Clinical outcomes of radiation therapy as an alternative to radical surgery for early-stage rectal cancer

Abstract

Purpose

Total mesorectal excision (TME) is the standard treatment for early rectal cancer, but radiation therapy (RT) with or without chemotherapy may be considered as an alternative organ-preserving option. This study evaluated the oncologic outcomes, patterns of failure, salvage treatment efficacy, and TME–free survival (TME-FS) following RT without radical surgery after local excision or biopsy in selected patients.

Materials and Methods

We retrospectively analyzed 40 patients with cT1–2N0 rectal adenocarcinoma who received RT after local excision or biopsy without subsequent radical surgery between 2007 and 2023. Indications for RT included clinical T1 disease with high-risk features, clinical T2 disease, or tumors located close to the anal verge. The median RT dose was 50.2 Gy, and 29 patients received chemotherapy. Survival outcomes, recurrence patterns, salvage treatments, toxicities, and TME-FS were assessed.

Results

The median follow-up was 38 months. The 5-year overall survival rates were 87.5% for the cT1 group and 85.6% for the cT2 group, and the 5-year cancer-specific survival rate was 100% in both groups. The 5-year recurrence-free survival rates were 79.3% and 62.8%, respectively. Overall, nine patients experienced treatment failure, and four patients ultimately underwent salvage TME. The 5-year TME-FS were 87.5% in the cT1 group and 78.6% in the cT2 group. No grade 2 or higher late toxicities were observed.

Conclusion

RT without radical surgery may be a feasible treatment option for cT1–2N0 rectal cancer patients who are unsuitable for, or decline, radical surgery. Close surveillance and timely salvage treatment remain essential, particularly for those with cT2 disease.

Introduction

While total mesorectal excision (TME) remains the standard treatment for early rectal cancer, it carries a risk of significant morbidity, including anastomotic leakage, genitourinary or sexual dysfunction, and the potential necessity of a permanent stoma—all of which can compromise patient quality of life and functional outcomes [1,2]. Consequently, there has been an increasing interest in organ-preserving strategies. Local excision, such as transanal excision (TAE) or endoscopic resection, offers advantages like lower procedural morbidity and superior preservation of anorectal function [1,3].

However, local excision alone is linked to higher locoregional recurrence rates and inferior oncologic outcomes, particularly in patients with high-risk features. These high-risk features include positive resection margins, lymphovascular invasion (LVI), poorly differentiated histology, and deep submucosal invasion [2-5]. Therefore, adding radiation therapy (RT), with or without chemotherapy, may be considered as an alternative organ-preserving treatment option in selected patients [5-7].

To date, limited data exist regarding the efficacy of RT as an alternative to completion radical surgery in this clinical setting. Furthermore, the feasibility of RT without radical surgery, the potential for effective salvage treatment, and the long-term success of TME avoidance have not been sufficiently investigated [5,8]. Therefore, we retrospectively evaluated the oncologic outcomes, patterns of failure, and feasibility of salvage treatment in patients with cT1–2N0 early-stage rectal cancer who were unsuitable for or declined radical surgery. Additionally, we analyzed TME–free survival (TME-FS) to assess the long-term viability of this organ-preserving strategy.

Materials and Methods

1. Patients

This retrospective study included patients diagnosed with cT1–T2N0 rectal adenocarcinoma who received RT, with or without concurrent chemotherapy, after local excision or biopsy without subsequent radical surgery at Yonsei Cancer Center and Gangnam Severance Hospital between 2007 and 2023. Patients with carcinoma in situ, double primary cancers, distant metastasis at diagnosis, or a history of prior pelvic RT were excluded. Consequently, a total of 40 patients were included in the current study. Clinical data were retrieved from electronic medical records and a departmental database, with ethical approval granted by the Institutional Review Board (IRB number: 9-2025-0142).

Regarding the initial local procedures, 18 patients (45.0%) underwent TAE, while 22 patients (55.0%) underwent endoscopic procedures, including endoscopic mucosal resection (EMR), endoscopic submucosal dissection (ESD), polypectomy, or biopsy alone. In this study, local excision modalities were categorized as TAE versus other approaches. TAE was considered a surgical full-thickness excision, whereas EMR and ESD were classified separately because they are endoscopic mucosal or submucosal resections rather than full-thickness excisions, despite involving actual tumor removal.

High-risk features were defined as the presence of at least one of the following factors: positive resection margin, LVI, poorly differentiated histology, or deep submucosal invasion (sm3; invasion to the lower third of the submucosa). Patients who underwent biopsy only could not be assessed for pathologic high-risk features; instead, their indication for treatment was clinical T2 disease or a tumor located close to the anal verge. All patients had one of the following indications for further treatment: clinical T1 disease with high-risk features (22 patients, 55.0%), clinical T2 disease (16 patients, 40.0%), or clinical T1 disease without other high-risk features but with tumors located close to the anal verge (2 patients, 5.0%). Consequently, all patients were initially considered candidates for additional transabdominal resection. However, they were deemed unsuitable for surgery due to significant medical comorbidities or declined surgery based on personal preference. Therefore, RT, with or without concurrent chemotherapy, was administered as an alternative to radical surgery in this cohort.

Clinical T stage was determined through a comprehensive assessment of pelvic magnetic resonance imaging (MRI) and/or pathologic findings from biopsy, endoscopic resection, or local excision. Clinical N stage was evaluated with abdominopelvic computed tomography (CT), pelvic MRI, or positron emission tomography–computed tomography (PET-CT). Patients were classified as cN0 when no radiologically suspicious lymph node enlargement was detected on imaging studies. After RT, patient follow-up was performed at the discretion of the physician, generally at 3- to 6-month intervals. Follow-up evaluations included physical examination, serum carcinoembryonic antigen (CEA), endoscopic evaluation, and imaging studies such as pelvic MRI, abdominopelvic CT, and PET-CT. Recurrence was assessed primarily based on imaging findings and interpreted with serum CEA levels together. The follow-up interval and surveillance modalities were not standardized within this cohort because of the retrospective nature of the study.

2. Treatment

All patients were treated with RT without radical surgery with a median total dose of 50.2 Gy. Specifically, 39 patients (97.5%) received a total dose of 45.0–59.4 Gy in 25–33 fractions, and one patient (2.5%) received short-course RT with 27.5 Gy in 5 fractions. RT was performed with 3D-conformal radiation therapy in 13 patients (32.5%) or intensity-modulated radiation therapy in 27 patients (67.5%). The clinical target volume included the rectal tumor bed and mesorectum with margins, pelvic lymph nodes, with or without the inguinal lymph node area, depending on tumor location [9]. In one patient, only the mesorectum was treated at the physician's discretion. Concurrent chemotherapy was administered to 29 patients (72.5%). Of these, 28 patients received 5-fluorouracil (5-FU)–based regimen, including intravenous 5-FU in 12 patients (30.0%), oral capecitabine in 15 patients (37.5%), and doxyfluridine in one patient (2.5%). The remaining one patient (2.5%) received the FOLFOX (folinic acid, fluorouracil, and oxaliplatin) regimen.

3. Endpoints

Overall survival (OS) was defined as the time from the start of RT to death from any cause or last follow-up. Cancer-specific survival (CSS) was defined as the time from the start of RT to the date of death from rectal cancer or the last follow-up. Recurrence-free survival (RFS) was defined as the time from the start of RT to the date of first treatment failure or last follow-up. Locoregional recurrence–free survival (LRFS) was defined as the time from the start of RT to the occurrence of locoregional recurrence or last follow-up. Distant metastasis-free survival (DMFS) was defined as survival without distant metastasis from the start date of RT to the last follow-up. TME-FS was defined as the time from the start of RT to the date of salvage TME operation after cancer recurrence or the last follow-up. Patients without events were censored at the date of last follow-up. If a patient was lost to follow-up after recurrence, recurrence-related endpoints were recorded as events at the date of recurrence, and survival endpoints were censored at the date of last clinical contact. Deaths from causes other than rectal cancer were censored in CSS analysis.

4. Statistical analysis

All statistical analyses were performed using SPSS, version 29.0 (IBM Corp., Armonk, NY, USA). Survival rates were calculated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. To identify prognostic factors, univariate analyses were performed using a Cox proportional hazards model. Hazard ratios and 95% confidence intervals were reported. Toxicities were defined according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0.

Results

1. Patient characteristics

A total of 28 patients (70.0%) were male, and 12 patients (30.0%) were female, with a median age of 70.5 years (range, 35 to 93) (Table 1). The median tumor size was 1.95 cm (range, 0.4 to 8.0), and the median distance from the anal verge was 4 cm (range, 0 to 12). The clinical T stage was classified as cT1 in 24 patients (60.0%), and cT2 in 16 patients (40.0%). Regarding pathologic features, 24 patients (60.0%) had positive resection margins, eight patients (20.0%) had LVI, and two patients (5.0%) presented with poorly differentiated tumors. In terms of the initial procedures, TAE was performed in 18 patients (45.0%), nine patients (22.5%) underwent EMR or ESD, 11 patients (27.5%) underwent biopsy only, and two patients (5.0%) underwent polypectomy. All patients received RT without radical surgery, with a median total dose of 50.2 Gy (range, 27.5 to 59.4), and 29 patients (72.5%) received concurrent chemotherapy.

Table 1.

Patients characteristics

Characteristic	No. of patients (%) (n = 40)
Sex
Male	28 (70.0)
Female	12 (30.0)
Age (year)
Median	70.5
Range	35–93
ECOG PS
0	8 (20.0)
1	32 (80.0)
CEA (ng/mL)
Median	2.165
Range	0.69–27.56
Tumor size (cm)
Median	1.95
Range	0.4–8.0
Tumor location (cm from AV)
Median	4
Range	0–12
<5 cm	22 (55.0)
≥5 cm	17 (42.5)
Unknown	1 (2.5)
Clinical T stage
cT1	24 (60.0)
cT2	16 (40.0)
Local excision
Endoscopic mucosal resection	3 (7.5)
Endoscopic submucosal dissection	6 (15.0)
Transanal excision	18 (45.0)
Biopsy only	11 (27.5)
Polypectomy	2 (5.0)
Depth of SM invasion (mm)
Median	2
Range	0.2–6.0
<1 mm	3 (7.5)
≥1 mm	16 (40.0)
Invades muscularis propria	6 (15.0)
Unknown	15 (37.5)
Differentiation
Well differentiated	10 (25.0)
Moderately differentiated	26 (65.0)
Poorly differentiated	2 (5.0)
Unknown	2 (5.0)
Lymphovascular invasion
No	19 (47.5)
Yes	8 (20.0)
Unknown	13 (32.5)
Perineural invasion
No	20 (50.0)
Yes	0 (0)
Unknown	20 (50.0)
Resection margin
Negative	3 (7.5)
Positive	24 (60.0)
Unknown	13 (32.5)
Radiation therapy total dose (Gy)
Median	50.2
Range	27.5–59.4
Concurrent chemotherapy
No chemotherapy	11 (27.5)
5-FU	12 (30.0)
Xeloda	15 (37.5)
Fluridin	1 (2.5)
FOLFOX	1 (2.5)

ECOG PS, Eastern Cooperative Oncology Group performance status; CEA, carcinoembryonic antigen; AV, anal verge; SM, submucosal; 5-FU, 5-fluorouracil; FOLFOX, folinic acid, fluorouracil, and oxaliplatin.

2. Survival analysis

The median follow-up was 38 months (range, 2 to 162). OS, CSS, RFS, LRFS, and DMFS are presented in Fig. 1A–E. The 3- and 5-year OS were 89.7% and 86.5%, respectively, and the CSS was 100%. The 3- and 5-year RFS, LRFS, DMFS, and TME-FS were 79.7% and 71.8%, 88.2% and 80.2%, 91.3% and 86.5%, and 90.8% and 84.8%, respectively.

Fig. 1.

Survival and recurrence outcomes according to clinical T stage: (A) overall survival, (B) cancer-specific survival, (C) recurrence-free survival, (D) locoregional recurrence–free survival, and (E) distant metastasis-free survival.

When stratified by clinical T stage, patients with cT2 disease showed lower survival rates than those with cT1 disease, although the differences were not statistically significant. The 5-year OS were 87.5% in the cT1 group and 85.6% in the cT2 group, and the 5-year CSS was 100% in both groups. The 5-year RFS were 79.3% in the cT1 group and 62.8% in the cT2 group, the 5-year LRFS were 88.4% and 70.2%, and the 5-year DMFS were 90.4% and 82.5%, respectively. Stage-specific survival curves are shown in Fig. 1.

Univariate Cox proportional hazard analyses were performed to evaluate potential prognostic factors, including age, sex, tumor location, clinical T stage, LVI, type of local excision, and chemotherapy status (Table 2). Resection margin status was excluded because only three patients had a negative margin, while 24 patients had a positive and 13 patients had unknown margin status. Multivariate analysis was not performed due to the limited number of events. None of the evaluated variables were significantly associated with either LRFS or RFS.

Table 2.

Prognostic factors related to LRFS and RFS

	LRFS		RFS
	HR (95% CI)	p-value	HR (95% CI)	p-value
Age (year)
<70	Ref		Ref
≥70	0.533 (0.098–2.913)	0.468	0.292 (0.060–1.406)	0.125
Sex
Male	Ref		Ref
Female	2.085 (0.420–10.359)	0.369	3.002 (0.804–11.201)	0.102
Tumor locationa)
<5 cm	Ref		Ref
≥5 cm	0.629 (0.115–3.451)	0.593	0.346 (0.071–1.681)	0.188
Clinical T stage
T1	Ref		Ref
T2	2.874 (0.525–15.739)	0.224	1.848 (0.495–6.895)	0.361
LVI
Negative	Ref		Ref
Positive	0.803 (0.083–7.788)	0.850	0.457 (0.053–3.950)	0.477
Unknown	1.101 (0.182–6.661)	0.917	0.962 (0.226–4.082)	0.958
Local excision
TAE	Ref		Ref
Others	1.945 (0.356–10.632)	0.443	0.726 (0.194–2.707)	0.633
Chemotherapy
No	Ref		Ref
Yes	0.628 (0.114–3.451)	0.593	0.598 (0.148–2.413)	0.470

LRFS, locoregional recurrence–free survival; RFS, recurrence-free survival; HR, hazard ratio; CI, confidence interval; Ref, reference; LVI, lymphovascular invasion; TAE, transanal excision.

^a)One patient with unknown tumor location was excluded from the analysis.

TME-FS is presented in Fig. 2. The 5-year TME-FS were 87.5% in the cT1 group and 78.6% in the cT2 group (p = 0.184). Overall, four patients (10.0%) ultimately underwent salvage TME surgery after recurrence, consisting of one patient in the cT1 group and three patients in the cT2 group (Table 3).

Fig. 2.

Total mesorectal excision–free survival according to clinical T stage.

Table 3.

Characteristics of patients with treatment failure

Sex	Age (year)	cT stage	Excision	Resection Margin	RT dose (Gy)	Disease-free interval (month)	Failure site	Salvage treatment	Months after recurrence	Last status
Female	64	T1	TAE	0.5 mm	54	24	Lung, liver	FOLFIRI + metastasectomy	131	NED
Female	58	T1	TAE	FFT	45	25	Lt. lateral wall, mesorectal region LN, EMVI	Follow-up lost	15	Follow-up lost
Male	68	T1	ESD	Positive	50.4	41	Upper anal canal	APR + FOLFOX	25	Alive with disease
Male	55	T1	TAE	Unknown	50.4	17	Inguinal LN	CCRT (5-FU) + local RT to inguinal LN	3	Alive with disease
Male	71	T2	Biopsy only		50	10	Rectum (local failure)	APR	26	NED
Female	77	T2	Biopsy only		59.4	21	Rectum (local failure)	Ultra LAR	16	Intercurrent death
Male	58	T2	TAE	FFT	50.4	50	Inguinal LN, external and internal iliac chain LN, lung, scrotum subcutaneous lesion, axillary LN	A-FOLFIRI + local RT to inguinal LN	10	Died of disease
Female	60	T2	Biopsy only	0.8 mm	50	33	Rectum (Local failure)	APR + A-FOLFOX	33	Died of disease
Female	56	T2	TAE	0.3 mm	50.4	30	Lung	Metastasectomy	122	NED

RT, radiation therapy; TAE, transanal excision; FOLFIRI, folinic acid, fluorouracil, and irinotecan; NED, no evidence of disease; FFT, free from tumor; LN, lymph node; EMVI, extramural venous invasion; ESD, endoscopic submucosal dissection; APR, abdominoperineal resection; FOLFOX, folinic acid, fluorouracil, and oxaliplatin; CCRT, concurrent chemoradiotherapy; 5-FU, 5-fluorouracil; LAR, low anterior resection.

3. Patterns of failure

Patterns of failure stratified by clinical T stage are shown in Fig. 3 and detailed characteristics of patients with treatment failure are listed in Table 3. Failure patterns were categorized as local failure, regional failure, and distant metastasis. Local failure was defined as recurrence within the rectum or mesorectal area, whereas regional failure included recurrence within the pelvic nodal chains. Overall, nine patients (22.5%) experienced treatment failure; four failures in cT1 group, and five in cT2 group. In the cT1 group (Fig. 3A), local recurrence alone was observed in one patient, locoregional recurrence in one patient, and distant metastasis in two patients. In the cT2 group (Fig. 3B), local recurrence alone was the most common failure pattern, observed in three patients. Distant metastasis alone was observed in one patient, and one patient developed simultaneous locoregional and distant failure involving multiple sites.

Fig. 3.

Patterns of failure in patients with (A) clinical T1 disease and (B) clinical T2 disease.

Among patients with recurrence, systemic therapy, metastasectomy, and salvage TME were used as salvage treatment. At the last follow-up, three patients had no evidence of disease, two were alive with disease, one was alive with unknown disease status, two had died of disease, and one had died of an intercurrent cause (Table 3).

4. Toxicity

Overall, 26 patients (65.0%) experienced acute toxicities, which were predominantly low grade. Grade 1 toxicities, including anal pain, cystitis, diarrhea, and proctitis, were observed in 23 patients, while only three patients (7.5%) developed grade 3 perianal skin reactions. There were no grade 2 or higher late toxicities beyond 3 months after the completion of RT.

Discussion and Conclusion

In this study, we demonstrated that RT without radical surgery with or without concurrent chemotherapy following local excision is a feasible organ-preserving alternative to radical surgery for cT1–2N0 rectal cancer patients who were medically unsuitable for or declined radical surgery. This approach was associated with acceptable disease control, favorable CSS, and a high probability of avoiding subsequent TME. Importantly, our findings highlight that organ preservation should be assessed not only by the initial recurrence rate, but also in terms of salvageability and long-term avoidance of radical surgery.

Previous studies have shown that local excision alone is associated with a higher local recurrence rate compared to conventional radical resection. Up to 20% of patients with clinical T0-2 rectal cancer may harbor occult lymph node metastasis, which can lead to recurrence after treatment with local excision alone [10]. The American Society of Colon and Rectal Surgeons guideline reports that the local recurrence rates following local excision vary from 7%–21% for T1 lesions and from 26%–47% for T2 lesions, and recommends radical surgical resection for the patients with adverse pathologic features [11]. In a retrospective study, Al-Sawat et al. [12] reported that local excision with adjuvant RT significantly reduces the risk of complications and stoma formation, without significant differences in OS or disease-free survival (DFS) in high-risk T1 rectal cancer patients. Furthermore, a meta-analysis reported that adjuvant (chemo)radiation therapy could serve as an organ-preserving alternative to completion TME for high-risk early T1–2 rectal cancer following transanal removal, with a local recurrence rate of 5%, which was similar to the rate after TME (4%) [13]. However, a single-institution retrospective study by Lee et al. reported no significant difference in 5-year RFS and 5-year OS between patients treated with adjuvant RT after local excision and local excision alone in early rectal cancer patients with deep submucosal invasion [14]. Taken together, previous evidence indicates that local excision alone may be associated with a substantial risk of recurrence in patients with high-risk features, while the benefit of RT remains variable across studies. Therefore, further evaluation of RT-based organ-preserving strategies in carefully selected patients is warranted.

The 5-year LRFS in our study were 88.4% and 70.2% in the cT1 and cT2 groups, respectively. Overall recurrence also appeared to be more frequent in patients with cT2 disease than in those with cT1 disease. In the cT1 group, four of 24 patients (16.7%) experienced recurrence, including one local recurrence, one locoregional recurrence, and two distant metastases. In contrast, five of 16 patients (31.3%) with cT2 disease experienced recurrence, including three local failures, one distant failure, and one combined locoregional and distant failure. Although these comparisons are limited by the small sample size and were not statistically significant, these findings suggest that disease control was favorable in patients with cT1 disease, whereas cT2 tumors may carry a greater risk of recurrence, particularly local failure. Therefore, patients with cT2 disease may require more stringent selection criteria for organ-preserving treatment and closer post-treatment surveillance.

We also evaluated TME-FS in this study as a clinically relevant endpoint reflecting the durability of organ preservation. In our cohort, only four patients (10.0%) ultimately required salvage TME, and the 5-year TME-FS was 87.5% in the cT1 group and 78.6% in the cT2 group. These findings suggest that, for carefully selected patients, RT without radical surgery with or without concurrent chemotherapy may allow a substantial proportion of patients to avoid radical transabdominal surgery. However, this endpoint does not fully represent the organ preservation. Three patients required abdominoperineal resection with permanent stoma formation and one patient underwent ultra-low anterior resection. Therefore, TME-FS should not be interpreted as a colostomy-free survival. In this study, TME-FS was used to evaluate how long radical surgery could be safely avoided or delayed after RT as an alternative to radical surgery.

Recurrence after local excision-based treatment should be interpreted in the context of salvageability and subsequent oncologic outcomes. A previous study of patients undergoing local excision for rectal cancer reported substantially higher 5-year OS among patients who underwent salvage surgery for recurrence compared with those managed nonoperatively (84.7% vs. 44.2%, p < 0.001) [15], supporting the importance of active salvage management when recurrence occurs. Consistent with the favorable survival outcomes observed in our study, Suzuki et al. [16] reported that local excision followed by chemoradiation therapy achieved a 5-year DFS rate of 86% and a 5-year OS of 92% in selected patients with T1 rectal cancer. In our cohort, although recurrence was not uncommon, with 5-year RFS rates of 79.3% and 62.8% in the cT1 and cT2 groups, the ultimate survival outcomes remained favorable, demonstrating 5-year OS of 87.5% and 85.6% in the cT1 and cT2 groups and a 5-year CSS of 100% in both groups. Furthermore, among the nine patients who experienced recurrence, three had no evidence of disease at the last follow-up after receiving salvage treatment. These findings suggest that recurrence should not be considered uniformly unsalvageable or inevitably associated with poor disease-specific survival, although timely detection and appropriate salvage treatment remain essential.

This study has several limitations. First, because of its retrospective design, this study was subject to potential selection bias. In addition, the follow-up intervals and surveillance modalities were not standardized, and quality of life and functional outcomes, including bowel function and low anterior resection syndrome, were not assessed using validated questionnaires. Moreover, late toxicities may also be underreported because the data were collected retrospectively. Second, several pathological variables, including resection margin status, LVI, perineural invasion, and depth of submucosal invasion, were unknown in a high proportion of patients because many patients underwent only biopsy or endoscopic local excision. Consequently, the prognostic impact of these pathological factors could not be adequately evaluated. Third, no clinicopathological factor was significantly associated with either LRFS or RFS in our exploratory univariate analysis; however, given the relatively small sample size and limited number of events, the statistical power was low, and these prognostic analyses were insufficient to identify predictors of recurrence. Fourth, the patient cohort was heterogeneous regarding clinical stage, initial local procedure, and chemotherapy regimen. Fifth, this was a single-arm study without a control group, so it cannot show whether RT adds additional benefit over local excision or biopsy alone. Finally, although both 3-year and 5-year outcomes were presented, the median follow-up duration was relatively short, and the 5-year estimates should be interpreted cautiously.

In conclusion, RT without radical surgery, with or without concurrent chemotherapy following local excision or biopsy, may be a feasible organ-preserving treatment option for cT1–2N0 rectal cancer patients who are medically unsuitable for or decline radical surgery. Despite a non-negligible recurrence rate, particularly within the cT2 subgroup, oncologic outcomes were favorable, with excellent 5-year CSS and a high TME-FS. A significant portion of patients with recurrence were successfully managed with salvage treatments, supporting the potential role of close surveillance and timely salvage intervention in this strategy.