Limited wedge resection for T1 colon cancer (LIMERIC-II trial) – rationale and study protocol of a prospective multicenter clinical trial

A segmental colon resection with lymph node dissection is associated with significant morbidity (24%) and mortality (2%) [1], regardless of T-stage. In T1 colorectal cancer (CRC), three independent histological features have been identified to be risk factors for lymph node metastases (LNM), i.e. (lympho-) vascular invasion, poor (high-grade) differentiation (G3) and high-grade tumor budding (Bd2-Bd3) [2‐7]. A completely removed (R0) T1 CRC can be classified as low risk T1 when histological examination shows none of the high risk factors for LNM. Prior retrospective studies report a very low risk of LNM (1%) for low risk T1 CRC [8‐11]. As this negligible risk does not outweigh the significant morbidity and mortality of an segmental resection, patients with low risk T1 CRC are sufficiently treated with a local resection.

As upfront risk stratification is not possible, all patients with suspected T1 CRC could receive initial local treatment according to current national guidelines. An additional segmental resection with lymph node dissection is only indicated if histological examination shows high risk factors for LNM or when an incomplete local resection (R1) is performed. It has been proven that secondary surgery after initial local endoscopic resection (or an attempt) has no negative effects on the development of LNM or cancer recurrence [8, 12‐16]. Therefore a primary local en-bloc resection strategy in order to prevent unnecessary upfront major surgery for low risk T1 CC patients is justified. Moreover, patients who underwent local endoscopic therapy have similar quality of life (QoL) without increased fear of cancer recurrence compared to patients who underwent a segmental resection [17].

In contrast to previous insights, the presence of deep invasion is no longer recognized as high risk factor for LNM. A suspected deeply-invasive T1 colon carcinoma (CC) is therefore not a strong indicator for a segmental resection [18]. Unfortunately, the therapeutic options to perform a ‘’full thickness’’ local resection in the colon are scarce. The R0-resection rate and safety of an endoscopic submucosal dissection (ESD) in the colon is significantly compromised in case of deeply invasive lesions and therefore not recommended in these cases [19, 20]. This means that the endoscopic Full Thickness Resection (eFTR) is currently the only available technique to remove deeply-invasive T1 colon carcinomas (CC) up to 15 mm [21].

Recently, we have shown that Colonoscopy-Assisted Laparoscopic Wedge Resection (CAL-WR) is an effective and safe technique to remove endoscopically unresectable benign colonic lesions (LIMERIC-trial [22]). The technical success rate of CAL-WR was 93% and a R0-resection was achieved in 91% of the patients with low morbidity (6%) and no mortality. Subsequently, CAL-WR was introduced as a feasible ‘’full-thickness’’ resection technique for the removal of suspected T1 CCs [23], however long-term oncological outcomes of CAL-WR are yet unclear. In particular for large deep-invasive lesions, CAL-WR could fill the gap between the current endoscopic resection techniques and major oncologic surgery, providing more patients an minimal invasive organ-preserving treatment option.

The aim of this study is to prospectively register the effectiveness, safety, and costs of CAL-WR as primary treatment of suspected (deep-invasive) T1 CC and to provide insight in its long-term oncological outcomes.

In this prospective multicenter clinical trial, patients with suspected T1 CC (after assessment by an expert panel) will receive CAL-WR as initial treatment in order to prevent unnecessary segmental resections for low risk T1 CC.

The substantially rising incidence of stage I CRC emphasizes the urgent need for a more widely applicable local resection technique in order to prevent unnecessary major oncologic surgery. This prospective multicenter study aims to give insight in the effectiveness, safety, and costs of CAL-WR as local resection technique for T1C CCs. Furthermore, patients will be followed for a total duration of five years, allowing us to evaluate the short and long-term oncological outcome of CAL-WR as primary resection strategy.

The biggest challenge in this study is to include patients with T1 CC as accurately as possible. Identification of malignant morphologic features such as depression, excavation and spontaneous bleeding, combined with evaluation of the vascular pattern using the Hiroshima Classification, is currently used in standard clinical practice to differentiate between adenomas and T1 CCs. In a large multicenter study in the Netherlands, Backes et al. achieved a sensitivity of 78.7% with a specificity of 94.2% using this standardized assessment [27]. Optical discrimination between deep submucosal invasive T1 and T2 CC is also challenging. To minimize the inclusion of benign or ≥ T2 lesions, we implemented an expert panel of five dedicated gastroenterologist to reassess all eligible patients. In order to reduce inter-observer variation, each independent member of the expert panel, as well as the referring gastroenterologist, will estimate the chance of T1 malignancy using international scoring systems with low-threshold use of the risk stratification developed by Backes et al. [27]. Moreover, our expert panel will be trained to macroscopically differentiate between deep submucosal invasive CCs and ≥ T2 CCs by using a scoring system composed of five endoscopic findings which was developed by Koyama et al. [28]. These 5 endoscopic findings (deep depression, demarcated depressed area, fold convergency, erosion or white plaque, and Borrmann type 2 or 3 tumor) combined in a scoring system resulted in a sensitivity and specificity of respectively 82% and 83% in their development cohort. Despite the precautions taken in this study protocol to limit inclusion of adenomas and ≥ T2 colon carcinomas, it shall be unavoidable to include purely T1 CCs. This margin is taken into account in our sample size calculation.

In contrast to earlier insights, deep submucosal invasion (T1b) is not considered as independent risk factor for lymph node metastasis [18]. In the absence of other histological risk features (i.e. poorly differentiation, high grade tumor budding, (lympho-)vascular invasion), patients with deep submucosal invasive T1 colon carcinomas have no longer an indication for major oncologic surgery. Therefore, deep submucosal invasion was not included as criterion to advice an additional oncologic resection after radical CAL-WR in this protocol. Prior endoscopic resection or attempt is in this study handled as exclusion criterion to guarantee an unbiased sample population. However, outside study context CAL-WR is also suited to perform scar excisions of irradical endoscopically removed low-risk pT1 colon carcinoma. In the recent prospective clinical trial by Leicher et al. [22, 32], a technical success rate of 90% was achieved with R0 resection rates variating from 86%-92% for irradical resected low-risk T1 colon carcinomas, however numbers were small (n = 24).

In the study of Leicher et al., macroscopic recurrent tissue was found during follow up colonoscopy in four cases of which three cases were initial radical (R0) resections. The indication to perform CAL-WR in these patients were either a difficult location of the polyp or a non-lifting polyp. In all these patients, initial and recurrence histology was benign. Considering this recurrence rate, all patients with definite benign histology in this study, will receive a follow-up colonoscopy 1 year after CAL-WR, regardless of current surveillance guidelines.

A significant part of patients with suspected T1 CC are not eligible for removal by eFTR due to the size of the lesion. Expertise with ESD in the proximal colon is often not available in non-academic hospitals. Moreover, ESD leads more often to an incomplete resection in case of deep submucosal invasion [19]. Our hypotheses is that CAL-WR will safely expand the local therapeutic armamentarium that is currently available for T1 colon carcinomas. In particular for deep-invasive CCs, CAL-WR could fill in the gap between endoscopic minimally invasive local resection techniques and major oncologic surgery, preventing unnecessary major oncologic surgery.