Abstract
Background
This study aimed to compare the effects of different treatment methods on the postoperative results and long-term prognosis of patients with 1-2 cm rectal neuroendocrine tumors (R-NETs), and provide evidence-based guidance for the treatment decision-making of this specific tumor size.
Methods
The retrospective cohort study of this single-center included 736 patients with pathologically diagnosed R-NETs who were treated from January 2013 to August 2024. According to the treatment received, the patients were divided into endoscopic resection group (ER), local surgery group (LS) and radical surgery group (RS). The primary endpoint was recurrence free survival (RFS).
Results
A total of 736 R-NETs patients were enrolled. For the controversial 1–2 cm tumor, there was no significant difference in the long-term RFS between the LS group and the ER group (P = 0.280), and the postoperative hospitalization days of the LS group was relatively short (median 2 days vs. 3 days, P < 0.001). Sensitivity analysis using propensity score matching (PSM) further confirmed these outcomes in postoperative hospitalization days (P < 0.001).
Conclusion
For 1–2 cm R-NETs, recurrence events were rare in both treatment groups, and local surgery was associated with a shorter postoperative hospital stay.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s00384-026-05247-y.
Keywords: Neuroendocrine Tumors, Rectal Neoplasms, Endoscopic Resection, Transanal Endoscopic Surgery, Prognosis
Introduction
Neuroendocrine neoplasms (NENs) are a group of heterogeneous tumors, which can be pathologically divided into well-differentiated neuroendocrine tumors (NETs) and poor-differentiated neuroendocrine carcinomas (NECs), with NETs demonstrating higher clinical incidence [1, 2]. The gastrointestinal tract is the main primary site of NETs, of which rectal neuroendocrine tumors (R-NETs) account for 12%−27% of all gastrointestinal NETs, making it a primary research focus in this disease [3, 4]. In recent years, due to the widespread development of colorectal cancer screening projects and the progress of endoscopic technology, especially in Asian countries, the incidence of NETs has been on the rise [5–8]. Although most R-NETs show chronic growth and good prognosis, a subgroup demonstrates aggressive behavior, with tumors size larger than 2 cm showing higher propensity for lymph node or distant metastasis [9, 10].
At present, the clinical management strategy of R-NETs mainly depends on the size of the tumor for risk stratification. Common treatment methods include endoscopic resection, local surgery and radical surgery [11]. According to the guidelines of the European Neuroendocrine Tumor Society (ENETS), endoscopic resection is recommended for lesions smaller than 1 cm, radical surgery is recommended for lesions larger than 2 cm [12]. However, for medium-sized R-NETs (1–2 cm), the best treatment plan is still controversial, because there is no conclusion on the reliable comparative data on the efficacy of endoscopic resection and local surgery [13–18]. In view of the above considerations, this study uses large-scale single-center cohort data to explore this key clinical problem. We systematically compared the clinical efficacy and safety of endoscopic resection and local surgery in 1–2 cm R-NETs patients, aiming to provide evidence-based basis for the optimization of R-NETs treatment decision-making.
Materials and methods
Data collection
The study of this single center has continuously included 770 cases of R-NETs patients diagnosed with pathology in Renji Hospital, affiliated with Shanghai Jiao Tong University School of Medicine from January 2013 to August 2024. This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement. Through systematic review of electronic medical records, we collected the demographic characteristics, clinical symptoms, pathological characteristics, treatment methods, perioperative parameters and follow-up data of patients. The exclusion criteria include: (1) combined with other active malignant tumors; (2) loss to follow-up; (3) non-resected tumors. Consequently, after excluding 4 cases with other malignant tumors, 23 lost to follow-up and 7 patients without tumor resection, a total of 736 patients were included in the final analysis (Fig. 1).
Fig. 1.
Patient Selection Flowchart
Resection Modality
According to the method of resection received by the patient, they were divided into three groups: endoscopic resection group (ER), local surgery group (LS) and radical surgery group (RS). The ER group includes operations such as endoscopic submucosal dissection (ESD), endoscopic mucosal resection (EMR), snare polypectomy, cold forceps polypectomy (CFP), and argon plasma coagulation (APC). The LS group comprised procedures including transanal minimally invasive surgery (TAMIS) and transanal endoscopic microsurgery (TEM). Radical resection referred to low anterior resection (LAR) with regional lymph node dissection. The anatomical range of lymphadenectomy mainly involved the mesorectal regional lymph nodes. Treatment decisions were based on preoperative evaluation, including cross-sectional imaging, endoscopic ultrasound (EUS) assessment. RS was generally considered for patients with suspicious regional lymph nodes or evidence of deeper invasion, or higher tumor grade on preoperative evaluation. In contrast, lesions without these features were more commonly treated with ER or LS, particularly among patients with 1–2 cm tumors. For lymph node staging, N stage in the RS group was determined by postoperative pathological examination because lymph node dissection was routinely performed in these patients. In the ER and LS groups, where lymph node dissection was not routinely performed, N stage was assessed clinically based on preoperative imaging (CT or MRI) and/or EUS.
Follow-up and outcome
Postoperative surveillance was conducted through routine outpatient visits according to clinical practice at our center. Colonoscopy was generally recommended within the first year after treatment and subsequently at 1–2 year intervals to monitor for local recurrence. Imaging examinations were arranged when clinically indicated. Patient prognosis information was obtained through a combination of medical records and telephone follow-up. The main endpoint of this study was recurrence free survival (RFS), which is defined as the time interval from the beginning of initial treatment to the end of tumor recurrence or follow-up. Tumor recurrence is defined as the reappearance of any tumor confirmed by endoscopy or imaging examination at least 6 months after the end of the initial treatment, including local rectal recurrence, regional lymph node recurrence or distant metastasis of the liver, lungs and other organs [19, 20].
Statistical analysis
Categorical variables were expressed as frequencies and percentages, and compared using Chi-square or Fisher's exact tests. For categorical variables with three subgroups showing statistically significant differences, pairwise comparisons were further conducted using the z-test with Bonferroni correction. Non-normally distributed continuous variables were presented as medians with interquartile ranges (IQR) and compared using the Mann–Whitney U test. To balance baseline characteristics, 1:1 Propensity Score Matching (PSM) was performed using the nearest neighbor matching method with a caliper of 0.2. To construct the Love plot, SMDs in the unmatched cohort were recalculated including only the ER and LS groups prior to matching, ensuring that balance assessments before and after matching were based on the same analytical sample. Survival was evaluated using Kaplan–Meier curves and log-rank tests. Analyses were conducted using R software (v4.2.1), with two-sided P < 0.05 considered significant.
Results
Patient baseline characteristics according to the excision method
This analysis included 736 patients underwent ER, LS or RS for details of baseline characteristics (Table 1). The median follow-up time for the overall cohort was 48 months (range: 12–141 months). The cohort includes 608 ER cases (82.60%; APC: n = 3, CFP: n = 24, snare polypectomy: n = 31, EMR: n = 224, ESD: n = 326), 106 cases of LS (14.40%; TAMIS: n = 94, TEM: n = 12) and 22 RS cases (3.40%; all of LAR). There was no significant difference between groups in terms of gender (P = 0.612), age (P = 0.112) or incidental discovery (P = 0.624). However, tumor characteristics differed significantly (P < 0.001). The proportion of tumor size < 1 cm in the ER group was significantly higher proportions (89.46% vs. 10.36% vs. 0.18%). T stage (T2: 66.67%; T3 + T4: 100.00%) and G3 (100.00%) were exclusively seen in the RS group. The positive rates of perineural or lymphovascular invasion (PNI/LVI) varied across groups (ER: 56.52%, LS: 13.04%, RS: 30.43%), while lymph node metastasis (N stage) dominated in the RS group (85.00% vs. 5.00% vs. 10.00%, P < 0.001).
Table 1.
Baseline characteristics of the patient via resection modality
| Variables | Total (n = 736) | Endoscopic Resection (n = 608) |
Local Surgery (n = 106) |
Radical Surgery (n = 22) |
P |
|---|---|---|---|---|---|
| Basic Information | |||||
| Sex, n (%) | 0.612 | ||||
| Male | 377 (51.22) | 314 (83.29) | 54 (14.32) | 9 (2.39) | |
| Female | 359 (48.78) | 294 (81.89) | 52 (14.48) | 13 (3.62) | |
| Age, n (%) | 0.112 | ||||
| < 45 | 257 (34.92) | 222 (86.38) | 32 (12.45) | 3 (1.17) | |
| 45–59 | 256 (34.78) | 209 (81.64) | 39 (15.23) | 8 (3.12) | |
| ≥ 60 | 223 (30.30) | 177 (79.37) | 35 (15.70) | 11 (4.93) | |
| Reason for presentation, n (%) | 0.624 | ||||
| Incidental | 409 (55.57) | 340 (83.13) | 59 (14.43) | 10 (2.44) | |
| Symptomatic | 327 (44.43) | 268 (81.96) | 47 (14.37) | 12 (3.67) | |
| EUS evaluation, n (%) | 0.088 | ||||
| No | 448 (60.87) | 360 (80.36) | 71 (15.85) | 17 (3.79) | |
| Yes | 288 (39.13) | 248 (86.11) | 35 (12.15) | 5 (1.74) | |
| Pathological Features | |||||
| Size, n (%) | < 0.001* | ||||
| < 1 cm | 560 (76.09) | 501 (89.46) a | 58 (10.36) b | 1 (0.18) c | |
| 1–2 cm | 162 (22.01) | 105 (64.81) a | 43 (26.54) b | 14 (8.64) c | |
| ≥ 2 cm | 14 (1.90) | 2 (14.29) a | 5 (35.71) b | 7 (50.00) c | |
| Grade, n (%) | < 0.001* | ||||
| G1 | 699 (94.97) | 592 (84.69) a | 98 (14.02) b | 9 (1.29) c | |
| G2 | 35 (4.76) | 16 (45.71) a | 8 (22.86) b | 11 (31.43) c | |
| G3 | 2 (0.27) | 0 (0.00) a | 0 (0.00) a | 2 (100.00) b | |
| PNI/LVI, n (%) | < 0.001* | ||||
| Negative | 690 (93.75) | 582 (84.35) a | 100 (14.49) a | 8 (1.16) b | |
| Positive | 46 (6.25) | 26 (56.52) a | 6 (13.04) a | 14 (30.43) b | |
| T stage, n (%) | < 0.001* | ||||
| T1 | 712 (96.74) | 606 (85.11) a | 103 (14.47) b | 3 (0.42) c | |
| T2 | 15 (2.04) | 2 (13.33) a | 3 (20.00) b | 10 (66.67) c | |
| T3 + T4 | 9 (1.22) | 0 (0.00) a | 0 (0.00) a | 9 (100.00) b | |
| N stage, n (%) | < 0.001* | ||||
| N0 | 716 (97.28) | 607 (84.78) a | 104 (14.53) b | 5 (0.70) c | |
| N1 | 20 (2.72) | 1 (5.00) a | 2 (10.00) b | 17 (85.00) c | |
| M stage, n (%) | |||||
| M0 | 728 (98.91) | 608 (83.52) a | 104 (14.29) b | 16 (2.20) c | < 0.001* |
| M1 | 8 (1.09) | 0 (0.00) a | 2 (25.00) b | 6 (75.00) c |
EUS Endoscopic Ultrasonography, PNI/LVI perineural or lymphovascular invasion. Different superscript letters (a, b, c) indicate statistically significant differences between columns by pairwise z-test with Bonferroni correction (adjusted P < 0.05). *Statistically significant difference (P < 0.05)
Given the critical role of tumor size in clinical decision-making, we noticed that ER was the main treatment for patients with a tumor diameter of less than 1 cm (89.46%), while for patients with a tumor diameter of more than 2 cm, RS is the main treatment (50.00%) in our cohort. For patients with tumor diameters between 1–2 cm, the proportion of patients underwent LS was higher than tumor diameter less than 1 cm (26.54% vs. 10.36%).
Treatment options for 1–2 cm rectal neuroendocrine tumors
Considering the subjectivity inherent in treatment selection for 1–2 cm tumors, we further analyzed this subgroup. Patients who underwent Radical Surgery (RS) generally showed more advanced pathological features, including a higher proportion of T2-T4 tumors and N1 disease. Therefore, subsequent analyses were restricted to the Endoscopic Resection (ER) and Local Surgery (LS) groups. Baseline characteristics of the RS group (n = 14) are presented separately in Supplementary Table S1. Pairwise comparison showed that there was no statistical difference in the baseline characteristics of the ER group and the LS group (P > 0.05) (Table 2).
Table 2.
Baseline Characteristics of Patients with 1–2 cm Tumors Treated with Endoscopic Resection or Local Surgical Excision
| Variables | Total (n = 148) |
Endoscopic Resection (n = 105) |
Local Surgery (n = 43) |
P | SMD |
|---|---|---|---|---|---|
| Sex, n (%) | 0.440 | 0.140 | |||
| Male | 65 (43.92) | 44 (41.90) | 21 (48.84) | ||
| Female | 83 (56.08) | 61 (58.10) | 22 (51.16) | ||
| Age, n (%) | 0.315 | 0.278 | |||
| < 45 | 40 (27.03) | 29 (27.62) | 11 (25.58) | ||
| 45–59 | 49 (33.11) | 38 (36.19) | 11 (25.58) | ||
| ≥ 60 | 59 (39.86) | 38 (36.19) | 21 (48.84) | ||
| Reason for presentation, n (%) | 0.492 | 0.125 | |||
| Incidental | 90 (60.81) | 62 (59.05) | 28 (65.12) | ||
| Symptomatic | 58 (39.19) | 43 (40.95) | 15 (34.88) | ||
| EUS evaluation, n (%) | 0.222 | 0.225 | |||
| No | 92 (62.16) | 62 (60.78) | 30 (29.41) | ||
| Yes | 56 (37.84) | 43 (71.67) | 13 (21.67) | ||
| Grade, n (%) | 0.233 | 0.205 | |||
| G1 | 134 (90.54) | 97 (92.38) | 37 (86.05) | ||
| G2 | 14 (9.46) | 8 (7.62) | 6 (13.95) | ||
| G3 | 0 (0.00) | 0 (0.00) | 0 (0.00) | ||
| PNI/LVI, n (%) | 0.331 | 0.173 | |||
| Negative | 136 (91.89) | 98 (93.33) | 38 (88.37) | ||
| Positive | 22 (8.11) | 7 (6.67) | 5 (11.63) | ||
| T stage, n (%) | 0.147 | 0.248 | |||
| T1 | 143 (96.62) | 103 (98.10) | 40 (93.02) | ||
| T2 | 5 (3.38) | 2 (1.90) | 3 (6.98) | ||
| T3 + T4 | 0 (0.00) | 0 (0.00) | 0 (0.00) | ||
| N stage, n (%) | 0.498 | 0.108 | |||
| N0 | 146 (98.65) | 104 (99.05) | 42 (97.67) | ||
| N1 | 2 (1.35) | 1 (0.95) | 1 (2.33) | ||
| M stage, n (%) | 0.291 | 0.218 | |||
| M0 | 147 (99.32) | 105 (100.00) | 42 (97.67) | ||
| M1 | 1 (0.68) | 0 (0.00) | 1 (2.33) |
EUS Endoscopic Ultrasonography, PNI/LVI perineural or lymphovascular invasion, SMD standardized mean difference
No statistically significant differences were observed between the two groups in positive margin rate, total medical costs, and total hospitalization days (P > 0.05). However, the postoperative hospitalization days of the LS group was significantly shorter than that of the ER group (2.00 days vs. 3.00 days, P < 0.001). In terms of safety, there were no postoperative complications in both groups (Table 3). In this subgroup, the median follow-up time was 45.5 months (range: 14–138 months), during which 7 cases of recurrence were observed: 2 cases were local rectal recurrence, 1 case was sigmoid colon recurrence, 3 cases were lymph node metastasis, and 1 case was combined lymph node and liver metastasis. The survival analysis showed that there was no significant difference in the recurrence outcome of the two groups of patients (P = 0.280) (Fig. 2). The estimated 5-year RFS rate was 97.8% (95% CI, 94.8%−100.0%) in the ER group and 91.5% (95% CI, 82.7%−100.0%) in the LS group.
Table 3.
Comparison of Postoperative Outcomes between Endoscopic Resection and Local Surgical Excision in Patients with 1–2 cm Tumors
| Variables | Total (n = 148) |
Endoscopic Resection (n = 105) |
Local Surgery (n = 43) |
P |
|---|---|---|---|---|
| Margin, n (%) | 0.757 | |||
| Negative | 135 (91.22) | 95 (90.48) | 40 (93.02) | |
| Positive | 13 (8.78) | 10 (9.52) | 3 (6.98) | |
| Salvage treatment, n (%) | > 0.999 | |||
| No | 12 (92.31) | 9 (90.00) | 3 (100.00) | |
| Yes | 1 (7.69) | 1 (10.00) | 0 (0.00) | |
| Total Cost (¥) | 0.626 | |||
| Median (Q₁, Q₃) |
15,721 (9,892, 22,232) |
16,144 (9,190, 21,886) |
13,719 (11,310, 25,573) |
|
| Total hospitalization days | 0.354 | |||
|
Median (Q₁, Q₃) |
4.00 (4.00, 5.00) |
4.00 (4.00, 5.00) |
4.00 (2.00, 7.00) |
|
| Postoperative hospitalization days | < 0.001* | |||
|
Median (Q₁, Q₃) |
3.00 (2.00, 4.00) |
3.00 (3.00, 4.00) |
2.00 (2.00, 3.00) |
|
| Complication, n (%) | —a | |||
| No | 148 (100.00) | 105 (100.00) | 43 (100.00) | |
| Bleeding | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| Fever | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| Obstruction | 0 (0.00) | 0 (0.00) | 0 (0.00) |
aChi-square test was not applicable as no positive events were observed in either group. *Statistically significant difference (P < 0.05)
Fig. 2.
Kaplan–Meier curve for recurrence-free survival comparing endoscopic resection and local surgical excision in patients with 1–2 cm tumors
Sensitivity analysis based on propensity score matching (PSM)
To further verify our findings and minimize selection bias, a propensity score matching (PSM) analysis was performed for the 1–2 cm subgroup. We matched patients in the LS group with those in the ER group at a 1:1 ratio. After matching, a total of 80 patients were included in the sensitivity analysis. No statistically significant differences in baseline characteristics were observed between the two groups. Most standardized mean differences (SMDs) were below 0.1, and all were below 0.2, which is generally considered acceptable thresholds for covariate balance (Table 4, Fig. S1). Consistent with the results of the primary analysis, LS group remained superior in terms of shorter postoperative hospitalization days (P < 0.001; Table 5). Furthermore, only four recurrence events were observed during the follow-up period (two in the ER group and two in the LS group). This corresponds to an identical incidence proportion of 5.00% (2/40) in both groups, yielding a risk difference of 0.00%. Given the extremely low number of recurrence events, the log-rank test was not performed.
Table 4.
Baseline Characteristics of Patients with 1–2 cm Tumors According to Resection Method After PSM
| Variables | Total (n = 80) |
Endoscopic Resection (n = 40) |
Local Surgery (n = 40) |
P | SMD |
|---|---|---|---|---|---|
| Sex, n (%) | > 0.999 | 0.000 | |||
| Male | 36 (45.00) | 18 (45.00) | 18 (45.00) | ||
| Female | 44 (55.00) | 22 (55.00) | 22 (55.00) | ||
| Age, n (%) | 0.884 | 0.111 | |||
| < 45 | 19 (23.75) | 9 (22.50) | 10 (25.00) | ||
| 45–59 | 24 (30.00) | 13 (32.50) | 11 (27.50) | ||
| ≥ 60 | 37 (46.25) | 18 (45.00) | 19 (47.50) | ||
| Reason for presentation, n (%) | 0.813 | 0.053 | |||
| Incidental | 53 (66.25) | 27 (67.50) | 26 (65.00) | ||
| Symptomatic | 27 (33.75) | 13 (32.50) | 14 (35.00) | ||
| EUS evaluation, n (%) | 0.813 | 0.053 | |||
| No | 53 (66.25) | 26 (65.00) | 27 (67.50) | ||
| Yes | 27 (33.75) | 14 (35.00) | 13 (32.50) | ||
| Grade, n (%) | > 0.999 | 0.000 | |||
| G1 | 70 (87.50) | 35 (87.50) | 35 (87.50) | ||
| G2 | 10 (12.50) | 5 (12.50) | 5 (12.50) | ||
| G3 | 0 (0.00) | 0 (0.00) | 0 (0.00) | ||
| PNI/LVI, n (%) | > 0.999 | 0.089 | |||
| Negative | 73 (91.25) | 37 (92.50) | 36 (90.00) | ||
| Positive | 7 (8.75) | 3 (7.50) | 4 (10.00) | ||
| T stage, n (%) | > 0.999 | 0.132 | |||
| T1 | 77 (96.25) | 38 (95.00) | 39 (97.50) | ||
| T2 | 3 (3.75) | 2 (5.00) | 1 (2.50) | ||
| T3 + T4 | 0 (0.00) | 0 (0.00) | 0 (0.00) | ||
| N stage, n (%) | > 0.999 | 0.000 | |||
| N0 | 78 (97.50) | 39 (97.50) | 39 (97.50) | ||
| N1 | 2 (2.50) | 1 (2.50) | 1 (2.50) | ||
| M stage, n (%) | > 0.999 | 0.000 | |||
| M0 | 80 (100.00) | 40 (100.00) | 40 (100.00) | ||
| M1 | 0 (0.00) | 0 (0.00) | 0 (0.00) |
EUS Endoscopic Ultrasonography, PNI/LVI perineural or lymphovascular invasion, SMD standardized mean difference.
Table 5.
Comparison of Postoperative Outcomes between Endoscopic Resection and Local Surgical Excision in Patients with 1–2 cm Tumors After PSM
| Variables | Total (n = 80) |
Endoscopic Resection (n = 40) |
Local Surgery (n = 40) |
P |
|---|---|---|---|---|
| Margin, n (%) | > 0.999 | |||
| Negative | 73 (91.25) | 36 (90.00) | 37 (92.50) | |
| Positive | 7 (8.75) | 4 (10.00) | 3 (7.50) | |
| Salvage treatment, n (%) | —a | |||
| No | 7 (100.00) | 4 (100.00) | 3 (100.00) | |
| Yes | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| Total Cost (¥) | 0.708 | |||
|
Median (Q₁, Q₃) |
17,845 (11,310, 24,684) |
19,063 (10,410, 24,477) |
13,937 (11,310, 25,573) |
|
| Total hospitalization days | 0.351 | |||
|
Median (Q₁, Q₃) |
5.00 (4.00, 6.00) |
5.00 (4.00, 5.00) |
4.00 (2.00, 7.00) |
|
| Postoperative hospitalization days | < 0.001* | |||
|
Median (Q₁, Q₃) |
3.00 (2.00, 4.00) |
4.00 (3.00, 4.00) |
2.00 (2.00, 3.00) |
|
| Complication, n (%) | —a | |||
| No | 80 (100.00) | 40 (100.00) | 40 (100.00) | |
| Bleeding | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| Fever | 0 (0.00) | 0 (0.00) | 0 (0.00) | |
| Obstruction | 0 (0.00) | 0 (0.00) | 0 (0.00) |
aChi-square test was not applicable as no positive events were observed in either group. *Statistically significant difference (P < 0.05)
Discussion
This study systematically analyzed the clinical characteristics of 736 cases with R-NETs, focusing particularly on clinical outcomes in cases with 1–2 cm tumors, revealing the indications and prognostic significance of different treatment strategies. According to the current guidelines of the European Society of Neuroendocrine Oncology (ENETS), tumor size is a key factor in selecting treatment modalities for R-NETs [12]. For tumors with a diameter of less than 1 cm, G1 and limited to the submucosal layer, the risk of lymphatic vascular invasion is extremely low, and endoscopic resection has been widely accepted as a standard treatment. For tumors with a diameter of more than 2 cm or tumors with high-risk characteristics such as muscle invasion, deep infiltration or lymph node metastasis, radical surgery is recommended [12, 21]. For controversial 1–2 cm tumors, several studies have suggested that more aggressive surgical approaches may be considered in selected patients with high-risk features, including muscularis invasion or nodal involvement [22, 23]. Conversely, other studies indicate that for 1–2 cm tumors without high-risk features, endoscopic resection can achieve favorable long-term outcomes, with salvage surgery considered if pathology reveals positive margins [15, 24].
In this study cohort, the size of the tumor also significantly affected the treatment decision-making. ER dominates in the tumor diameter < 1 cm subgroup, and RS was the main treatment of the subgroup with a diameter ≥ 2 cm. For medium-sized 1–2 cm tumors, the treatment choice was more subjective. Our data revealed that T2/T3 invasion and G3 histology were exclusively observed in the RS group, whereas the ER and LS groups overwhelmingly comprised T1 and G1 tumors, suggesting that beyond tumor size, pathological grade and depth of invasion should also be taken into consideration. Indeed, the comparably low recurrence rates between ER and LS further support this notion: for 1–2 cm R-NETs with G1, T1, negative LVI, and N0 status by imaging, both local resection modalities achieve adequate tumor control. Regarding the shorter postoperative hospital stay associated with LS (median 2 vs. 3 days), although statistically significant, this modest difference may largely reflect center-specific perioperative protocols, and its generalizability requires further validation. Taken together, we propose the following stratified approach, in which ER or LS is appropriate for G1, T1, negative LVI, N0 lesions; radical resection is indicated for T2 or deeper invasion, confirmed nodal metastasis, or G3; and patients with G2, positive LVI, or suspicious lymphadenopathy warrant multidisciplinary evaluation, with a low threshold for radical resection if any of these features are confirmed. Notably, the long-term recurrence free survival rate of LS and ER is similar, which is consistent with the research results of Jin et al. and Son et al. [13, 25]. Compared to ER, the LS group was associated with a shorter postoperative hospital stay, while postoperative complication rates and recurrence events were low in both groups during follow-up. This difference may be mainly related to the procedure type and perioperative workflow at our center. Specifically, in this study, LS referred to transanal local procedures, which also represent minimally invasive local treatment with limited procedural trauma. In addition, patients in the ER group were usually managed with a more conservative postoperative protocol, including temporary fasting, prophylactic treatment, and monitoring for delayed bleeding or perforation. In contrast, patients in the LS group followed a shorter postoperative pathway and could usually be discharged earlier if no obvious discomfort or complications occurred. Differences in perioperative course have also been variably reported in previous studies [13, 15, 26, 27]. After PSM in the 1–2 cm subgroup, recurrence events were rare and occurred in equal numbers in both groups. Therefore, these observations should be considered descriptive, and further validation in larger cohorts is needed to clarify long-term oncological outcomes.
In addition, the relatively low proportion of patients undergoing radical surgery with lymph node dissection in the 1–2 cm subgroup was related to the absence of high-risk features on preoperative evaluation. Most tumors in this subgroup did not demonstrate radiologic or endoscopic evidence of nodal involvement or deep invasion. However, two patients in the ER/LS groups had clinically suspected lymph node metastasis before treatment. One patient recorded in the ER group had suspicious lymph node metastasis on initial imaging at local hospital but underwent ESD, and subsequently received radical surgery at our hospital, with postoperative pathology confirming lymph node metastasis. The other patient, recorded in the LS group, refused radical surgery despite suspicious lymph node metastasis on imaging. Thus, these two cases were managed under specific clinical circumstances rather than as routine treatment choices.
A noteworthy finding in our cohort was the disparity in preoperative endoscopic ultrasound (EUS) utilization. In the overall cohort, 86.11% of patients in the ER group underwent EUS examination, whereas the corresponding proportion was only 12.15% in the LS group. A similar trend was also observed in the 1–2 cm subgroup (71.67% vs. 21.67%). In fact, EUS is a key tool for assessing the depth of tumor infiltration and lymph node status [28], and the ENETS guidelines also recommend the application of EUS [12]. The observed difference in EUS utilization between groups may be related to variations in referral pathways, clinical workflow, or temporal practice patterns. Therefore, further standardization of preoperative evaluation strategies may help optimize staging accuracy and treatment selection, particularly for tumors larger than 1 cm.
There are some limitations in this study, which may lead to potential deviations in the research results. First, the single-center patient cohort leads to selection bias, especially fewer cases with a tumor diameter > 1 cm or those with muscularis propria invasion, which reflects improved early diagnosis in China. Second, the extremely low number of recurrence events in both the overall 1–2 cm subgroup (n = 7) and the matched cohort (n = 4) limits the statistical power of Kaplan–Meier and log-rank analyses. This should be considered when interpreting these survival results. Third, although the maximum follow-up reached 138 months and 5-year RFS was estimated using Kaplan–Meier analysis, the median follow-up time of 45.5 months may still be insufficient to fully capture late recurrences in this indolent disease. Continued long-term follow-up and larger multicenter studies are warranted to further clarify long-term oncological outcomes. Finally, the 1–2 cm subgroup had unbalanced sample sizes and limited G2 cases, which restricted the analysis based on tumor grade. These limitations highlight the important factors that need to be considered when interpreting research results and designing future research.
Supplementary Information
Below is the link to the electronic supplementary material.
Abbreviations
- NENs
Neuroendocrine neoplasms
- NETs
Neuroendocrine tumors
- NECs
Neuroendocrine carcinomas
- R-NETs
Rectal neuroendocrine tumors
- ER
Endoscopic resection
- LS
Local surgery
- RS
Radical surgery
- ESD
Endoscopic submucosal dissection
- EMR
Endoscopic mucosal resection
- APC
Argon plasma coagulation
- TAMIS
Transanal minimally invasive surgery
- TEM
Transanal endoscopic microsurgery
- LAR
Low anterior resection
- EUS
Endoscopic Ultrasonography
- PNI/LVI
Perineural or lymphovascular invasion
- RFS
Recurrence free survival
- IQR
Interquartile ranges
- PSM
Propensity Score Matching
Authors contributions
Concept and design: Hua Xiong, Wan Du; Material preparation, data collection and analysis: Wanrong Zhou, Qi Cheng, Xiaolin Lin, Xialu Hong, Ziyu Tai; Follow-up: Wanrong Zhou, Xiaolin Lin; Writing—original draft preparation: Wanrong Zhou; Writing—review and editing: Hua Xiong, Xiuying Xiao, Wan Du; Funding acquisition: Hua Xiong; Supervision: Hua Xiong, Xiuying Xiao, Wan Du.
Funding
This work was supported by grants from National Natural Science Foundation of China (82472898).
Data availability
The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.
Declarations
Ethical approval
This study has approved by the Medical Ethics Committee of Renji Hospital (Approval NO. LY2024-048-B).
Conflict of interest
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.
Wanrong Zhou and Qi Cheng These authors contributed equally.
Contributor Information
Wan Du, Email: duwan2017@126.com.
Xiuying Xiao, Email: xiaoxiuying2002@163.com.
Hua Xiong, Email: xionghua@renji.com.
References
- 1.Cives M, Strosberg JR (2018) Gastroenteropancreatic neuroendocrine tumors. CA Cancer J Clin 68:471–487. 10.3322/caac.21493 [DOI] [PubMed] [Google Scholar]
- 2.Rindi G, Mete O, Uccella S et al (2022) Overview of the 2022 WHO Classification of Neuroendocrine Neoplasms. Endocr Pathol 33:115–154. 10.1007/s12022-022-09708-2 [DOI] [PubMed] [Google Scholar]
- 3.Taal BG, Visser O (2004) Epidemiology of neuroendocrine tumours. Neuroendocrinology 80:3–7. 10.1159/000080731 [DOI] [PubMed] [Google Scholar]
- 4.Maione F, Chini A, Milone M et al (2021) Diagnosis and management of rectal neuroendocrine tumors (NETs). Diagnostics 11:771. 10.3390/diagnostics11050771 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fan J-H, Zhang Y-Q, Shi S-S et al (2017) A nation-wide retrospective epidemiological study of gastroenteropancreatic neuroendocrine neoplasms in China. Oncotarget 8:71699–71708. 10.18632/oncotarget.17599 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lim C-H, Lee IS, Jun BY et al (2017) Incidence and clinical characteristics of gastroenteropancreatic neuroendocrine tumor in Korea: a single-center experience. Korean J Intern Med 32:452–458. 10.3904/kjim.2015.232 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.JNETS Project Study Group, Masui T, Ito T et al (2020) Recent epidemiology of patients with gastro-entero-pancreatic neuroendocrine neoplasms (GEP-NEN) in Japan: a population-based study. BMC Cancer 20:1104. 10.1186/s12885-020-07581-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Scherübl H (2009) Rectal carcinoids are on the rise: early detection by screening endoscopy. Endoscopy 41:162–165. 10.1055/s-0028-1119456 [DOI] [PubMed] [Google Scholar]
- 9.Li Y-W, He Y-P, Liu F-Q et al (2021) Grade G2 rectal neuroendocrine tumor is much more invasive compared with G1 tumor. Front Oncol 11:646536. 10.3389/fonc.2021.646536 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sohn B, Kwon Y, Ryoo S-B et al (2017) Predictive factors for lymph node metastasis and prognostic factors for survival in rectal neuroendocrine tumors. J Gastrointest Surg 21:2066–2074. 10.1007/s11605-017-3603-y [DOI] [PubMed] [Google Scholar]
- 11.Frydman A, Srirajaskanthan R (2024) An update on the management of rectal neuroendocrine neoplasms. Curr Treat Options Oncol 25:1461–1470. 10.1007/s11864-024-01267-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rinke A, Ambrosini V, Dromain C et al (2023) European Neuroendocrine Tumor Society (ENETS) 2023 guidance paper for colorectal neuroendocrine tumours. J Neuroendocrinol 35:e13309. 10.1111/jne.13309 [DOI] [PubMed] [Google Scholar]
- 13.Jin R, Bai X, Xu T et al (2023) Comparison of the efficacy of endoscopic submucosal dissection and transanal endoscopic microsurgery in the treatment of rectal neuroendocrine tumors ≤ 2 cm. Front Endocrinol 13:1028275. 10.3389/fendo.2022.1028275 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Brand M, Reimer S, Reibetanz J et al (2021) Endoscopic full thickness resection vs. transanal endoscopic microsurgery for local treatment of rectal neuroendocrine tumors - a retrospective analysis. Int J Colorectal Dis 36:971–976. 10.1007/s00384-020-03800-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Park SS, Kim BC, Lee D et al (2021) Comparison of endoscopic submucosal dissection and transanal endoscopic microsurgery for T1 rectal neuroendocrine tumors: a propensity score-matched study. Gastrointest Endosc 94:408-415.e2. 10.1016/j.gie.2021.02.012 [DOI] [PubMed] [Google Scholar]
- 16.Feng Y, Liu L (2024) Endoscopic therapy versus surgical resection for elderly patients with colorectal neuroendocrine tumors. Surg Endosc 38:6356–6367. 10.1007/s00464-024-11220-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fields AC, Saadat LV, Scully RE et al (2019) Local excision versus radical resection for 1- to 2-cm neuroendocrine tumors of the rectum: a national cancer database analysis. Dis Colon Rectum 62:417–421. 10.1097/DCR.0000000000001210 [DOI] [PubMed] [Google Scholar]
- 18.Chen Q, Chen J, Huang Z et al (2022) Comparable survival benefit of local excision versus radical resection for 10- to 20-mm rectal neuroendocrine tumors. Eur J Surg Oncol 48:864–872. 10.1016/j.ejso.2021.10.029 [DOI] [PubMed] [Google Scholar]
- 19.Pattarajierapan S, Khomvilai S (2022) Recurrence after endoscopic resection of small rectal neuroendocrine tumors: a retrospective cohort study. Ann Coloproctol 38:216–222. 10.3393/ac.2021.00017.0002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chung HG, Goh MJ, Kim ER et al (2021) Recurrence pattern and surveillance strategy for rectal neuroendocrine tumors after endoscopic resection. J Gastroenterol Hepatol 36:968–973. 10.1111/jgh.15231 [DOI] [PubMed] [Google Scholar]
- 21.Ma X-X, Wang L-S, Wang L-L et al (2023) Endoscopic treatment and management of rectal neuroendocrine tumors less than 10 mm in diameter. World J Gastrointest Endosc 15:19–31. 10.4253/wjge.v15.i2.19 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Choi JS, Kim MJ, Shin R et al (2024) Risk factor analysis of lymph node metastasis for rectal neuroendocrine tumors: who needs a radical resection in rectal neuroendocrine tumors sized 1–2 cm? Ann Surg Oncol 31:2414–2424. 10.1245/s10434-023-14829-x [DOI] [PubMed] [Google Scholar]
- 23.Li C, Yu M, Liu W et al (2024) Long-term outcomes of 1–2 cm rectal neuroendocrine tumors after local excision or radical resection: a population-based multicenter study. Heliyon 10:e28335. 10.1016/j.heliyon.2024.e28335 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sui Q, Lin J, Peng J et al (2017) Local surgical excision versus endoscopic resection for rectal carcinoid: a meta-analysis. J Cancer 8:3969–3974. 10.7150/jca.21476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Son J, Park IJ, Yang D-H et al (2022) Oncological outcomes according to the treatment modality based on the size of rectal neuroendocrine tumors: a single-center retrospective study. Surg Endosc 36:2445–2455. 10.1007/s00464-021-08527-6 [DOI] [PubMed] [Google Scholar]
- 26.Yu JX, Russell WA, Ching JH et al (2019) Cost effectiveness of endoscopic resection vs transanal resection of complex benign rectal polyps. Clin Gastroenterol Hepatol 17:2740-2748.e6. 10.1016/j.cgh.2019.02.041 [DOI] [PubMed] [Google Scholar]
- 27.Nam MJ, Sohn DK, Hong CW et al (2015) Cost comparison between endoscopic submucosal dissection and transanal endoscopic microsurgery for the treatment of rectal tumors. Ann Surg Treat Res 89:202. 10.4174/astr.2015.89.4.202 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Park SB, Kim DJ, Kim HW et al (2017) Is endoscopic ultrasonography essential for endoscopic resection of small rectal neuroendocrine tumors? WJG 23:2037. 10.3748/wjg.v23.i11.2037 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not openly available due to reasons of sensitivity and are available from the corresponding author upon reasonable request.