Introduction

Esophageal squamous cell carcinoma (ESCC) represents a significant global health challenge, particularly in regions where risk factors such as tobacco use, alcohol consumption, and dietary habits are prevalent1,2. ESCC primarily arises from the esophageal lining and often presents at advanced stages because of subtle early symptoms, leading to poor prognosis and limited treatment options3,4. Despite substantial advancements in surgical techniques, radiotherapy, and systemic chemotherapy, survival outcomes for ESCC remain suboptimal, especially for unresectable or metastatic cases5,6. Furthermore, patients with ESCC frequently face high rates of recurrence even after aggressive multimodal treatments, which resulted in miserable prognosis7. Thus, how to extend recurrence free survival is crucial in these patients. In the past, doublet cytotoxic chemotherapy was used to reduce recurrence. However, several clinical trials had demonstrated that there were no additional survival benefits after doublet cytotoxic chemotherapy. In the contrary, the toxicities associated with intensive chemotherapy not only diminish the patient’s quality of life but also limit the feasibility of these approaches in many cases8,9.

In recent years, maintenance treatment with metronomic chemotherapy has garnered attention as a promising approach for managing cancers, including head and neck cancers and breast cancers10. This strategy involves administering chemotherapeutic agents at low, minimally toxic doses continuously, aiming to inhibit tumor angiogenesis, modulate immune responses, and prevent tumor regrowth11. Among the agents studied for metronomic chemotherapy, Tegafur-Uracil (UFUR), an oral prodrug of 5-fluorouracil (5-FU), has shown particular promise due to its favorable toxicity profile after sustaining drug exposure12,13. UFUR treatment works by providing continuous low-dose exposure to 5-FU, which is effective in targeting rapidly dividing tumor cells with minimal systemic toxicity14. This approach is especially advantageous for elderly or frail patients who cannot tolerate high-dose conventional chemoradiotherapy15. UFUR’s oral administration also offers enhanced convenience and adherence compared to intravenous chemotherapeutics16. However, little was known regarding the maintenance role of UFUR in patients with ESCC. Hence, this study aims to investigate the prognostic impact of UFUR maintenance on survival in patients with ESCC following definitive concurrent chemoradiotherapy (dCCRT).

Materials and methods

Study design and patient selection

This multicenter retrospective cohort study was conducted to evaluate the impact of UFUR maintenance on survival outcomes in patients with ESCC following dCCRT. Patients diagnosed with histologically confirmed esophageal squamous cell carcinoma in 2020–2023 were retrospectively reviewed. Inclusion criteria were: (1) patients who were treated with dCCRT as the primary treatment, (2) patients who had disease control after dCCRT, including complete response (CR), partial response (PR), stable disease (SD), and (3) patients who were unfit for surgery or refused surgery after dCCRT. Patients with incomplete medical records, progressive disease after dCCRT, metastatic disease at diagnosis, or significant comorbidities contraindicating systemic therapy were excluded. Patients were stratified into two groups based on whether they received UFUR maintenance or not after dCCRT. Baseline demographic, clinical, and pathological data were extracted from electronic medical records. Variables included age, gender, body mass index (BMI), tumor (T) stage, nodal (N) status, overall stage (based on the American Joint Committee on Cancer 8th edition), and histological subtype. dCCRT response was assessed with computed tomogram every 3 months and based on RECIST (Response Evaluation Criteria in Solid Tumors) version 1.1 which categorized as complete response (CR), partial response (PR), or stable disease (SD). This study was conducted following the ethical principles of the Declaration of Helsinki and was approved by the Institutional Review Board of E-Da Hospital (EMRP-114-014), the Institutional Review Board of Chang Gung Memorial Hospital (20250305B0), the Institutional Review Board of China Medical University Hospital (CMUH114-REC2-078), and the Institutional Review Board of Tri-service General Hospital (B202405200). Due to the retrospective design of the study, the requirement for informed consent was waived by the Institutional Review Board of E-Da Hospital. All patient data were anonymized prior to analysis to ensure confidentiality.

Treatment protocol

All patients received dCCRT comprising external beam radiotherapy with a total dose of 50–66 Gy delivered in fractions of 2 Gy, along with concurrent chemotherapy with platinum 80 mg/m² on day 1 plus 5-fluorouracil 800 mg/m² for four days on day 1 to day 4 every 4 weeks. Chemotherapy was given up to 6 cycles or tumor progression or intolerance. Following completion of CCRT, patients in the UFUR (+) group were prescribed UFUR maintenance with 400 mg per day until disease progression, unacceptable toxicity, or patient refusal. The UFUR (-) group did not receive additional systemic therapy after CCRT. To diminish selection bias, patients who did not received standard CCRT protocol or standard UFUR dosage were excluded from our study.

Outcomes

The primary endpoints were overall survival (OS) and progression-free survival (PFS), defined as the time from the completion of dCCRT to death from any cause or disease progression, respectively. Secondary endpoints included recurrence rate, local recurrence-free survival (LRFS), and distant metastasis-free survival (DMFS). Recurrence was classified as locoregional or distant based on imaging and pathological confirmation. Subgroup analyses were conducted to evaluate the effect of UFUR maintenance based on disease stage (Stage I–II vs. Stage III) and response to CCRT (CR/PR vs. SD). Interaction terms were included in the Cox model to assess the consistency of UFUR’s effects across subgroups.

Statistical analysis

Baseline characteristics between the UFUR (+) and UFUR (-) groups were compared using chi-square tests for categorical variables and independent t-tests for continuous variables. Kaplan-Meier survival curves were generated for OS, PFS, LRFS, and DMFS, with differences between the groups assessed using the log-rank test. Multivariate analysis was performed using Cox proportional hazards regression to identify independent predictors of survival, adjusting for gender, age, BMI, T stage, N stage, disease stage, CCRT response, and UFUR maintenance. Hazard ratios (HRs) and 95% confidence intervals (CIs) were reported. All statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). A p-value < 0.05 was considered statistically significant.

Results

Clinical characteristics

A total of 198 patients diagnosed with histologically confirmed esophageal squamous cell carcinoma in 2020–2023 were included in the analysis with 63 patients in UFUR (+) group and 135 in UFUR (-) group. The baseline clinical characteristics of esophageal cancer patients who underwent dCCRT are summarized in Table 1. Stratification by UFUR maintenance, patients in both groups revealed no significant differences in demographic, pathological, or tumor staging variables, ensuring a balanced comparison between the groups. Gender distribution was similar, with 89% male in the UFUR (+) group compared to 84% in the UFUR(−) group (p = 0.432). The majority of patients in both groups were over 60 years old, accounting for 75% in UFUR (+) versus 68% in UFUR (−) (p = 0.259). Although more patients in the UFUR (+) group had a BMI > 23.5 than those in the UFUR (-) group (47% vs. 33%), this trend did not reach statistical significance (p = 0.129). The initial tumor (T) stage and lymph node (N) stage were also comparable, with 40% of UFUR (+) and 36% of UFUR (−) patients presenting with T1-T2 tumors and 63% of UFUR (+) and 67% of UFUR (−) patients categorized as N0-N1 (p = 0.513 and p = 0.667, respectively). Similarly, the overall cancer stage showed no significant difference, with 63% of UFUR (+) and 65% of UFUR (−) patients having Stage III disease (p = 0.767). Squamous cell carcinoma was the predominant histological subtype in both groups (97% in UFUR (+) vs. 96% in UFUR (−), (p = 0.885). As for treatment, cycles and dose intensity of dCCRT were insignificant between UFUR (+) and UFUR (-). Objective response rate to dCCRT was slightly higher in the UFUR(−) group than in the UFUR (+) group (92% vs. 84%), but this difference was not statistically significant (p = 0.215).

Table 1 Clinical characteristics of 198 patients with esophageal cancer post definitive chemotherapy, stratified by UFUR maintenance or not.
Full size table

Oncologic outcomes

The oncologic outcomes of 198 esophageal cancer patients, stratified by UFUR maintenance, were summarized in Table 2, which highlighted the significant benefits of UFUR in reducing recurrence and improving survival. The median duration of UFUR were 10.5 months. All patients with UFUR(+) group received standard dosage of UFUR (400 mg per day) without dose adjustment. At the end of this study, 55% of patients in UFUR (+) discontinued UFUR maintenance because of tumor recurrence, while 45% of our patients in UFUR (+) continued UFUR maintenance. Patients in the UFUR (+) group demonstrated superior clinical outcomes compared to those in the UFUR (−) group. In terms of survival, patients in the UFUR (+) group experienced significantly improved survival outcomes. Kaplan-Meier survival analyses of survival were plotted in Fig. 1. The median PFS in the UFUR (+) group was 28.3 months, which was nearly twice the PFS of 15.2 months observed in the UFUR (−) group (p = 0.014) (Fig. 1A). Similarly, the median OS was significantly prolonged in the UFUR (+) group with 37.3 months compared to 18.6 months in the UFUR (−) group (p = 0.002) (Fig. 1B). In terms of recurrence pattern, the median LRFS and DMFS were 26.6 m versus 13.4 m (p = 0.002) and 33.3 m versus 19.0 m (p = 0.010) for UFUR (+) and UFUR (-) group, respectively (Fig. 1C and D). These findings underline the efficacy of UFUR maintenance in enhancing both disease control and overall survival. The recurrence rate was significantly lower in the UFUR (+) group, where 55% of patients experienced recurrence, compared to 74% in the UFUR (−) group (p = 0.009). This reduction was primarily due to a marked decrease in distant metastasis rates, with 44% of UFUR (+) patients developing distant metastases compared to 63% in the UFUR (−) group (p < 0.001). In contrast, the lifelong local recurrence rate was similar between the groups, accounting for 11% in each group (p = 1.000), indicating that UFUR maintenance primarily impacts systemic disease progression rather than locoregional recurrence. As for treatment related adverse events (TRAE), no patients had grade 3–5 TRAE and no patients had dose interruption or reduction during treatment. There were 65% of our patient with UFUR (+) had grade 1–2 TRAE. The most common grad 1–2 TRAE were stomatitis (32%), followed by fatigue (25%), anorexia (14%), anemia (11%), leukopenia (8%), nausea (6%) and vomiting (3%).

Table 2 Oncologic outcomes of 198 patients with esophageal cancer, stratified by UFUR maintenance.
Full size table
Fig. 1
figure 1

Kaplan-Meier survival analysis of 198 patients with esophageal cancer stratified by UFUR maintenance. (A) Progression-free survival, (B) Overall survival, (C) Local recurrence-free survival, (D) Distant metastasis-free survival.

Full size image

Subgroup analysis evaluated the stage-specific impact of UFUR maintenance on overall survival (OS). Kaplan-Meier survival analyses of stage-specific survival were plotted in Fig. 2, stratified by UFUR (+) and UFUR (-). Among patients with Stage I–II disease (Fig. 2A), median PFS of the UFUR (+) group was not reached (NR) at the end of our study, while median PFS of the UFUR (−) group was 20.5 months (p = 0.009). Similarly, for Stage III patients (Fig. 2B), the UFUR (+) group exhibited a significantly prolonged median PFS as compared with the UFUR (-) group, accounting for 20.8 months versus 12.0 months (p = 0.027). Furthermore, among patients with Stage I–II disease (Fig. 2C), median OS of the UFUR (+) group was not reached (NR) at the end of our study, while median OS of the UFUR (−) group was 32.7 months (p = 0.006). For Stage III patients (Fig. 2D), the UFUR (+) group exhibited a significantly prolonged median OS as compared with the UFUR (-) group, accounting for 32.2 months versus 18.0 months (p = 0.017). The impact of UFUR maintenance was also investigated with responses of previous dCCRT. The Kaplan-Meier survival analysis was plotted in Fig. 3. In the CR + PR group, UFUR (+) patients had a significantly longer median OS of 38.5 months compared to 17.8 months in the UFUR (−) group (p = 0.002, Fig. 3A), while in the SD group, UFUR(+) patients achieved a significantly better median OS of 10.5 months versus 3.0 months in the UFUR (−) group (p < 0.001, Fig. 3B). Moreover, in the CR + PR group, UFUR (+) patients had a significantly longer median OS of 45.9 months compared to 22.0 months in the UFUR (−) group (p = 0.003, Fig. 3C), while in the SD group, UFUR(+) patients achieved a significantly better median OS of 20.6 months versus 7.1 months in the UFUR (−) group (p < 0.001, Fig. 3D).

Fig. 2
figure 2

Kaplan-Meier survival analysis of 198 patients with esophageal cancer stratified by disease stage and UFUR maintenance. (A) Progression-free surival of patients with stage I–II disease, (B) Progression-free surival of patients with stage III disease, (C) Overall surival of patients with stage I–II disease, (D) Overall surival of patients with stage III disease.

Full size image
Fig. 3
figure 3

Kaplan-Meier survival analysis of 198 patients with esophageal cancer stratified by UFUR maintenance and response of concurrent chemoradiotherapy. (A) Progression-free survival of patients achieving complete or partial response (CR + PR), (B) Progression-free survival of patients with stable disease (SD), (C) Overall survival of patients achieving complete or partial response (CR + PR), (D) Overall survival of patients with stable disease (SD).

Full size image

Multivariate analysis

The multivariate analysis of PFS and OS in patients with ESCC was presented in Table 3. UFUR (tegafur/uracil) maintenance emerged as a significant predictor of improved survival, alongside tumor stage, nodal status, and response to CCRT. Patients receiving UFUR maintenance had significantly better outcomes compared to those who did not. UFUR maintenance was associated with a 44% reduction in the risk of progression (hazard ratio [HR]: 0.56, 95% confidence interval [CI]: 0.37–0.86, p = 0.007) and a 50% reduction in the risk of death (HR: 0.50, 95% CI: 0.32–0.78, p = 0.002). Overall cancer stage was also an independent factor associated with survival. Patients with stage I-II had significant better outcomes than patients with stage III, in terms of RFS (HR: 0.62, 95% CI: 0.41–0.92, p = 0.022) and OS(HR: 0.65, 95% CI: 0.42–0.99, p = 0.046). The response to CCRT was a critical determinant of survival. Patients achieving complete or partial response (CR + PR) had a substantially lower risk of recurrence (HR: 0.24, 95% CI: 0.15–0.40, p < 0.001) and mortality (HR: 0.43, 95% CI: 0.26–0.73, p = 0.002) compared to those with stable disease (SD).

Table 3 Multivariate analysis with survival of esophageal cancer post definitive chemotherapy.
Full size table

Discussion

To our best knowledge, there were no literatures discussing how to extend survival in those ESCC patients following dCCRT. Checkmate 577 was a phase III randomized control trial to demonstrate the adjuvant role of nivolumab for ESCC patients after neoadjuvant CCRT and radical esophagectomy17. Given the high recurrent rate and miserable prognosis with ESCC patients following dCCRT18, there was an urgent need to explore an effective maintenance. Our study is a pilot study to explore the maintenance role of UFUR in patients with ESCC following dCCRT. All survival outcomes were evaluated without the use of salvage therapy. Esophageal preservation was achieved in all patients who remained recurrence-free during follow-up. The results of this study demonstrated that UFUR maintenance significantly improves oncologic outcomes in patients with esophageal cancer by reducing overall recurrence and distant metastases, prolonging progression-free survival and overall survival, and enhancing distant metastasis-free survival. Subgroup demonstrate the substantial survival benefits of UFUR maintenance across all disease stages and all treatment responses of dCCRT. Multivariate analysis presented that UFUR maintenance, favorable CCRT response and early tumor stage were significant predictors of longer survival in ESCC patients. These findings emphasize the role of UFUR in mitigating systemic disease progression and improving survival outcomes, making it an useful strategy for the post-treatment management of ESCC. In addition to its clinical efficacy, UFUR offers notable economic advantages compared with other currently reported adjuvant therapies, such as immune checkpoint inhibitors or systemic chemotherapy. UFUR is an orally administered regimen with substantially lower medication and hospitalization costs, making it more accessible and feasible in real-world clinical settings. This cost-effectiveness further underscores the practical value of UFUR as an adjuvant option for patients after definitive chemoradiation.Thus, our results underscored the importance of integrating UFUR maintenance into treatment strategies to enhance long-term outcomes. Future research should focus on optimizing the duration and dosing of UFUR maintenance and exploring its integration with other therapeutic modalities, such as immunotherapy and targeted therapies. Additionally, identifying biomarkers predictive of response to UFUR maintenance could enable a more personalized approach to treatment, maximizing its efficacy and tolerability.

Tri-modality therapy including neoadjuvant chemoradiotherapy followed by radical surgery and dCCRT are both standard treatments for patients with ESCC. Several literatures had confirmed that tri-modality therapy provided a better survival than dCCRT19,20. Hence, dCCRT should be reserved for highly selected patients, such as fragile patients21. Notably, high recurrent rate and easily distant metastasis were observed after dCCRT, even for those patients with complete response after dCCRT22. A nation-wide questionnaire survey of institutions in Japan certified by the Japanese Esophageal Society to investigate outcomes of primary thoracic esophageal cancer patients initially treated by chemoradiotherapy with complete response diagnoses found a quarter of patients developed recurrent disease, mostly locoregional, after complete response23. Another retrospective study determined the patterns of recurrence and OS in patients achieving clinical complete response after treatment with dCCRT for esophageal cancer and showed that three- and five-year risk of recurrence were respectively 40% and 45% 24. A latest retrospective study to investigate the recurrence patterns and therapeutic outcomes after dCCRT for 71 patients with ESCC and found overall recurrence was 68% with median time from dCCRT to recurrence was 8.3 months25. These data all suggested ESCC patients were easily to develop tumor recurrence after dCCRT. Hence, maintenance is warranted to prolong recurrent-free survival as well as OS.

Studies in HNSCC have demonstrated the efficacy of UFUR maintenance strategy has improved survival in patients after dCCRT26. In ESCC, similar principles have been explored. A retrospective study investigating the combination of apatinib and S-1/capecitabine for ESCC patients with residual disease post-dCCRT provides insights into the benefits of integrating maintenance therapies27. Apatinib, a VEGFR-2 inhibitor, enhances anti-angiogenic effects, while S-1 or capecitabine, oral prodrugs of 5-FU, sustain cytotoxic activity against tumor cells28. The study revealed improved disease control rates, PFS, and OS, with a manageable safety profile. Another phase II study of Sintilimab has shown promising clinical efficacy and a manageable safety profile as maintenance after CCRT for local/regional recurrent ESCC29. However, apatinib and sintilimab were mainly available in China, which make it difficult to access for patients with ESCC. Based on these data of maintenance, it was suggested that UFUR, as an alternative oral 5-FU regimen, could be further investigated in ESCC for similar benefits. UFUR is an oral form chemotherapy and minimizes severe adverse effects, making it suitable for long-term use even in elderly and frail patients30,31. Oral administration of UFUR also simplifies treatment logistics and promotes patient adherence. Lower toxicity and sustained tumor control might ensure better patient outcomes and quality of life. The introduction of UFUR for ESCC patients following dCCRT represents a significant step toward personalized cancer treatment. By prioritizing efficacy, tolerability, and personalized care, these regimens pave the way for a more patient-centric approach to managing ESCC. Future clinical trials should focus on further validating UFUR’s role and optimizing its integration into multimodal treatment strategies.

Our study is a retrospective study and has several inevitable limitations. UFUR maintenance or not were decided at the discretion of physicians and patients, rather than randomized control. This might be the major bias in this study. The dose intensity of dCCRT and duration of UFUR maintenance were inconsistent in our patients. This will also have biases. Meanwhile, variable subsequent treatment, heterogeneity of our patients and different follow-up interval may also limit the power of our study. Given that, our study firstly explore the maintenance role of UFUR in patients with ESCC following dCCRT. To date, there are no prospective randomized controlled trials with larger cohorts focusing on UFUR maintenance for ESCC. In spite of a retrospective study with selection bias inherent to any retrospective studies, our study remains clinically valuable and provides clinical implication for physicians who treated patients with ESCC following dCCRT.

Conclusion

In conclusion, UFUR maintenance significantly improves survival outcomes in patients with ESCC following dCCRT, particularly through enhanced systemic disease control and prolonged overall survival. The strong correlation between UFUR maintenance and improved overall survival underscores its potential as a key component in the personalized treatment of esophageal cancer. These findings support the integration of UFUR into standard treatment protocols and highlight the need for further research to optimize its use in this challenging disease.