Esophageal cancer (EC) is an aggressive malignancy and one of the leading causes of cancer-related mortality globally. According to the American Cancer Society, EC accounts for 1% of all cancers diagnosed in the United States (1). The primary histologies for EC include esophageal squamous cell carcinoma (ESCC) and esophageal Adenocarcinoma (EAC), and others. ESCC is typically more common in Asia and parts of Africa, whereas EAC is rising rapidly in Western countries due to lifestyle and metabolic factors (2, 3). Despite advancements in early detection and treatment modalities, the overall prognosis remains poor, especially when diagnosed at an advanced stage. Currently, the mainstay of treatment for EC is multimodal therapy combining chemotherapy, radiotherapy, surgery, and immunotherapy.

For patients with resectable diseases, Neoadjuvant Chemoradiation (NCR) followed by surgery has become the standard of care. Alternatively, Definitive Chemoradiation (DCR) is the treatment of choice for patients who cannot tolerate surgery or have tumors at the proximal or Cervical esophagus, where post-surgical morbidity and mortality is higher (3).

Our study provides contemporary real-world data on ESCC using the National Cancer Database (NCDB), focusing on the comparative survival outcomes of NCR followed by surgery versus DCR alone. We aim to confirm the existing trial-based evidence and explore treatment outcomes based on tumor location, sex, and age. Our findings reinforce the survival benefit of NCR combined with surgical resection in appropriately selected patients, particularly those with locally advanced thoracic esophageal squamous cell carcinoma, to contribute to the ongoing discourse on personalized, multimodal treatment strategies.

This retrospective analysis accessed information from the National Cancer Database (NCDB) between 2004 and 2020 for patients with esophageal cancer. The NCDB is a database that compiles data from over 1,500 Commission on Cancer-accredited facilities across the United States. It contains de-identified, HIPAA-compliant patient-level data and is accessible to investigators from accredited programs. Access to such data is only available via an application process and Participant User Data file (PUF) agreement. The study was conducted after obtaining approval by the Cleveland Clinic’s Institutional Review Board (IRB) as Exempt Human Subject Research (IRB #22-153). The waiver of informed consent to participate and publish the data in an online open-access publication was granted in the application due to minimal risk research involving human subjects. This study was conducted per the regulations of the ethical review committee and the guidelines of the journal.

To our knowledge, this retrospective cohort analysis is among the largest, matched, real-world study focusing exclusively on ESCC, demonstrating that neoadjuvant chemoradiation followed by surgery confers a survival advantage over definitive chemoradiation, with outcomes further stratified by tumor location. This work confirms and reaffirms the results of existing clinical trials and clinical practices in the management of ESCC.

Neoadjuvant chemoradiation is established standard of care for the management of ESCC (4, 5). Randomized trials such as CROSS and NEOCRTEC5010, along with multiple cohort studies have demonstrated superiority in survival outcomes for patients who receive NCR plus surgery in patients with resectable lower and midthoracic locally advanced ESCC (6–11).The Cross trial reported long-term overall survival rate, improved R0 resection (R0 = clear negative margins), and a pathological complete response (pCR) (6). This survival advantage could be driven by several mechanisms. Neoadjuvant chemoradiation serves as a backbone, allowing for downstaging and reduction of the disease burden. In the CROSS trial, patients who received NCR plus surgery had a significantly higher R0 resection rate (92% vs. 69%, respectively). Tumor downstaging and size reduction are well-established predictors of long-term overall survival (12, 13) Adding surgical intervention would provide a synergistic effect, reducing local recurrence and mitigating the relatively high risk of ESCC locoregional spread through submucosal invasion (14). Furthermore, NCR provides a means to eradicate micro-metastatic disease, reducing early recurrences and improving disease-free survival (15).

NCR and DCR employ similar chemoradiotherapy backbone, with surgery serving as the key differentiating factor. In their meta-analysis, Kamarajah et al. compared NCR followed by surgery to DCR in locally advanced ESCC, reporting improved overall survival with the NCR group (16). The regimen consists of a combination of platinum and fluoropyrimidine (Cisplatin + 5-FU, or variants such as Cisplatin + Capecitabine). Less frequently, it can include docetaxel. Stahl et al., in their meta-analysis, compared radiotherapy dosages between NCR and DCR. NCR utilizes a lower dose of radiotherapy, ranging from 40 to 50.4 Gy, whereas DCR typically uses higher dosages, ranging from 50 to 66 Gy. Although DCR regimens typically use higher radiation doses, evidence of a corresponding survival benefit remains inconsistent, and surgery may contribute meaningfully to long-term disease control (17).

Tumor location is essential in determining the optimal treatment modality. Surgery provided a survival advantage for patients with thoracic esophageal tumors (C15.3), with a median overall survival of 35.15 months compared to 19.38 months (p-value = 0.0070). However, this advantage was not observed in proximal tumors (C15.0). The analysis suggested a survival advantage for patients managed without surgery, HR 0.68 (p-value = 0.5884). The findings must be interpreted with caution given the very small sample size (n=19). These results are expected since tumors located more proximally carry a greater surgical risk. NCCN guidelines and Waters, Reznik, and Deboever’s study highlighted that DCR is the standard of care for cervical and high thoracic ESCC, as surgery would pose a higher risk of morbidity and mortality due to the proximity of vital anatomical structures (7, 18, 19).

Furthermore, our analysis found that female sex was independently associated with a survival advantage consisted with prior reports (20, 21). This may be attributed to hormonal, behavioral, and biological factors (20). Importantly, Kalff et al. in their population-based study explain this interesting observation, attributing it to differences in tumor biology and a usual fewer comorbidity compared to males. Notably, they also highlight the existence of unconscious bias in treatment decisions, whereby under-treatment could mask the true female survival advantage (21).

Moreover, literature exists comparing chemotherapy stand alone vs. Chemoradiation therapy. In the NeoRes-I clinical trial, it compared two neoadjuvant modalities. While CRT was associated with a higher pathological complete response (pCR) rate, no significant differences were observed in overall survival (OS) or progression-free survival (PFS) between the two groups. Notably, the CRT group experienced a higher rate of postoperative mortality (22). This could be due to radiotherapy toxicity, inflammation, and other complications. Wang et al. demonstrated in their review paper how radiotherapy could lead to lymphopenia and adversely affect outcomes. They suggested strategies to mitigate radiation-induced lymphopenia, such as using different techniques and the possibility of using proton beam therapy (23). Similarly, the JCOG1109 (NExT) trial demonstrated that although chemoradiation achieved higher pCR, the triplet chemotherapy regimen, Cisplatin + 5-Fluorouracil (CF) + Docetaxel (D) (NeoCF+D), provided a significantly better 3-year overall survival in locally advanced ESCC patients, particularly in fit populations (24). In contrast, the FFCD 9102 and FFCD 9901 trials showed no OS benefit of surgery over DCR in early-stage disease, NCR did not improve survival in these early-stage cohorts, reinforcing the role of DCR in this specific setting (25, 26). The differences in results may be attributed to variations in tumor histology and disease stage. Our study concentrated on Esophageal squamous cell carcinoma (ESCC), whereas the FFCD trials included mixed histologies. Additionally, we had a cohort consisting of patients with locally advanced thoracic ESCC, which might benefit from a multimodal treatment approach. In contrast, the FFCD trials enrolled patients with early-stage disease. Furthermore, Stahl et al. emphasized that definitive chemoradiotherapy remains a curative option in ESCC, with local control being the primary benefit of adding surgery (27).

It is important to highlight the rapidly evolving role of immunotherapy in the management of esophageal cancer, particularly its integration into the adjuvant setting. Multiple clinical trials have been conducted, and others are ongoing to improve outcomes in patients with esophageal cancer. The CheckMate 577 trial found that adjuvant Nivolumab significantly improves DFS in patients with residual disease after NCR followed by surgery (28). Further analysis has shown that patients with ypN+ disease (pathologically node-positive status after surgery) and PD-L1 combined positive score (CPS ≥5) experienced the most significant benefit. At the same time, outcomes were less favorable in ypN0 (no residual nodal disease) or PD-L1 low subgroups (29). In addition, Panda and Schumacher reinforced the crucial role of lymphadenectomy and accurate staging within the evolving Immuno-Oncology era (30). Multiple trials, such as MATTERHORN, KEYNOTE-585, and ATTRACTION-5, are investigating the benefits of checkpoint inhibitors in the neoadjuvant or perioperative setting (31–33). More studies are shifting towards precision oncology, focusing on minimal residual disease (ctDNA) and incorporating biomarkers (34).

Our work is one of the largest real-world NCDB analyses that compares NCR vs. DCR exclusively in ESCC, providing stratification by tumor localization and sex. This data use propensity score matching and multivariate Cox regression to balance existing confounders. However, we acknowledge the existence of certain limitations inherent in the retrospective design, including potential selection bias, unmeasured confounders, and immortal time bias are possible. NCDB does not provide other metrics of survival, such as performance status, disease progression, recurrence status, and details regarding chemotherapy regimens. Although peri-operative mortality are available, our endpoint was long term overall survival of the two treatment modalities. In the subgroup analysis of C15.0 Cervical ESCC, the cohort is small (N = 19) and should be interpreted with caution. Additionally, NCDB includes de-identified data from community and academic centers with varying levels of surgical expertise and radiation protocols, which may impact generalizability. Despite these limitations, our data is consistent with the results of clinical trials and practices.

Our findings reaffirm that NCR and surgery, when feasible, offer survival benefits for patients with ESCC, especially for locally advanced tumors in the thoracic esophagus. NCR should remain central in the multidisciplinary management of this aggressive malignancy.