Local Excision

CRITICAL ELEMENTS

Complete Resection of an Intact Specimen with Negative Bowel Wall and Deep Margins

1. COMPLETE RESECTION OF AN INTACT SPECIMEN WITH NEGATIVE BOWEL WALL AND DEEP MARGINS

Recommendation: Local excision is appropriate for T1 rectal cancers without high risk features. When local excision is used, the primary goal is complete excision of the tumor with negative margins. Excision margins should be at least 5-10 mm along the bowel wall and >1 mm along the deep margin. Transanal excision of lesions close to the anal verge is acceptable; for more proximal lesions, either transanal endoscopic microsurgery or transanal minimally invasive surgery are preferable.

Type of Data: Several retrospective observational studies and case series.

Grade of Recommendation: Strong recommendation, moderate-quality evidence.

Rationale

For decades, local excision of small rectal tumors has been performed by using a transanal endoluminal approach, primarily by utilizing standard surgical lighting, instruments, and retractors. For well-selected lesions close to the anal verge, especially in thinner patients, this is readily accomplished and remains an efficient, cost-effective technique for local resection.³³ However, poor exposure of more proximal lesions, especially in patients with an unfavorable body habitus, may result in suboptimal excision.³⁴ The resulting specimens are more likely to be fragmented, and the surgical margins are more likely to be inadequate.

Irrespective of the specific approach (transanal excision [TAE], transanal endoscopic microsurgery [TEM], or transanal minimally invasive surgery [TAMIS]) chosen for local excision of rectal cancer, a complete excision of the tumor with negative bowel wall and deep margins should be performed. To ensure complete resection, the excision should aim to achieve a 10 mm margin along the mucosal and the fullthickness excision should be performed to the level of the peri-rectal fat. The specimen should be removed intact to enable histopathologic examination and determination of the true margin status. Excision that is performed piecemeal cannot be assured to be complete, and the margins cannot be adequately assessed.

Historically, the Kraske transcoccygeal or York-Mason transsphincteric approach was applied to improve exposure to overcome the limited upward reach provided by the conventional TAE technique. Newer techniques including TEM and, more recently, TAMIS, have overcome the shortcomings of standard TAE without the added morbidity associated with the transsphincteric or transcoccygeal approaches. These approaches have made TAE more precise with less fragmentation of the surgical specimen and have decreased the rate of local recurrence relative to standard TAE.⁸⁵,¹³³,¹³⁴,¹³⁵,¹³⁶,¹³⁷,¹³⁸,¹³⁹,¹⁴⁰

TEM was pioneered by Gerhard Buess in the mid-1980s. TEM utilizes a closed system that insufflates carbon dioxide into the rectum, enabling rectal distension and exposure (the “pneumorectum”). A fiberoptic high-definition scope with a wide, magnified field of view is inserted through a rectoscope with a 4-cm diameter. The rectoscope is regulated by an endosurgical unit that provides insufflation, irrigation, suction, and pressure monitoring. Long-shafted instruments, inserted through the working ports, are capable of grasping, dissecting, electrocauterizing, and suturing tissues.

TAMIS utilizes a variety of technologies initially developed for laparoscopic surgery (e.g., fiberoptic laparoscopes, multichannel single-port platforms) that similarly enable creation of a pneumorectum, provide high-quality images, and utilize ordinary long-shafted laparoscopic instruments to allow for excision of the lesion, suction and/or irrigation, and control of bleeding points.¹⁴¹,¹⁴²,¹⁴³ A number of variations have been utilized in attempts to enhance the safety and capability of TAMIS and to reduce costs.

In general, TEM and TAMIS have complication rates similar to those associated with traditional TAE. However, potential disadvantages associated with the newer platforms include the need for specialized training to learn the techniques and increased cost. The complication rates and functional results of TAE are difficult to compare with those of the newer platforms because the data are older and reflect a time when there was less emphasis on and sophistication in the measurement of outcomes.

TEM and TAMIS appear to have a number of clear oncologic advantages over TAE.¹⁴⁴ These include a substantially lower rate of positive margins, a diminished risk of tumor fragmentation, and a decrease in local recurrence rates. The incidence of positive or indeterminate margins with TAE has been two to eight times greater than with TEM.¹⁴⁵

Because the platforms, techniques, and instrumentation continue to evolve in the burgeoning field of transanal endoscopic surgery, a meaningful comparison of TEM with TAMIS and all of its iterations is of limited utility.¹⁴⁶,¹⁴⁷,¹⁴⁸ Nonetheless, it appears
that one advantage of TAMIS is the ability to use or adapt readily available laparoscopic equipment and instrumentation, whereas TEM requires specialized equipment. This difference could mean a cost savings for TAMIS and shorten the learning curve relative to TEM, because the equipment is already available and familiar to the surgeon.

Complication rates with TEM range from 0% to 29%¹³⁴,¹⁴⁹,¹⁵⁰,¹⁵¹,¹⁵²,¹⁵³,¹⁵⁴,¹⁵⁵,¹⁵⁶,¹⁵⁷ and are usually minor and self-limited (urinary retention/infection, bleeding, and suture line dehiscence).¹⁵⁸,¹⁵⁹,¹⁶⁰,¹⁶¹,¹⁶²,¹⁶³,¹⁶⁴,¹⁶⁵,¹⁶⁶,¹⁶⁷,¹⁶⁸,¹⁶⁹

Operative efficiency with TEM continues to improve up to 16 cases,¹⁷⁰ but the overall learning curve appears to be quite steep, with improved outcomes reported after performance of 35 cases.¹⁷¹,¹⁷² The conversion rate from TEM to an abdominal approach in one large series (n = 693) was 4.3%.¹⁷² The best available evidence strongly suggests that TEM has minimal or no long-term impact on continence or functional outcomes (including urologic and sexual dysfunction) as measured by clinical, manometric, and validated quality-of-life metrics.¹⁶¹,¹⁷³,¹⁷⁴,¹⁷⁵,¹⁷⁶,¹⁷⁷ The addition of radiation therapy to TEM (or other forms of TAE) has a substantial impact on the complication rate and functional results.¹⁷⁸,¹⁷⁹

A pooled analysis of 33 TAMIS-related reports included a 4.3% rate of positive margins and a 4.1% incidence of tumor fragmentation.¹⁸⁰ Concern has been expressed about the ability to close full-thickness defects in more proximal tumors with peritoneal entry by using the TAMIS platform.¹⁸¹ The overall complication rate with TAMIS was 7.4%, with minor, self-limited bleeding and urinary retention representing more than half of the reported complications. A learning curve of 20 cases has been suggested, generally shorter than that reported with TEM. The data on the functional outcomes with TAMIS are much more limited than with TEM; however, there does not appear to any significant adverse impact on quality of life or continence scores.¹⁸²,¹⁸³,¹⁸⁴

Key Question: Local Excision

In patients with early rectal cancer, does local excision affect recurrence, survival, and quality of life relative to proctectomy?

INTRODUCTION

Like so many issues in the management of rectal cancer, the most appropriate treatment of early stage tumors continues to be controversial¹⁸⁵,¹⁸⁶,¹⁸⁷ and sometimes even contentious. For patients with early rectal cancer, both local excision and proctectomy are accepted treatments, but the precise indications and the relative trade-offs for any given patient may be difficult to evaluate.¹⁸⁸,¹⁸⁹ On the one hand, proctectomy assures a wide surgical margin and provides for excision and histopathologic examination of the mesorectal lymph nodes.¹⁹⁰ Patients may be staged with confidence and receive adjuvant therapy as appropriate. Further, the risk of local recurrence should be exceedingly small in the setting of early stage disease, thereby simplifying posttreatment follow-up and surveillance.¹⁹¹

However, proctectomy is a major abdominal operation with potential for perioperative morbidity or need for temporary or permanent intestinal stoma. There is an appreciable risk for anastomotic leak or adverse functional outcomes that may transform the temporary fecal diversion into a lifelong ostomy. Quality of life can be affected with urinary, sexual, and bowel dysfunction.¹⁵⁷ Laparoscopic and robotic approaches along with advanced sphincter-saving techniques have reduced the short-term impact of surgery and have decreased the rate of permanent colostomies; however, it seems inappropriate to describe a proctectomy as “minimally invasive,” irrespective of the access approach chosen for radical excision, considering the incidence of potential functional impairments.

In comparison with proctectomy, local excision is associated with a lower incidence of major complications and less short- and long-term disability.¹⁹² Newer transanal endoscopic surgical approaches to local excision (e.g., TEM or TAMIS with the use of conventional laparoscopic instrumentation through a transanal access port) have made TAE far more precise with less fragmentation of the surgical specimen and have decreased the rate of local recurrence relative to standard TAE.⁸⁵,¹³³,¹³⁴,¹³⁵,¹³⁶,¹³⁷,¹³⁸,¹³⁹,¹⁴⁰ The rectal reservoir is preserved, yielding a superior functional outcome. However, one must acknowledge the very real risk of occult positive lymph nodes, resulting in undertreatment, because of both incomplete resection and missed opportunity for referral for adjuvant therapy.¹⁹³,¹⁹⁴

An alternative approach to decrease the risk or impact of undertreatment is to incorporate adjuvant or neoadjuvant chemoradiotherapy when tumors are identified as highrisk early cancers.¹⁸⁴,¹⁹⁵,¹⁹⁶,¹⁹⁷,¹⁹⁸,¹⁹⁹ However, the oncologic efficacy of this approach remains uncertain, and adjuvant and/or neoadjuvant therapy may substantially increase the complication rate associated with local excision as well as impair functional outcome.²⁰⁰,²⁰¹,²⁰² Furthermore, the risk of local recurrence is higher with local excision than with proctectomy, resulting in the need for more frequent and potentially costly follow-up.¹⁴⁵,²⁰³

Based on these concerns, we sought to address the following overarching question: In patients with early rectal cancer, does local excision affect recurrence, survival, and quality of life relative to proctectomy? A systematic review of the published literature was performed with the aim of answering the following specific questions:

Compared with patients who undergo radical excision of rectal adenocarcinoma, do patients who undergo local excision experience fewer postoperative complications?
Compared with the resected specimens of patients who undergo radical excision of rectal adenocarcinoma, are specimens from patients who undergo local excision more likely to have clear margins on pathologic examination?
Compared with patients who undergo radical excision of rectal adenocarcinoma, do patients who undergo local excision have better functional outcomes at 1 year?
Compared with patients who undergo radical excision of rectal adenocarcinoma, do patients who undergo local excision experience higher local recurrence rates?
Compared with patients who undergo radical excision of rectal adenocarcinoma, do patients who undergo local excision experience reduced overall survival rates?

METHODOLOGY

We performed a systematic review of Medline, Scopus, CINAHL, Embase, and Cochrane Library databases for articles published January 1, 2000, to December 31, 2015, to reflect the modern era of TAE for rectal cancer that includes TEM and TAMIS. The search terms included a combination of the following keywords, each set combined with the others by “AND”:

rectal cancer OR rectal neoplasm

surgery OR operation OR excision OR resection

transanal OR local OR TEMS OR TAMIS

Searches were limited to English language and peer-reviewed studies of adults aged ≥18 years. All citations were identified (n = 6,467) and combined into a single library. Duplicates (n = 2,242) were excluded, resulting in 4,225 final abstracts for review (Fig. 6-1).

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines,²⁰⁴ we predetermined exclusion criteria as follows: we reviewed all abstracts and excluded those published only in the gray literature, as well as commentary, review articles, or articles that otherwise contained no primary data or no primary analysis of secondary data (e.g., clinical registries or claims). We excluded articles based on data accrued prior to 2000 and case reports with fewer than 30 patients, unless there was a compelling reason for inclusion. In the case of multiple publications reporting on the same dataset, we included only the most recently published analysis, unless additional relevant data were available only within earlier publications. We excluded studies with <1 or 2 years of follow-up in the specific searches related to specific questions 3 or 4 and 5. We then searched bibliographies of included manuscripts by hand for additional publications that might fulfill inclusion criteria, but we found no additional articles. Applying exclusion criteria to abstracts yielded a final total of 148 articles for full review (Fig. 6-1).²⁰⁴ After application of inclusion and exclusion criteria to the full article review, 132 additional articles were excluded for a final total of 16 articles for data extraction.

FIGURE 6-1 Studies selected for full review based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses related to the key question regarding local excision for rectal cancer.²⁰⁴

We developed a data abstraction tool to capture detailed data related to study design, methodologic rigor (e.g., minimization of selection and attrition bias, appropriate statistical analysis), results, and conclusions. We then summarized the evidence in response to the predefined questions. Discordant views were managed by reviewing original sources and discussion a second time, and, if necessary, resolving selections through adjudication by the authors of this review.

FINDINGS

Sixteen articles met all criteria for data extraction in the systematic review (Tables 6-1 and 6-2).¹⁶¹,¹⁶²,¹⁶⁷,¹⁸⁶,¹⁸⁷,²⁰¹,²⁰⁵,²⁰⁶,²⁰⁷,²⁰⁸,²⁰⁹,²¹⁰,²¹¹,²¹²,²¹³,²¹⁴ All the articles included contained a comparison group of patients who had undergone radical excision of rectal adenocarcinoma. Most were retrospective studies from single institutions, but there were a few articles from large clinical registries.

The study questions focused on short-term operative outcomes, mid-term functional outcomes, and long-term oncologic outcomes. We framed the short-term outcome questions on desirable outcomes: fewer complications, clear or negative margins, and better functional outcomes. We framed the long-term oncologic questions on undesirable outcomes: local recurrence rate and reduced overall survival.

TABLE 6-1 Local Excision Studies: Study Patients and Short-Term Results

	Surgical Technique
Author, Year; Type of Data, Study Period	LE	TME	Complications	Margins
Lee et al,¹⁶¹ 2003; Sungkyunkwan University database, 1994-2000	74 N	100	4.1% LE vs. 48%	NR
Nascimbeni et al,²⁰⁷ 2004; Mayo Clinic database, 1979-1995	70 TNS	74		NR
Endreseth et al,²⁰⁸ 2005; Norwegian Rectal Cancer Project, 1993-1999	35 TAE	256		NR
Ptok et al,²⁰⁹ 2007; Colon/Rectal Cancer (Primary Tumor) Study Group, 2000-2001 (a multi-institutional prospective observational study)	85 TAE, 35 TEM	359	9.2% LE vs. 22.8% (P < 0.001)^*	NR
Folkesson et al,²¹⁰ 2007; Swedish Rectal Cancer Registry, 1995-2001	643^†	7,016	11.5% LE vs. 35.4%	NR
You et al,²⁰¹ 2007; National Cancer Database, 1994-1996	765 TNS	1,359	5.6% LE vs. 14.6% (P < 0.001)^*	NR
Hazard et al,²¹¹ 2009; Surveillance, Epidemiology, and End Results (SEER) program, 1998-2003	573, 283+^‡	3,464		NR
De Graaf et al,¹⁶² 2009; Dutch TME Trial Subgroup, a prospective, nonrandomized trial, 1996	80 TEM	75	5.1% LE vs. 64% (P < 0.001)^*	NR
Peng et al,²¹² 2011; Shanghai Cancer Center, 1992-2005 (a Fudan University database)	58 TAE	66		NR
Lezoche et al,²⁰⁵ 2012; Prospective randomized trial for patients with T2 rectal cancer treated with neoadjuvant chemoradiotherapy, 1997-2004	50 TEM	50	2% LE vs. 6% (P = 0.250)	All R0
Allaix et al,¹⁶⁷ 2012; University of Torino database, 1999-2009	32 TEM, 9 TEM+^**	33	14.6% LE vs. 37.1% (P = 0.046)	Favors TME
Bhangu et al,¹⁸⁷ 2013; SEER program, 1998-2009	3,715 TNS	9,547		NR
Stitzenberg et al,¹⁸⁶ 2013; National Cancer Database, 1998-2010	34,697 TNS	76,756		Negative margin: LE 76% vs. TME 95% (P <0.001)^*
Saraste et al,²¹³ 2013; Swedish Rectal Cancer Registry, 1995-2006	448 TNS	3,182		NR
Elmessiry et al,²¹⁴ 2014; Cleveland Clinic Florida, 2004-2012	47 TAE, 27 TEM	79	0% LE vs. 21.6% (P = 0.009)^*	Positive margin: LE 13.5% vs. TME 0% (P = 0.001)^*
Lezoche et al,²⁰⁶ 2014; Sapienza University of Rome database, 2008-2010	32 TEM	18		NR
Note: LE, local excision; TME, total mesorectal excision – TME refers to various types of radical surgery, including open and laparoscopic low anterior resection and abdominoperineal resection. N, number of patients in each arm of the study; NR, not reported; TAE, transanal excision; TEM, transanal endoscopic microsurgery; TNS, technique not specified; R0, indication of clear margins. Not all reports made all data available.
^Statistically significant. ^†TAE, TEM, polypectomy, and fulguration were combined in the LE cohort. ^‡In this study, 573 patients had LE alone, and 283 patients had LE with radiation. The technique was not specified. ^*In this study, 32 patients had TEM only, and 9 patients had TEM plus neoadjuvant radiation.