Endoscopic Interventions in Barrett’s Esophagus: Do the Dollars Make Sense?

Advances in the development of endoscopic therapies for Barrett’s esophagus have resulted in the emergence of an important paradigm shift for management of early neoplasia and represent an opportunity to alter the natural history of the disease. Clinical incorporation of these endoscopic modalities may have significant implications for disease management and health care delivery from a cost perspective. This article reviews the current literature on the cost analyses of commonly used Barrett endoscopic interventions and summarizes the overall cost-effectiveness of these treatments as compared with surveillance or surgery.

Advances in the development of endoscopic therapies for Barrett’s esophagus (BE) have resulted in the emergence of a variety of treatment options for this condition, particularly regarding management of early neoplasia. The published results of these modalities have demonstrated promising efficacy and safety. Long-term clinical data have been collected on some techniques but are still awaited on others. These surgery-sparing options have dramatically changed the treatment paradigm of BE and have potentially important implications for disease management and health care delivery from a cost perspective.

The major endoscopic treatment modalities used today include photodynamic therapy (PDT), endoscopic mucosal resection (EMR), multipolar electrocoagulation (MPEC), argon plasma coagulation, radiofrequency ablation (RFA), and cryotherapy. This article reviews the current literature on the cost analyses of these commonly used Barrett endoscopic interventions and summarizes the overall cost-effectiveness of these treatments as compared with surveillance or surgery. Understanding the issues that contribute to the economics of BE is critical for health care professionals, affected patients and their families, and those whose responsibilities require decision-making based on costs imposed on either patients or health care organizations.

Natural history of Barrett’s esophagus and cost implications

The natural history of BE must be factored into cost evaluations of its various endoscopic interventions. Based on a simulation model confirmed by data from the US Surveillance, Epidemiology and End Results (SEER) cancer registry, an estimated prevalence for BE in the general population of 5.6% (5.49%–5.70%) accurately reflects reported incidence rates for esophageal adenocarcinoma (EAC). The overall prevalence of BE in the setting of chronic reflux is as high as 10%. Moreover, the annual risk of developing EAC in the setting of BE is 0.5% to 1.0%. These general statistics influence clinical management decisions for varying degrees of BE neoplasia from endoscopic screening to surveillance to therapies. Although it is generally believed that EAC develops in the setting of BE through a progressive dysplastic sequence, the data regarding rates of cancer progression for varying degrees of dysplasia are inherently flawed due to issues, such as spontaneous regression, sampling error, interobserver histologic assessment variability, and overestimation of effect in observational studies. Despite these limitations, the presence and extent of dysplasia within the BE segment provides the ability to risk-stratify patients based on its severity, which has implications on deciding whether or not to subject a patient to a medical intervention.

Screening for BE and surveillance for dysplasia and cancer remain controversial issues. Current guidelines advise that the highest yields for diagnosing the condition are in patients ages greater than 50 years, patients who are white, and patients who have long-standing heartburn. In addition, despite debates about cost-benefit profiles and cost-effectiveness, society guidelines continue to support the use of surveillance programs for BE patients both with and without dysplasia at varying frequency intervals.

Even the mere diagnosis of BE without dysplasia can have profound effects on patients’ life insurance premiums due to insurers’ concerns about abbreviated life expectancy of policy holders due to an increased risk of developing EAC. In a survey study of 20 US-based life insurance companies (10 in southern California and 10 in North Carolina), investigators found that, for a base case, an otherwise healthy 43-year-old man with no BE, the yearly preferred life insurance averaged $1255. For the same-aged individual with BE as a pre-existing condition, the mean cost of policies offered was $2731 ( P <.001). In the case of a 36-year-old woman with no documented medical conditions, the rate ranged from $472 to $551. With the addition of a BE diagnosis, the mean rate rose by 177%, to $1434, with a range of $1144 to $1896. Thus, endoscopic treatments aimed at eradication of BE must also study the issues surrounding overall risk reduction for EAC and recurrence of BE in the postablation esophagus. This is paramount, because these factors have bearing on health care and insurance costs, and potential for patient labeling in society due to the rationale that mere eradication of BE may only temporarily eliminate the BE disease state and may not confer long-term protection against disease-related complications.

Economics of Barrett’s esophagus endoscopic therapies

The majority of cost-effectiveness analyses pertaining to BE have focused on screening or surveillance programs. With the emergence of endoscopic therapies demonstrating acceptable safety and efficacy profiles, recent data have emerged comparing costs related to these modalities to either surveillance or surgery alternatives. Initial cost studies focused on endoscopic ablation with PDT, which was the first nonsurgical modality demonstrated to have durable and efficacious results.

To delineate the best management strategy for high-grade dysplasia (HGD), researchers have used decision analytic modeling to systematically assess various clinical options. In one study, the four main arms included no preventative strategy, elective esophagectomy, endoscopic ablation (PDT in this study), and surveillance endoscopy for a base case healthy 50-year-old white man with an initial diagnosis of BE with HGD. The most effective strategy was endoscopic ablation, yielding 15.5 discounted quality-adjusted life years (QALYs) compared with 15.0 for surveillance endoscopy and 14.9 for surgery. The most inexpensive option, as expected, was having no preventative strategy. This choice had an average cost per QALY of $54. It resulted, however, in high rates of cancer. Endoscopic surveillance was less expensive and more effective than esophagectomy. Additionally, through a phenomenon called extended dominance, although the total costs of ablation were greater than surveillance, an additional life year was less expensive to purchase through ablation than surveillance. The change in the ICER moving from no therapy to ablation was a reasonable $25,621 per QALY. Through sensitivity analyses, researchers found that when yearly rates of progression from HGD to cancer were greater than 30%, esophagectomy became the most cost-effective strategy. These findings helped set the stage for the financial justifications for ablative technology use in the setting of BE with HGD.

Another group also studied the most cost-effective strategy to manage BE HGD and any residual disease present after endoscopic treatment. Four strategies were evaluated: esophagectomy, endoscopic surveillance, PDT followed by esophagectomy for residual HGD, and PDT followed by endoscopic surveillance for residual HGD. They found that esophagectomy cost $24,045, with life expectancy of 11.82 QALY. PDT followed by endoscopic surveillance for residual HGD was the most effective strategy, with a life expectancy of 12.31 QALY. It also incurred the greatest lifetime cost ($47,310), however, resulting in an incremental cost-effectiveness of $47,410 per QALY.

As the efficacy of various ablative technologies began to gain recognition during the first decade of the twenty-first century, a landmark cost-utility analysis was published to help guide future cost assessment studies. In a decision analytic model created to examine a BE patient population with a mean age of 50 years, separate subanalyses were conducted for patients with no dysplasia, low-grade dysplasia (LGD), or HGD. The management strategies compared were endoscopic surveillance, endoscopic surveillance with ablation for incident dysplasia, immediate ablation followed by endoscopic surveillance in all patients or limited to patients in whom metaplasia persisted, and esophagectomy. The ablation modalities included in the model were RFA, APC, MPEC, and PDT.

The primary outcome of the study was measurement of the incremental cost per QALY between the analyzed management strategies, also known as the incremental cost-effectiveness ratio (ICER). This ratio is defined as the cost difference when moving from a less expensive but less effective strategy, to a more expensive but more effective strategy, divided by the change in QALYs between these two strategies. The study showed that endoscopic ablation for HGD could increase life expectancy of patients by 3 quality-adjusted years at an incremental cost of less than $6000 compared with no intervention. In the base case analysis of 50-year-old patients with BE and HGD, endoscopic ablation with PDT, RFA, or APC with postablation endoscopic surveillance was calculated to extend life by 3.2 QALYs compared with performing no surveillance. MPEC did not have any data reported for use in BE with HGD. APC and RFA demonstrated cost increases of $19,000 to $20,000 per person, and PDT costs were $30,000 greater than no surveillance. Comparing RFA with performing no surveillance demonstrated an acceptable ICER of $5839 per QALY gained. The ICER calculated with PDT, however, required an overwhelming $32 million per QALY gained.

Patients with LGD or no dysplasia could also be optimally managed with ablation, but it was expensive to continue surveillance after eradication of metaplasia. Ablation was preferred to surveillance if ablation permanently eradicated greater than 28% of LGD or 40% of nondysplastic metaplasia. The investigators concluded that endoscopic ablation could be the preferred strategy for managing patients with BE and HGD. The unknown long-term effectiveness of ablation for BE, however, with either LGD or no dysplasia influenced the overall cost-effectiveness of this approach for these disease states in an unfavorable manner.

With respect to management of stage 0 or 1 esophageal cancer (adenocarcinoma or squamous cell cancer), a systematic review of the SEER database has shown the long-term survival of endoscopic resection with or without ablation using PDT or thermal therapy is comparable with that of surgery. Concerning BE-related early cancer (T1 with mucosal or minimal submucosal infiltration), a decision tree model compared EMR of the cancerous lesion with RFA of the remaining Barrett segment to esophagectomy. During the 5-year interval of the study, endoscopic therapy cost $17,000. and yielded 4.88 QALYs, compared with $28,000 and 4.59, respectively, for esophagectomy. The overall outcome was not changed by varying the recurrence rates of cancer or BE metaplasia after endoscopic therapy. Even under the most optimal circumstances favoring esophagectomy, such as 2% operative mortality rate, no reduced quality of life (QOL) after esophagectomy, and a low 5-year survival rate after recurrence associated with endoscopic ablation, the risk of positive lymph nodes still needed to exceed 25% before esophagectomy became the preferred treatment option. The threshold for risk of lymph node invasion in the setting of submucosal invasion for which it is significant enough to warrant the greater invasiveness and cost involved with esophagectomy over endoscopic resection remains to be clearly defined.

The extent of reduction of EAC after BE ablation therapy also deserves more scrutiny, because this may have bearing on long-term and overall costs. In a pooled analysis of natural history of BE and ablation of BE articles, the rates of cancer in patients undergoing ablation and from the natural history were calculated. Ablation was associated with a reduction in cancer incidence with the greatest benefit observed in the BE HGD subgroup, although these findings may have been limited by the heterogenous nature of the included studies.

Increasing attention has been given to endoscopic ablation of LGD and nondysplastic BE due to concerns about potential neoplastic progression. A recent study reported that approximately 85% of 147 patients who carried the diagnosis of LGD before expert pathology review actually had no dysplasia. For patients who had a consensus diagnosis of LGD, the cumulative risk of progression to HGD or cancer was 85% in 109.1 months as compared with 4.6% in 107.4 months for patients down-staged to nondysplastic BE ( P <.0001) on the basis of expert review. Thus, BE with LGD is an overdiagnosed entity with an underestimated neoplastic potential. In addition to the presence of LGD, the extent of LGD is a significant risk factor for the development of esophageal adenocarcinoma. Even though the presence of HGD is associated with a significantly greater relative risk for development of cancer, the extent of HGD is not an independent risk factor for progression to cancer.

With these issues in mind, endoscopic ablation has been proposed for LGD and nondysplastic BE. In a Markov model, three competing strategies were evaluated in a hypothetical 50-year-old cohort with nondysplastic BE, allowing for the natural history to be modeled for various health and disease states associated with a different set of utilities ( Fig. 1 ). These three strategies were (1) no surveillance, (2) surveillance, and (3) endoscopic ablation. The model was biased against ablation with a conservative estimate of complete response and continued standard surveillance even after complete ablation. All potential complications were considered, and an incomplete histologic response after ablation was presumed to have the same neoplastic risk of progression as a case of untreated BE. Patients in the surveillance arm who demonstrated nondysplastic BE would undergo repeat endoscopy every 3 years. Those who demonstrated LGD would have surveillance endoscopy performed every 1 year until no dysplasia was detected. Those with focal HGD would have repeat EGD and biopsy by endoscopic surveillance every 3 months. All patients with persistent, diffuse or multifocal HGD and cancers were considered for surgical resection. Endoscopic surveillance was allowed up to the age of 80 years. The model was constructed to allow for misdiagnosis of histologic specimens, based on available published false-positive and false-negative rates. The ablative strategy yielded the highest QALYs and was more cost-effective than surveillance. The incremental cost to gain an extra QALY with the strategy of ablation was $48,626 compared with the endoscopic surveillance strategy. By current standard of cost-effective medical interventions, a strategy with an incremental cost of $50,000 or less per QALY gained is deemed acceptable in terms of society’s willingness to pay (WTP). Endoscopic ablation was the dominant of the three strategies with respect to average net health benefits (NHBs) when the WTP exceeded $60,000 ( Fig. 2 ).