Cracking the code of early esophageal cancer detection

Esophageal cancer has long posed a significant challenge for clinicians due to its tendency for late-stage diagnosis. By the time symptoms appear and the esophageal lumen becomes obstructed, curative treatment options are often no longer possible. Several screening programs, particularly in Asia, have been introduced to address this issue. However, they come with inherent limitations and risks. In recent years, non-invasive detection methods, particularly liquid biopsies, have generated growing interest. The potential to detect early-stage cancer has been a longstanding goal in oncology. This context highlights the importance of the study by Li et al. (1). The authors propose a novel detection method that enhances the analysis of cell-free DNA (cfDNA) by incorporating multiple genomic features, which have already shown promise in other cancers (2,3).

The authors performed whole-methylome sequencing (WMS) on plasma cfDNA and analysed four key genomic features: global methylation density (MD), fragmentation patterns (FSI), copy number alterations in chromosome arms (CAFF), and methylation and fragmentation in accessible regulatory elements (ATAC). The first method focused on global MD, a crucial process for regulating gene expression. In cancer, abnormal methylation patterns are frequently observed, with global hypomethylation leading to the activation of oncogenes and hypermethylation causing the silencing of tumor suppressor genes. However, while methylation-based detection can be effective, it faces limitations, particularly due to the low quantity of circulating DNA and tumor heterogeneity, which can result in a higher rate of false negatives.

The second method addressed the analysis of cfDNA fragmentation patterns, specifically focusing on fragment sizes. Conventional methylation detection techniques can damage DNA, making them unsuitable for concurrent analysis of fragmentation. To overcome this, the authors employed a refined method of methylation analysis that preserves the integrity of the DNA, allowing for additional analyses. They divided the genome into 100-kilobase sections and measured the ratio of short fragments to long fragments, referred to as FSI. This approach allows for a more comprehensive analysis of cfDNA, capturing both methylation and fragmentation information, which is critical for improving detection sensitivity.

The third method employed by the authors aimed at detecting chromosomal instability by analysing copy number alterations using the CAFF technique. Chromosomal instability, which manifests as gains or losses of specific chromosome segments, is a well-established hallmark of cancer. The CAFF method assesses these variations by examining cfDNA fragment coverage across chromosome arms, allowing for the identification of copy number imbalances.

Finally, the authors explored the methylation and fragmentation patterns in ATAC. These regulatory elements are crucial for controlling gene expression, as they permit the binding of proteins that regulate gene activity. In cancer, the accessibility of these regions is often disrupted, leading to abnormal gene regulation.

By profiling both methylation and fragmentation in these accessible regions, ATAC enhances the understanding of how these regulatory disruptions may serve as potential biomarkers for early cancer detection.

The study aimed to evaluate this non-invasive approach for the early detection of esophageal squamous cell carcinoma (ESCC) through blood-based analysis. Plasma samples were obtained from 168 patients with ESCC and 251 healthy, age- and sex-matched individuals. WMS was employed to examine methylation, fragmentation patterns, and chromosomal alterations in cfDNA. The cohort was divided into a training set and a test set to validate the method’s effectiveness. By applying machine learning, the researchers trained a model to identify cancer-specific DNA features, enhancing its detection accuracy within the training cohort before assessing its performance on the test cohort.

The study’s results demonstrated widespread hypomethylation in ESCC samples, with methylation loss increasing with disease progression. The MD model achieved an area under the curve (AUC) of 0.931 [95% confidence interval (CI): 0.895–0.966] in the training cohort and 0.923 (95% CI: 0.886–0.961) in the test cohort. FSI also correlated with disease stage, with AUCs of 0.913 (95% CI: 0.866–0.959) in training and 0.899 (95% CI: 0.856–0.943) in the test cohort. CAFF identified consistent abnormalities in chromosome arms 3p, 3q, and 8q in ESCC, with AUCs of 0.850 (95% CI: 0.790–0.909) in training and 0.854 (95% CI: 0.798–0.910) in testing. Additionally, ATAC highlighted accessibility and methylation changes in squamous cell-specific regulatory regions, with AUCs of 0.933 (95% CI: 0.901–0.964) in training and 0.912 (95% CI: 0.873–0.950) in testing.

By combining the four features, the AUC, sensitivity, and specificity metrics were calculated for the entire cohort, which included both early and late-stage ESCC cases. While the model showed impressive overall performance, with a specificity of 98.4% across all stages, the sensitivity differed by disease stage. For early-stage ESCC (stages 0-II), sensitivity was lower, at 70.0% in training and 69.0% in testing, whereas for late-stage ESCC, the sensitivity was notably higher, reaching 86.0% and 84.3% in training and testing, respectively.

This article presents a significant advancement in non-invasive ESCC detection through comprehensive cfDNA analysis. By integrating DNA methylation, fragmentation patterns, and chromosomal alterations, the researchers developed a highly specific early detection model. In previous studies, the sensitivity and specificity of similar approaches have been reported around 87% and 95%, respectively (4). With its specificity of 98.4%, the method developed by Li et al. significantly reduces the likelihood of false positives, making it more reliable for population screening. This high specificity ensures that those without the disease are accurately excluded, thus minimizing unnecessary anxiety and further testing, a crucial factor when screening for a rare condition like ESCC.

However, while the study shows promise, it has limitations. It focuses exclusively on ESCC, a predominant type in East Asia, and does not address esophageal adenocarcinoma (EAC), more common in Western populations. Given the differences in biology between ESCC and EAC, this method’s applicability to adenocarcinoma remains to be tested. Additionally, sensitivity for early-stage ESCC, although improved, is still suboptimal for effective early detection. Further studies with more diverse populations, including patients with adenocarcinoma and benign esophageal conditions, are necessary to validate the method’s broader use. Given that this study was conducted in a single-center cohort, replication in independent, multi-center cohorts with a wider spectrum of patients is essential to confirm the robustness and generalizability of these findings. Additionally, the smoking status of patients—a well-established factor influencing DNA methylation in the esophagus—was not explicitly discussed in the original manuscript. Future research should consider stratifying results based on smoking history to assess its potential confounding effect. The test may perform best in populations with a high prevalence of ESCC. However, for high-risk individuals, conventional methods like esophagogastroduodenoscopy might still offer a more practical and cost-effective solution. Combining this cfDNA-based approach with tissue analysis for recurrence surveillance could enhance monitoring of ESCC patients. By integrating liquid biopsy with direct tumor tissue profiling, clinicians could track both the tumor’s molecular characteristics and the presence of residual or recurrent cancer. This would allow for earlier detection of relapse and more tailored treatment strategies.

This study represents a significant step forward in non-invasive cancer detection, offering a cost-effective and robust method for early ESCC detection. The next step for this approach would likely involve expanding its use to larger, more diverse populations and evaluating its effectiveness across different cancer types, particularly EAC. This method could be particularly useful for post-operative surveillance, offering a non-invasive way to monitor recurrence and guide personalized treatments.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Esophagus. The article has undergone external peer review.

Peer Review File: Available at https://aoe.amegroups.com/article/view/10.21037/aoe-24-39/prf

Funding: None.

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://aoe.amegroups.com/article/view/10.21037/aoe-24-39/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Li Y, Liu B, Zhou X, et al. Genome-Scale Multimodal Analysis of Cell-Free DNA Whole-Methylome Sequencing for Noninvasive Esophageal Cancer Detection. JCO Precis Oncol 2024;8:e2400111. [Crossref] [PubMed]
2. Foda ZH, Annapragada AV, Boyapati K, et al. Detecting Liver Cancer Using Cell-Free DNA Fragmentomes. Cancer Discov 2023;13:616-31. [PubMed]
3. Mathios D, Johansen JS, Cristiano S, et al. Detection and characterization of lung cancer using cell-free DNA fragmentomes. Nat Commun 2021;12:5060. [PubMed]
4. Liu J, Dai L, Wang Q, et al. Multimodal analysis of cfDNA methylomes for early detecting esophageal squamous cell carcinoma and precancerous lesions. Nat Commun 2024;15:3700. [Crossref] [PubMed]