protective effect from Aspirin and PPIs
  • AspECT

Histology

• Looking for intestinal metaplasia has high false negative rate
  • Need 8 biopsies for 70% pick-up of IM
    • 4 biopsies only picks up 38%
• Histological Findings
  • Metaplasia 90%
  • Indeterminate for dysplasia
  • Low Grade Dysplasia 9%
  • High Grade Dysplasia 1%
  • Intramucosal carcinoma

Histology - Metaplasia

• Intestinal type epithelium characterized by presence of Goblet Cells

Histology – Low Grade Dysplasia

• Loss of cellular differentiation
• Loss of goblet cells
• Difficult to classify:
  • 10% progress to HGD
  • 60% static
  • 20% stable

Histology – High Grade Dysplasia

• Distinct cytological changes
• Should be confirmed by two pathologists
• Pleomorphic nuclei
• Loss of nuclear polarity
• Nuclei become oriented apically, rather than basally
• Abnormal crypt architecture
• Progression
  • Approx 40-50% progress to adenocarcinoma in 5 years
  • Approx 10% in 12 months
  • Occult Ca may already be present in ≈ 40-50% at time of operation for HGD
  • Includes ≈ 10% invasive CA
  • Therefore all require complete staging and treatment

Histology – Intramucosal Adenocarcinoma

• Invasion through BM into Lamina Propria
• But not into submucosa

Classification

• Traditional
  • Short segment (< 3cm) vs Long Segment (> 3cm)
• Histological
  • Metaplasia
  • LGD
  • HGD
• Prague Criteria
  • Validated method of disease classification based on endoscopic appearance

Prague Criteria

• Validated method of disease classification based on endoscopic appearance
• 3 measurements
  • Depth of GOJ
  • Depth of circumferential extent
  • Depth of maximal extent
• Islands don’t count
• Reported as
  • Prague C(x) and M(x)

Investigations

Endoscopy

• Tongues of salmon-pink epithelium in lower oesophagus
• Above GOJ, as measured by the proximal extent of the gastric folds
• Biopsy Protocol (Seattle Protocol)
  • Systematic assessment of oesophagus from GOJ, extending to the squamo-columnar junction
  • Visible lesions target biopsied (Cancer until proven otherwise!)
  • 4 quadrant biopsies every 2cms of macroscopically normal mucosa, or every 1cm if previous dysplasia
• New techniques for image-focused biopsy can be added
  • Narrow band imaging (do before biopsies!)
  • Chromoendoscopy
  • Enhance epithelial architecture or to stain columnar cells
    • Will increase accuracy of Bx
    • Increases HGD / intramucosal cancer yield

GOJ Landmarks on Endoscopy

• Most proximal extent of gastric folds (under minimal insufflation)
  • Most consistent/reproducible
  • Consensus: best to use
• Inferior limit of straight palisade vessels
  • Terminate when hit the LOS
  • Used more in Asia
  • May be useful in difficult cases
• Hiatal Pinch – not reliable because moves with hiatus hernia
• LOS – poorly reproducible
• Squamo-columnar junction wont be GOJ in Barrett’s! (duh)

Management

• Biopsy results
  • Metaplasia only and Indefinite
  • Low Grade Dysplasia
  • High Grade Dysplasia
  • Ulceration
    • Give high dose PPI and re-scope 6-8/52
    • If ulceration in HGD – 80% chance of cancer
• Medical Management
• Endoscopic Management
• Surgical Management

Management of Barrett’s Metaplasia

• Acid suppression

  • PPI or anti-reflux therapy

• Surveillance depends on length of columnar lined oesophagus and presence of intestinal metaplasia

• Short segment <3cm - 3-5

• Long segment > 3cm 2-3

• No evidence for ablation for metaplasia only

• If biopsy shows Indefinite for dysplasia

  • High dose PPI
  • Rescope in 6 months

• No dysplasia

  • Short segment <3cm - 3-5
  • Long segment > 3cm 2-3

• If dysplasia, manage as per LGD/HGD algorithms

Management of Barrett’s LGD

• Repeat scope in 6 months
• Confirm histology with 2 GI pathologists
  • 75-85% downgrade after further review
• If confirmed low grade dysplasia and persistent on repeat scope
  • Maximal PPI
  • Refer UGI Centre for MDT discussion + Endoscopic Ablation
• If just observed (no rx) - Surveillance
  • Repeat scope in 6/12 and remain on PPI
• SURF Trial
  • Endoscopic ablation (RFA) vs surveillance in confirmed LGD
  • Reduced risk of progression to HGD/Ca if ablate (1%) cf surveillance (26%)
  • No change survival
  • Sustained clearance of BE in 91%
  • 11% stricture
• If dysplasia resolves
  • Can go back to 2-3 yearly scopes

Management of Barrett’s HGD

• Confirm with 2 GI pathologists and discuss at Upper GI MDM
• Repeat biopsies (Seattle criteria)
  • 4 quadrant 1cm apart with large number of biopsies
• Stage patients
• May require diagnosis endoscopic resection of most affected part
• Management – Treatment, not surveillance
  • RFA to HGD
  • EMR to suspicious areas
    • EMR works as an excisional/ therapeutic biopsy to confirm not invasive
    • Selected cases may require surgical excision – see below
  • NB: If ulcerated lesion in area of HGD
    • Likely Invasive Ca (80%), unlikely suitable for endoscopic management

Management of Intramucosal Cancer

• Confirm with 2 GI pathologists
• Refer UGI centre
• Discuss at Upper GI MDM
  • Patient factors
  • Barrett’s factors
    • Length, multifocal, associated hiatus hernia
    • Differentiation
    • LVI
• Risk of LN involvement 2%
• EMR for T1a lesions
  • Curative management, or at least good staging biopsy
  • Equivalent overall survival with surgery
  • Higher recurrence rates
    • But majority still able to be treated endoscopically
  • Significantly reduced morbidity
• Requires intensive treatment and follow-up over multiple sessions
• NB: T1b lesions not amenable to endoscopic management – need surgery

Medical Management

• Long term, High dose PPI
  • Improvement in metaplasia
• Treat all LGD with 8-12 weeks of high dose PPI suppression
• Ranitidine
  • Not as effective in improvement in columnar lined oesophagus
• Aspirin as chemoprevention
  • COX inhibitor, acts to down regulate the levels of prostaglandins
  • Prostaglandins responsible for
    • Resistance to apoptosis
    • Increased angiogenesis
    • Enhanced invasion of the esophageal mucosa in Barrett’s

AspECT Trial

• Prospective, randomised trial investigating the protective effect of esomeprazole and aspirin for Barrett’s
  • 1x1x1x1 design
    • Esomperazole 40mg PO BD vs 20mg OD
    • Aspirin 300/325mg PO OD vs No Aspirin
  • 84 centres, UK and Canada
  • 2557 patients, Median F/U 8.9 yrs
  • Endpoints
    • All-cause mortality
    • Oesophageal adenocarcinoma
    • High grade dysplasia
  • Results
    • High dose PPI significantly better than Low dose
    • Aspirin not significantly better than no aspirin
      • But was when patients on NSAIDs taken out of study
    • High dose PPI with Aspirin the best Low dose PPI without Aspirin the worst
  • NNT
    • 34 for PPI
    • 43 for Aspirin

Endoscopic Management

• Aims
  • Confirm dysplasia
  • Remove all nodules or HG dysplasia with EMR
  • Biopsy all lesions (don’t ablate lesions)
  • Ablate all metaplasia
  • Preserve swallow, minimal morbidity
• Treatment
  • Endoscopic mucosal resection (EMR)
    • Goal
      • Remove mucosa and submucosa
      • Works as curative treatment or biopsy
    • Complications
      • Bleeding 5%, Perforation 1-2% (most managed endoscopically)
      • High risk of stricture for circumferential resection (20%)
    • Remission in 80-90% pts with HGD
      • Can re-treat for recurrence
  • Endoscopic ablation (do 3 months post EMR)
    • Thermal (APC, multipolar electrocautery, laser therapy and cryotherapy)
      • Effective but less uniform in depth and higher risk of stricture
    • Chemical (Photodynamic Therapy – PDT and Radiofrequency Ablation - RFA)
      • Deliver even therapy of consistent depth
      • Can treat large circumferential areas
    • RFA
      • Good for ablating the remaining Barrett’s mucosa after any nodules/HGD removed with EMR
      • Complete therapy in 90% for LGD and 80% HGD at 12 months

Radiofrequency Ablation (Halo)

• Circumferential balloon/plate
• Bipolar energy produces heat – get thermal ablation of superficial tissue
• Regenerates with neo-squamous mucosa
• Usually requires multiple sessions, continued follow-up
• Success for Dysplasia – 91%
• Success for Metaplasia – 78%
  • Durable at 5yrs
• Risks:
  • Stricture – 6% (note up to 33% with EMR)
  • Pain – 4%
  • Bleeding/perforation – 1%
  • Buried glands – rare 0.9%
    • Mucosa gets buried and miss sub-squamous Barrett’s
  • Be careful with long segments (may require surgery instead), multifocal area, hiatus hernias
  • Post-ablation recurrence – 7-10%
• Recurrence of dysplasia – 1-2%
• Needs on-going close surveillance on PPIs post

Surgical Management

Fundoplication for Barrett’s

• For symptom control of reflux only
• Can maintain endoscopic surveillance after this
• At best, a reduction in the length, but minimal regression of dysplasia
  • No reduction in rates of Oesophageal Ca
• Lap fundoplication & Barrett’s
  • Regression of intestinal metaplasia is reported in 14-47% after lap anti-reflux surgery and up to 59% if short-segment Barrett’s is included
  • No evidence for the prevention of progression to Ca

Oesophagectomy for Barrett’s

• Has role in treatment for HGD or Intramucosal cancer
• Requires MDT discussion in specialised UGI centre
• Increasing length of Barrett’s reduces success of endoscopic methods
• May consider for
  • Multifocal disease
  • Recurrence
  • Associated Hiatus Hernia
  • Patient preference (don’t want multiple treatment sessions)
  • Concerns re: compliance with follow-up
  • Young or fit patient
  • Long segment
• Up to 40% of pts with HGD harbour invasive component
  • With over half of these invading submucosa
• Factors increasing risk:
  • HGD at multiple levels
  • Macroscopically abnormal or ulcer

Prognosis / Natural Hx

• Risk of progression to malignancy
  • ≈ 0.5% per year
• Patients with high grade dysplasia
  • ≈ 50% progression to CA over 5yrs
• Follow-up controversial
  • Evidence about benefit however has not been shown to demonstrate survival benefit
  • Needs to be conducted on patients that would be fit enough to proceed to treatment if detected.

Surveillance

• See Above

Screening

• Not currently recommended
• Low absolute risk in individuals with chronic reflux
• Most individuals with Barrett’s die from other causes
• Not cost effective and benefit does not outweigh risk of endoscopy
• But in high risk patients this may be indicated - >50+, male, smoker, first degree relative, hernia