Treatment response prediction of neoadjuvant chemotherapy for rectal cancer by deep learning of colonoscopy images.

In current clinical practice, several treatment methods, including neoadjuvant therapy, are being developed to improve overall survival or local recurrence rates for locally advanced rectal cancer. The response to neoadjuvant therapy is usually evaluated using imaging data collected before and after preoperative treatment or postsurgical pathological diagnosis. However, there is a need to accurately predict the response to preoperative treatment before treatment is administered. The present study used a deep learning network to examine colonoscopy images and construct a model to predict the response of rectal cancer to neoadjuvant chemotherapy. A total of 53 patients who underwent preoperative chemotherapy followed by radical resection for advanced rectal cancer at the Osaka University Hospital between January 2011 and August 2019 were retrospectively analyzed. A convolutional neural network model was constructed using 403 images from 43 patients as the learning set. The diagnostic accuracy of the deep learning model was evaluated using 84 images from 10 patients as the validation set. The model demonstrated a sensitivity, specificity, accuracy, positive predictive value and area under the curve of 77.6% (38/49), 62.9% (22/33), 71.4% (60/84), 74.5% (38/51) and 0.713, respectively, in predicting a poor response to neoadjuvant therapy. Overall, deep learning of colonoscopy images may contribute to an accurate prediction of the response of rectal cancer to neoadjuvant chemotherapy.

Surgery for Rectal Cancer With Mixed Reality of 3D Model During Operation: A Case Report.

Background/Aim: Recently, robotic surgery for rectal cancer has become a common minimally invasive surgery. In addition, the technology of augmented and mixed reality is applied in various living environments, including medicine. We successfully performed robotic surgery for rectal cancer with three-dimensional (3D) images as mixed reality (MR) using HoloLens2. Case Report: The patient was diagnosed with rectal cancer by colonoscopy and a positron-emission computed-tomography scan, and we performed robot-assisted anterior resection. The operator used HoloLens2 and performed the surgery while visualizing 3D images of pelvic anatomy with the location of the rectal cancer as hologram. The operation was performed completely and safely, and she was discharged 11 days after surgery with no postoperative complications. Conclusion: This case presents the usefulness of a MR system offering organ visualization as hologram during surgery.

Fibroblast Activation Protein and Tertiary Lymphoid Structure in Colorectal Cancer Recurrence.

BACKGROUND/AIM: Fibroblast activation protein (FAP) is known to have prognostic significance in colorectal cancer (CRC). However, FAP and tertiary lymphoid structures (TLSs) have not been associated with each other in predicting the prognosis of CRC recurrence. PATIENTS AND METHODS: FAP expression was evaluated by real-time reverse transcription polymerase chain reaction in 195 CRC patients at Osaka International Cancer Institute (first data set). Immunohistochemistry (IHC) was then performed to stain FAP at the invasive margin (IM) and in the central tumour (CT) in 159 CRC patients at Osaka University Hospital (second data set). Consecutive slides were used to evaluate the presence of TLSs in 159 CRC patients from Osaka University Hospital. RESULTS: The high FAP mRNA expression group (n=82) was associated with poor recurrence-free survival (RFS) compared with the low FAP expression group (n=83) (p=0.004). In the second data set, patients with high FAP expression in CT and TLS absence (n=49) showed significantly poorer RFS compared with those with low expression of FAP in CT and presence of TLSs (n=101) (p=0.002). CONCLUSION: FAP in the CT combined with TLSs was shown to have significant prognostic value in predicting CRC recurrence after curative resection.

Diagnosis of Depth of Submucosal Invasion in Colorectal Cancer with AI Using Deep Learning.

The submucosal invasion depth predicts prognosis in early colorectal cancer. Although colorectal cancer with shallow submucosal invasion can be treated via endoscopic resection, colorectal cancer with deep submucosal invasion requires surgical colectomy. However, accurately diagnosing the depth of submucosal invasion via endoscopy is difficult. We developed a tool to diagnose the depth of submucosal invasion in early colorectal cancer using artificial intelligence. We reviewed data from 196 patients who had undergone a preoperative colonoscopy at the Osaka University Hospital and Osaka International Cancer Institute between 2011 and 2018 and were diagnosed pathologically as having shallow submucosal invasion or deep submucosal invasion colorectal cancer. A convolutional neural network for predicting invasion depth was constructed using 706 images from 91 patients between 2011 and 2015 as the training dataset. The diagnostic accuracy of the constructed convolutional neural network was evaluated using 394 images from 49 patients between 2016 and 2017 as the validation dataset. We also prospectively tested the tool from 56 patients in 2018 with suspected early-stage colorectal cancer. The sensitivity, specificity, accuracy, and area under the curve of the convolutional neural network for diagnosing deep submucosal invasion colorectal cancer were 87.2% (258/296), 35.7% (35/98), 74.4% (293/394), and 0.758, respectively. The positive predictive value was 84.4% (356/422) and the sensitivity was 75.7% (356/470) in the test set. The diagnostic accuracy of the constructed convolutional neural network seemed to be as high as that of a skilled endoscopist. Thus, endoscopic image recognition by deep learning may be able to predict the submucosal invasion depth in early-stage colorectal cancer in clinical practice.

DEB-TACE combined with hepatic artery infusion chemotherapy might be an affordable treatment option for advanced stage of HCC.

Alternative treatment modalities are necessary because of the low response rates and unsuitability of molecular-targeted agents (MTA) and/or immune checkpoint inhibitors (iCIs) in HCC patients. Therefore, we analyzed whether drug-eluting beads (DEB)-transcatheter arterial chemoembolization (TACE) with low-dose-FP (Ultra-FP) therapy could improve the efficacy and safety of treatment in difficult-to-treat HCC patients, especially those with advanced stage HCC. From November 2017 to April 2021, 118 consecutive patients with non-resectable difficult-to-treat HCC were included in this study. All patients were treated with Ultra-FP therapy. After the weak DEB-TACE procedure, we administered low-dose FP for 2 weeks followed by resting for 4 weeks. The numbers of HCC patients CR/PR/SD/PD induced by Ultra-FP therapy were 36/52/17/13 (Modified RECIST) patients, respectively. The objective response rate of Ultra-FP therapy was 74.6% (88/118 patients). Tumor marker reduction was observed in 81.4% (96/118 patients). The objective response rate (ORR) in the HCC patients with portal vein tumor thrombosis (PVTT) was 75% (18/24 patients). Median overall survival (mOS) of all included HCC patients was 738 days. The mOS of HCC patients with PVTT (-)/PVTT (+) was 816 days/718 days. The proportion of patients based on ALBI grade system was not significantly different between pre- and after 3 course Ultra-FP therapy. Ultra-FP therapy might be an affordable treatment option for difficult-to-treat advanced HCC. ORR and overall survival after receiving Ultra-FP therapy were remarkable in comparison to various kinds of systemic therapy including MTA and iCIs.

Combined Inflammation and Nutrition Factors Reinforce the Prognostic Prediction for Stage III Colorectal Cancer Patients.

BACKGROUND/AIM: Previous studies have shown that postoperative adjuvant chemotherapy improves overall survival in patients with stage III colorectal cancer (CRC). However, adjuvant chemotherapy may not be necessary for some patients. This study aimed to develop a new nutritional-inflammation score, which would be useful in identifying a favorable prognosis group among stage III CRC patients. PATIENTS AND METHODS: This retrospective study included 262 patients with stage III CRC who underwent curative surgery and were divided into two groups: a training set (TS) of 162 patients and a validation set (VS) of 100 patients. In the TS, clinicopathological factors were tested using a Cox regression model, and a new prognostic model was developed. RESULTS: Multivariate analyses in TS revealed that lymph node metastasis (N2) (p=0.002), low albumin (p=0.017), high monocyte counts (p=0.008), and low platelet counts (p=0.018) were independent risk factors for disease free survival (DFS). The Kansai prognostic score (KPS) was assessed by 1 point each for <3.5 g/dl albumin level, >450 monocyte counts, and <1.65×105 platelet counts. Using KPS, DFS and overall survival (OS) were validated in VS. The C-indices of KPS to predict DFS and OS in TS were 0.707 and 0.772. It was validated in VS that the C-indices of KPS to predict DFS and OS were 0.618 and 0.708, respectively. A high KPS was a significant predictor of DFS and OS. CONCLUSION: KPS serves as a new model for the prognosis of patients with stage III CRC.

The Geriatric Nutritional Risk Index as a Prognosis Predictor in Patients With Rectal Cancer Receiving Neoadjuvant Chemotherapy.

BACKGROUND/AIM: There are few reports on the clinical significance of the geriatric nutritional risk index (GNRI) in patients with locally advanced rectal cancer who undergo preoperative chemotherapy (NAC, neoadjuvant chemotherapy) followed by radical resection; this study examined the relationship between preoperative GNRI, postoperative complications, and prognosis in these patients. PATIENTS AND METHODS: Fifty-seven patients with rectal cancer who underwent radical resection after NAC at Osaka University Hospital between November 2011 and May 2018 were included. The GNRI was calculated as follows: GNRI= [1.489×serum albumin level (g/l)]+[41.7×present/ideal body weight (kg)]. Patients were classified into high (GNRI ≥96.74; n=36) and low GNRI (GNRI <96.74; n=21) groups, based on the results of the receiver operating characteristic curve analysis. RESULTS: The Kaplan-Meier analysis showed that the low GNRI group had a significantly poorer cancer-specific survival (CSS) and a poorer overall survival tendency than the high GNRI group. In the univariate analysis, venous invasion, lymphatic vessel invasion, and low GNRI were significantly correlated with CSS; depth of tumor invasion, lymph node metastasis, and lymphatic vessel invasion were significantly correlated with disease-free survival (DFS). In the multivariate analysis, there were no significantly poor prognostic factors for CSS and DFS. CONCLUSION: Preoperative GNRI may be a useful predictor for recurrence and poor prognosis in elderly patients with rectal cancer who undergo radical resection after NAC. Further studies and accumulation of cases should investigate the relationship between preoperative GNRI and prognosis after NAC in elderly patients.

[A Case of Long-Term Disease Control after Administration of Oxaliplatin for Esophagogastric Junction Squamous Cell Carcinoma].

There are few reports of using oxaliplatin(L-OHP)for esophageal squamous cell carcinoma.We report a case of long-term disease control after administration of L-OHP for esophagogastric junction squamous cell carcinoma(EG).A woman in her 40s was diagnosed with EG(cT3, cN2, cM0, cStage Ⅲ).She received thoracoscopic, laparoscopic-assisted, subtotal esophagectomy with 3-field dissection after 3 courses of preoperative chemotherapy with DCF(docetaxel hydrate, cisplatin, 5- fluorouracil).Reconstruction was achieved using a retrosternally shifted gastric tube and transesophageal gastro-tubing. Pathological examination showed EG(ypT3, ypN2, ypM0, ypStage Ⅲ)(chemotherapy evaluation: Grade 1a).After 12 months, para-aortic lymph node recurrence(#112aoP, #16a2lat)was observed on a follow-up CT examination.First, we administered 5 courses of chemotherapy with SOX(S-1 100mg/m2 day 1-14 and L-OHP 100 mg/m2 day 1).Recurrent lymph nodes shrunk slightly, and there were no new lesions.Subsequently, there was no other adverse event except for Grade 1 chemotherapy-induced neuropathy.Second, we administered 3 courses of chemotherapy with FOLFOX(5-FU 400mg/m2, L-OHP 85mg/m2, Leucovorin 200 mg/m2 day 1, 5-FU 1,600mg/m2/46 hr)in combination with radiotherapy(total 60 Gy/ 30 Fr).Recurrent lymph nodes shrunk slightly, and there were no new lesions.Subsequently, there was no other adverse event except for Grade 1 chemotherapy-induced neuropathy.No new recurrence was observed 19 months after the first recur- rence.The patient continues to receive weekly nab-PTX(85mg/m2).A regimen that includes L-OHP is useful in the treatment of advanced or recurrent esophageal squamous cell carcinoma.

The effect of the microstructure of beta-tricalcium phosphate on the metabolism of subsequently formed bone tissue.

The response of bone cells to a newly developed porous beta-tricalcium phosphate composed of rod-shaped particles (RSbeta-TCP), beta-TCP composed of conventional non-rod-shaped particles (Cbeta-TCP), and hydroxyapatite (HA) was analyzed using in vivo implantation and in vitro osteoclastogenesis systems. Implantation of the materials into the rabbit femur showed that RSbeta-TCP and Cbeta-TCP were bioresorbable, but HA was not. Up to 12 weeks after the implantation, bioresorption of RSbeta-TCP and Cbeta-TCP accompanied by the formation of new bone occurred satisfactorily. At 24 weeks post-implantation, most of the RSbeta-TCP had been absorbed, and active osteogenesis was preserved in the region. However, in the specimens implanted with Cbeta-TCP, the amount of not only the implanted Cbeta-TCP but also the newly formed bone tissue decreased, and bone marrow dominated the region. The implanted HA was unbioresorbable throughout the experimental period. When osteoclasts were generated on RSbeta-TCP, Cbeta-TCP, or HA disks, apparent resorption lacunae were formed on the RSbeta-TCP and Cbeta-TCP, but not HA disks. Quantitation of the calcium concentration in the culture media showed an earlier and more constant release of calcium from RSbeta-TCP than Cbeta-TCP. These results showed that the microstructure of beta-TCP affects the activity of bone cells and subsequent bone replacement.