Preoperative ctDNA Detects Occult Nodal Metastasis and Predicts Recurrence in Early-Stage Esophageal Squamous Cell Carcinoma

Highlights

Preoperative circulating tumor DNA (ctDNA) detection is a powerful predictor of pathologic nodal upstaging in clinical stage I (T1b) and T2N0 esophageal squamous cell carcinoma (ESCC).

In patients with T2N0 disease, ctDNA positivity demonstrated a positive predictive value (PPV) for occult nodal metastasis ranging from 88.89% to 100%, significantly outperforming traditional guideline-based risk factors.

The presence of preoperative ctDNA was associated with a four-fold increase in the risk of recurrence and death, independent of conventional pathologic staging.

Incorporating ctDNA into clinical risk models improved the area under the receiver operating characteristic curve (AUC) for predicting nodal metastasis from approximately 0.66 to 0.91.

The Clinical Challenge of T2N0 Esophageal Cancer

Esophageal squamous cell carcinoma (ESCC) remains a highly aggressive malignancy with a propensity for early lymphatic spread. For patients presenting with clinical stage II (T2N0) disease, the decision to proceed with neoadjuvant chemoradiotherapy (nCRT) versus upfront surgery is a subject of ongoing debate. Current guidelines, such as those from the National Comprehensive Cancer Network (NCCN), suggest nCRT for T2N0 patients with high-risk features, including tumor size ≥3 cm, lymphovascular invasion (LVI), or poor histological differentiation. However, these criteria are often subjective or difficult to assess accurately via preoperative biopsy and imaging (CT or PET-CT), which have limited sensitivity for detecting microscopic nodal involvement.

Consequently, many patients staged as cN0 undergo upfront surgery only to be pathologically upstaged to pN+, while others may receive intensive neoadjuvant therapy unnecessarily. There is an urgent clinical need for objective, molecular biomarkers that can more accurately identify patients at high risk of occult metastasis and systemic recurrence at the time of diagnosis.

Study Design and Methodology

In a study published in JAMA Surgery, Hong and colleagues evaluated the utility of tumor-informed ctDNA sequencing for preoperative risk stratification in ESCC. This cohort study utilized data from two independent institutions in Seoul, Korea: the Samsung Medical Center (SMC; n = 50) and Yonsei University Severance Hospital (YUSH; n = 24). The study population specifically targeted patients with clinical stage I (T1b) or T2N0 ESCC who underwent radical esophagectomy and lymph node dissection without receiving neoadjuvant therapy.

The researchers employed a tumor-informed approach, where whole-exome sequencing of the primary tumor tissue was performed to identify patient-specific somatic mutations. Subsequently, personalized digital PCR or targeted sequencing assays were used to detect these mutations in preoperative plasma samples. The primary outcomes measured were pathologic nodal upstaging and survival metrics, including recurrence-free survival (RFS) and overall survival (OS).

Key Findings: ctDNA and Nodal Upstaging

The detection of preoperative ctDNA was significantly more frequent in T2N0 disease than in T1b disease, reflecting the correlation between tumor burden and DNA shedding. In the SMC cohort, 54.0% of patients were ctDNA-positive, while 37.5% were positive in the YUSH cohort.

The most striking finding was the association between ctDNA and occult nodal metastasis. Among patients with T2N0 disease, ctDNA positivity was an exceptionally strong predictor of nodal upstaging. In the SMC cohort, the PPV for nodal metastasis was 100% (95% CI, 71.51-100). In the YUSH cohort, the PPV was 88.89% (95% CI, 51.75-99.72). Multivariable analysis revealed that preoperative ctDNA detection was associated with a nearly 20-fold increase in the odds of nodal metastasis (OR, 19.98; 95% CI, 3.90-211.42; P < .001).

Furthermore, when ctDNA status was added to traditional risk models (which include LVI, differentiation, and tumor size), the predictive accuracy improved dramatically. The AUC rose from 0.66 to 0.91 in the SMC cohort and from 0.67 to 0.89 in the YUSH cohort, suggesting that liquid biopsy provides critical biological information that imaging and histopathology cannot capture.

Prognostic Significance: Survival Outcomes

The presence of ctDNA before surgery also served as a harbinger of poor long-term outcomes. Over a median follow-up of 37.7 months, patients who were ctDNA-positive experienced significantly worse survival compared to those who were ctDNA-negative:

Recurrence-Free Survival (RFS)

ctDNA-positive patients had a Hazard Ratio (HR) of 4.15 (95% CI, 1.54-11.22; P = .005) for recurrence.

Overall Survival (OS)

ctDNA-positive patients had a Hazard Ratio (HR) of 4.02 (95% CI, 1.50-10.74; P = .006) for death.

These findings suggest that preoperative ctDNA detection is not merely a marker of local nodal spread but also a surrogate for subclinical micrometastatic disease, which drives systemic recurrence even after radical surgical resection.

Mechanistic Insights and Clinical Implications

The biological plausibility of these findings rests on the understanding that ctDNA shedding into the bloodstream is a function of both tumor volume and invasive potential. In ESCC, the presence of detectable ctDNA in early-stage disease likely indicates that the tumor has already breached vascular or lymphatic barriers, even if these changes are not yet visible on cross-sectional imaging.

From a clinical perspective, these results suggest a paradigm shift in the management of T2N0 ESCC. Patients who are ctDNA-positive preoperatively should likely be prioritized for neoadjuvant therapy escalation, as their risk of occult nodal disease is nearly certain. Conversely, for ctDNA-negative patients, the risk of nodal metastasis is significantly lower, potentially allowing for a more streamlined surgical approach or de-escalated adjuvant surveillance.

Expert Commentary and Limitations

While the study provides robust evidence for the use of ctDNA, several considerations remain. The tumor-informed sequencing approach, while highly sensitive, requires access to tumor tissue and a longer turnaround time compared to tumor-agnostic assays. Additionally, the study was conducted in a high-prevalence region for ESCC in South Korea; therefore, the generalizability to Western populations, where esophageal adenocarcinoma is more prevalent, requires further validation.

The relatively small sample size, particularly in the validation cohort, necessitates larger, prospective interventional trials. Such trials should determine whether changing treatment based on ctDNA status (e.g., giving nCRT to ctDNA-positive T2N0 patients who would otherwise have gone to surgery) actually improves overall survival.

Conclusion

Preoperative ctDNA detection represents a significant advancement in the risk stratification of clinical N0 ESCC. By accurately identifying occult nodal metastasis and predicting recurrence risk, ctDNA can serve as a precision medicine tool to guide neoadjuvant treatment decisions. As liquid biopsy technology continues to mature, its integration into standard preoperative workups may soon become essential for optimizing outcomes in esophageal cancer surgery.

References

Hong TH, Jeong JG, Park SY, et al. Preoperative Circulating Tumor DNA Detection and Risk Stratification in Esophageal Squamous Cell Carcinoma. JAMA Surg. 2026; doi:10.1001/jamasurg.2025.6755.