Ablation for Patients With Barrett or Dysplasia

Which Patients With Barrett Should Have Ablation?

For a clinician to consider performing a significant intervention on a patient, one must be secure the diagnosis is BE and when dysplastic changes are reported, that this is a true reflection of a potential neoplastic process. Chapter 31 has addressed the issues related to the definition of Barrett intestinal metaplasia that, when present, will lead to endoscopic surveillance. If there is a diagnosis of dysplasia at endoscopy, the management implications change for the patient and the clinician. However, pathologic consensus of the histologic diagnosis can be a problem, and the dysplastic changes may not be present on subsequent biopsies, particularly LGD. The diagnosis of HGD, confirmed by a second pathologist with gastrointestinal (GI) expertise, carries a clear risk for progression to adenocarcinoma such that all these patients should be considered for definitive treatment of the neoplastic lesion as well as ablation of the whole BE segment. The alternative is an esophagectomy. The presence of visible lesions (nodules or ulceration) in a segment of HGD carries significant implications. If nodules are present, there is a 2.6 times potential for progression to EAC, and if ulceration is present, the risk of the presence of EAC in a high-grade dysplastic segment has been reported to be 80%, compared with 52% if there was no ulceration. Endoscopic resection of these abnormalities, if possible, offers better pathologic staging and complete resection and should be performed before attempts at BE ablation. Patients with a diagnosis of IMC that has been endoscopically completely removed will need ablation of the residual BE segment.

Patients with confirmed LGD in BE will require more regular endoscopic surveillance, and there is evidence to support considering this group for complete BE ablation. However, the diagnosis of LGD needs to be secure. There has been a meta-analysis of the outcomes from endoscopic therapies for LGD, which assessed 37 studies comprising 521 patients. The multiple techniques of ablation used provided complete eradication (CE) of dysplasia in 88.9% and intestinal metaplasia (IM) in 67.8%, with a pooled incidence of progression to cancer of 3.9 (95% CI, 1.27 to 9.1) per 1000 patient-years. The authors concluded that there was likely to be histologic overdiagnosis of LGD. RFA was the most safe and effective option in this group, but ablation did not eliminate the risk of progression to HGD or EAC.

The potential for “overdiagnosis” of LGD and the consequences have been highlighted by a number of studies. In a group of patients diagnosed with LGD by community pathologists, where subsequent review by expert GI pathologists occurred, the diagnosis was downgraded in 85% of patients to NDBE or indefinite dysplasia. In the “downgraded group” the risk of progression to HGD or EAC was 0.49% per patient per year, compared with the group confirmed to have LGD where the progression was 13.4%. In a Dutch study, the percentage of diagnoses downgraded from LGD to indeterminate or NDBE was 73%, where expert pathologists reviewed the histology of 293 referred patients. At a median follow-up of 39 months, 21 of the 75 (27%) confirmed LGD patients progressed to HGD/EAC. The rate of progression was measured at 9.1% per patient-year compared with the pathologically reviewed NDBE group who had a rate of progression of 0.6 per patient-year.

The multicenter European SURF trial randomized 136 patients with LGD to receive RFA ablation to the BE or surveillance, with both groups receiving proton pump inhibitor (PPI) therapy. There was CE of the dysplasia in 93% of the RFA group and 28% of the controls and CE of the IM in 88% of the RFA group compared with zero in the controls. At 3-year follow-up, the rate of progression to HGD was 1.5% after RFA compared with 26.5% in the surveillance group, and the rate of progression to EAC was 1.5% compared with 8.8%, respectively. They reported a neoplasia progression rate of 8% per year. The rate of progression of LGD confirmed by an expert pathologist was further assessed in a cohort study of 170 patients, with 45 undergoing RFA ablation and 125 surveillance with median follow-up of more than 2 years. The rate of progression to HGD or EAC in the RFA group was 0.77% compared with 6.6% in the surveillance group. The measured risk of progression after RFA had a hazard ratio of 0.08 (95% CI, 0.01 to 0.61). In the surveillance group, following a multivariate analysis, independent factors associated with the risk of progression were nodularity in the BE and multifocal dysplasia.

For LGD to be considered confirmed, guidelines recommend a second pathologist with GI expertise review the biopsies, and for a second endoscopy and biopsy after 6 months to reassess the BE and confirm the continued presence of the LGD. In this group of patients with confirmed LGD, the data for a higher risk for progression to HGD and IMC are clear. Using this definition for LGD, guidelines in the United States recommend endoscopic ablation to be appropriate for patients with confirmed LGD. In the UK the published guidelines in 2014 were updated in 2016 when reviewed evidence supported the role of RFA to ablate confirmed LGD. The factors that have been associated with higher rates of progression of LGD to HGD/EAC include male gender, NDBE present for more than 10 years, length of BE (>3 cm), persistent esophagitis, multifocal dysplasia, and the presence of nodules in the BE mucosa. In a consensus statement from the BOBCAT group, it was agreed that there was “moderate quality evidence” to support the ablation of the high-risk LGD group. The criteria for considering ablating this group were consensus of the diagnosis by two expert pathologists, persistence of the LGD over time, multifocal dysplasia, and longer BE segments.

Issues Relating to Nondysplastic Barrett Esophagus and Ablation

With the improvement in the outcomes of the techniques for ablative therapy, the pendulum has swung to consider the role of this therapy for patients with NDBE. Reasons stated include anxiety to the potential for malignant progression that may not be identified from the random sampling, performed at the surveillance endoscopy. The patient perspective was highlighted in a study assessing patient preference in the management of NDBE where the options, hypothetically, were put for either chemoprevention with aspirin and 3 to 5 yearly surveillance endoscopies or endoscopic ablation. The patients preferred ablation.

For ablation to be a realistic option in asymptomatic patients with NDBE, the treatment must be safe, be effective, have durable long-term results, and be cost effective. The safest, most effective treatment is RFA. In a cohort study of patients with NDBE who had RFA and regular follow-up with treatment of residual or recurrent IM, the complete regression rate was 70% at 1 year and 92% at 5 years. This group of patients required multiple endoscopies and the need for more intense surveillance than recommended for nontreated NDBE. The impact of this approach on the patients' longevity and quality of life along with the cost effectiveness is yet to be clearly defined.

With respect to cost effectiveness of BE ablation, in 2004, a study examined the management of HGD comparing endoscopic surveillance, endoablation (using PDT), and esophagectomy. Endoablation with PDT was shown to be the most effective strategy. A recent literature review concluded that endoscopic therapy for dysplastic BE, using PDT or RFA, was cost effective compared with esophagectomy. One single-institution cohort study reported the cost of PDT to be 5 times that of RFA. Using a Markov model, assessing patients with HGD, RFA with continued surveillance is more cost effective than endoscopic surveillance and esophagectomy when a cancer develops.

For patients with LGD, if confirmed by two pathologists and the diagnosis remains stable on repeated endoscopies when assessing quality-adjusted life years (QALYs), RFA is more cost effective than continued regular surveillance and performing the RFA when HGD develops. Although the American Society of Gastroenterology (ASGE) has recommended ablation is an “option” for NDBE, the evidence suggests this is not cost effective, due to the low rate of progression to adenocarcinoma in this group of patients.

At this time, for NDBE there is no evidence that supports the routine use of ablative therapies. In the future, for ablation to be considered, there will need to be a number of factors better defined. They may include selection of patients with longer life expectancy than average BE patients; identifying patients with higher risk for progression than the average patient without dysplasia, most likely through a panel of biomarkers; identifying which patients will do well with RFA and which group has a higher risk of failing; and finally creating a situation whereby minimal surveillance is required, as well as a reduced need to “touch up” the ablated segments.

Ablative Techniques: Methods, Outcomes, and Complications

The aims of ablation are to eliminate all the intestinal metaplasia/dysplasia and provide a uniform depth of treatment to allow squamous reepithelialization with minimal residual IM. Assessment of the outcomes of these procedures relates to the efficacy of the eradication of the dysplastic and NDBE, as well as the durability with respect to the long-term eradication of the at-risk mucosa, along with the rate of adverse effects from the procedure.