Transanal minimally invasive surgery (TAMIS) for local excision of benign and malignant rectal neoplasia: a 7-year experience

Proctectomy with total mesorectal excision (TME) remains the standard of care for management of rectal cancer [1, 2]. Despite providing effective local tumour control, proctectomy with TME is associated with a significant risk of peri-operative morbidity, and a 3% risk of mortality [3]. From a quality of life (QOL) perspective, a significant proportion of patients will also require a temporary or permanent stoma, and for patients who undergo a low anterior resection, the incidence of low anterior resection syndrome is up to 45% [4, 5].

The widespread implementation of national screening programmes has led to a significant increase in rates of detection of early rectal neoplasms [6, 7]. This, coupled with advances in the scope and efficacy of modern neoadjuvant therapies, has resulted in a significant increase in the proportion of rectal cancers potentially amenable to local excision [8]. Consequently, there has been a growing trend towards implementation of less invasive procedures and rectum-preserving techniques.

There are two predominant rectum sparing approaches to transanal excision: transanal endoscopic microsurgery (TEM) and transanal minimally invasive surgery (TAMIS). Gerhard Buess first described TEM in 1984. It involves the use of a set of highly specialised instruments applied through a fixed metallic platform placed in the anal canal [9]. Atallah et al. first described transanal minimally invasive surgery (TAMIS) in 2010 [10]. TAMIS allows the use of standard laparoscopic cameras and instruments used in collaboration with a single-use flexible transanal access platform [11]. Some benefits of TAMIS over TEM include a lower associated cost of implementation, easier setup and greater procedural flexibility [12]. It is also assumed to have a more shallow learning curve due to the use of familiar laparoscopic cameras and instruments [11, 12]. At present, TAMIS is only recommended with curative intent for patients with pre-cancerous polyps; T1 rectal cancers; or small T2 rectal cancers, usually in the setting of clinical trials [13].

Current available evidence suggests local excision confers a QOL benefit over TME, whilst from an oncological perspective we await outcomes from ongoing randomised trials including the STAR-TREC and TESAR trials [14‐18]. At present, much of the evidence pertaining to local excision in this context relates to TEM. Our unit, a tertiary referral rectal cancer centre, was an early adopter of TAMIS local excision in 2016. The objectives of this study are to present the clinical and midterm oncological outcomes for 168 consecutive patients who have undergone TAMIS LE of benign or malignant neoplasia at our institution, since implementation of the TAMIS LE programme.

The implementation of a TAMIS LE programme has revolutionised the management of early rectal neoplasia in our institution. Many patients who previously were managed with TME are now offered a LE alternative in TAMIS which has been associated with acceptable peri-operative and oncological outcomes.

Previous studies have used measures such as tumour fragmentation and margin positivity rate to evaluate excision quality. 95.2% of our excisions were performed with negative margins and 96.4% were submitted without fragmentation. In a review of 200 consecutive TAMIS LE procedures, Lee et al. reported a negative resection margin rate of 93% whilst 95% of tumours were submitted without fragmentation [20]. Meanwhile, some smaller studies have reported contrasting results, Haugvik et al. found that 22% of patients had a positive resection margin following TAMIS LE of benign rectal lesions, whilst 31% of specimens in this study were too fragmented to allow reporting of marginal status [21]. Mohamed et al. report findings from a cohort of 42 patients who underwent TAMIS LE; the positive margin rate was 4% in this study; however, 19% of specimens again were submitted in fragments [22]. From a technical perspective, 79.2% of our patients underwent full thickness excision, compared to only 14% in the Mohamed et al. group [22]. The authors hypothesize that by carrying out full thickness excisions as standard, the likelihood of fragmenting the tumour is significantly reduced. It is also notable that in both our study and the study by Lee et al., surgical volume was significantly higher than in the other aforementioned studies [20‐22]. In our single surgeon study, 168 TAMIS procedures were performed over 7 years, whilst in the findings reported by Mohamed et al., 42 excisions were performed by 2 surgeons over a 5 and a half–year period; similarly in the study by Haugvik et al., 51 procedures were performed over a 4-year study period [21, 22]. There is a likely association between surgical volume and excision quality in this context.

Our recurrence rate was 6.4% for malignant lesions, with only 1 instance of mucosal recurrence. Lee et al. reported a local recurrence rate of 6% for patients with malignant disease treated with TAMIS LE [20]. These oncological outcomes compare favourably with data obtained from TEM series. Moore et al. reported a local recurrence rate of 8% for malignant lesions, whilst Christoforidis et al. reported a 15% 5-year local recurrence rate after TEM [23, 24]. A critical juncture in the management of patients post TAMIS LE is how to proceed when initial primary histology reveals high-risk pathological features. In our single experience with mucosal recurrence, the patient was offered further treatment due to lymphovascular invasion on primary histology, but declined. Similarly in the series by Lee et al., 3 of 6 patients who developed recurrence declined further treatment initially despite the presence of high-risk pathological features. Four further patients in our cohort proceeded to immediate salvage proctectomy, due to the presence of lymphovascular invasion or poor differentiation on their initial pathology; all 4 had positive definitive outcomes, none requiring an unnecessary permanent ostomy. Apart from 1 case of mucosal recurrence, we also report 2 incidences of nodal recurrence and a singular instance of distant metastatic disease, diagnosed 3 years post TAMIS LE. None of these 3 patients was in receipt of scheduled neoadjuvant or adjuvant systemic therapy as part of their management, and a key question is whether it could have made a positive impact in this context? At present, there remains no definitive consensus on the use of systemic chemotherapy and radiotherapy in combination with local excision [25, 26]. Data from the TREC study suggests short-course radiotherapy followed by TEM may achieve high levels of organ preservation, with relatively low morbidity and indications of improved quality of life over radical resection; however, larger randomised studies, such as the ongoing STAR-TREC and TESAR trials, are needed to more precisely determine oncological outcomes following different organ preservation treatment schedules [17‐19].

The overall post-operative complication rate across our TAMIS cohort was 8.3%. Lee et al. reported post-operative morbidity in 11% of patients [20]. A further large multi-centre study which included 428 patients compared outcomes post TEM and TAMIS, and found no significant difference in post-operative complication rate amongst the 2 methods of local excision (TEM 11% vs TAMIS 9% P = 0.477) [27]. The most common post-operative complication encountered across our cohort was post-operative haemorrhage. The most significant complication in our cohort occurred in a patient who received neoadjuvant radiotherapy. Similarly in the series by Lee et al., patients who received radiotherapy in advance of surgery were more likely to encounter wound-related complications or impaired post-operative functionality [20].

Selecting appropriate patients for local excision can be difficult and is limited significantly by the efficacy of radiological investigative techniques in determining pre-operative depth of local mural invasion. MRI was associated with a modest sensitivity of 86.7%, and a specificity of only 19.1% in this study. Similarly, a previous systematic review found endoscopic rectal ultrasound to be associated with a sensitivity of only 50% and a specificity of 89% in determining occult T1 rectal cancers pre-operatively [28]. These findings are important and suggest clinicians and multi-disciplinary forums should interpret pre-operative radiological investigations with caution before proceeding with LE.

The authors acknowledge that this study has important limitations. Most notably, the retrospective nature of data analysis can introduce selection bias that may affect the study’s veracity. This study is also limited because all TAMIS operations were performed by a single-experienced colorectal surgeon at a single high-volume institution, which may limit the generalizability of the results.

For carefully selected patients, TAMIS LE of rectal neoplasia is a valid option with low morbidity and comparable midterm oncological outcomes.