Advances in Screening for Barrett’s Esophagus and Esophageal Adenocarcinoma

Abstract

Esophageal adenocarcinoma (EAC), the primary form of esophageal cancer in the US, is a lethal cancer with exponentially increasing incidence. Screening for Barrett’s esophagus (BE), the only known precursor to EAC, followed by endoscopic surveillance to detect dysplasia and early stage EAC, and subsequent endoscopic treatment (to prevent progression of dysplasia to EAC and treat early stage EAC effectively), is recommended by several society guidelines. Sedated endoscopy (the primary current tool for BE screening), is both invasive and expensive, limiting its widespread use. In this review, we aim to provide a comprehensive review of recent innovations in the nonendoscopic detection of BE and EAC. These include swallowable cell sampling devices, combined with protein and epigenetic biomarkers (which are now guideline endorsed as alternatives to sedated endoscopy), tethered capsule endomicroscopy, emerging peripheral blood-sampled molecular biomarkers, and exhaled volatile organic compounds. We also summarize progress and challenges in assessing BE and EAC risk, which is an important complementary component of the process for the clinical implementation of these innovative non-endoscopic tools and propose a new paradigm for the strategy to reduce EAC incidence and mortality.

Introduction

In 2023, there will be an estimated 16,120 deaths from esophageal cancer and 21,560 new cases of esophageal cancer in the United States (US).¹ Esophageal adenocarcinoma (EAC), the primary form of esophageal cancer in the US, has exponentially increased in incidence over the past four decades.² Unfortunately, EAC still has a dismal 5-year survival rate of only 20%, partly due to the propensity for metastasis as the tumor is located in close proximity to lymphatic drainage in the submucosa, and obstructive symptoms (dysphagia and weight loss) develop only once the cancer has reached an advanced stage. Barrett’s esophagus (BE), the only known precursor to EAC, is characterized by a metaplastic change from the normal squamous to intestinal epithelium in the esophagus in the setting of chronic gastroesophageal reflux disease (GERD) (Figure 1). A diagnosis of BE is established when endoscopy illustrates columnar mucosa that extends ≥1 cm above the gastroesophageal junction and biopsies demonstrate specialized intestinal metaplasia, containing goblet cells. While estimates vary, the prevalence of BE in the US general population generally ranges between 0.04% (for long segment BE) and 5% (for BE of any length).^2–4 EAC develops from BE in a metaplasia-dysplasia-carcinoma pathway, progressing from non-dysplastic BE (NDBE) to BE with low-grade dysplasia (LGD) or high-grade dysplasia (HGD) and then to EAC (Figure 1). BE patients have a 3-5% risk of developing EAC over their lifetime.^{2, 5}

Figure 1.

Diagnostic features of Barrett’s esophagus: (A) Endoscopic images and histopathology of non-dysplastic Barrett’s esophagus (B) Endoscopic images and histopathology of Barrett’s esophagus with low-grade dysplasia (C) Eendoscopic images and histopathology of Barrett’s esophagus with high-grade dysplasia (D) Endoscopic images and histopathology of adenocarcinoma

Risk Factors for Barrett’s Esophagus

There are several known risk factors for BE, including chronic GERD, age >50 years, family history of BE or EAC, central obesity, male sex, white race, tobacco use, and presence of hiatal hernia.⁶ GERD is often considered the hallmark risk factor for BE as it is thought to be the main driver behind the metaplastic changes observed in BE. A systematic review and meta-analysis demonstrated that the summary odds ratio (OR) (95% confidence interval (CI)) for the association of GERD with BE was 2.90 (1.86-4.54); GERD symptoms increased odds of long-segment BE five-fold but did not show an association with short-segment BE.⁷ While GERD is often considered the primary risk factor for BE, it is important to note that 40-50% of patients with BE or EAC do not report experiencing chronic GERD symptoms.^{6, 8} This may be, in part, due to known hyposensitivity of BE patients to perceiving GERD symptoms.⁹ Older age (>50 years) is another risk factor for BE [OR (95% CI) 1.6 (1.1-2.4)] with the prevalence of BE increasing with advancing age in endoscopic studies.¹⁰ A family history of BE or EAC has been reported to be the strongest risk factor for BE in some studies, with one study demonstrating a twelve-fold increased risk.¹¹ Central obesity is another reflux independent risk factor for BE (OR 1.98, 95% CI 1.52-2.57); in general, waist circumference is a better predictor of central adiposity as compared to body mass index.¹² Men have a higher risk of BE compared to women. In a systematic review and meta-analysis, the pooled male to female sex ratio for BE was 1.96 (1.77-2.17).¹³ The risk of EAC developing in BE in women is comparable to the risk of breast cancer in men. Caucasian race is associated with nearly a four-fold higher risk of BE compared to African Americans (OR 3.62, 95% CI 2.77-4.73).¹⁴ The incidence of EAC in Caucasians is double that of Hispanics and four-fold higher than Asians.¹⁵ Cigarette smoking is also associated with an increased risk of BE (OR 1.42, 95% CI 1.15-1.76), and the risk increases with a greater number of pack-years smoked.¹⁶ The presence of a hiatal hernia, even when adjusted for GERD and body mass index, was associated with an increased risk of BE (OR 3.94, 95% CI 3.02-5.13).¹⁷

Current Screening Recommendations for Barrett’s Esophagus

Screening for BE, the only known precursor lesion of EAC, is endorsed by multiple professional societies (Table), including the American College of Gastroenterology,¹⁸ American Gastroenterological Association (AGA),¹⁹ European Society of Gastrointestinal Endoscopy,²⁰ American Society for Gastrointestinal Endoscopy,²¹ British Society of Gastroenterology,²² and the American College of Physicians.²³ These guidelines do not recommend screening the general population for BE. Instead, screening is suggested in those with chronic reflux and other risk factors, namely age >50 years, male sex, White race, central obesity, current or prior tobacco use, and first-degree relative(s) with BE or EAC. In a systematic review and meta-analysis, the prevalence of BE with GERD and the presence of any risk factor was 12.2% (95% CI 10.2-14.6%).²⁴ The risk of BE also increases with increasing number of risk factors present.²⁴ More recently, the AGA Clinical Practice Update suggested that screening may be considered in individuals with or without GERD as long as they have at least 3 total risk factors.¹⁹ This latest best practice advice underscores that GERD may not be considered an essential risk factor for BE screening.

Table.

Guidelines for Barrett’s esophagus screening by national and international organizations

Organization (Publication Year)	Screening Recommendations
American College of Gastroenterology (2022)18	A single screening endoscopy is suggested for patients with chronic GERD symptoms and 3 or more additional risk factors for BE, including male sex, age >50 years, White race, tobacco smoking, obesity, and family history of BE or EAC in a first-degree relative.
	A swallowable, non-endoscopic capsule device combined with a biomarker is an acceptable alternative to endoscopy for screening for BE.
American Gastroenterological Association (2022)19	Screening with standard upper endoscopy may be considered in individuals with at least 3 established risk factors for BE and EAC, including individuals who are male, non-Hispanic white, age >50 years, have a history of smoking, chronic GERD, obesity, or a family history of BE or EAC.
	Non-endoscopic cell-collection devices may be considered as an option to screen for BE.
European Society of Gastrointestinal Endoscopy (2020)20	Endoscopy screening may be considered only in those with long-standing GERD symptoms (i. e., > 5 years) and multiple risk factors (age ≥ 50 years, white race, male sex, obesity, first-degree relative with BE or EAC).
American Society for Gastrointestinal Endoscopy (2019)21	If screening endoscopy for BE is performed, a screening strategy is suggested that identifies an at-risk population, defined as individuals with a family history of EAC or BE (high risk) or patients with GERD plus at least 1 other risk factor (moderate risk).
British Society of Gastroenterology (2014)22	Endoscopic screening can be considered in patients with chronic GERD symptoms and multiple risk factors (at least three of age 50 years or older, white race, male sex, obesity). However, the threshold of multiple risk factors should be lowered in the presence of family history including at least one first-degree relative with Barrett’s or EAC.

^aBE, Barrett’s esophagus; EAC, esophageal adenocarcinoma; GERD, gastroesophageal reflux disease

Rationale and Challenges with Screening

The principal rationale for performing BE screening is that if the screening examination yields a diagnosis of BE, then patients can be placed in an endoscopic surveillance program to allow for timely detection of BE dysplasia or early-stage EAC amenable to endoscopic treatment, preventing EAC or enabling effective treatment of early-stage EAC.²⁵ Specifically, if a patient is found to have dysplastic BE or early-stage EAC during surveillance, endoscopic eradication therapy consisting of endoscopic resection or ablation can be utilized to reduce disease progression with excellent long-term outcomes as shown in multiple randomized controlled studies.²⁶ For intramucosal (T1a) EAC, endoscopic eradication therapy has equivalent overall survival compared to surgical management (esophagectomy) along with significantly less morbidity and no mortality.^{27, 28} More recently, the ASPECT trial demonstrated that high-dose proton pump inhibitor and aspirin, especially in combination, prolonged survival and progression to EAC or HGD, suggesting a role for chemoprevention when NDBE is identified.²⁹ BE screening has also been shown to be cost-effective in multiple modeling studies.^30–32

The gold standard for BE screening is a careful endoscopic examination using esophagogastroduodenoscopy (EGD). If the endoscopy shows features suggestive of BE, biopsies can be taken during the procedure to establish the diagnosis. Unfortunately, the current endoscopic approach to BE screening has several limitations. At present, only 10-15% or patients with risk factors for BE undergo endoscopic screening.³³ Even though 60-90% of the patients have BE present at EAC diagnosis, up to 90% of patients with EAC are diagnosed outside surveillance programs, often when alarm symptoms arise.³⁴ This represents a substantial missed opportunity for early detection.

As outlined above, almost all society guidelines place a strong emphasis on the presence of chronic GERD when selecting patients for BE screening. However, this symptom lacks sensitivity and specificity for identifying patients with BE and EAC.³⁵ Nearly half the patients with BE report absence of GERD symptoms and, therefore, may not undergo screening under current criteria.³⁶ This leads to reduced sensitivity for the detection of BE, as shown in many studies.^{35, 37} In addition, some unanswered questions remain, including the impact of endoscopic screening on EAC mortality and potential harms of endoscopic screening, given the low absolute risk of BE progression to EAC in those without dysplasia (approximately 0.12% per year).³⁸ Despite evidence of cost-effectiveness when used for BE screening (compared to no screening),³⁰ EGD is also expensive and invasive with challenging access given the limited number of trained endoscopists. Ideally, the approach to patient selection for BE screening should be individualized based on risk and probability of harboring BE and EAC, which theoretically can be calculated using risk prediction models.^39–42 However, these tools, utilizing GERD symptoms and other demographic and clinical characteristics, are not often used in clinical practice given their modest accuracy and discriminatory ability; only a few prediction tools have been externally validated. Additionally, there has been slow dissemination and adoption of these prediction models among busy primary care and gastroenterology clinical practices. Several additional challenges in estimating BE and EAC risks using prediction tools remain, including how to select the optimal risk threshold to initiate BE screening, which can vary depending on the incidence of BE/EAC in the targeted population.

More recently, there have been significant advancements in the use of minimally invasive nonendoscopic tools for BE detection, which address several limitations of the conventional endoscopic approach.

Methods

A detailed electronic literature search was conducted in the PubMed, MEDLINE, and Google Scholar databases from their inception until March 2023 to capture relevant publications on the topic of interest. The search terms included Barrett’s esophagus, Barrett’s esophagus screening, Barrett’s neoplasia, biomarkers, capsule based imaging, capsule endoscopy, dysplasia, early detection, endoscopic surveillance, esophageal adenocarcinoma, esophageal cancer, exhaled volatile organic compounds, gastroesophageal reflux, innovations, methylation biomarkers, risk stratification, sampling device, swallowable cell sampling device, tethered capsule endomicroscopy, and unsedated transnasal endoscopy. Studies were restricted to those published in English.

Advances and Innovations in Screening for Barrett’s Esophagus

I. Swallowable Cell Sampling Devices combined with Biomarkers

Non-endoscopic cell collection devices are a novel alternative option for BE screening and allow collection of cells from the entire esophagus for analysis with biomarkers. These non-endoscopic cell sampling devices include Cytosponge^™ [Medtronic, Minneapolis, MN] and EsophaCap^® [PavMed, NY, NY], which utilize string-based, capsule-enclosed polyurethane foam. The EsoCheck^™ [PavMed, NY, NY] device is an expandable and retractable silicone balloon catheter, silicon catheter. All three devices are able to collect esophageal cytology samples for analysis of BE and EAC associated biomarkers such as Trefoil Factor 3 (a protein marker) and methylated DNA markers (MDMs). While these tests are not yet widely available, they are anticipated to be less expensive compared to endoscopy, do not require sedation or anesthesia, and can be performed in the office setting by a non-physician provider (i.e., nurse) in less than 10 minutes.³⁴ Recent guidelines consider these tests acceptable alternatives to endoscopy for BE screening in those with multiple risk factors.^{18, 19} Similar to the fecal immunochemical test and multitarget stool DNA tests for colon cancer screening which require a colonoscopy when positive, these minimally invasive tests for BE would require a confirmatory EGD when positive. Additional studies to delineate the positive and negative predictive value of these tests in screening populations are ongoing.

1. Cytosponge and Trefoil Factor 3 (TFF)

The Cytosponge (Figure 2) consists of a 30 mm piece of polyurethane foam which is compressed within a gelatin capsule, attached to a surgical suture. The capsule is swallowed by the patient with sips of water and dissolves within 5-8 min of reaching the stomach, thereby releasing the compressible 30 mm spherical sponge. The expanded sponge is then slowly pulled back using the attached suture, collecting cells from the gastric cardia and esophagus during withdrawal. The sponge is placed into a buffer; samples are then centrifuged to concentrate a cell pellet, which is analyzed with immunohistochemical (IHC) staining for TFF3, a protein marker expressed in the goblet cells characteristic of BE.^43–47

Figure 2.

Swallowable esophageal cell collection device administration and withdrawal.

The Cytosponge TFF3 device has been tested in the United Kingdom and the United States in several clinical trials. The device has been shown to be safe, well-tolerated, and have good patient acceptability when used in a primary care setting.⁴⁵ In one study, its sensitivity and specificity for BE detection were 80% and 92%, respectively.⁴⁷ The sensitivity was higher at 87% for patients with longer BE segments ≥ 3 cm. A recent large multicenter randomized controlled trial by Fitzgerald et al.⁴⁶ that included 13,514 patients with chronic GERD symptoms from UK primary care practices, showed that the Cytosponge-TFF3 testing improved the rate of BE detection ten-fold when compared to standard care, in which the patients only received endoscopic evaluation as per primary care physician judgment. The Cytosponge TFF3 testing led to BE diagnosis in 140 patients, including 9 with dysplasia & 5 early-stage EAC, while the standard care group had BE diagnosis in only 13 patients, including 3 with advanced-stage EAC. This trial showed the potential of such a test in shifting the spectrum of BE diagnosed to identify those with dysplasia and early cancer, wherein endoscopic treatment could be more impactful. The most common adverse event was sore throat (51%), and only one procedure-related adverse event was reported (detachment of the sponge, requiring endoscopic removal).

2. EsophaCap and DNA Methylation Biomarkers

The EsophaCap [PavMed, NY, NY] or “sponge on a string” (SOS) utilizes the same concept as the Cytosponge but has a thicker cord and slightly smaller 25 mm sponge. The device has been used in combination with an MDM panel to detect BE from characteristic epigenetic changes, rather than goblet cells. MDMs are aberrantly methylated genes which have been extensively studied as potential markers for metaplasia and BE-related dysplasia.⁴⁸ The quantitative nature of MDMs assessment (using PCR based assays) helps eliminate the subjectivity of IHC interpretation by expert pathologists as may be the case when other BE-associated protein markers (e.g., TFF3) are used. In a pilot study, the combination of EsophaCap and 2-marker MDM panel was found to have sensitivity and specificity of 100% for BE detection.⁴⁹ In a multisite case-control study utilizing the EsophaCap with a 5-MDM panel for BE detection, the sensitivity and specificity for BE detection were 92% and 94%, respectively.⁵⁰ Additionally, 91% of the patients swallowed the EsophaCap device successfully, and 95% of the patients preferred it over endoscopy. The use of 5-MDM panel was later validated in a multicenter case-control study using training and independent test cohorts. The sensitivity and specificity of 5-MDM panel for BE diagnosis were 93% and 90%, respectively, in the training set and 93% and 93%, respectively, in the test set.⁵¹ The study also showed that the accuracy of a 3-MDM panel was similar to the 5-MDM panel. Although the 3-MDM panel will ultimately require validation, this study highlights the potential of 3-MDM panel as a BE diagnostic tool, which could help save time and reduce the total cost of testing. Other investigators have also combined MDMs with a smaller diameter (20 mm) EsophaCap device for the non-endoscopic detection of BE.⁵²

3. EsoCheck

EsoCheck [PavMed, NY, NY] is a swallowable, inflatable balloon-based esophageal sampling device. Similar to Cytosponge and EsophaCap, the device is swallowed and reaches the stomach. The balloon is then inflated with air, and slowly withdrawn, allowing the textured balloon surface to capture cells for sampling. Five centimeters above the gastroesophageal junction, the balloon is deflated and inverted to prevent potential contamination from squamous mucosa in the proximal esophagus during the capsule retrieval. In a pilot study by Markowitz et al.,⁵³ the EsoCheck device had a sensitivity and specificity of 90% and 92%, respectively, for BE detection when combined with a 2-marker DNA panel (methylated CCNA1 and VIM). A large multicenter, single-arm study on its efficacy and safety compared to standard endoscopy is currently underway.

II. Imaging-Based Screening Methods

1. Unsedated Transnasal Endoscopy (uTNE)

Unsedated transnasal endoscopy (uTNE) is performed using an ultrathin endoscope (diameter <6 mm), which is introduced through the nasal cavity without the need for intravenous sedation and can be performed in an office setting. Several studies have shown that uTNE for BE screening, is both safe and well-tolerated and offers similar diagnostic yield as EGD.^{54, 55} A randomized controlled trial by Sami, et al.⁵⁶ of 459 patients compared participation rates and clinical effectiveness of sedated EGD versus uTNE. This trial showed that uTNE had numerically higher participation rates (47% versus 40%)) and similar effectiveness compared to standard EGD with significantly shorter procedure and recovery times. Despite being acknowledged by society guidelines for several years, uTNE is currently not widely used as an alternative for BE screening given the lack of provider experience, preference and perceived poor patient tolerance.²¹

2. Capsule-Based Imaging

a. Esophageal Capsule Endoscopy (ECE)

Utilizing the same principles as the video capsule endoscopy used for suspected small bowel bleeding, a dedicated esophageal capsule (PillCam^™ ESO) [Medtronic] was developed to allow direct visualization of the esophagus. The swallowable capsule has a camera at both ends and captures images at a combined frame rate of 35 frames/second in the 3^rd generation capsule (PillCam^™ ESO3). A meta-analysis⁵⁷ examining the diagnostic accuracy of ECE for BE detection showed that pooled sensitivity and specificity of ECE for diagnosing BE were 77% and 86%, respectively. The main limitation of ECE is an inability to obtain esophageal biopsies and potential for smaller lesions to be missed due to lack of air insufflation. The movement of ECE also relies on esophageal peristalsis leading to unpredictable capsule transit time. Several innovations have been developed to address these limitations. For instance, the use of a magnetized capsule allows control of the capsule using an external magnetic field.⁵⁸ There are also efforts to develop wireless biopsy devices using a magnetized capsule and untethered microgripper, which have been studied in animal models.⁵⁹ Ultimately, further technological advancement is still required to overcome these limitations, and its use in clinical practice is still not recommended.

b. Tethered Capsule Endomicroscopy (TCE)

Tethered capsule endomicroscopy (TCE) (Figure 3) is based on principles of optical coherence tomography (OCT)⁶⁰ which is capable of capturing high-resolution, cross-sectional images using interferometric microscopic imaging technique.⁶¹ Abnormalities in surface signal intensity and gland architecture are suggestive of BE.⁶² After the tethered capsule (25 mm) is swallowed, cross-sectional microscopic OCT images are constructed in real-time during its transit through the entire esophagus and can be viewed on a monitor.

Figure 3.

(A) OCT tethered capsule endomicroscopy capsule and (B) Portable OCT imaging system (C) Exhaled volatile organic compound handheld device. (From Gut with permission).⁶⁸ OCT, optical coherence tomography.

Several preliminary studies have demonstrated the potential of this imaging modality for BE screening.^63–65 A recent multicenter prospective study⁶⁶ consisting of 147 patients with biopsyproven BE showed that TCE is safe and can be completed with a short procedure time (average imaging duration of 5.6 ± 1.9 minutes). The main limitation of TCE preventing its widespread use is the lack of high-quality correlative studies validating the correlation between abnormalities on OCT images and histopathologic samples. Overall, this technology is still under development and further studies are needed to validate this technique for clinical use.

III. Exhaled Volatile Organic Compounds

Evaluation of volatile organic compounds (VOCs) from exhaled breath samples is another novel non-invasive approach for BE screening. VOCs are gaseous end products that are generated during a cellular catabolic process. Cancerous and precancerous cells have been shown to exhibit different VOC profiles, given their significant difference in metabolic requirements compared to normal cells. Subtle changes in VOC profiles can be detected using two main currently available chemical analytical techniques, including mass spectrometric analysis and a hand held electronic nose (e-nose) device [Aeonose, Zutphen, The Netherlands],⁶⁷ which measure electronic conductivity profiles specific to VOCs and translate them into “breath prints” for disease states (Figure 3). The handheld device needs a patient to breathe in and out of the device for a total of 5 min.

A proof-of-concept case control study by Peters, et al.⁶⁸ showed that the VOC profiles of patients with BE significantly differed from non-BE patients. Utilizing the electronic nose device, this screening method was found to have sensitivity and specificity of 91% and 74%, respectively, for BE detection. Moreover, the VOC profiles of BE patients could be differentiated from patients with GERD with sensitivity and specificity of 64% and 74%, respectively. Given its minimally invasive nature, the acceptability of this technology is excellent. While this pilot study displayed great promise of exhaled VOCs for BE screening, the clinical application of VOCs in this setting needs to overcome several challenges in terms of variability caused by diet, medications, microbiome, and the lack of a standard sampling protocol. Hence further validation is still needed to assess its generalizability.

IV. Peripheral Blood-sampled Molecular Biomarkers

Circulating microRNAs (miRNAs) have recently been studied for potential use in cancer identification, including EAC.⁶⁹ Circulating miRNAs are small non-coding RNA molecules (approximately 20-25 nucleotides in length) which can be measured in serum. Aberrant expression of miRNAs is postulated to play a vital role in neoplastic progression to EAC.^70–72 Expression levels of specific miRNAs are upregulated in patients with BE and EAC and can differentiate healthy controls from BE/EAC patients.^{73, 74} One study⁷³ reported that a combination of 4 circulating miRNAs (miRNA 95-3p, 136-5p, 194-5p, and 451a) had a sensitivity and specificity of 78% and 86%, respectively, for distinguishing patients with BE from controls. A combination of 3 different circulating mRNAs (miRNA 133a-3p, 382-5p, and 451a) was able to differentiate EAC patients from controls with sensitivity of 86% and specificity of 80%. Additional prospective studies, with implementation into real clinical settings, are needed for further validation of these circulating miRNAs biomarkers.

Recent advancements in genomics techniques have also explored the possibility of ‘liquid biopsy’ using circulating tumor DNA (ctDNA) as a non-invasive diagnostic tool for EAC.⁷⁵ ctDNA is a single or double-stranded DNA released by tumor cells into the bloodstream and carries the mutation of the original tumor. Another study⁷⁶ highlighted the potential role of novel cell-free MDMs in plasma for the detection of EAC and esophageal squamous cell carcinoma (ESCC). In this study, 12 MDMs with the highest discrimination in tissue were selected and a 5-MDM panel was able to accurately detect 74% of esophageal cancer (74% of EAC and 78% of ESCC). Sensitivity for early-stage disease was modest. Additional studies are ongoing to further develop these markers.

Advances in Assessing Risk of BE and EAC

Several studies have investigated the identification of individuals at higher risk for BE and EAC by creating risk prediction models that integrate risk factors, such as male sex, age, cigarette smoking, BMI/measures of central obesity, and GERD symptoms. To date, the notable risk prediction tools include the Michigan BE pREdiction Tool [M-BERET],⁷⁷ the Gerson tool,³⁹ the Locke tool,⁷⁸ the Thrift tool,⁴⁰ the Nord-Trøndelag Health Study (HUNT) tool,⁴¹ and the Kunzmann tool.⁴² However, only two of these prediction tools have been validated in independent populations: M-BERET and Thrift tool. The M-BERET includes on GERD symptoms, age, waist-to-hip ratio, and cigarette smoking. This tool was found to have a reasonable predictive performance with area under the receiver operating characteristic curve (AUROC) of 0.72, which was significantly better than the model using GERD symptoms alone (0.72 vs 0.61, p<0.01).⁷⁹ The Thrift tool was developed based on a large population-based study in Australia to help predict the presence of BE in patients with GERD symptoms. This tool, which relies on age, sex, smoking status, BMI, education, and frequency of use of acid suppressant medications, performed moderately well with AUROC 0.61 in the external validation dataset.⁴⁰ A study by Rubenstein, et al.⁸⁰ prospectively validated and compared these tools for predicting the presence of BE and demonstrated that these models were significantly better in identifying individuals with BE than GERD frequency and duration alone, although with modest accuracy, with AUROC ranging between 0.66-0.70. When applied to a cohort of patients filling out symptom questionnaires and providing anthropometric measurements in the 1960s, the Kunzman score was able to predict incident EAC and gastroesophageal junction adenocarcinoma (GEJAC) with reasonable accuracy, particularly in patients with the highest quartile score.

Despite these data, several of these scores require the use of multiple variables (some of which are not routinely available in the medical record) to be manually inputted into an online score generator by the provider, rendering this approach less than optimal.⁸¹ Additionally, the optimal risk threshold to offer screening remains unknown and may require the use of modeling studies as empiric studies to address this issue may be impractical to conduct. It is conceivable that the threshold scores to offer screening with minimally invasive devices may be lower than the threshold required to recommend invasive endoscopy. It is also important to note that findings from these studies are limited to a very specific population (White non-Hispanic individuals) which limits the generalizability of these models.

Next Steps in early detection of Esophageal adenocarcinoma and its precursors

With the emergence of minimally invasive, office-based non-endoscopic tools over the past decade, significant advancements have been made in BE screening. While promising, additional research is necessary to fill gaps in our knowledge related to these tools before their widespread adoption. A conceptual flow diagram summarizing an approach to risk assessment for BE/EAC, early detection using non-endoscopic technologies, and management as guided by histology is presented in Figure 4. This hypothetical model helps visualize how this streamlined process can improve BE/EAC detection rate and ultimately reduce EAC-related mortality. It will still be essential to ascertain how often false negatives and false positives occur with the use of these tests in screening populations and better understand their impact on detection rates of EAC and EAC-related mortality. The degree of acceptance of these modalities by primary care providers in the office setting and patients is another consideration. Furthermore, additional data are needed comparing outcomes using these strategies compared to conventional management. In a modeling study which compared the cost effectiveness of these minimally invasive methods and endoscopy, capsule sponge-biomarker combination approaches were deemed to be most cost effective compared to no screening, or screening by endoscopy.³⁰ While BE is not uncommon, progression to EAC in BE patients is a far from universal event. Determining a better risk stratification approach to identify patients most likely to progress to EAC and prioritize their follow up and management is critical. For instance, the TissueCypher assay [Castle Bioscience, Inc, Pittsburgh, PA], utilizing a multiplexed fluorescence imaging platform that analyzes multiple morphometric features from the tissue specimen, has been developed to predict the risk of progression to HGD and EAC by classifying patients into low, intermediate and high risk of progression over 5 years.⁸² A multicenter study from the United States showed that a high-risk test class is a strong independent predictor of neoplastic progression to HGD/EAC (OR 6.0, 95% CI 1.2-7.2) in patients with BE.⁸³ With ongoing research, we will be able to better learn where these tools fit in clinical practice, but they appear to have the potential to improve risk stratification when used in carefully selected patients with and ultimately improve clinical outcomes in patients with BE and EAC.

Figure 4:

Hypothetical model summarizing the emerging paradigm of risk assessment for Barrett’s esophagus/esophageal adenocarcinoma, early detection of Barrett’s esophagus using non-endoscopic technologies and pathways of subsequent management (dysplasia detection, risk stratification and endoscopic therapy). BE, Barrett’s esophagus; EAC, esophageal adenocarcinoma; EET, endoscopic eradication therapy; EGD; esophagogastroduodenoscopy; EHR, electronic health record; HD-WLE, high definition white light endoscopy; HGD, high grade dysplasia; LGD, low grade dysplasia; NDBE, non dysplasitc Barrett’s esophagus; TCE, tethered capsule endomicroscopy; VOC, volatile organic compound.

Conclusion

Over the past few decades, there has been a plethora of advancements related to BE and EAC diagnosis and management. Despite these successes, the overall incidence of EAC continues to rise, and survival rates remain dismal. Multiple national and international societies suggest consideration of BE screening in patients with multiple risk factors. While endoscopy has long been considered the primary tool for BE screening, recent advances and innovation have made available minimally invasive, non-endoscopic, and cost-effective screening tools, such as swallowable cell sampling devices combined with biomarkers. Other tests such as tethered capsule endomicroscopy, exhaled volatile organic compounds, and liquid biopsy molecular biomarkers are in development. While promising, these technologies are in various stages of development and have varying levels of evidence supporting their use. In coordination with progress in other dimensions such as BE risk assessment, detection of dysplasia and risk stratification, these novel tools can pave the way for a reduction in the incidence and mortality of a lethal malignancy.

Abbreviations

BE

Barrett’s esophagus

EAC

Esophageal adenocarcinoma

ECE

Esophageal Capsule Endoscopy

GERD

Gastroesophageal reflux disease

HGD

High-grade dysplasia

LGD

Low-grade dysplasia

MDMs

Methylated DNA markers

NDBE

Non-dysplastic BE

TCE

Tethered capsule endomicroscopy

TTF

Trefoil Factor 3

uTNE

Unsedated Transnasal Endoscopy

VOCs

Volatile organic compounds