Implementation of Surgical Safety and Intraoperative Metastasis Identification Through Deep Learning: Multicentric Video Collection for Minimally Invasive Sentinel Lymph Node Dissection in Uterine Malignancies (LYSE)

The loco-regional metastatic or non-metastatic status of lymph nodes (LN) is critical for the therapeutic management of most cancer patients. Indeed, the presence or absence of lymphatic metastasis is essential for the accurate staging of the disease and strongly influence the prognosis and adjuvant treatment regimens. An important revolution in oncological surgery has been the introduction of the concept of sentinel lymph node (SLN) biopsy to reduce the complications of extensive loco-regional lymphadenectomies. SLN identification through ICG- based near-infrared fluorescence (NIR) cervical injection and its dissection is now recommended by European guidelines to stage uterine malignancies (endometrial and cervical cancers). However, SLN procedures have several limitations. In 11.2% of cases intra- or postoperative complications are reported due to anatomical structures injuries (vessels, nerves and lymphatic channels disruptions). Common mistakes, especially when the learning curve is not completed (at least 40 procedures), include mapping failure (25%) and removal of second/third-level nodes and/or empty nodes packets (8-14%). Additionally the intraoperative accuracy of frozen section is still far to be adequate with only the 65% of SLN metastasis detection.

These limitations are a result of the lack of precision of current SLN localization and analysis as well as of the overall difficulty of visualizing lymph nodes and other critical structures in the retroperitoneum.

Currently, studies on the safety of surgical procedures are based on perioperative clinical information and postoperative reports written by the surgeons themselves. Today, videos guiding minimally invasive surgical interventions allow for objective documentation of the procedure and provide opportunities to explore solutions for enhancing safety in the operating room. With an increasing use of endoscopic systems across different specialties, there is a need for standardization of training, assessment, testing and sign-off as a competent surgeon in order to improve patient safety. In laparoscopic lymph node dissection in endometrial and cervical cancer, a standardize stepwise approach to the procedure is highly recommended, by identifying key anatomic landmarks and structures, in various scenarios, that could prevent vascular, nervous and ureters injuries and enhance the mapping rate. Therefore, quantifying and studying intraoperative events such as the rate of achieving the right space dissections and anatomic structures visualization as a recommended step for safety and proficiency, would enable the examination of how best to implement guideline recommendations and seek new solutions to reduce operative risks. These videos could be utilized to train and validate artificial intelligence (AI) algorithms, with the potential to assist surgeons in the operating room and make the procedures safer. Additionally, the visual information (ICG intensity) could hide data that the AI can analyze and correlate with anatomopathological reports. By the integration of AI tool with laparoscopic/robotic platform it is possible to enhance MIS video streams in real time with surgical phases detection, events recognition, ICG signal intensity, anatomical structure identification and auto-targeting

Study Overview

• Status: Recruiting
• Conditions: Cervical Cancer; Artificial Intelligence; Endometrial Cancer; Deep Learning
• Intervention / Treatment: Procedure: Sentinel lymph node sampling

Detailed Description

Surgery accounts for approximately two-thirds of all hospital complications. About 75% of these complications occur during surgical interventions, and half of them are considered preventable. Traditionally, studies on surgical safety have relied mainly on perioperative information and written reports from the surgical operators themselves. Nowadays, videos that guide the surgeon during minimally invasive procedures can be easily recorded, providing objective documentation of surgical interventions. Studying surgical videos and quantifying intraoperative events offer a promising new perspective on the causes of adverse events, with the potential to develop solutions for reducing surgical risks and facilitating the implementation of international safety recommendations in surgery .

Furthermore, recent advancements in Artificial Intelligence (AI) have shown great success in image understanding, raising the possibility of developing algorithms for automatic and large-scale analysis of surgical videos. These AI algorithms could also analyze surgical videos in real-time and provide decision support in the operating room.

The assessment of lymph node status is of crucial importance in uterine malignancies (endometrial and cervical cancer). The rate of positive lymph nodes in apparently early stage cancers is far to be low (10% endometrial and cervical 15% ). Indeed, the prognosis and adjuvant treatment regimens are strongly influenced by the presence of nodal involvement. Systematic extensive lymphadenectomies are often performed for staging, diagnosis of skip metastases and to define the radiation field when adjuvant radiotherapy treatments are required. Nevertheless, pelvic and/or para- aortic lymphadenectomy can lead to significant short-term and long-term complications.

What makes the burden of surgical morbidity even more difficult to bear is that in most cases, regardless of the type of pelvic malignancy, the lymph nodes are free from metastasis. This means that the majority of patients undergo an unnecessary, risky, and burdensome procedure that has no proven impact on their survival. To solve this issue in early stage endometrial and cervical cancers, evaluation of the sentinel lymph node (SLN) has acquired a valuable role..The evaluation of the first lymph node or group of lymph nodes draining the initial lesion and therefore the first station to collect neoplastic cells in case of tumor nodal spread allows node-negative patients to be spared from the surgical comorbidities associated with systematic lymphadenectomy. However even if only SLN is removed, the rate of surgical complications is still high. Capozzi et al reported a 11.2% complications rate related to intra- operative injuries of vessels, nerves, and ureters and to post-operative evidence of lymphoceles and lymphedemas due to wrong anatomy dissection and lymphatic channels disruption. It is reported a 25% of SLN mapping failure according to patients and cancer characteristics and variability in the surgical steps. In 2021 Moloney et al and in 2023 Bizzarri et al published a Delphi consensus for SLN surgical steps in endometrial and cervical cancer to improve the clinical outcomes, decrease surgical complications and bias in multicentric prospective trials. To date a learning curve of 40 procedures is demonstrated to be needed for successful bilateral mapping. Furthermore, in some case it is reported the inadvertent failure to harvest nodal tissue in 14% of cases (higher rate in obese patients with endometrial cancer) with empty packets that may lead to overtreatment for patient safety. Sentinel node frozen section analysis tough then is not routinely performed due to its disadvantage in terms of time consumption and not accurate results for micro-metastasis detection (65% of detection accuracy) thereby losing part of the techniques potential.

In laparoscopic cholecystectomy, the critical view of safety (CVS) has been described to reduce the risk of bile duct injury. We aim to adapt concepts of "culture of safety" to laparoscopic SLN node dissection by identifying key anatomic landmarks and structures, in various scenarios, that could prevent intra- operative injuries and enhance the mapping rate by following the standardized surgical steps.

Therefore, quantifying and studying intraoperative events such as the rate of achieving the recommended surgical steps, would enable the examination of how best to implement guideline recommendations and seek new solutions to reduce operative risks. Furthermore, these videos could be utilized to train and validate artificial intelligence (AI) algorithms, with the potential to assist surgeons in the operating room helping in the training to reach faster the learning curve and improve the procedural safety. Infact, recent AI algorithms capable of automatically assessing and documenting the critical view of safety (CVS) with short videos could be utilized to support the implementation of guidelines and make laparoscopic SLN dissection safer.

Analyzing videos of laparoscopic SLN dissection performed at referral centers could help assess the implementation of a stepwise approach to the procedure and the development of a critical view of safety (CVS) along with other guideline recommendations, as well as develop new solutions for safe laparoscopic SLN dissection. The steps to identification and the "Critical view of safety" are fully described in the study. Our project emphasizes the importance of achieving these views of safety before proceeding to the next step of the procedure. The next step involves analysing the intensity of the fluorescent signal and correlating it, as no-one has ever done, with the histological diagnosis of the sentinel lymph node to assess the possibility of correlation between intensity and metastatic nature.

Additionally, AI algorithms for safe laparoscopic SLN dissection could be tested on the collected videos from multiple hospital centers to adequately study the generalizability of these algorithms before clinical translation.

Purposes and objective of the clinical study:

The objective of this study is to establish the rate of critical view of safety (CVS) for safe laparoscopic SLN dissection in a multicentre cohort of surgical videos to reduce complications, the surgical learning curve and correlate ICG intensity with the lymph node metastatic nature

• Study Type: Observational
• Enrollment (Estimated): 100

Contacts and Locations

• Study Contact: Name: Matteo PAVONE, MD; Phone Number: 00390630151; Email: matteo.pavone@guest.policlinicogemelli.it

Study Locations

• Italy
  • Roma, Italy
    • Recruiting
    • Fondazione Policlinico Universitario A. Gemelli IRCCS
    • Contact: Matteo Pavone, MD; Phone Number: 00390630151
  • Rome, Italy
    • Recruiting
    • Fondazione Policlinico Universitario A. Gemelli IRCCS

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Women undergoing laparoscopic or robotic sentinel lymph node dissection for endometrial or cervical cancer

Description

Inclusion Criteria:

• Women undergoing MIS sentinel lymph node dissection for endometrial or cervical cancers
• Availability of video
• Age >18 years
• Willingness to participate in the study and to provide informed consent

Exclusion Criteria:

• Previous pelvic radiotherapy treatments
• Severe endometriosis or other conditions able to alter the pelvic anatomy

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Evaluate the rate of achieving critical view of safety (CVS) according to standard agreement recommendations in Laparoscopic SLN dissection procedures performed at centers involved in the study.	The rate of Critical views of safety achievement will be evaluated in a video-based assessment by experts gyn surgeons and labeled according to the annotation protocol. The inter-rather agreement will be therefore also evaluated	24 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Develop an Artificial intelligence tool	able to reduce perioperative complications ( graded according to Clavien-Dindo scale)	24 months
Develop an Artificial intelligence tool	able to reduce the rate of empty packets (expressed in percentage and evaluated with pathology as gold standard)	24 months
Develop an Artificial intelligence tool	able to implement the surgical learning curve measured as the rate of mapping failure and of empty packets	24 months
Develop an Artificial intelligence tool	able to define the presence of micro and macro-metastasis correlating with ICG fluorescence (evaluated with pathology as a gold standard and measured in percentages)	24 months

Collaborators and Investigators

• Sponsor: Fondazione Policlinico Universitario Agostino Gemelli IRCCS
• Collaborators: IHU Strasbourg; University Hospital, Strasbourg
• Investigators: Principal Investigator: Matteo PAVONE, MD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; IHU Strasbourg; IRCAD Strasbourg; Icube Strasbourg;; Principal Investigator: Nicolò BIZZARRI, MD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; Principal Investigator: Lise LECOINTRE, MD, PhD, University Hospitals of Starsbourg; Icube Strasbourg; IHU Strasbourg; Study Chair: Denis QUERLEU, MD, PhD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; Study Chair: Nicolas PADOY, PhD, IHU Strasbourg; Study Director: Giovanni SCAMBIA, MD, PhD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; Study Chair: Francesco FANFANI, MD, PhD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; Study Chair: Cherif AKLADIOS, MD, PhD, University Hospitals of Strasbourg; Study Chair: Pietro MASCAGNI, MD, PhD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; Study Chair: Valentina IACOBELLI, MD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy; Study Chair: Andrea ROSATI, MD, Fondazione Policlinico Universitario A. Gemelli, IRCCS, Rome Italy

Study record dates

Study Major Dates

• Study Start (Actual): September 17, 2024
• Primary Completion (Estimated): September 30, 2026
• Study Completion (Estimated): September 30, 2027

Study Registration Dates

• First Submitted: September 24, 2024
• First Submitted That Met QC Criteria: September 27, 2024
• First Posted (Actual): October 1, 2024

Study Record Updates

• Last Update Posted (Actual): October 2, 2024
• Last Update Submitted That Met QC Criteria: October 1, 2024
• Last Verified: October 1, 2024

More Information

Terms related to this study

• Keywords: endometrial cancer; sentinel lymph node; artificial intelligence; cervical cancer; deep learning; computer vision; indocyanin green
• Additional Relevant MeSH Terms: Neoplasms; Urogenital Neoplasms; Neoplasms by Site; Uterine Neoplasms; Genital Neoplasms, Female; Uterine Cervical Diseases; Uterine Diseases; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Urogenital Diseases; Genital Diseases; Genital Diseases, Female; Uterine Cervical Neoplasms; Endometrial Neoplasms
• Other Study ID Numbers: ID 6672

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No