The outcome of transanal endoscopic microsurgery and adjuvant radiotherapy in patients with high-risk T1 rectal cancer.

Introduction: Transanal endoscopic microsurgery (TEM) is considered the technique of choice for adenoma and low-risk T1 rectal cancer. The adequacy of such treatment for high-risk T1 tumours, however, is still controversial. The aim of the study is to evaluate our results with local excision of high-risk T1 cancers. and
Methods: Demographic, clinical data pertaining to patients undergoing TEM for T1 rectal cancer between 1999 and 2015 was retrospectively collected. Long-term outcomes were assessed for the entire cohort. Patients were classified into two groups: favourable and high-risk cancer features.
Results: Three hundred and fifty-five TEM procedures were recorded in the study period. Forty-three patients were included in the present study. There were 20 females and 23 males, the median age was 69 ± 9. The median tumour distance from the anal verge was 6 cm (range 1-13 cm). Post-operative histopathology showed well/moderately differentiated T1 adenocarcinoma in 30 patients and poorly differentiated in 13. The overall survival for patients with favourable and high-risk features groups were 93.5% and 77%, respectively, while the local recurrence (LR) was 3.5% and 23.1%, respectively. Nine patients with high-risk features received adjuvant radiotherapy; one (11.1%) of them experienced LR. Conclusions: Local excision by TEM augmented by adjuvant radiotherapy may be a feasible alternative for selected patients with high-risk T1 rectal cancer. The addition of radiotherapy seems to decrease the rates of LR.

INTRODUCTION

Local excision by transanal endoscopic microsurgery (TEM) is the treatment of choice for T1 rectal adenocarcinoma with favourable prognostic features.[123] Favourable features include free margins, no lymphovascular invasion, well to moderately differentiated tumours and superficial submucosal invasion Sm1, Sm2. High-risk features that may preclude local excision include positive margins, lymphovascular invasion, tumour budding, poorly differentiated tumours or deep submucosal invasion Sm3.[45] Tumours with high-risk features, on the other hand, harbour occult nodal metastases in 20% of cases. Proctectomy, i.e. anterior resection (AR) or abdominoperineal resection (APR) with total mesorectal excision (TME) is therefore recommended.[123] Proctectomy is the procedure of choice also if the tumour is >30% of rectal wall circumference, >3 cm in size, or if enlarged lymph nodes are documented in pre-treatment imaging.

In high-risk patients or those who deny proctectomy, local excision and adjuvant radiotherapy may be considered. It has been reported that radiotherapy may decrease local recurrence (LR) of rectal cancer.[6789101112] Pre-operative radiotherapy is well-established treatment for patients with locally advanced rectal cancer.[13] Nevertheless, it has not shown a major benefit in patients with T1 or T2 cancer undergoing proctectomy.[89] Whether it may benefit those undergoing local excision, has not been well investigated. In particular, patients with high-risk T1 cancer, who are not candidates for proctectomy, adjuvant radiotherapy may positively impact LR and overall survival (OS).

The aim of this study is to report outcomes of series of patients undergoing TEM for T1 rectal cancer and to assess the efficacy of adjuvant radiotherapy in those with high-risk features.

MATERIALS AND METHODS

All patients who underwent local excision of rectal cancer, either by transanal excision, TME or transanal minimally invasive surgery, between the years 1999 and 2015, were identified from the departmental database. Based on the post-operative histopathology report, patients with T1 rectal adenocarcinoma were isolated and consist of the study group. For the purposes of this study, ‘rectum’ was defined as 15 cm from the anal verge, measured by rigid proctoscopy (RP).

Our routine practice for patients with adenocarcinoma of the rectum includes physical examination by digital rectal examination (DRE), RP, transrectal ultrasound (TRUS), abdominal and chest computerised tomography. Based on pre-operative workup, patients were categorised into favourable T1 or high-risk T1 groups. In general, patients with pre-operative tumour staging of favourable T1N0M0 are amenable for local excision. In high-risk group; i.e., tumour >3 cm, poorly differentiated or mucin-producing, presence of lymphovascular invasion, tumor budding or deep invasion of the submucosa, an AR or APR with TME is usually performed.

Data pertaining to patients and pre-operative tumor characteristics, including endoscopic (size, height) and radiologic (tumour stage) findings were retrieved. Furthermore, intra-operative findings; tumor site, height and size were collected.

Data collection included intra- and post-operative complications and length of hospital stay. Complications were defined as local (post-operative bleeding or pelvic sepsis) or systemic (pulmonary, cardiovascular or urinary i.e., infection or retention).

Post-operative histopathologic reports were reviewed, tumour stage, differentiation, resection margins, lymphovascular invasion, tumour budding and the depth of submucosal invasion were determined.

Patients with involved surgical margins underwent local re-excision to achieve free margins.

Patients with high-risk pathological features were referred to proctectomy and TME.

Adjuvant radiotherapy was offered to patients who denied proctectomy or were high risk for radical surgery. In general, radiation was administered during 5 weeks period to the tumour bed, with a goal dose of 5040 Gy.

On completion of the surgical and radiation treatments, all patients were scheduled for follow-up, every 3 months during the first 2 years, and every 6 months thereafter. Follow-up included DRE, endoscopic, radiologic and tumour marker tests.

Long-term oncologic outcomes including OS, disease-free survival and LR were calculated for the entire cohort, and then for patients with favourable and high-risk features. Oncologic outcomes for patients who received adjuvant radiotherapy were assessed separately.

Statistical analysis

Categorical variables were summarised as frequency (%), and quantitative variables as mean ± standard deviation and median interquartile range (minimum; 25th%; median; 75th%; maximum). Associations with quantitative variables analysed by t-test and associations with categorical variables analysed with Chi-square test or exact-Fischer test. Survival outcomes were calculated using the Kaplan–Meier method. A P < 0.05 was considered statistically significant.

RESULTS

Three hundred and fifty five patients underwent local excision of the rectal lesion during the study period. Forty-three patients underwent full-thickness excision of T1 rectal tumour using the TEM technique. There were 20 females and 23 males. The mean age was 69.5 ± 9 years. The median tumour distance from the anal verge was 6 cm (range 2–10 cm). Pre-operative TRUS revealed T1N0 in 32 patients and carcinoma in situ (TisN0) in eight patients, and for three data were not available. No distant metastases were noted in the pre-operative systemic workup.

As mentioned, the final pathology reports of the TEM specimens revealed T1 rectal cancer. High-risk features were seen in 13 patients and favourable ones in 30 patients.

Patients’ demographics and tumor characteristics were comparable between the favourable features T1 group and the high-risk features group [Table 1].

Table 1

Patient and tumour characteristics in malignant and favourable tumour groups

	T1 with favourable features (n=30)	T1 with high risk features (n=13)	P
Age (years), mean±SD	68.2±8.5	72.1±7.6	0.14
Gender (male/female)	18/12	5/8	0.31
ASA score	2.2±1.1	3.1±0.9	0.013
Tumour distance from anal verge (cm), median (range)	8 (1-13)	5 (1-8)	0.041
Tumour site
Anterior	12	6	0.82
Posterior	9	3
Lateral	9	4
Tumour size	2 (1-7)	2 (1-7)	0.82
Pre-operative biopsy
Adenocarcinoma	23	10	0.69
Adenoma with high grade dysplasia	5	2
Tubular/villous adenoma	2	1

SD: Standard deviation, ASA: American Society of Anaesthesiologists

Patients with high-risk features were offered radical rectal surgery with TME. Of them, two patients underwent AR and they were found to have no tumour in the final pathology. Nine patients were high surgical risk or refused further surgical intervention, therefore, adjuvant radiotherapy, 5040 Gy for 5 weeks, was administered. All of them completed radiation courses.

The last two patients denied further treatment, neither proctectomy nor radiation. One of them agreed to local re-excision, due to close margins. Repeated TEM was performed and free margins were obtained.

Thirty patients with favourable cancer features underwent local excision. Close margins were documented in three patients, 2 underwent re-excision. One refused further treatment and was lost to follow-up.

Histopathological findings of rectal cancer were documented pre-operatively in 33 patients. Another 10 patients underwent surgery for pre-operative diagnosis of adenoma (with or without high-grade dysplasia) and were found to have T1 rectal cancer on the final pathology report.

Post-operative histopathology findings showed well to moderately differentiated T1 rectal adenocarcinoma in 30 patients, and poorly differentiated and/or mucin-producing in 13 patients.

Submucosal invasion was assessed in 30 patients. Sm3 invasion (deep invasion) was noticed in four patients. All of them presented with either poorly differentiated or mucin-producing tumours.

Free margins (>3 mm) were obtained in 39 patients while close ones (<3 mm) were seen in four. Of the latter, three had favourable cancer features. Two of them underwent re-excision for margins extension. The third patient refused further intervention. In the last, where a poorly differentiated tumour was reported, a proctectomy was recommended. It was aborted as per the patient request. The patient again underwent local re-excision only.

Early post-operative outcomes were favourable in almost all patients. The mean length of hospital stay was 2.5 (±1) days. Overall post-operative morbidity was 5%. It includes one patient with non-specific fever and one patient with urinary retention. No perioperative mortality was documented.

After a median follow-up of 32 months, the long-term OS of and LR rate of the entire cohort were 88% and 13.8%, respectively. The OS of patients with favourable T1 rectal cancer was 93.5% and LR was 3.5%. The OS and LR in patients with high-risk T1 rectal cancer were 77% and 23.1%, respectively. Of note, long-term cancer-related mortality was reported in one patient only. Difference in LR between favourable and high-risk groups was remarkable but did not reach statistical significance (P = 0.15), probably due to the small sample size [Figure 1].

Figure 1

Local recurrence in patients with favorable and high risk features (A: high risk, B: favorable)

As mentioned, a total of 29 patients with favourable T1 rectal cancer were available to follow-up. They underwent local excision with free resection margins. During the follow-up period, one patient experienced LR. An APR was performed and subsequent adjuvant chemotherapy was administered.

In the subgroup of patients with high-risk features, three patients experienced LR. The first refused any additional intervention after local excision; the second underwent local re-excision to obtain free margins but denied proctectomy or radiotherapy. The third patient denied proctectomy but agreed to adjuvant radiotherapy. Local and distant recurrences were documented in the first patient 10 months after surgery, while the latter two patients presented with LR after 18 and 10 months respectively [Figure 2].

Figure 2

Follow up data of patients with T1 rectal cancer undergoing surgery. FU: follow up, APR: Abdominoperineal resection, AR: Anterior resection

As discussed, the subgroup of patients with high-risk features was treated with local excision and adjuvant radiotherapy. All patients received high-dose radiation (5040 Gy). All of them tolerated and completed the treatment. One out of, nine patients presented with LR after 10 months. Proctectomy was subsequently performed. LR rate after 32 months was 11.2% [Figures 2 and 3].

Figure 3

Local recurrence in patients with favorable and high risk features, with and without radiation (A: High risk without radiation, B: Favorable, C: High risk with radiation)

DISCUSSION

Radical rectal resection with TME is the treatment of choice for high-risk T1 rectal cancer. A major surgery that is associated with considerable morbidity and mortality. Local excision by TEM is provided in patients who decline major surgery or frail patients given their high surgical risk. Although TEM carries a low and acceptable morbidity rate, it still regarded an oncological compromise in the T1 rectal cancer, since perirectal lymph nodes remain intact.

Previous reports have found no difference between local excision and TME surgery in terms of LR rate or survival of patients with T1 rectal cancer.[78] However, Stornes et al.[9] reported that the LR rate in T1 rectal cancer was 14.5% in patients who underwent TEM, which was significantly higher than the LR rate (1.4%) in patients who underwent TME (P < 0.001). The 5-year OS rate was significantly lower in the local excision and TEM group than the standard surgery.[910]

Radiation therapy has long been used together with radical surgery for rectal cancer to control LR.[11] The same rationale led to the use of radiotherapy following local excision when pathological features are unfavourable and local excision may have not been a sufficient treatment.

In actual fact, Borstlap et al.[12] performed a meta-analysis of 14 studies and found that the LR rate in patients with T1 rectal cancer who additionally received radiotherapy or chemoradiotherapy after local excision (5%) was similar to that in patients who underwent TME (4%).

Our study discusses the benefits of adjuvant radiotherapy in patients with high-risk T1 cancer. In particular, we evaluated local control outcomes since groups were significantly heterogenous in terms of the American Society of Anaesthesiologists score, thus comparison of survival outcomes may be inadequate.

After a median follow-up of 36 months, the overall LR rate of T1 cancer in patients treated with local excision was 13.8%. It was significantly greater than that reported for comparable groups undergoing conventional proctectomy.[1314] However, when classifying patients into two subgroups, either favourable or high-risk T1 cancer, the outcomes significantly differed between subgroups. LR in the favourable and high-risk groups was 3.5%, and 23.1%, respectively. Differences did not reach statistical significance probably due to the small sample size. It may also be argued that tumours with favourable features were located high in the rectum thus associated with lower LR rate. Similar long-term outcomes have been previously reported as well.[1516]

Recent data have suggested that the risk of LR of T1 rectal cancer, in particular, those with high-risk features, is worse than initially thought.[67] High-risk T1 cancer is associated with high lymph node metastases rate.[17] Conventional proctectomy with TME is therefore the treatment of choice.

When local excision is provided, assuming that adequate surgery is inappropriate, adjuvant radiotherapy should be considered.[7812] We assume that radiation may favourably impact local control of high-risk T1 cancer since LR decreased from 23% to 11.2% in the entire high-risk group and the radiation group, respectively. A total of four patients experienced LR in our study, three of them had had high-risk cancer features, but only one of them received radiation therapy. We believe that adjuvant radiotherapy may be considered in high surgical risk patients with T1 high-risk cancer features. It is, however, not an alternative treatment in fit patients. Large-scale studies assessing the efficacy of adjuvant radiation are warranted.

The role of post-operative radiotherapy for rectal cancer has been widely studied. It decreases the risk of local failure by 30%–40% in patients with advanced cancer but is not yet well-established for early rectal tumours.[181920] Nevertheless, radiation has been shown to improve local control in patients who have the residual microscopic disease. Radiation is less effective in those with gross residual disease.[21] Thus, treating residual microscopic disease, i.e. involved margin or microscopic lymph node metastases, would be the mechanism by which radiation improved outcomes in high-risk T1 cancer patients in our study. Therefore, if appropriate surgical treatment cannot be provided, radiotherapy may be an alternative.

It may be argued that radiation might affect anal sphincter function and patients’ quality of life.[2223] In addition the anorectal functions could be affected after TEM.[24] Therefore, patients should be counselled regarding the pros and cons of radiotherapy with preserving the rectum comparing to operative outcomes of radical rectal resection.

Interestingly, one out of four patients with recurrence presented with combined local and distant recurrence, while the other three experienced LR only. The later three underwent either AR or APR and are free of disease. It seems that most cases with LR may be treated by resection of the rectum with minor long-term sequalae.[25]

The present study has several limitations; the retrospective nature and small sample size groups. Furthermore, long accrual periods contribute to variability in the perioperative care and diagnostic modalities as they evolve over the years, as the tumour budding, a new feature of high-risk T1 tumours, has not been routinely assessed at our institute until recently. Furthermore, the submucosal invasion has been evaluated in 30 patients only.

CONCLUSIONS

Adjuvant radiotherapy may improve oncologic outcomes in high-risk T1 rectal cancer patients undergoing local excision. The treatment of choice in this subgroup of patients remains radical rectal surgery with TME. However, in high surgical risk patients, or in patients declining, adjuvant radiation therapy may decrease the risk of LR, and should be considered as an alternative for radical surgery in selected patients.

Drs Khoury, Issa, Duek, Khalefah, and Mrs Dauod have no conflicts of interests that may affect the results or bias its interpretation.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.