Long-term outcomes of robot-assisted laparoscopic surgery versus conventional laparoscopic surgery for rectal cancer: single-center, retrospective, propensity score analyses.

Although the short-term outcomes of robot-assisted laparoscopic surgery (RALS) for rectal cancer are well known, the long-term oncologic outcomes of RALS compared with those of conventional laparoscopic surgery (CLS) are not clear. This study aimed to compare the long-term outcomes of RALS and CLS for rectal cancer using propensity score matching. This retrospective study included 185 patients with stage I-III rectal cancer who underwent radical surgery at our institute between 2010 and 2019. Propensity score analyses were performed with 3-year overall survival (OS) and relapse-free survival (RFS) as the primary endpoints. After case matching, the 3-year OS and 3-year RFS rates were 86.5% and 77.9% in the CLS group and 98.4% and 88.5% in the RALS group, respectively. Although there were no significant differences in OS (p = 0.195) or RFS (p = 0.518) between the groups, the RALS group had slightly better OS and RFS rates. 3-year cumulative (Cum) local recurrence (LR) and 3-year Cum distant metastasis (DM) were 9.7% and 8.7% in the CLS group and 4.5% and 10.8% in the RALS group, respectively. There were no significant differences in Cum-LR (p = 0.225) or Cum-DM (p = 0.318) between the groups. RALS is a reasonable surgical treatment option for patients with rectal cancer, with long-term outcomes similar to those of CLS in such patients.

[Significance of Colorectal Cancer Resection for the Elderly].

INTRODUCTION: Elderly patients requiring surgical treatment is increasing in Japan, and while surgical treatment is expected to be effective even in the very elderly, there is a lack of evidence for the safety and efficacy of surgical resection due to problems with perioperative management and operative tolerance. We therefore retrospectively examined the short-term and long-term outcomes of colorectal cancer surgery for the very elderly at our hospital. SUBJECTS: The study included 14 cases of colorectal cancer in the very elderly who underwent radical resection at our hospital between January 2010 and March 2020. RESULTS: The mean age was 92 years, PS; 1/2=8/6, ASA-PS; 2/3/4=8/4/2, primary site was C/A/T/S/R= 2/5/2/2/3, pStage; 1/2/3=1/9/4, and only 1 case of decompression with ileus tube due to obstructive symptoms was treated before surgery. All patients underwent radical surgery. Median blood loss was 61 mL, median operation time was 190.5 min, and median postoperative hospital stay was 16 days. 5 patients had CD≥2 complications. All patients did not receive adjuvant chemotherapy, and recurrence in was observed 3 patients. CONCLUSION: Surgical treatment of very elderly patients seems to be acceptable under appropriate patient selection.

[A Case of Pathologically Complete Response after Preoperative Chemotherapy in a Locally Advanced Rectal Cancer].

The case is a 78-year-old male. The chief complaint was melena and weight loss. After careful examination, the patient was diagnosed with advanced rectal cancer, and 3 courses of capecitabine plus oxaliplatin therapy were performed as preoperative chemotherapy. He underwent robot-assisted laparoscopic rectal resection, D3 lymphadenectomy, lateral lymphadenectomy, and temporary colostomy, and was discharged on hospital day 15. Postoperative pathological diagnosis showed only ulcerative lesions in the rectum, and malignant cells could not be confirmed. After postoperative adjuvant chemotherapy, the patient is alive without recurrence on an outpatient basis. There are many reports that it is slightly lower than radiotherapy. Therefore, it is important to select a more appropriate preoperative treatment, and the concentration of future cases is recognized.

Long-term outcomes of robot-assisted versus minimally invasive esophagectomy in patients with thoracic esophageal cancer: a propensity score-matched study.

BACKGROUND: Recently, robot-assisted minimally invasive esophagectomy (RAMIE) has gained popularity worldwide. Some studies have compared the long-term results of RAMIE and minimally invasive esophagectomy (MIE). However, there are no reports on the long-term outcomes of RAMIE in Japan. This study compared the long-term outcomes of RAMIE and MIE. METHODS: This retrospective study included 86 patients with thoracic esophageal cancer who underwent RAMIE or MIE at our hospital from June 2010 to December 2016. Propensity score matching (PSM) was employed, incorporating co-variables such as confounders or risk factors derived from the literature and clinical practice. These variables included age, sex, body mass index, alcohol consumption, smoking history, American Society of Anesthesiologists stage, comorbidities, tumor location, histology, clinical TNM stage, and preoperative therapy. The primary endpoint was 5-year overall survival (OS), and the secondary endpoints were 5-year disease-free survival (DFS) and recurrence rates. RESULTS: Before PSM, the RAMIE group had a longer operation time (min) than the MIE group (P = 0.019). RAMIE also exhibited significantly lower blood loss volume (mL) (P < 0.001) and fewer three-field lymph node dissections (P = 0.028). Postoperative complications (Clavien-Dindo: CD ≥ 2) were significantly lower in the RAMIE group (P = 0.04), and postoperative hospital stay was significantly shorter than the MIE group (P < 0.001). After PSM, the RAMIE and MIE groups consisted of 26 patients each. Blood loss volume was significantly smaller (P = 0.012), postoperative complications (Clavien-Dindo ≥ 2) were significantly lower (P = 0.021), and postoperative hospital stay was significantly shorter (P < 0.001) in the RAMIE group than those in the MIE group. The median observation period was 63 months. The 5-year OS rates were 73.1% and 80.8% in the RAMIE and MIE groups, respectively (P = 0.360); the 5-year DFS rates were 76.9% and 76.9% in the RAMIE and MIE groups, respectively (P = 0.749). Six of 26 patients (23.1%) in each group experienced recurrence, with a median recurrence period of 41.5 months in the RAMIE group and 22.5 months in the MIE group. CONCLUSIONS: Compared with MIE, RAMIE led to no differences in long-term results, suggesting that RAMIE is a comparable technique.

Robotic True D3 Lymph Node Dissection With Superior Mesenteric Vein-Taping Technique for Right-Sided Colon Cancer.

BACKGROUND: D3 is unaffected by anatomic factors even when the ileocolic artery runs along the dorsal side of the superior mesenteric vein. Complete "true D3" lymph node dissection in minimally invasive surgery for right-sided colon cancer could be beneficial for certain patients with lymph node metastases. IMPACT OF INNOVATION: The study aimed to determine the safety and feasibility of robotic true D3 lymph node dissection for right-sided colon cancer using a superior mesenteric vein-taping technique. TECHNOLOGY, MATERIALS, AND METHODS: The superior mesenteric vein was slowly and gently separated from the surrounding tissues and taped. Lifting the tape with the robotic third arm and fixing it in place using rock-stable tractions provides a good surgical view, which cannot otherwise be obtained. As a result, the ileocolic artery that branches from the superior mesenteric artery can be accurately exposed. Handling of the taping then enables expansion to a different surgical view. As the lymph nodes are originally concealed on the dorsal side of the superior mesenteric vein, this technique provides a good view for lymph node dissection. The root of the ileocolic artery was clipped and separated, and true D3 was thus completed. PRELIMINARY RESULTS: Fourteen patients underwent robotic true D3 lymph node dissection for right-sided colon cancer. No Clavien-Dindo classification grade II or higher intraoperative or postoperative complications were observed. The 30-day mortality rate was 0%. CONCLUSIONS: Our robotic true D3 lymph node dissection with superior mesenteric vein-taping technique is considered safe and feasible; it might be a promising surgical procedure for treating advanced right-sided colon cancer. FUTURE DIRECTIONS: Even when the ileocolic artery runs along the dorsal aspect of the superior mesenteric vein, the technique seems promising for facilitating robotic D3 lymph node dissection.

Three-dimensional Computed Tomography Evaluation of the Vasculature of Right-side Colon Cancer.

BACKGROUND/AIM: From an oncological perspective, central ligation of the feeding vessel is an important approach to consider when performing colon cancer surgery. This study aimed to use three-dimensional computed tomography (3D-CT) to clarify the vascular anatomy for performing central vascular ligation to improve the accuracy of minimally invasive surgery (MIS) performed to treat advanced right-side colon cancer. PATIENTS AND METHODS: This descriptive study was conducted at one institution and targeted 92 patients with right-side colon cancer whose vascular anatomy was evaluated with 3D-CT before surgery between January 2014 and December 2020 at Tokyo Medical University Hospital. RESULTS: In 49 patients (53.3%), the ileocolic artery was ventral to the superior mesenteric vein (SMV), whereas in 43 patients (46.7%), it was dorsal to the SMV. The right colic artery was present in 31 patients (33.7%). The middle colic artery was present in all patients (100%). A common duct type was present in 80 patients (87.0%). Branching directly from the superior mesenteric artery without a common duct was observed in 12 patients (13.0%). Twenty-one patients (22.9%) had an accessory superior mesenteric artery. CONCLUSION: The vascular structure of the right-side colon is highly complex. Conducting 3D-CT evaluations of the vessel anatomy is very useful for surgeons who conduct MIS, and is considered to enable central ligation to be performed safely and improve the quality of surgery, which will benefit patients.

Learning Curve of Intracorporeal Anastomosis in Laparoscopic Colectomy for Right Side Colon Cancer: A Cumulative Sum Analysis.

BACKGROUND/AIM: Recently, laparoscopic colectomy with intracorporeal anastomosis for colon cancer gained popularity due to the evolution of the laparoscopic linear stapler device and improved techniques from laparoscopic surgeons. However, there are technical difficulties associated with intracorporeal anastomosis. The aim of the study was to clarify the number of cases that are required for laparoscopic surgeons to master the technique of intracorporeal anastomosis in right side colon cancer. PATIENTS AND METHODS: In this retrospective single-center study, 51 consecutive patients who underwent intracorporeal overlap anastomosis, between July 2018 and March 2020, by one laparoscopic surgeon were selected. Clinicopathological and perioperative data were obtained from our database. The learning curves of intracorporeal anastomosis time (IAT) were created using the cumulative sum (CUSUM) method. RESULTS: The CUSUM score for IAT increased as the number of operative cases progressed, up to the 20th case (Phase 1), after which it started to decrease (Phase 2). Compared to the initial learning phase (Phase 1), the master phase (Phase 2) had a significantly faster IAT (p<0.001), significantly decreased incidence of organ/space surgical site infection (p=0.009), and significantly decreased postoperative hospital stay (p=0.021). CONCLUSION: Twenty cases were required for a laparoscopic surgeon to achieve expertise when conducting intracorporeal anastomosis in laparoscopic colectomy for right side colon cancer. It was suggested that proficiency in intracorporeal anastomosis may contribute to a reduction in the incidence of organ/space surgical site infections and postoperative hospital stay.

Robotic versus laparoscopic approach for minimally invasive lateral pelvic lymph node dissection of advanced lower rectal cancer: a retrospective study comparing short-term outcomes.

PURPOSE: The importance of lateral pelvic lymph node dissection (LLND) for advanced low rectal cancer is gradually being recognized in Europe and the USA, where some patients were affected by uncontrolled lateral pelvic lymph node (LLNs) metastasis, even after total mesorectal excision (TME) with neoadjuvant chemoradiotherapy (CRT). The purpose of this study was thus to compare robotic LLND (R-LLND) with laparoscopic (L-LLND) to clarify the safety and advantages of R-LLND. METHODS: Sixty patients were included in this single-institution retrospective study between January 2013 and July 2022. We compared the short-term outcomes of 27 patients who underwent R-LLND and 33 patients who underwent L-LLND. RESULTS: En bloc LLND was performed in significantly more patients in the R-LLND than in the L-LLND group (48.1% vs. 15.2%; p = 0.006). The numbers of LLNs on the distal side of the internal iliac region (LN 263D) harvested were significantly higher in the R-LLND than in the L-LLND group (2 [0-9] vs. 1 [0-6]; p = 0.023). The total operative time was significantly longer in the R-LLND than in the L-LLND group (587 [460-876] vs. 544 [398-859]; p = 0.003); however, the LLND time was not significantly different between groups (p = 0.718). Postoperative complications were not significantly different between the two groups. CONCLUSION: The present study clarified the safety and technical feasibility of R-LLND with respect to L-LLND. Our findings suggest that the robotic approach offers a key advantage, allowing significantly more LLNs to be harvested from the distal side of the internal iliac region (LN 263D). Prospective clinical trials examining the oncological superiority of R-LLND are thus necessary in the near future.

[Short-Term and Long-Term Results of Robot-Assisted Rectal Resections Performed in Our Department].

BACKGROUND: We introduced the da Vinci robotic surgical system in 2006 for the first time in Japan, and have been performing robot-assisted rectal cancer surgeries since 2010, after receiving approval from the hospital's Ethics Review Committee in 2009. Here we report the long-term and short-term outcomes of robot-assisted rectal cancer surgeries performed in our department. METHODS: Target patients were those who underwent robot-assisted radical rectal resection for rectal cancer; 165 patients in the short term(2010-2021), and 49 patients in the long term(2010-2016). Data were retrospectively analyzed, and Kaplan-Meier curves were used for the survival analysis. RESULTS: The short-term results are summarized in Table 1. The long-term results were as follows: 5-year overall survival rate, 90.8%; 5-year recurrence-free survival rate, 90.6%; 5-year cumulative local recurrence rate, 7.3%; 5-year cumulative distant metastasis rate, 9.4%. CONCLUSION: In our department, 11 years have passed since we began performing robotic rectal surgeries, and the short- and long-term results have generally been acceptable.

Robotic total mesorectal excision for rectal cancer based on the theory of fundamental use of surgical energy.

BACKGROUND: Improving oncological curability and preserving urinary function must be established in robotic total mesorectal excision (TME) for rectal cancer. To achieve this, it is important to avoid nerve injury by sharp dissection of the avascular plane by the monopolar device and thermal spread. The aim of this study was to improve the robotic TME quality by focusing on the theory of fundamental use of surgical energy (FUSE) of the monopolar device and investigating the surgical procedure. METHODS: In this single-center retrospective study, 26 consecutive patients who underwent robotic TME for rectal cancer at Tokyo Medical University Hospital between June 2022 and August 2022 were included. All surgeries were performed by FUSE-certified surgeons in accordance with FUSE theory, which was, thermal effect = current density (current/area) squared × tissue resistance × contact time. RESULT: The median age of the patients was 64 years (range 40-79 years), and 17 patients were male. The median operative time was 287 min (range 229-430 min); median bleeding volume, 22 ml (range 5-223 ml); and the median number of harvested lymph nodes, 17 (range 4-40). The conversion rate to open surgery was 0%. A breakdown of Clavien-Dindo classification Grade ≥ II post-operative complications was as follows: surgical site infection, one patient (3.8%); ileus, one patient (3.8%); and urinary dysfunction, one patient (3.8%). No surgery-related or in-hospital deaths occurred. The pathological positive resection margin was not observed. CONCLUSIONS: Robotic TME for rectal cancer based on the theory of FUSE can be safely performed, making it a promising surgical procedure. It is suggested that robotic surgeons acquire surgical operation skills with monopolar devices based on the principles of FUSE, which may lead to an increased quality of robotic TME.

[Preoperative Prognostic Factors in CRS plus HIPEC for Patients with Peritoneal Dissemination of Colorectal Cancer].

Peritoneal dissemination of colorectal cancer has the poorest prognosis among metastatic sites, with an average overall survival of less than 6 months. Various treatment methods have been reported for these patients, and recently there have been several reports showing the usefulness of cytoreductive surgery(CRS)combined with hyperthermic intraperitoneal chemotherapy (HIPEC). However, the studies on this treatment are limited. In this study, we retrospectively reviewed cases of CRS plus HIPEC. Twenty-one patients who underwent CRS plus HIPEC at Tokyo Medical University Hospital and Toda Central General Hospital between August 2014 and December 2017 were included in this study. The long-term and short-term survival groups were analyzed separately, and predictors of preoperative treatment efficacy were examined. The surgical approach was open in 16 cases and laparoscopic in 5 cases. Complete resection was achieved in 10 of these patients. Postoperative complications occurred in 6 patients. There were no deaths within 90 days of surgery. The median overall survival was 17.0 months, and the 1-year survival rate was 65%. Median progression-free survival was 11.0 months. In a multivariate analysis predicting long-term versus short-term survival groups, sex, primary tumor location, and P factor were independent predictors of treatment response. CRS plus HIPEC therapy is considered an effective treatment option. The predictors of preoperative treatment response include sex, primary tumor location, and P factor.

Predicting the prognosis of lower rectal cancer using preoperative magnetic resonance imaging with artificial intelligence.

BACKGROUND: There are various preoperative treatments that are useful for controlling local or distant metastases in lower rectal cancer. For planning perioperative management, preoperative stratification of optimal treatment strategies for each case is required. However, a stratification method has not yet been established. Therefore, we attempted to predict the prognosis of lower rectal cancer using preoperative magnetic resonance imaging (MRI) with artificial intelligence (AI). METHODS: This study included 54 patients [male:female ratio was 37:17, median age 70 years (range 49-107 years)] with lower rectal cancer who could be curatively resected without preoperative treatment at Tokyo Medical University Hospital from January 2010 to February 2017. In total, 878 preoperative T2 MRIs were analyzed. The primary endpoint was the presence or absence of recurrence, which was evaluated using the area under the receiver operating characteristic curve. The secondary endpoint was recurrence-free survival (RFS), which was evaluated using the Kaplan-Meier curve of the predicted recurrence (AI stage 1) and predicted recurrence-free (AI stage 0) groups. RESULTS: For recurrence prediction, the area under the curve (AUC) values for learning and test cases were 0.748 and 0.757, respectively. For prediction of recurrence in each case, the AUC values were 0.740 and 0.875, respectively. The 5-year RFS rates, according to the postoperative pathologic stage for all patients, were 100%, 64%, and 50% for stages 1, 2, and 3, respectively (p = 0.107). The 5-year RFS rates for AI stages 0 and 1 were 97% and 10%, respectively (p < 0.001 significant difference). CONCLUSIONS: We developed a prognostic model using AI and preoperative MRI images of patients with lower rectal cancer who had not undergone preoperative treatment, and the model could be useful in comparison with pathological classification.

[A Case of Rectal Cancer with Solitary Lymph Node Metastasis in the Stomach after Hepatectomy for Liver Metastasis].

A 44-year-old woman had undergone a laparoscopic low anterior resection and lymph node dissection for rectal cancer (pT4aN2aH0P0M0, pStage Ⅲc)in 20XX. Six months postoperatively, a CT scan revealed recurrent liver metastasis. She underwent surgery and adjuvant chemotherapy. Three years after the initial surgery, her liver metastasis recurred again, and the patient underwent another cycle of surgical treatment and adjuvant chemotherapy. Five years after the initial surgery, a lesion was found in a gastric lesser curvature lymph node. Gastric kyphosis lymph node dissection was performed under the suspicion of a solitary lymph node metastasis. The resected lymph node was diagnosed as a medium-differentiated adenocarcinoma, with findings consistent with a lymph node metastasis from the initial rectal cancer. Postoperative adjuvant chemotherapy was administered. No recurrence was noted 6 years and 6 months after the initial surgery. Rectal cancer rarely metastasizes to the gastric lymph nodes in a solitary fashion. We describe a case of a solitary gastric regional lymph node metastasis observed after the resolution of previous liver metastases.

Retroperitoneal hibernoma resected via laparoscopic surgery.

Hibernomas are extremely rare, benign tumors of brown fat origin with no specific symptoms. Surgery is the only treatment option, and because a definitive preoperative diagnosis is often not obtained, open surgery is usually chosen. In this case, we performed laparoscopic surgery on a 33-year-old woman with retroperitoneal hibernoma. As in most cases, a definitive diagnosis had not been preoperatively made; therefore, we laparoscopically removed the retroperitoneal tumor of unknown pathology as a diagnostic treatment. We chose laparoscopic surgery because of the magnifying effect of the laparoscope and to minimize scarring. The surgery was uneventful, with a procedure time of 280 minutes and a blood loss of 20 mL. The postoperative course was uneventful with no complications or recurrence. We conclude that laparoscopic surgery may be a viable option for hibernomas.

Predictive modelling for high-risk stage II colon cancer using auto-artificial intelligence.

BACKGROUND: Recently, stratification of high-risk stage II colon cancer (CC) and the need for adjuvant chemotherapy have been the focus of attention. The aim of this retrospective study was to define high-risk factors for recurrent stage II CC using Prediction One auto-artificial intelligence (AI) software and develop a new predictive model for high-risk stage II CC. METHODS: The study included 259 consecutive pathological stage II CC patients undergoing curative resection at our institution between January 2000 and December 2016. Prediction One software with five-fold cross-validation was used to create a predictive model and receiver operating characteristic (ROC) curve. Predictive accuracy of AI was evaluated using the area under the ROC curve (AUC). We also evaluated the importance of variables (IOV) using a method based on permutation feature importance (IOV > 0.01 defined high-risk factors) to evaluate disease-free survival (DFS). RESULTS: The median observation period was 6.1 (range = 0.3-15.8) years. Thirty-seven patients had recurrence (14.3%); the AUC of the AI model was 0.775. Preoperative carcinoembryonic antigen > 5.0 ng/mL (IOV = 0.047), venous invasion (IOV = 0.014), and obstruction (IOV = 0.012) were high-risk factors contributing to cancer recurrence. Patients with 2-3 high-risk factors had lower 5-year DFS than those with 0-1 factor (87.4% vs 62.7%, p < 0.001). CONCLUSIONS: We developed a new predictive model that could predict recurrent high-risk stage II CC with high probability using auto-AI Prediction One software. Patients with ≥ 2 of the aforementioned factors are considered to have high risks for recurrent stage II CC and may benefit from adjuvant chemotherapy.

Significance of Zinc Replacement Therapy After Pancreaticoduodenectomy.

BACKGROUND/AIM: Recently, a decrease in serum zinc levels and the need for zinc preparations have been reported in the perioperative period of gastrointestinal surgery. In this study, we examined treatment outcomes among patients supplemented with zinc after pancreaticoduodenectomy (PD) and evaluated the significance of zinc replacement therapy. PATIENTS AND METHODS: From June 2020 to April 2021, 56 patients who received zinc acetate hydrate (50 mg/day) from postoperative day 3 after PD in our department were retrospectively reviewed. Patients' characteristics and preoperative as well as postoperative data, including serum zinc levels and surgical results at 1 month were reviewed. RESULTS: Preoperative zinc deficiency was present in 86.1% (46/56) of the patients. Moreover, despite zinc supplementation, 17.8% (10/56) of patients had postoperative zinc deficiency. A comparison between the low zinc level group (Zn <80 µg/dl) and the normal zinc level group (Zn ≥80 µg/dl) after surgery showed siginificant differences among patients with malignant diseases (vs. benign diseases, p=0.044), those undergoing open surgery (vs. minimally invasive surgery, p=0.036), and those with intraoperative blood loss ≥346 ml (vs. <346 ml: p=0.041) in the univariate analysis. Multivariate analysis revealed that zinc deficiency was significantly associated with open surgery [odds ratio (OR)=15.885, 95% confidence interval (CI)=1.77-142.01, p=0.013] and intraoperative blood loss (OR=9.329, 95% CI=1.50-57.74, p=0.016). CONCLUSION: In patients undergoing open PD for pancreatic cancer, zinc preparations of 50 mg may not be sufficient and further supplementation may be necessary.

Preliminary Evaluation of a Novel Artificial Intelligence-based Prediction Model for Surgical Site Infection in Colon Cancer.

BACKGROUND/AIM: There are few studies on artificial intelligence-based prediction models for colon cancer built using clinicopathological factors. Here, we aimed to perform a preliminary evaluation of a novel artificial intelligence-based prediction model for surgical site infection (SSI) in patients with stage II-III colon cancer. PATIENTS AND METHODS: The medical records of 730 patients who underwent radical surgery for stage II-III colon cancer between 2000 and 2018 at our institute were retrospectively analyzed. Kaplan-Meier curves were used to examine the association between SSI and oncological outcomes (recurrence-free survival time). Next, we used the machine learning software Prediction One to predict SSI. Receiver-operating characteristic curve analysis was used to evaluate the accuracy of the artificial intelligence model. RESULTS: The prognosis in terms of recurrence-free survival time was poor in patients with SSI (p=0.005, 95% confidence interval=4892.061-5525.251). The area under the curve of the artificial intelligence model in predicting SSI was 0.731. CONCLUSION: As SSI is an important prognostic factor associated with oncological outcomes, the prediction of SSI occurrence is important. Based on our preliminary evaluation, the artificial intelligence model for predicting SSI in patients with stage II-III colon cancer was as accurate as the previously reported model derived through conventional statistical analysis.

Artificial intelligence predicts lymph node metastasis or risk of lymph node metastasis in T1 colorectal cancer.

BACKGROUND: The treatment strategies for colorectal cancer (CRC) must ensure a radical cure of cancer and prevent over/under treatment. Biopsy specimens used for the definitive diagnosis of T1 CRC were analyzed using artificial intelligence (AI) to construct a risk index for lymph node metastasis. METHODS: A total of 146 T1 CRC cases were analyzed. The specimens for analysis were mainly biopsy specimens, and in the absence of biopsy specimens, the mucosal layer of the surgical specimens was analyzed. The pathology slides for each case were digitally imaged, and the morphological features of cancer cell nuclei were extracted from the tissue images. First, statistical methods were used to analyze how well these features could predict lymph node metastasis risk. A lymph node metastasis risk model using AI was created based on these morphological features, and accuracy in test cases was verified. RESULTS: Each developed model could predict lymph node metastasis risk with a > 90% accuracy in each region of interest of the training cases. Lymph node metastasis risk was predicted with 81.8-86.3% accuracy for randomly validated cases, using a learning model with biopsy data. Moreover, no case with lymph node metastasis or lymph node risk was judged to have no risk using the same model. CONCLUSIONS: AI models suggest an association between biopsy specimens and lymph node metastases in T1 CRC and may contribute to increased accuracy of preoperative diagnosis.