Impact of combined endoscopic therapy on progression-free survival in patients with primary tracheal cancers: a multivariate-adjusted retrospective cohort study Authors: Jiayao Wang, Yahan Xu, Tao Wang Published: 2025 ABSTRACT AbstractObjectiveThe purpose of this study was to assess the benefits and safety of combining endoscopic interventional therapy with surgery in patients with primary tracheal cancer.Patients and methodsThis single-center retrospective cohort study included 39 patients with primary tracheal malignancies treated between January 2015 and August 2020. They were divided into a surgery-only group (Group A, n = 21) and a surgery plus endoscopic therapy group (Group B, n = 18). The primary endpoints were progression-free survival (PFS) and 5-year survival rate. Participants were followed up every three months, which included CT scans and assessments of treatment efficacy. Multivariate Cox regression models were employed to adjust for potential confounding factors.ResultsAdenoid cystic carcinoma (ACC) was the predominant histology in the entire cohort (64.1%). Median PFS was significantly longer in the combined therapy group compared to the surgery-only group (55.05 months vs. 35.8 months, P = 0.0276). In the subgroup analysis for ACC, the combination therapy group consistently showed a trend toward superior outcomes, though the differences did not reach statistical significance. The 5-year survival rates were 66.7% and 57.1% in the combined therapy and surgery-only groups, respectively. All treatment-related adverse events were mild to moderate, with no serious adverse events reported.ConclusionSurgery combined with endoscopic intervention significantly improves PFS in patients with primary tracheal carcinoma and shows a trend toward superior 5-year survival rate without increasing severe toxicity. This represents a safe and effective treatment strategy.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12931-025-03459-4.Keywords: Primary tracheal cancer, Surgery, Endoscopic therapy, A cohort study Introduction Primary tracheal tumors are uncommon but usually malignant, accounting for about 0.2% of all malignancies in adults [1–3].The estimated incidence of tracheal cancer is about 0.1 cases per 100,000 people per year [4, 5], and its pathological types are mainly squamous cell carcinoma (SCC) and adenoid cystic carcinoma (ACC), accounting for about 60%~70% of tracheal malignant tumors [6, 7], while carcinoid carcinoma, adenocarcinoma, neuroendocrine carcinoma and so on are relatively rare [3, 8]. Although SCC has historically been considered the most common subtype worldwide, a recent meta-analysis including 733 patients with malignant tracheal tumors found that ACC was the most common histological subtype, accounting for 33.7% (247 cases), followed by SCC (228 cases, 31%) [7]; these findings differ from earlier SEER database analyses, in which ACC accounted for only 16% of cases [9]. This may reflect regional differences in disease biology or diagnostic patterns, although population-level confirmation is still lacking. Surgical resection is deemed the most effective therapy for primary tracheal cancer, often involving tracheectomy with end-to-end anastomosis [5, 10]. However, the trachea’s unique anatomy limits resectable lengths to ≤ 4 cm in most cases, requiring complex tissue release procedures for longer resections [5, 10, 11]. These anatomical constraints frequently result in margin compromises and local recurrence rates of 30–40% even after R0 resection [12]. And primary tracheal cancer often presents with non-specific clinical symptoms, which may be absent in early stages [13]. Symptoms such as dyspnea, cough, and hemoptysis typically become apparent only when the tumor obstructs 50% of the tracheal lumen [14–16], resulting in up to 70% misdiagnosis rates and frequent advanced-stage diagnoses [3, 12]. The 5-year survival rate for patients with surgically resected tracheal cancer is 50%, and only 10% for non-surgical patients [15, 17]. At present, except for surgery, the treatment of primary tracheal malignancies mainly includes radiotherapy, chemotherapy and interventional therapy [6, 18]. While postoperative adjuvant radiotherapy has been widely adopted - particularly for cases with non-R0 resection [17, 18], its efficacy remains suboptimal due to the tendency of ACC to spread along submucosal and perineural planes, its high likelihood of local recurrence, and its inherent radioresistance [6, 19–21]. This is particularly relevant in chinese populations where ACC predominance has been reported [22, 23]. Endoscopic techniques are promising as surgical adjuncts, especially for ACC which requires long-term local control [20, 24, 25]. However, robust evidence for combined surgical-endoscopic approaches remains lacking [17, 26]. This study aimed to evaluate the efficacy and safety of endoscopic intervention combined with surgery for primary malignant tracheal tumors. By employing a multivariate adjustment approach, we aimed to assess the association between this combined treatment strategy and progression-free survival (PFS), thereby providing a basis for developing personalized treatment plans for future patients. Patients and method Patients and study design This single-center cohort study included all primary tracheal cancer patients who underwent surgery at Tongji Hospital Affiliated to Tongji Medical College, Huazhong University of Science and Technology. The institutional ethics review board approved this study. The study protocol was conducted per the Declaration of Helsinki. Informed consent was waived due to the retrospective nature of this study. Primary tracheal cancer patients who received surgery between Jan 2015 and Aug 2020 were categorized into surgery alone group (group A) and the combination therapy group (surgery + endoscopic therapy, group B). Informed consent of surgery or/and endoscopic therapy was obtained before these procedures. Patient inclusion criteria were as follows: (a) age of 18–85 years; (b) chest CT, bronchoscopy and other examinations confirmed as primary tracheal malignancy; (c) lesions are located from the first tracheal cartilage ring to the carina and include lesions involving the carina crest, left and right main bronchus; and (d) the patient is conscious and able to cooperate with the treatment. Patient exclusion criteria were as follows: (a) adjacent malignancies involving the trachea (pulmonary malignancy, thyroid malignancy, laryngeal malignancy, esophageal malignancy, etc.); (b) exogenous airway compression; (c) history of other malignancies; and (d) poor adherence to treatment. The pathological subtype of the tissue is confirmed after the pathological tissue is obtained by surgery or bronchoscopy. Positron emission tomography or CT is used to assess the location, number, size, and local and distant metastases of tumors. As there is no universally accepted TNM stage for primary malignant tumors, the staging proposed by Bhattacharya in 2004 based on tumor size, extent of infiltration, and regional lymph nodes was used [13]. Treatment All patients included in this study were treated surgically. Among these, two patients received preoperative chemotherapy, and one patient received preoperative radiotherapy. The accepted surgical modality for tracheal malignancy is tracheal resection combined with end-to-end anastomosis. Owing to the unique anatomy of the trachea, the extent of resectable trachea is constrained. It is widely accepted that if the tracheal resection length is under 4 cm, a direct end-to-end anastomosis is feasible. However, if the resection exceeds 4 cm, it becomes necessary to mobilize surrounding tracheal tissue to facilitate a tension-free anastomosis, potentially involving the resection and repair of adjacent areas. The surgical approach is chosen based on the tumor location, such as a transverse cervical incision with or without partial sternotomy, or via thoracotomy. Exploratory procedures without resection and simple tracheostomy are not considered surgical treatment. The precise surgical plan is tailored by the surgeon according to the situation, with the goal of achieving negative margins on both resection sides. In addition to surgical treatment, all patients in Group B underwent combined bronchoscopic interventional therapy, with the core objective of maintaining or restoring airway patency. All interventional procedures were performed using rigid or flexible bronchoscopes. Based on therapeutic goals and technical characteristics, the primary interventional approaches employed were categorized into bronchoscopic tumor resection (intraluminal debulking) and stent placement (Table 1). For airway obstruction caused by intraluminal tumor proliferation, bronchoscopic tumor resection (intraluminal debulking) was primarily employed. Specific techniques are selected based on intraoperative tumor morphology: high-frequency electrocautery loop resection is used for pedunculated or larger tumors; laser and plasma ablation are suitable for treating broad-based or highly vascular tumors; cryotherapy may employ cryosectioning to resect larger tumors or cryofusion to address superficial and residual lesions; for extrinsic stenosis or collapsed airway walls, stent placement is performed. Silicone or metal stents are implanted to provide mechanical support to the narrowed segment, restoring airway patency. The specific techniques involved in this study include cryotherapy, laser ablation, plasma ablation, radiofrequency loop ablation, radioactive seed implantation, and stent placement. Table 1Baseline clinical characteristics of patients enrolled in this studyVariablesOverall(n=39)Group A(n=21)Group B(n=18)P-valueAge(years)49.82±11.76747.67±11.67652.33±11.6920.222Gender0.748 Men22(56.4%)11(52.4%)11(61.1%) Women17(43.6%)10(47.6%)7(38.9%)Somking history0.196 Yes13(33.3%)5(23.8%)8(44.4%) No26(66.7%)16(76.2%)10(55.6%)Comorbidity Hypertension7160.035 Diabetes3030.089 Cardiocerebrovascular diseases8350.432 Pulmonary diseases9360.255Symptom No symptom330 Cough231013 Blood in the sputum or hemoptysis1587 Chest tightness, shortness of breath, difficulty breathing1468 Others321Tumor subtypes>0.999 Squamous cell carcinoma853 Adenoid cystic carcinoma251312 Other633Location0.117 Upper trachea1075 Mid trachea952 Lower trachea17611 Trachea (unknown)330 Tumor diameter (cm)2.4±1.152.5±12±2.350.345Preoperative tumor marker levels CEA (ng/ml)2.11±1.32.38±1.171.58±0.9930.278 NSE11.63±3.9112.24±5.4712.22±2.1140.791 SCC-Ag (ng/ml)0.7±0.450.8±0.880.7±0.20.225Endoscopic interventional therapy Bronchoscopic stent placement/6 Bronchoscopic tumor resection/16 R0 Resection27(69.2%)14(66.7%)13(72.2%)0.742Postoperative treatments Targeted therapy7(17.9%)4(19%)3(16.7%)0.591 Immunotherapy7(17.9%)2(9.5%)5(27.8%)0.144 Chemoradiotherapy28(71.8%)16(76.2%)12(66.7%)0.723Recurrence Local recurrence13(33.3%)10(47.6%)3(16.7%)0.043 Distant recurrence11(28.2%)5(23.8%)6(33.3%)0.38Frequencies and percentages are reported for categorical variables, mean ± SD are reported for the normally distributed continuous variables and median ± IQR are reported for the non-normally distributed continuous variables. Group A: Surgery alone group; Group B: Combined group Baseline clinical characteristics of patients enrolled in this study Frequencies and percentages are reported for categorical variables, mean ± SD are reported for the normally distributed continuous variables and median ± IQR are reported for the non-normally distributed continuous variables. Group A: Surgery alone group; Group B: Combined group Follow-up and assessments After surgery, all patients were monitored until death, loss to follow-up, or withdrawal from the study. The final follow-up of this study was on August 15, 2025. Chest CT scans were scheduled every three months and treatment response was assessed using Efficacy Evaluation Criteria in Solid Tumors version 1.1 [27] and classified as complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). The primary endpoints of this study are progression-free survival (PFS) and 5-year survival rate. Overall survival (OS) is defined as the time interval from the start of surgery to death or the last follow-up visit (August 15, 2025). For patients who died during the follow-up period, OS was calculated as the time interval from the day of surgery to death. For those lost to follow-up, OS is from surgery to the last known follow-up. PFS is the interval from surgery to documented progression, encompassing local advancement and/or distant metastases. For patients without death or progression, the last clinical assessment date serves as the censored date. Statistical analysis For data analysis, SPSS 26.0 and GraphPad Prism were employed. Data normally distributed are presented as mean ± SD, while skewed data are shown as median ± IQR. The t-test was applied for comparing continuous variables with normal distribution, and non-parametric tests were used for those with non-normal distribution. Fisher’s exact test was utilized for categorical variables. The Kaplan-Meier method assessed median PFS and OS between groups. Variables with a p-value < 0.1 in univariate analysis were advanced to multivariate Cox proportional hazards analysis. This analysis identified predictors for PFS and OS, with a p-value < 0.05 indicating statistical significance. Results Patient characteristics A total of 39 patients were studied, divided into group A (n = 21) and group B (n = 18), with a median age of 50 years. The specific enrollment process is illustrated in Fig. 1. The study included 56.4% of men (n = 22) and 43.6% of women (n = 17), of whom 33.3% were current or former smokers. Histologically, 8 (20.5%) had squamous cell carcinoma (SCC) and 25 (64.1%) had adenoid cystic carcinoma (ACC); R0 resection (microscopically negative margins) was achieved in 14 patients (66.7%) in Group A and 13 patients (72.2%) in Group B; 15 patients (38.5%) had stage III and 12 (30.8%) had stage IV (Table 2). Table 1 details the demographic data, showing a significant difference in hypertension prevalence (P = 0.035), while no statistically significant differences were observed in other baseline characteristics. Statistical findings indicate significant differences in recurrence patterns between the combined therapy group and the surgery-only group. The local recurrence rate was significantly lower in the combined therapy group compared to the surgery-only group (16.7% vs. 47.6%, P = 0.043); while no significant difference was observed in distant metastasis rates (33.3% vs. 23.8%, P = 0.38). Group A received surgery alone, while group B underwent combined surgery and endoscopic treatment. The specific endoscopic intervention protocols for group B are detailed in the Methods section and summarized in Table 1. Postoperatively, 7 patients received immunotherapy, 7 patients received targeted therapy, and 28 patients received chemotherapy, radiotherapy, or chemoradiotherapy. A total of 7 patients (17.9%) were lost to follow-up, primarily due to discontinuation of treatment at our hospital and loss of contact information. For these patients, survival time was censored at the date of their last known clinical evaluation. Patient selection flowchart Table 2Staging of malignant tumors in the trachea (Bhattachyryya's method, 2004)Overall(n=39)Group A(n=21)Group B(n=18)T T1422 T21284 T318711 T4321 Tx220N N0271314 N1633 Nx651M M0331617 M1651Stage Stage I422 Stage II534 Stage III1578 Stage IV1284 Unknown110 Staging of malignant tumors in the trachea (Bhattachyryya's method, 2004) Complications The safety analysis of this study indicates that all treatment-related adverse events were mild to moderate in severity and manageable. Endoscopy- and surgery-related adverse events (cough, sputum production, sore throat) resolved rapidly with symptomatic treatment. Radiotherapy-related adverse events included radiation stomatitis/parotitis, while chemotherapy-related adverse events comprised nausea/vomiting and bone marrow suppression, both managed according to standard protocols. Additionally, two postoperative tracheal fistulas were successfully repaired via endoscopic intervention, with no treatment-related serious adverse events observed. Clinical outcomes Clinical outcomes for both groups are detailed in Table 3. During follow-up, the median PFS was 35.8 months for group A and 55.05 months for group B, showing a significant difference (P = 0.0276; Fig. 2A). However, after stratification analysis based on ACC, no statistically significant difference in PFS was observed between the two groups (P = 0.606; Fig. 2B). No significant differences were observed between the two groups in terms of 5-year overall survival rate (57.1% vs. 66.7%, P = 0.1865; Fig. 3A) or overall survival time (P = 0.1705; Fig. 3B). This study conducted subgroup analyses based on diameter, TNM stage, pathological type, and whether R0 resection was achieved. The results showed no statistically significant differences between the groups (P < 0.05) (Figs. 4 and 5). Table 3Clinical outcomes between primary tracheal patients in groups A and BOverall(n=39)Group A(n=21)Group B(n=18)Status Survival351718 Death330 Unknow1101-years survival rate(%)92.3%85.7%100.0%Status Survival221111 Death1082 Unknow725 3-years survival rate(%)74.4%61.9%88.9%Status Survival17107 Death1596 Unknow725 5-years survival rate(%)61.5%57.1%66.7% Clinical outcomes between primary tracheal patients in groups A and B A Comparison of median PFS between primary tracheal cancer patients in groups A and B; (B) Comparison of median PFS in patients with primary tracheal cancer in groups A and B after stratified analysis of ACC A Comparison of median OS between primary tracheal cancer patients in groups A and B; (B) Comparison of 5-year survival rate in patients with primary tracheal cancer in groups A and B A Comparison of PFS between adenoid cystic carcinoma and non-adenoid cystic carcinoma(P = 0.0147); (B) Comparison of PFS between TNM Stage I + II and TNM Stage III + IV groups(P = 0.2864); (C) Comparison of PFS between the group with tumors larger than 3 cm in diameter and the group with tumors no larger than 3 cm in diameter(P = 0.6601); (D) Comparison of PFS between the R0 resection group and the non-R0 resection group(P = 0.846) A Comparison of 5-year survival rate between adenoid cystic carcinoma and non-adenoid cystic carcinoma (P = 0.2421); (B) Comparison of 5-year survival rate between TNM Stage I + II and Stage III + IV groups (P = 0.3881); (C) Comparison of 5-year survival rate between tumors larger than 3 cm and those no larger than 3 cm in diameter (P = 0.7539); (D) Comparison of 5-year survival rate between R0 resection and non-R0 resection groups (P = 0.0756) To further clarify the efficacy differences among various endoscopic techniques, we subdivided the combined therapy group into two subgroups—“surgery combined with endoscopic tumor resection” and “surgery combined with stent placement”—for in-depth analysis (Supplementary Material 1). Results demonstrated that the surgery combined with endoscopic tumor resection subgroup exhibited a significant PFS benefit compared to the surgery-only group (median PFS: 66.6 months vs. 35.8 months; HR = 0.392, 95% CI: 0.165–0.929; P = 0.0139). Predictors of progression-free survival for primary tracheal cancer In the analysis of factors affecting PFS, univariate Cox regression results showed that pathological type (ACC vs. non-ACC, HR = 0.381, P = 0.028), targeted therapy (HR = 2.969, P = 0.038), and immunotherapy (HR = 11.553, P < 0.001) were significantly associated with PFS. Multivariate Cox regression revealed that pathological type (ACC vs. non-ACC, HR = 0.099, P = 0.013) and targeted therapy (HR = 6.538, P = 0.023) were independent predictors of PFS. Immunotherapy did not maintain independent statistical significance in multivariate analysis (HR = 7.016, P = 0.148). Detailed results of univariate and multivariate analyses are presented in Table 4. Additionally, to further investigate the impact of endoscopic intervention on PFS, the inclusion of combined endoscopic therapy was incorporated into the Cox regression analysis (Supplementary Material 2). Table 4Cox proportional hazard regression analysis of survival in patientsVariablePFSHR95%CI for HRPLowerUpperUnivariate COX regression Gender(male vs female)1.5510.6483.7120.324 Age (>60 vs <60y)0.7010.2052.390.57 Smoking history (Yes vs No)0.670.2711.6590.387 Hypertension (Yes vs No)1.1150.4083.0520.832 Diabetes (Yes vs No)0.4240.0563.2170.407 Cardiocerebrovascular diseases (Yes vs No)1.0920.4012.9740.863 Pulmonary diseases (Yes vs No)1.0640.3892.9070.904 Pathology(ACC vs non-ACC)0.3810.1610.9 0.028 Diameter(>3cm vs <3cm)0.7740.2732.1910.629 TNM(III-IV vs I-II)0.5840.2411.4180.235 R0 resection(Yes vs No)0.9160.3812.20.844 Targeted therapy(Yes vs No)2.9691.0648.283 0.038 Immunotherapy(Yes vs No)11.5532.93245.517 <0.001 Chemoradiotherapy(Yes vs No)0.390.1291.1780.095Multivariable Cox regression Age (>60y vs <60y)1.0710.2155.3390.934 Smoking history (Yes vs No)0.7230.1922.7210.631 Hypertension (Yes vs No)0.340.0522.2070.258 Pathology(ACC vs non-ACC)0.0990.0160.62 0.013 Diameter(>3cm vs <3cm)0.6140.142.6940.518 TNM(III-IV vs I-II)1.7940.31710.1510.509 Targeted therapy(Yes vs No)6.5381.29832.935 0.023 Immunotherapy(Yes vs No)7.0160.598.3880.148 Chemoradiotherapy(Yes vs No)0.8410.0967.3330.875 R0 resection(Yes vs No)0.1810.0181.8310.148OS Overall survival, HR Hazard ratio, CI Confidence interval, PFS Progression-free survival. The bold values are less than 0.05. Cox proportional hazard regression analysis of survival in patients OS Overall survival, HR Hazard ratio, CI Confidence interval, PFS Progression-free survival. The bold values are less than 0.05. Discussion Primary tracheal tumors are rare, constituting 0.1–0.4% of all malignant tumors [6, 24], and only 0.2% of respiratory tract tumors [28]. While 90% of tracheal tumors in adults are malignant, in children, they are rarer and mostly benign (10–30% are malignant) [24]. These tumors can arise from various structures like respiratory epithelium, salivary glands, and connective tissue. Histologically, SCC and ACC make up about two-thirds of adult primary tracheal tumors [3, 8, 24]. SCC is traditionally considered the most common type of primary tracheobronchial tumor, followed by ACC [29, 30]. However, recent Chinese studies and a meta-analysis have indicated a shift, with ACC becoming the most common type among primary tracheal tumors [7, 22, 23], aligning with the pathological classification of patients in this study. Based on existing research, we propose the following hypothesis regarding the relative increase in ACC incidence: its core mechanism likely involves certain factors selectively promoting decisive molecular events in ACC development. Studies have clearly demonstrated that the MYB-NFIB gene fusion caused by the 6q-9q chromosomal translocation is a specific molecular marker for ACC [20, 21]. This fusion alters MYB protein overexpression and related gene activity, promoting tumorigenesis [21, 31]. This fusion directly drives tumorigenesis by altering the overexpression of the MYB protein and the activity of related genes [31, 32]. Concurrently, other studies have independently identified common mutated genes such as LPAR3 and ALPI in ACC, along with significantly mutated genes involved in the NOTCH signaling pathway [20, 33]. Furthermore, ACC exhibits a unique “immunologically cold” tumor microenvironment characterized by PD-L1 expression negativity and low infiltration of CD3 + and CD8 + T cells [34]. Therefore, a plausible hypothesis is that the rising proportion of ACC may reflect the combined effects of certain internal and external factors (such as diagnostic technology and aging), which have increased the detection opportunities or growth conditions for tumors possessing specific combinations of molecular events (MYB fusions, LPAR3/ALPI mutations, Notch pathway involvement) and immune-evasive characteristics at the population level. Of course, the precise relationships among these molecular events and their exact roles in disease progression remain to be elucidated by future studies. There is evidence suggesting that combining endoscopic therapy with radiotherapy may offer long-term survival benefits for patients with histologically unresectable ACC [24, 25]. Repeated endoscopic tumor resection can also maintain airway patency in cases of local recurrence with minimal side effects [24]. Beyond biopsy, endoscopy offers treatments like laser resection, plasma ablation, cryosurgery, electric snaring, photodynamic therapy, radioactive particle stent implantation, and stent placement for tracheal cancer, aiming to relieve symptoms and improve airway function [6, 24]. In this study, surgery combined with endoscopic intervention significantly improved patients’ PFS. The median PFS in the combined therapy group was 55.05 months, significantly superior to the 35.8 months observed in the surgery-only group (P = 0.0027; Fig. 2A). In the stratified analysis of ACC patients, although the difference in PFS between the two groups did not reach statistical significance (P = 0.0606), the survival curves (Fig. 2B) demonstrated that the PFS rate in the combined therapy group remained consistently higher than that in the surgery-only group throughout the observation period. This suggests that with extended follow-up, this treatment modality may confer more clinically meaningful survival benefits for ACC patients. ACC exhibits unique biological behavior, primarily characterized by perineural infiltration and submucosal spread. This results in a significantly higher risk of local recurrence compared to distant metastasis [6, 15], which is also a key reason for the limited efficacy of surgery alone. In this study, significant differences were observed in recurrence patterns between the combined therapy group and the surgery-only group: the local recurrence rate was significantly lower in the combined therapy group (16.7% vs. 47.6%, P = 0.043), while no statistically significant difference was found in distant metastasis rates between the two groups (33.3% vs. 23.8%, P = 0.38) (Table 1). No significant improvement in overall survival was observed (P = 0.1705) (Fig. 3A). These findings indicate that bronchoscopic intervention, as an adjunct to surgery, significantly enhances local disease control by achieving more thorough clearance of intraluminal tumors, managing resection margins, and ensuring airway patency. This advantage directly translates into clinical benefit, manifested as prolonged median PFS in the combined therapy group (55.05 months vs. 35.8 months, P = 0.0276). Concurrently, this study revealed limitations of the current combined treatment strategy: while optimizing local control, it failed to effectively suppress the formation of distant micrometastases. Consequently, for patients with established distant metastases, the advantage in local control could not be translated into long-term survival benefit. This study further confirms that endoscopic interventional techniques such as laser ablation can effectively eliminate residual micrometastases distributed along nerve bundles, thereby compensating for the limitations of surgery in achieving local control. Our subgroup analysis further corroborates this mechanism and reveals distinct therapeutic roles and clinical values among different endoscopic techniques (Supplementary Material 1): endoscopic tumor resection effectively delays local recurrence by directly reducing tumor burden and addressing micro-invasive foci, translating into PFS benefits. whereas stent placement, as an airway reconstruction measure, primarily maintains lumen patency through mechanical support without directly interfering with tumor progression. Consequently, the stent subgroup showed no PFS benefit, primarily reflecting the group’s clinical characteristics of heavier tumor burden and more advanced disease stage rather than technical failure. Together, these two techniques form an individualized treatment system tailored to different disease stages. Consequently, for ACC characterized primarily by local infiltration, combined endoscopic treatment with surgery achieves more thorough residual lesion clearance compared to surgery alone. Treatment strategies for ACC differ significantly from those for SCC. Due to the high risk of distant metastasis, systemic therapy is often prioritized for SCC [3, 6]; ACC, however, relies more on aggressive local treatment. Its relatively limited radiosensitivity also suggests that local radical resection holds greater clinical significance. Against this backdrop, endoscopic interventions demonstrate dual value in the comprehensive management of ACC: they serve as a crucial adjunct when achieving R0 resection is challenging, and they can also be employed during postoperative follow-up to monitor and manage local recurrence. The present study further substantiates the clinical efficacy of this combined strategy: the combined therapy group (Group B) demonstrated a 1-year survival rate of 100% (vs. 85.7% in Group A), with 3-year and 5-year survival rates also showing a trend of superiority over the surgery-only group (3-year survival rate: 88.9% vs. 61.9%; 5-year survival rate: 66.7% vs. 57.1%), with no serious adverse events reported during treatment. These findings provide reliable evidence for further promoting endoscopic intervention as a safe and effective adjunctive therapy for ACC, offering significant reference value for clinical practice in regions with high ACC incidence. Univariate Cox regression analysis revealed that the ACC histological subtype was a protective factor for PFS (HR = 0.381, 95% CI: 0.161–0.900, P = 0.028), consistent with the known biological behavior of this subtype characterized by slow growth and delayed metastasis. However, targeted therapy (HR = 2.969, 95% CI: 1.064–8.283, P = 0.038) and immunotherapy (HR = 11.553, 95% CI: 2.932–45.517, P < 0.001) were significantly associated with increased risk of disease progression. In multivariate Cox regression analysis, the protective effect of ACC (HR = 0.099, 95% CI: 0.016–0.620, P = 0.013) and the risk effect of targeted therapy (HR = 6.538, 95% CI: 1.298–32.935, P = 0.023) remained independent and significant. Notably, the high risk ratios observed for targeted and immunotherapy in the model likely reflect substantial “indication bias.” Comparative analysis of patient baseline characteristics supports this interpretation(Supplementary Material 3). Data indicate that patients receiving targeted or immunotherapy exhibited significantly higher rates of distant recurrence (54.55% vs. 17.86%, P = 0.044), overall recurrence rate (100% vs. 46.43%, P = 0.002), and bronchial stent placement rate (36.36% vs. 7.14%, P = 0.042). Moreover, the vast majority (90.91%) were refractory cases with prior failure of chemoradiotherapy. These characteristics collectively delineate a subgroup of advanced patients exhibiting more aggressive tumor biology and heavier disease burden. Thus, the high risk captured by statistical models reflects the inherent poor prognosis of this specific refractory cohort rather than directly asserting the ineffectiveness of the aforementioned treatments. In future prospective studies, balancing baseline risk across patient groups is crucial for accurately assessing the true efficacy of these novel therapies. Furthermore, this study’s exploration of the value of combined endoscopic therapy highlights the importance of controlling for confounding factors in statistical models. In univariate Cox regression analysis, endoscopic therapy showed only a non-significant trend toward reducing disease progression risk (HR = 0.528, 95% CI: 0.217–1.280, P = 0.157). However, in the multivariate Cox regression analysis, after adjusting for multiple potential confounders including pathological type, TNM staging, and targeted therapy, the hazard ratio decreased to 0.182 (95% CI: 0.025–1.355, P = 0.096). This shift suggests that the true effect of endoscopic therapy was masked by other factors (particularly histological type) in the univariate analysis. In clinical practice, ACC patients with a high propensity for local recurrence are more frequently recommended for endoscopic treatment, introducing confounding due to this non-random treatment allocation. After adjusting for this factor in multivariate analysis, the potential benefits of endoscopic intervention in reducing local recurrence (16.7% vs. 47.6%, P = 0.043) and prolonging PFS (55.05 months vs. 35.8 months, P = 0.0276) became more evident, providing supportive evidence for its adjunctive value in local control. This study systematically evaluated the efficacy of combined endoscopic surgery for primary tracheal malignancies based on an expanded sample size (n = 39). Results demonstrated a median PFS advantage in the combined therapy group (55.05 months vs. 35.8 months, HR = 0.464, P = 0.0276). It should be noted that this study has certain limitations. The extremely low incidence of primary tracheal cancer limited further expansion of the sample size. Additionally, a 17.9% loss to follow-up rate may have potentially influenced the results. Sensitivity analysis was conducted, revealing that even under the most conservative assumption (all lost-to-follow-up patients experienced disease progression), the combined therapy maintained a trend toward benefit (HR = 0.645), partially supporting the robustness of the study conclusions(Supplementary Material 4). The current follow-up period remains insufficient to fully evaluate the long-term outcomes of slow-growing ACC, and the diversity of endoscopic intervention techniques may introduce selection bias. Future multicenter collaboration is needed to expand sample sizes, extend follow-up periods, and standardize treatment protocols to more accurately assess the impact of combined therapeutic strategies on long-term survival. Conclusion Combined surgical and endoscopic interventional therapy offers a safe and effective treatment strategy for primary tracheal carcinoma. This approach significantly improves PFS, particularly in patients with adenoid cystic carcinoma, without increasing the incidence of severe adverse events. Endoscopic treatment enables precise removal of residual lesions and local recurrences that are difficult to address surgically, thereby overcoming the limitations of surgery alone and providing tracheal carcinoma patients with a more promising personalized treatment option. Supplementary Information Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Supplementary Material 4. Authors’ contributions Conception and design of study: Tao Wang; Data analysis, table plotting and drafting of manuscript: Jiayao Wang, Yahan Xu. Funding None. Data availability The data used to support the findings of this study are available from the corresponding author upon request. Declarations Ethics approval and consent to participate This study has been reviewed and approved by the Ethics Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, and the ethics approval number is TJ-IRB202412190. Since this study is a retrospective observational study, the informed consent of patients has been waived according to the guidance and regulations of the ethics committee. During the research process, we strictly adhered to ethical principles to ensure the confidentiality of patient data and the scientific nature of the study. Consent for publication The authors consent to publication. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Supplementary Material 3. Supplementary Material 4. Data Availability Statement The data used to support the findings of this study are available from the corresponding author upon request.