Esophageal cancer (EC) represents a globally prevalent malignant neoplasm, ranking as the seventh leading cause of cancer-related mortality worldwide (1). Esophagectomy stands as a pivotal treatment for patients presenting with early-stage or locally advanced esophageal cancer (LAEC). However, surgery alone often results in postoperative recurrence or metastasis for LAEC patients, with rates approaching 50% (2). Recent advancements in neoadjuvant therapy have substantially enhanced prognostic outcomes for LAEC patients (3, 4). Especially following the introduction of immunotherapy, Neoadjuvant Immunochemotherapy (NICT) for LAEC patients has demonstrated distinct therapeutic advantages. According to research, NICT not only achieves comparable pathological complete response (pCR) rates to neoadjuvant chemoradiotherapy (NCRT) (5, 6) but also can mitigate surgical complexities (7, 8). Despite these advancements, the treatment paradigm of esophagectomy following NICT for LAEC still encounters challenges. Recent investigations indicate that the 1-year failure rates remain at 20% for LAEC patients, with local recurrence constituting over 60% of treatment failure patterns (9–11). Consequently, accurately predicting and identifying patients with a high risk of recurrence following surgery, and subsequently administering proactive postoperative interventions, are the urgent problems to be solved.

However, the research on the application of postoperative adjuvant therapy (PAT) after NICT in LAEC patients is limited. Only a few retrospective studies have explored the feasibility of PAT following NICT in LAEC patients (12–15). Some results have shown that PAT may enhance postoperative survival among patients (12, 13, 15) In particular, Feng et al. (13) identified ypT₊N₊ patients as a specific subgroup benefiting significantly from PAT, with an increase in the 3-year progression-free survival (PFS) rate by 14.6%. Given the current insufficient evidence, our study aims to identify LAEC patients who may benefit from PAT through prognostic nomogram-based risk stratification, thereby providing an evidence-based guidance for postoperative management.

In recent years, the application of NICT in LAEC patients has attracted increasing attention. Some clinical studies have confirmed that NICT demonstrates acceptable safety and promising postoperative pathological response rates (7, 8). However, approximately 20% of patients treated with NICT still experience failure within 1 year after surgery (9–11), and the postoperative Local recurrence (LR)rate can be as high as 64.9% (9). Our results were consistent with those reported in previous studies. The 1-year recurrence and/or metastasis rate was 17.2%. In addition, the pattern of failure in our study also observed a high rate of LR, accounting for 67.7% (63/93). The patients who experienced recurrence and/or metastasis after surgery will have a poorer postoperative survival rate (17). Prior research suggested that PAT following NCRT may enhance postoperative survival in patients with a high risk of recurrence, especially among those who did not achieve pCR (non-pCR) (18). However, Xie et al. (14) observed that the non-pCR patients derived no significant benefit from PAT following NICT. We hypothesize that this discrepancy may be attributed to the distinct biological mechanisms of immunotherapy, such as immune modulation and long-lasting immune memory effects (19). Meanwhile, studies have shown that pCR evaluation after NICT is affected by many factors (20, 21). Consequently, relying on postoperative pathology alone to identify patients who may benefit from PAT may not be sufficient. It is essential to establish a multidimensional predictive nomogram for the comprehensive screening of high-risk patients.

Our study revealed that tumor length, tumor thickness reduction rate, surgical method, the number of LNs dissected, and ypN-stage were independent prognostic factors affecting PFS. Notably, tumor length≥5cm has been found to have a notable correlation with the poorer prognosis of patients, which was consistent with the biological characteristics of esophageal squamous cell carcinoma cells (22). We consider that tumor length can directly reflect the longitudinal invasion range of the primary lesion, and indicate an increased risk of LN metastasis when it exceeds a certain length (23, 24). Previous research has demonstrated a close correlation between the dynamic imaging index of tumor thickness reduction rate and the effectiveness of neoadjuvant therapy^- (16, 25, 26). In our study, we identified 26% as the optimal cutoff value using the ROC curve, which is marginally higher than the 22% suggested in the NCT cohort by Matsumoto et al (25). We speculate that this difference may be attributed to variations in treatment intensity between NICT and NCT for primary tumors. Furthermore, it was discovered that patients with a high LN metastasis burden (ypN_2-3) after surgery are significantly associated with poor PFS, which aligns with prior research findings (14, 27). The inadequate tumor response rate and the high LN metastasis burden following NICT elevate the risk of postoperative recurrence and/or metastasis. This may be attributed to the low sensitivity of cancer cells in both the primary tumor and metastatic LNs to neoadjuvant therapy, failing to achieve significant tumor downstaging. Consequently, Wang et al. (28) proposed that preoperative chemoradiotherapy and targeted therapy could be administered to these patients to attain optimal clinical downstaging, ease surgical complexity, and enhance the survival rates of patients.

For surgical quality control in patients with LA-ESCC, our study showed that the choice of open surgery (Ivor-Lewis/Sweet esophagectomy) and the number of LNs dissected less than 24 were unfavorable factors affecting PFS. Previous studies (10, 29) have indicated that LN metastasis in LA-ESCC patients treated with NICT predominantly occurs in the upper mediastinal LNs, specifically in the mediastinal 2R and 4R stations. In clinical practice, compared with open esophagectomy, MIE-McKeown esophagectomy can achieve complete resection of upper mediastinal LNs, particularly those of the right recurrent laryngeal nerve chain (30). In addition, MIE-McKeown esophagectomy can also decrease the risk of postoperative pulmonary infection and other associated complications (31), and minimize the damage of surgical trauma on the patient’s immune system (32). For these reasons, MIE-McKeown esophagectomy is currently a widely used procedure for LAEC patients (33). For the number of LNs dissected, our results were consistent with those reported by Wu et al. (34), which recommended a minimum of 23 LNs to be dissected. Previous studies by Wang et al. (22) demonstrated that LAEC patients are at a high risk of LN metastasis. These metastatic LNs could potentially facilitate distant metastasis by inducing tumor-specific immune tolerance (35), potentially leading to an unfavorable prognosis for patients. Consequently, even after NICT, reducing LN dissection may heighten the risk of postoperative recurrence or metastasis for LA-ESCC patients.

In this study, a prognosis nomogram incorporating the aforementioned five predictors was developed to forecast the PFS of LA-ESCC patients treated with NICT. The C-index of the nomogram was 0.776, while the AUC value exceeded 0.80. Following internal validation, the model demonstrated robust predictive performance and clinical applicability. According to the nomogram, patients were divided into three risk groups (low, medium, and high), and there were significant differences in PFS among the groups (P<0.001). Utilizing nomogram risk stratification, we discovered that PAT only exhibited a notable trend of benefit for high-risk patients. Upon further analysis of the risk factors present in the high-risk patient group, it was discovered that all patients in this group exhibited the trait of positive lymph nodes (ypN₊). Additionally, 93% of these patients showed a tumor thickness reduction rate of ≤ 26%. This finding aligns with the benefit group characterized by ypT₊N₊ proposed by Feng et al. (13). Therefore, we consider that ypN₊ patients exhibiting a poor response to neoadjuvant therapy may constitute the subgroup that derives the most significant clinical benefit from PAT. Previous studies have demonstrated that the presence of tumor cells in lymph nodes not only indicates the metastatic potential of the primary tumor but also enables these cells to colonize and disseminate to distant organs (36). Tumor cells can modify the microenvironment of lymph nodes, leading to an increased accumulation of immune-tolerant or immunosuppressive cell populations within metastatic lymph nodes, which in turn facilitates immune evasion (37–39). Therefore, patients who demonstrate a high nodal metastatic burden following NICT are considered to harbor tumor cells with enhanced metastatic potential and diminished sensitivity to NICT. Despite undergoing surgery, minimal residual disease may persist. Therefore, we speculate that in the high-risk populations, intensified PAT may represent a critical strategy for controlling tumor progression and extending PFS.

At present, the evidence on the selection of PAT regimen after NICT for LA-ESCC patients is extremely limited, and a few studies (13–15) mention that PAT regimens involve immunotherapy combined with chemotherapy, chemotherapy, immunotherapy, and radiotherapy-based combination therapy. Interestingly, some studies found that the subgroup of patients who received postoperative adjuvant chemotherapy alone experienced the poorest outcome. The authors considered that chemotherapy may further impede an already compromised immune system’s capacity to effectively recognize and target cancer cells, ultimately resulting in tumor recurrence (14). Furthermore, given the primary failure mode of local recurrence following NICT, the potential significance of postoperative adjuvant radiotherapy deserves attention. Previous studies indicate that radiotherapy demonstrates a high rate of local control in both neoadjuvant therapy (29) and postoperative adjuvant therapy (40) for esophageal cancer. Consequently, incorporating postoperative adjuvant radiotherapy could potentially enhance the local control rate for LA-ESCC patients who have a high recurrence risk after surgery. Nevertheless, due to the limited number of cases involving postoperative adjuvant radiotherapy in this study (13 cases), further evidence is required.

Our study still has some limitations: 1) This was a retrospective study, which inherently introduces some degree of bias; 2)The fact that it was conducted at a single center without external validation, future research should involve multi-center collaboration to expand the sample size to validate the results; 3) A limited data set for postoperative adjuvant immunotherapy and radiotherapy, lacking further stratification of the treatment protocol; 4) A relatively small number of high-risk cases, making it essential to collect more data or conduct prospective studies for future verification.

The prognosis nomogram can effectively predict the PFS of LA-ESCC patients after NICT. Through survival risk stratification, patients in the high-risk group may benefit from PAT.