Multicentric Study About Pathological Risk Factors for Lymph Node Metastasis in Malignant Colorectal Polyps

Overview

Colorectal cancer screening showed an increased incidence of malignant colorectal polyps pT1 after endoscopic excision. Their management is not yet standardized, for the presence of histological features increasing early lymph node involvement. The literature has proposed several histopathological criteria, for which the risk of lymph node metastasis can vary (6-20%), but final data are not yet available. Aim 1.To collect data about patients undergoing an endoscopic polypectomy with histologic finding of pT1, retrospectively and prospectively, dividing both databases into two groups, endoscopic group (EG) and surgical group (SG) Aim 2. To analyze retrospectively which pathological criteria can increase the risk of lymph node metastasis and to elaborate a prognostic score for lymph node metastatic risk Aim 3. To verify prospectively the prognostic score capacity on predicting lymph node metastasis Aim 4. To calculate the disease free survival, overall survival, local recurrence rate and distal recurrence rate and verify if there is a difference between EG and SG According to literature, the most important histopathological criteria to establish the high risk of lymph node metastasis are: 1. Lateral margin of healthy tissue (high risk: \<1mm and piecemeal polypectomy) 2. Depth of submucosa invasion (high risk: \>1000 μM or sm2-sm3 for sessile polyps; Haggitt level 4 for pedunculated polyps) 3. Vascular invasion (high risk: presence) 4. Lymphatic invasion (high risk: presence) 5. Tumor budding (high risk: presence) 6. Tumor differentiation (high risk: grade G3-G4 or mucinous) A database will be used by all participating centres for collecting clinical and pathological data. All the analyses will be centralized by the PI. Uni-multivariate analyses will be conducted at the end of data collection for retrospective arm and at 2 years of follow-up for prospective arm. Impact: This study aimed to investigate pathological risk factors for lymph node metastasis in pT1 colorectal polyps after endoscopic polypectomy; their accurate identification could lead to improve their management, avoiding useless complementary surgery. Results could change clinical practice and reduce health-related costs.

The management of malignant colorectal tumors pT1 after polypectomy is not yet standardized, for the presence of histological features increasing early lymph node involvement. Three recent meta-analysis have identified some pathological criteria influencing the risk of lymph node metastasis in early colorectal cancer, but there is not agreement in the conclusions between these articles. Because this uncertainty, the decision making criteria for this patients can be different in every Institute. Furthermore, the majority of the literature refers to surgical patients, with lack in demonstrating, at long-term, the absence of lymph nodes metastasis in patients not operated after endoscopic polypectomy. Retrospective study: From a retrospective period of 6 years, with at list 2 years of follow-up, all patients who underwent endoscopic polypectomy for malignant colorectal polyp pT1, will be enrolled for the study. The patients will be divided into two groups: patients who did only the endoscopic treatment (EG), and patients who underwent complementary surgery (SG). The most important histopathological criteria to list in the database, when available (to establish the high risk of lymph node metastasis) are as follows: 1. Lateral margin of healthy tissue (high risk: \<1mm and piecemeal polypectomy) 2. Depth of submucosa invasion (high risk: \>1000 μM or sm2-sm3 for sessile polyps and Haggitt level 4 for pedunculated polyps) 3. Vascular invasion (high risk: presence); method of determination (Hematoxylin \& Eosin or immunohistochemical marker) 4. Lymphatic invasion (high risk: presence); method of determination (Hematoxylin \& Eosin or immunohistochemical marker) 5. Tumor budding (high risk: presence) 6. Tumor differentiation (high risk: grade G3-G4 or mucinous) Other data to be reported are as follows: 1. Demographic and clinical data for all patients 2. Decisional context (multidisciplinary group decision; single specialist decision; patient decision) and decision making criteria for complementary surgery for all patients 3. Type of surgical resection for SG, access (laparotomy, laparoscopy or transanal), site of lesion (right, transverse, left, sigmoid colon or rectum) 4. Histopathological criteria, other than the 6 high risk criteria above 5. Histopathological finding after surgical resection for SG 6. Postoperative data (hospital stay, morbidity with Clavien-Dindo's classification) for SG 7. Type and length of follow-up for all patients (EG and SG) (endoscopic surveillance, thoracic and abdominal CT scan, positron emission computed tomography (PET-CT), carcinoembryonic antigen (CEA) determination, clinical examination) Prospective study: According to the histological characteristics, the patients will be considered with low risk (without any histological criteria), or with high risk (with at least one of six criteria). The histopathological criteria to consider the malignant polyps at high risk of lymph node metastasis are as follows: 1. Lateral margin of healthy tissue \<1mm and piecemeal polypectomy 2. Depth of submucosa invasion \>1000 μM or sm2-sm3 for sessile polyps and Haggitt level 4 for pedunculated polyps 3. Presence of vascular invasion, determined with the immunohistochemical marker CD34 or, in alternative, with Hematoxylin \& Eosin 4. Presence of lymphatic invasion, determined with the immunohistochemical marker D2-40 or, in alternative, with Hematoxylin \& Eosin 5. Presence of tumor budding (low and high grade) 6. Tumor differentiation grade G3-G4 or mucinous Patients who will present at least one of these six criteria, will be considered at high risk for lymph node metastasis. Every Institute will take an independent decision about therapeutic strategy: only endoscopic treatment (EG), or complementary surgery (SG). To participate to the prospective study, is mandatory to obtain these 6 pathological characteristics. It is strongly recommended a multidisciplinary discussion for all patients, in presence of at least one surgeon, one endoscopist, one pathologist and one oncologist, as well as one radiotherapist for rectal lesions. The surgical resection for the colon can be performed with laparotomic, laparoscopic or robotic approach, based on single Institute surgical experience. For the rectum, the surgical approach (laparotomic, laparoscopic, robotic, transanal resection) will be decided by the surgical team in accordance with the multidisciplinary group, based on single Institute experience and patient clinical features. All patients (EG and SG) have to perform within one month from diagnosis or, in any way, before the surgery for SG, a thoracic and abdominal CT scan with contrast (for staging) and CEA determination. The histopathological finding to be reported after surgical resection for SG are as follows: number of lymph nodes, number of metastatic lymph nodes, margin distance from the site of polypectomy, local residual tumor. Follow-up program: * EG: colonoscopy at 6 and 12 months for colic lesions, recto-sigmoidoscopy at 6, 18 and 24 months and colonoscopy at 12 months for rectal lesions * SG: colonoscopy at 12 months for colic lesions, recto-sigmoidoscopy at 6, 18 and 24 months and colonoscopy at 12 months for rectal lesions * Thoracic and abdominal CT scan with contrast at 12 and 24 months * Clinical examination and CEA determination at 6, 12, 18 and 24 months Statistic analysis: Retrospective study: considering a 10% prevalence of pN+ patients with pT1 colorectal tumor, including the above 6 histopathological criteria as predictors in logistic regression model, the investigators estimate to retrospectively collect data from 600 patients. The continuous numerical data will be expressed as average and standard deviation or as median and interquartile range, if appropriate. A descriptive analysis for collected variables will be done and the pN+ prevalence will be calculated. A logistic recession model will be estimated, for evaluating the 6 histopathological criteria effect on presence of lymph node metastasis. Furthermore, the disease-free survival will be calculated with Kaplan-Meier method. A p-value \< 0.05 will be considered significant. Statistic analysis will be done with Software 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria). Prospective study: the investigators want to evaluate a sensibility of 0.80 for overall histopathological criterion (i.e., patients with at least one of six criteria are considered at high risk, patients without the six criteria are at low risk), in order that the maximum estimation error does not exceed 0.10, with a 95% confidence interval. Considering a 10% prevalence of pN+ patients with pT1 colorectal tumor, including the above 6 histopathological criteria as predictors in logistic regression model, the investigators estimate to prospectively collect data from 615 patients. The continuous numerical data will be expressed as average and standard deviation or as median and interquartile range, if appropriate. A descriptive analysis for collected variables will be done and the pN+ prevalence will be calculated. The diagnostic capacity of the overall histopathological criterion will be evaluated calculating sensibility, specificity, positive predictive value, negative predictive value and accuracy. A similar evaluation will be done for every single histopathological criterion with explorative purpose, if appropriate. A p-value \< 0.05 will be considered significant. Statistic analysis will be done with Software 3.3.2 (R Foundation for Statistical Computing, Vienna, Austria). Statement: With the retrospective observational study, the investigators can quickly have an extended overview about the current management of endoscopic pT1, to verify if there is, in fact, decisional heterogeneity, and to analyze the pathological criteria influence in lymph nodes metastasis. The limit of retrospective study, in the other side, is data heterogeneity, but this topic is recent, and it is normal not to have uniformity between different Institutes. This problem could be solved with the prospective study, where pathological criteria are included in the protocol itself. The strong points in this prospective project are: 1) uniformity in pathological analysis, 2) without any mandatory indication for complementary surgery, in order to maintain the observational nature of the study; 3) the multidisciplinary discussion for every patient, an aspect more and more required by hospital administrations, and 4) the uniformity in follow-up, that is a feature not yet clear in the literature for endoscopic pT1. Furthermore, this prospective study can be the base to create a national register, to set up future studies on this topic, as analysis of tumor molecular characteristics. The elaboration of this protocol followed the STROBE statement Guidelines and Explanation.