1
|
Wu S, Hong G, Xu A, Zeng H, Chen X, Wang Y, Luo Y, Wu P, Liu C, Jiang N, Dang Q, Yang C, Liu B, Shen R, Chen Z, Liao C, Lin Z, Wang J, Lin T. Artificial intelligence-based model for lymph node metastases detection on whole slide images in bladder cancer: a retrospective, multicentre, diagnostic study. Lancet Oncol 2023; 24:360-370. [PMID: 36893772 DOI: 10.1016/s1470-2045(23)00061-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/19/2022] [Revised: 02/01/2023] [Accepted: 02/02/2023] [Indexed: 03/08/2023]
Abstract
BACKGROUND Accurate lymph node staging is important for the diagnosis and treatment of patients with bladder cancer. We aimed to develop a lymph node metastases diagnostic model (LNMDM) on whole slide images and to assess the clinical effect of an artificial intelligence-assisted (AI) workflow. METHODS In this retrospective, multicentre, diagnostic study in China, we included consecutive patients with bladder cancer who had radical cystectomy and pelvic lymph node dissection, and from whom whole slide images of lymph node sections were available, for model development. We excluded patients with non-bladder cancer and concurrent surgery, or low-quality images. Patients from two hospitals (Sun Yat-sen Memorial Hospital of Sun Yat-sen University and Zhujiang Hospital of Southern Medical University, Guangzhou, Guangdong, China) were assigned before a cutoff date to a training set and after the date to internal validation sets for each hospital. Patients from three other hospitals (the Third Affiliated Hospital of Sun Yat-sen University, Nanfang Hospital of Southern Medical University, and the Third Affiliated Hospital of Southern Medical University, Guangzhou, Guangdong, China) were included as external validation sets. A validation subset of challenging cases from the five validation sets was used to compare performance between the LNMDM and pathologists, and two other datasets (breast cancer from the CAMELYON16 dataset and prostate cancer from the Sun Yat-sen Memorial Hospital of Sun Yat-sen University) were collected for a multi-cancer test. The primary endpoint was diagnostic sensitivity in the four prespecified groups (ie, the five validation sets, a single-lymph-node test set, the multi-cancer test set, and the subset for a performance comparison between the LNMDM and pathologists). FINDINGS Between Jan 1, 2013 and Dec 31, 2021, 1012 patients with bladder cancer had radical cystectomy and pelvic lymph node dissection and were included (8177 images and 20 954 lymph nodes). We excluded 14 patients (165 images) with concurrent non-bladder cancer and also excluded 21 low-quality images. We included 998 patients and 7991 images (881 [88%] men; 117 [12%] women; median age 64 years [IQR 56-72]; ethnicity data not available; 268 [27%] with lymph node metastases) to develop the LNMDM. The area under the curve (AUC) for accurate diagnosis of the LNMDM ranged from 0·978 (95% CI 0·960-0·996) to 0·998 (0·996-1·000) in the five validation sets. Performance comparisons between the LNMDM and pathologists showed that the diagnostic sensitivity of the model (0·983 [95% CI 0·941-0·998]) substantially exceeded that of both junior pathologists (0·906 [0·871-0·934]) and senior pathologists (0·947 [0·919-0·968]), and that AI assistance improved sensitivity for both junior (from 0·906 without AI to 0·953 with AI) and senior (from 0·947 to 0·986) pathologists. In the multi-cancer test, the LNMDM maintained an AUC of 0·943 (95% CI 0·918-0·969) in breast cancer images and 0·922 (0·884-0·960) in prostate cancer images. In 13 patients, the LNMDM detected tumour micrometastases that had been missed by pathologists who had previously classified these patients' results as negative. Receiver operating characteristic curves showed that the LNMDM would enable pathologists to exclude 80-92% of negative slides while maintaining 100% sensitivity in clinical application. INTERPRETATION We developed an AI-based diagnostic model that did well in detecting lymph node metastases, particularly micrometastases. The LNMDM showed substantial potential for clinical applications in improving the accuracy and efficiency of pathologists' work. FUNDING National Natural Science Foundation of China, the Science and Technology Planning Project of Guangdong Province, the National Key Research and Development Programme of China, and the Guangdong Provincial Clinical Research Centre for Urological Diseases.
Collapse
Affiliation(s)
- Shaoxu Wu
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Malignant Tumour Epigenetics and Gene Regulation, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Clinical Research Centre for Urological Diseases, Guangzhou, Guangdong, China
| | - Guibin Hong
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Abai Xu
- Department of Urology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China
| | - Hong Zeng
- Department of Pathology, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Xulin Chen
- Cells Vision Medical Technology, Guangzhou, Guangdong, China
| | - Yun Wang
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Yun Luo
- Sun Yat-sen Memorial Hospital and Department of Urology, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Peng Wu
- Department of Urology, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, China
| | - Cundong Liu
- Department of Urology, The Third Affiliated Hospital, Southern Medical University, Guangzhou, Guangdong, China
| | - Ning Jiang
- Department of Urology, Zhujiang Hospital, Southern Medical University, Guangzhou, Guangdong, China
| | - Qiang Dang
- Department of Urology, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, China
| | - Cheng Yang
- Department of Urology, The Third Affiliated Hospital, Southern Medical University, Guangzhou, Guangdong, China
| | - Bohao Liu
- Sun Yat-sen Memorial Hospital and Department of Urology, The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Runnan Shen
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Zeshi Chen
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Chengxiao Liao
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China
| | - Zhen Lin
- Cells Vision Medical Technology, Guangzhou, Guangdong, China
| | - Jin Wang
- Cells Vision Medical Technology, Guangzhou, Guangdong, China
| | - Tianxin Lin
- Department of Urology, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Malignant Tumour Epigenetics and Gene Regulation, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Clinical Research Centre for Urological Diseases, Guangzhou, Guangdong, China.
| |
Collapse
|
2
|
Increased NUSAP1 expression is associated with lymph node metastasis and survival prognosis in bladder urothelial carcinoma. Sci Rep 2022; 12:7003. [PMID: 35487972 PMCID: PMC9054847 DOI: 10.1038/s41598-022-11137-4] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/20/2022] [Accepted: 04/19/2022] [Indexed: 12/05/2022] Open
Abstract
The main route of metastasis of bladder urothelial carcinoma is through lymph nodes; however, its exact mechanism remains unclear. In this study, we found an association of nucleolar and spindle associated protein 1 (NUSAP1) expression with BUC tissues along with lymph node metastasis and the survival prognosis. A total of 178 pathological specimens following radical bladder cancer resection were obtained. NUSAP1 expression was analyzed by immunohistochemistry. We evaluated the correlation between clinicopathological characteristics and NUSAP1 expression. Logistic regression was used to determine the independent variables that influenced lymph node metastasis. Uni- and multi-factorial Cox regression methods were used to determine the prognostic value of NUSAP1 expression in urothelial carcinoma of the bladder. High expression of NUSAP1 in BUC was not significantly related to the patient’s gender, age, or tumor number (p > 0.05), however was significantly associated with pathological grade, tumor diameter, pathological stage, and lymph node metastasis (p < 0.05). Lymph node metastasis was significantly correlated with pathological stage, pathological grade, tumor number, tumor diameter, and NUSAP1 expression (p < 0.05); only NUSAP1 expression was an independent predictor of lymph node metastasis in BUC (OR:1.786, 95% CI 1.229–2.596, p = 0.002). In addition, high NUSAP1 expression was an independent prognostic predictor for BUC. In BUC, NUSAP1 showed high expression and was significantly associated with lymph node metastasis, pathological stage, pathological grade, and tumor diameter. NUSAP1 was an independent predictor of lymph node metastasis and prognosis in BUC; higher expression indicated poorer prognosis of BUC patients.
Collapse
|
3
|
Jena R, Shrivastava N, Sharma AP, Choudhary GR, Srivastava A. The Adequacy of Pelvic Lymphadenectomy During Radical Cystectomy for Carcinoma Urinary Bladder: A Narrative Review of Literature. Front Surg 2021; 8:687636. [PMID: 34222323 PMCID: PMC8247657 DOI: 10.3389/fsurg.2021.687636] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/29/2021] [Accepted: 05/19/2021] [Indexed: 11/24/2022] Open
Abstract
An adequate pelvic lymph node dissection (PLND) is an essential part of radical cystectomy for muscle invasive bladder cancer. However, the definition of what constitutes an adequate PLND is often shrouded in controversy. Various authors have defined different anatomic templates of PLND based on levels of pelvic lymph nodes. Some have suggested other surrogate markers of the adequacy of PLND, namely lymph node count and lymph node density. While individual studies have shown the efficacy and reliability of some of the above markers, none of them have been recommended forthright due to the absence of robust prospective data. The use of non-standardized nomenclature while referring to the above variables has made this matter more complex. Most of older data seems to favor use of extended template of PLND over the standard template. On the other hand, one recent randomized controlled trial (RCT) did not show any benefit of one template over the other in terms of survival benefit, but the study design allowed for a large margin of bias. Therefore, we conducted a systematic search of literature using EMBASE, Medline, and PubMed using PRISMA-P checklist for articles in English Language published over last 20 years. Out of 132 relevant articles, 47 articles were included in the final review. We have reviewed existing literature and guidelines and have attempted to provide a few suggestions toward a uniform nomenclature for the various anatomical descriptions and the extent of PLND done while doing a radical cystectomy. The results of another large RCT (SWOG S1011) are awaited and until we have a definitive evidence, we should adhere to these suggestions as much as possible and deal with each patient on a case to case basis.
Collapse
Affiliation(s)
- Rahul Jena
- Department of Urology, All India Institute of Medical Sciences, Jodhpur, India
| | - Nikita Shrivastava
- Department of Urology, All India Institute of Medical Sciences, Jodhpur, India
| | - Aditya Prakash Sharma
- Department of Urology, Postgraduate Institute of Medical Education and Research, Chandigarh, India
| | | | - Aneesh Srivastava
- Department of Urology and Renal Transplant, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India
| |
Collapse
|
4
|
Matsumoto R, Abe T, Takada N, Minami K, Harabayashi T, Nagamori S, Hatanaka KC, Yamashiro K, Kikuchi H, Osawa T, Maruyama S, Shinohara N. Oncologic outcomes of laparoscopic radical nephroureterectomy in conjunction with template-based lymph node dissection: An extended follow-up study. Urol Oncol 2020; 38:933.e13-933.e18. [PMID: 32534960 DOI: 10.1016/j.urolonc.2020.05.013] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/02/2020] [Revised: 04/14/2020] [Accepted: 05/11/2020] [Indexed: 11/29/2022]
Abstract
OBJECTIVES This study investigated the relapse pattern and oncologic outcomes after laparoscopic nephroureterectomy with template-based lymph node dissection (LND) in patients with clinically node-negative (cN0) upper urinary tract urothelial carcinoma. The frequency of lymph node metastasis, including micrometastases, was also evaluated. METHODS AND MATERIALS A total of 105 patients with cTa-3N0M0 upper urinary tract urothelial carcinoma were analyzed, all of whom underwent regional LND during laparoscopic nephroureterectomy. Of those patients, 96 (91%) underwent complete LND in accordance with an anatomical template-based rule. We collected patient characteristics, pathological data, and follow-up data from medical charts. Micrometastases were assessed by pan-cytokeratin immunohistochemistry. Nonurothelial recurrence-free survival and cancer-specific survival were estimated using the Kaplan-Meier method. RESULTS The median number of lymph nodes removed was 12 (range, 1-59). Lymph node metastasis was identified by routine pathological examination in 7 (7/105, 6.7%) patients. Pan-cytokeratin immunohistochemistry revealed micrometastases in 5 additional patients (pNmicro +: 5/105, 4.8%). Nonurothelial disease recurrence was observed in 21 (20%) patients at a median of 10 months (range: 1-33) after surgery. Distant metastasis was dominant (15/105, 14.3%), followed by locoregional recurrence (5/105, 4.8%) and both (1/105, 0.95%). The 5-year nonurothelial recurrence-free survival rates were 84.8% for pN0, 53.3% for pNmicro+, and 19.1% for pN+ (3-sample log-rank test, P < 0.0001). The 5-year cancer-specific survival rates were 95.0% for pN0, 53.3% for pNmicro+, and 23.8% for pN+ (P < 0.0001). CONCLUSIONS Our observation showed that template-based LND could contribute to precise disease staging and better local disease control probably by eliminating nodal disease, compared with previous studies. The survival impact and ideal management of pNmicro+ disease should be evaluated in a larger cohort.
Collapse
Affiliation(s)
- Ryuji Matsumoto
- Department of Urology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
| | - Takashige Abe
- Department of Urology, Hokkaido University Graduate School of Medicine, Sapporo, Japan.
| | - Norikata Takada
- Department of Urology, Hokkaido Cancer Center, Sapporo, Japan
| | - Keita Minami
- Department of Urology, Sapporo City General Hospital, Sapporo, Japan
| | | | | | - Kanako C Hatanaka
- Clinical Research and Medical Innovation Center, Hokkaido University Hospital, Sapporo, Japan
| | | | - Hiroshi Kikuchi
- Department of Urology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
| | - Takahiro Osawa
- Department of Urology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
| | - Satoru Maruyama
- Department of Urology, Hokkaido Cancer Center, Sapporo, Japan
| | - Nobuo Shinohara
- Department of Urology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
| |
Collapse
|
5
|
Crozier J, Papa N, Perera M, Stewart M, Goad J, Sengupta S, Bolton D, Lawrentschuk N. Lymph node yield in node-negative patients predicts cancer specific survival following radical cystectomy for transitional cell carcinoma. Investig Clin Urol 2017; 58:416-422. [PMID: 29124240 PMCID: PMC5671960 DOI: 10.4111/icu.2017.58.6.416] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/20/2017] [Accepted: 05/23/2017] [Indexed: 01/29/2023] Open
Abstract
Purpose To determine the oncological implications of increased nodal dissection in node-negative bladder cancer during radical cystectomy in a contemporary Australian series. Materials and Methods We performed a multicenter retrospective study, including more than 40 surgeons across 5 sites over a 10-year period. We identified 353 patients with primary bladder cancer undergoing radical cystectomy. Extent of lymphadenectomy was defined as follows; limited pelvic lymph node dissection (PLND) (perivesical, pelvic, and obturator), standard PLND (internal and external iliac) and extended PLND (common iliac). Multivariable cox proportional hazards and logistic regression models were used to determine LNY effect on cancer-specific survival. Results Over the study period, the extent of dissection and lymph node yield increased considerably. In node-negative patients, lymph node yield (LNY) conferred a significantly improved cancer-specific survival. Compared to cases where LNY of 1 to 5 nodes were taken, the hazard ratio (HR) for 6 to 15 nodes harvested was 0.78 (95% confidence interval [CI], 0.43-1.39) and for greater than 15 nodes the HR was 0.31 (95% CI, 0.17-0.57), adjusted for age, sex, T stage, margin status, and year of surgery. The predicted probability of cancer-specific death within 2 years of cystectomy was 16% (95% CI, 13%-19%) with 10 nodes harvested, falling to 5.5% (95% CI, 0%-12%) with 30 nodes taken. Increasing harvest in all PLND templates conferred a survival benefit. Conclusions The findings of the current study highlight the improved oncological outcomes with increased LNY, irrespective of the dissection template. Further prospective research is needed to aid LND data interpretation.
Collapse
Affiliation(s)
- Jack Crozier
- Department of Surgery, Austin Health, University of Melbourne, Melbourne, Australia
| | - Nathan Papa
- Department of Surgery, Austin Health, University of Melbourne, Melbourne, Australia
| | - Marlon Perera
- Department of Surgery, Austin Health, University of Melbourne, Melbourne, Australia
| | | | | | - Shomik Sengupta
- Department of Surgery, Austin Health, University of Melbourne, Melbourne, Australia.,Olivia Newton-John Cancer Research Institute, Melbourne, Australia
| | - Damien Bolton
- Department of Surgery, Austin Health, University of Melbourne, Melbourne, Australia.,Olivia Newton-John Cancer Research Institute, Melbourne, Australia
| | - Nathan Lawrentschuk
- Department of Surgery, Austin Health, University of Melbourne, Melbourne, Australia.,Olivia Newton-John Cancer Research Institute, Melbourne, Australia.,Department of Surgical Oncology, Peter MacCallum Cancer Centre, Melbourne, Australia
| |
Collapse
|