Category: 3. Business

Introduction

Scientific advances have significantly influenced the evolution of education and training in recent decades. Emerging technologies such as technology-enhanced learning and simulation-based training have played a crucial role in improving the learning experience of practitioners and have become essential in modern education systems [].

Traditionally, surgical training has mainly focused on gaining experience through a significant number of surgeries and direct involvement, in which trainees receive less supervision from experienced surgeons as they gain competence and eventually become capable of doing surgeries independently []. This model embodies the “see one, do one, teach one” approach []. An experienced surgeon first executes a procedure, which the trainee observes. Then, under supervision, the trainee replicates the process. Finally, upon achieving competence, the trainee is expected to instruct others on how to perform it. This approach underscores the importance of direct observation, practical experience, and the ability to transmit information and expertise to future generations of medical practitioners. However, it also raises inquiries regarding the diversity of learning experiences, the consistency of the skills acquired, and the stress that it places on seasoned surgeons and trainees to quickly comprehend and transmit complex procedures involving inherent risks [,]. Acquiring and improving skills in the field of medicine are complex processes that last throughout a physician’s career. Since the 1990s, ongoing discussions have focused on enhancing teaching practices [].

Researchers have developed various simulators and training platforms to address these challenges and the demands of an expanding spectrum of surgical operations []. These tools enable trainees to develop expertise in different surgical procedures and provide the benefit of unlimited practice opportunities, customizable difficulty levels, and cost-effective solutions that emulate the difficulties of actual surgery procedures [,]. Furthermore, these platforms offer a secure and interactive setting that promotes learning through experimentation, enabling risk-free practice. Nevertheless, there remains considerable potential to improve the effectiveness of these training setups [,].

As technological advancements continue, interest in incorporating artificial intelligence (AI) into medical training has also increased []. AI, with its capacity to emulate certain aspects of human cognition, has the potential to enhance educational outcomes and transform traditional methods of training and teaching []. It enables the creation of the next generation of autonomous systems to execute tasks usually performed by individuals, representing a substantial advancement in computer science. Furthermore, AI algorithms could assist in enhancing conceptual understanding, facilitating virtual practice, and offering analytical feedback on performance. Through the use of data-driven insights and predictive analytics, AI has the potential to revolutionize surgical training, offering customized and efficient learning pathways.

This scoping review aims to map and analyze current applications of AI in surgical training, assessment, and evaluation, identifying the most common surgical procedures, AI techniques, and training setups while highlighting gaps and opportunities for future research. The following research questions guided this study:

1. What are the specific surgical procedures where AI algorithms are most frequently applied in surgical training?
2. Which AI techniques have been used in surgical training and evaluation?
3. How are AI techniques being used to assess and improve surgical training?
4. How do AI applications in surgical training affect the learning curve of surgical residents and fellows?

The paper is organized as follows: the “Methods” section outlines the methodology used to carry out this scoping review. The “Results” section provides a comprehensive overview of the findings, shows additional findings, and identifies potential areas for opportunity. The “Discussion” section presents an outline of the research questions, shows additional findings, identifies potential areas for opportunity, acknowledges the limits of the current review, and concludes with final thoughts and directions for future research in the realm of AI in surgical education.

Although there are different definitions and approaches to what AI is, this study is particularly interested in Russell and Norvig’s [] approach to systems that act rationally, that is, systems that act intelligently and rationally, ideally in the best possible way given the available information. AI is a disruptive technology that is reshaping education, facilitating a shift toward more efficient teaching protocols []. It enables machines to imitate various complex human skills, and AI-based techniques are typically employed in the following areas:

• Expert systems “emulate the behavior of a human expert within a well-defined, narrow domain of knowledge” [].
• Intelligent tutoring systems emulate “model learners’ psychological states to provide individualized instruction. They… help learners acquire domain-specific, cognitive, and metacognitive knowledge” [].

AI can be subdivided into machine learning (ML), which further includes deep learning (DL). ML aims to “perform intelligent predictions based on a data set” []. It uses statistical, data mining, and optimization methods to design models that can identify patterns and make predictions with higher precision than human experts. In this field, there are 3 fundamental ML paradigms:

• Supervised learning uses input data and their matching labeled output to train models []. A labeled output is data that has been assigned labels to add context; consequently, the objective of supervised learning is to learn and predict outputs for unseen data based on the initial input-output pairs.
• Unsupervised learning involves working with unlabeled data []. The algorithms autonomously attempt to discern patterns and relationships within the data.
• Reinforcement learning uses an autonomous entity known as an agent, which learns to make decisions by performing activities inside an environment to reach a specific objective []. The feedback the agent receives in the form of rewards or penalties serves as a guide as it iteratively refines its strategy to achieve optimal performance.

Finally, DL is a branch of machine learning that uses artificial neural networks to replicate the sophisticated processes of the human brain []. Algorithms in this category learn to identify patterns and comprehend large datasets. DL is highly efficient because it can automatically extract and learn high-level characteristics from data, reducing the need for manual feature selection. It excels at handling complex tasks such as image and audio recognition, natural language processing, image generation, and data-driven prediction.

Numerous models have been developed within AI to address challenging problems and tasks across different sectors and research fields. Each approach provides certain advantages specific to the type of data to be processed and the analytical needs (see ).

These AI models highlight the potential of this technology in educational contexts. The United Nations Educational, Scientific and Cultural Organization indicates that digital technologies have the potential to complement, enrich, and transform education, aligning with the United Nations’ Sustainable Development Goal 4 (SDG 4) for education and providing universal access to learning []. Consequently, the integration of AI in surgical training could boost independence, self-study, engagement, and motivation.

Methods

Overview

This review adheres to the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews; see ) statement, designed for publications in the health and medical sciences []. The review process was organized following a structured protocol consisting of four stages: (1) planning, which involved establishing the criteria for the search and databases to be used; (2) conducting, which entailed performing the search and applying filters for the scoping review; (3) reporting, which included compiling the studies that met the criteria and were included in the review. During stages 1 and 2, the research papers were compiled, and the initial screening process was conducted, focusing solely on papers that fall within the scope of the review and were published in peer-reviewed scientific journals. Stage 3 consists of identifying the main characteristics that distinguish the contributions and unique features of each article that has passed the initial screening process. Subsequently, the necessary analysis was performed to present the summary of the research and compile tables and figures. The starting date for stages 1 and 2 of the scoping review was February 27, 2024, and it concluded on March 18, 2024.

Information Sources

A total of 3 databases were selected to search for relevant studies: PubMed, Scopus, and Web of Science. The inclusion of Web of Science and Scopus databases consolidates information from other sources, such as IEEE Xplore, ScienceDirect, and SpringerLink. Therefore, they expand the scope of accessible academic literature. These platforms also provide search and analytical tools, making it easier to find pertinent studies and analyze trends. By using the 3 databases, the review considered articles with different AI models beyond the limitation of just focusing on clinical trials. By implementing this procedure, the scope of the review is expanded, enabling the identification of significant manuscripts to identify areas of opportunity in the field.

Search Strategy

A total of 4 keywords related to AI concepts and 4 keywords related to surgical training were selected based on the research questions. The selected keywords were converted into search strings and processed to be compatible with the advanced search tool of each database. shows the search strings used in this scoping review.

Eligibility Criteria

Records retrieved from the initial search were examined to verify their compliance with the eligibility criteria and their alignment with the research questions ().

Data Charting and Synthesis

After the inclusion and exclusion criteria had been applied during screening, data were charted for each included study covering three dimensions: (1) the surgical procedure (eg, laparoscopy, minimally invasive surgery, neurosurgery, and arthroscopy), (2) the AI model (eg, support vector machine [SVM], convolutional neural network [CNN], deep neural network [DNN], long short-term memory [LSTM], and transformers), and (3) the training setup (eg, simulation platforms, box trainers, surgical video analysis, in-vivo settings, virtual reality, and da Vinci system). These variables structured the subsequent evidence synthesis and guided the organization of results by procedure, technique, and setup. In addition, bibliographic fields, including year of publication and type of publication, were also charted to support descriptive reporting in the Results section. This structured approach enabled a descriptive and narrative synthesis aimed at elucidating how AI contributes to educational outcomes and skill acquisition in surgical training.

Results

Search Results and Study Selection

presents the PRISMA-ScR flow diagram illustrating the complete selection process. The initial search identified 1400 records: 545 from PubMed, 288 from Web of Science, and 567 from Scopus, obtained using the search strings described in . After applying the publication date range from January 2020 to March 2024, a total of 461 records were excluded, leaving 939 for further screening. Duplicate removal eliminated 363 records, yielding 576 unique studies.

‎

Subsequent filtering was conducted in stages to ensure methodological rigor and relevance. Database parameters were adjusted to retain only peer-reviewed journal articles and conference proceedings, excluding 260 reviews and 36 editorials that did not meet the inclusion criteria. A total of 280 records proceeded to qualitative screening. During this stage, the relevance of each article to the review objectives was reassessed. This process excluded 76 studies that, despite meeting database filters, were secondary reviews, surveys, or editorials; 9 non-English papers, 7 papers focused on nonsurgical training, and 18 papers described simulator development or validation without AI integration. Additional exclusions comprised 1 duplicate, 3 studies addressing “Data Collection Systems,” 11 centered on “LLMs in Non-Surgical Education,” and 99 that did not provide sufficient information about AI-enhanced surgical training. This filtering process excluded 224 additional studies, leaving 56 studies for the final synthesis and analysis.

The characteristics of the 56 included studies are summarized in , organized across five domains: (1) surgical procedure (eg, laparoscopy, minimally invasive surgery [MIS], neurosurgery, and arthroscopy), (2) year of publication, (3) type of publication, (4) AI technique or model used (eg, SVM, CNN, DNN, LSTM, and transformers), and (5) training setup (eg, simulation platforms, box trainers, da Vinci system, surgical video analysis, and in vivo or virtual-reality environments). This structure enables direct comparison across specialties and methodological approaches, while supporting a descriptive and narrative synthesis of cross-cutting trends.

Across the included studies, MIS, neurosurgery, and laparoscopy represented the majority of AI applications. ML and DL techniques were the most frequently used computational approaches, while simulation environments and box trainers constituted the primary training configurations. Collectively, these trends indicate a primary emphasis on risk-managed training environments that leverage accessible kinematic and video data. However, heterogeneity in studies and limited standardization of outcome measures remain persistent challenges, underscoring the need for unified evaluation frameworks in the future.

Findings and Interpretation

Specific Surgical Procedures

The scoping review reveals the range of surgical procedures where AI algorithms are being used (see ). The analysis emphasizes the integration of AI in MIS skills (27%, 15/56) [-], neurosurgery (20%, 11/56) [-], and laparoscopy (16%, 9/56) [-] (see ). Moderate representation was observed in arthroscopy (5%, 3/56) [-], ophthalmology (5%, 3/56) [-], and robot-assisted surgery (5%,3/56) [-]. Several other domains appeared less frequently, including open surgery (4%, 2/56) [,], general surgery (4%, 2/56) [,], and surgery skills (4%, 2/56) [,]. Finally, isolated studies were identified in otolaryngology (2%, 1/56) [], orthopedics (2%, 1/56) [], plastic surgery (2%, 1/56) [], radiology (2%, 1/56) [], urology (2%, 1/56) [], and vascular surgery (2%, 1/56) [].

Functionally, most studies focused on automated skill assessment and learning-curve analysis, while comparatively few examined procedure guidance, workflow recognition, or decision support. This trend was especially evident in MIS and laparoscopy, which relied heavily on video-centric datasets and computer-vision models [-,-], and in neurosurgery, where virtual reality simulators provided standardized training environments and feedback mechanisms [-]. The specialty distribution appears to be driven by the availability of high-quality labeled data. Overall, the distribution of specialties indicates that AI integration aligns strongly with domains that generate structured, labeled, and reproducible data, such as endoscopic or robotic procedures. By contrast, open and specialty surgeries remain underrepresented, constrained by the limited standardization of datasets and variability in operative workflows. Future progress will depend on developing shared, procedure-specific repositories, cross-institutional benchmarks, and multimodal data capture beyond video and kinematic streams to enhance generalizability and educational impact [-].

AI Techniques Used

The scoping review identified a diverse set of AI techniques in surgical training (see ). The most frequent were ML (unspecified; 21%, 12/56) [,,,,,,,,,-], clustering (13%, 7/56) [,,,,,,], and CNNs (11%, 6/56) [,,,,,]. We also observed DL (unspecified; 11%, 6/56) [,,,,,] and SVMs (9%, 5/56) [,,,,], followed by neural networks (NNs; 7%, 4/56) [,,,] and AI (unspecified; 7%; 4/56) [,,,]. Additional categories included CNN+LSTM (4%, 2/56) [,], DNNs (4%, 2/56) [,], and fuzzy systems (4%, 2/56) [,]. Single-study categories (2%, 1/56) included regression analysis [], Markov chains [], tutoring system (unspecified) [], Bayesian network [], transformer [], and SVM+RF [].

From 2020 to 2024 (see ), ML (unspecified) appears every year, CNNs strengthen in 2021 and 2023, and DL (unspecified) is present in 2020 and 2022-2024. Sequential and hybrid models (CNN+LSTM and DNNs) clusters in 2022-2023. AI (unspecified) emerges from 2022 onward. Probabilistic and rule-based approaches (Bayesian networks, fuzzy systems, and Markov chains) and transformer/SVM+RF appear as single-study categories. Overall, the technique mix tracks data modality and availability (video and kinematics), reinforcing the need for shared multimodal repositories and standardized evaluation metrics to compare methods fairly and improve external validity.

In the analyzed studies, the number of publications increased from 12 in 2020 to 17 in 2023, with 11 in both 2021 and 2022, and 5 in 2024. The literature search concluded on March 18, 2024, which likely accounts for the lower count in 2024. These totals are summarized in the “Total per year” row of .

Application of AI Techniques

AI techniques have been applied across diverse training setups, enhancing both learning experiences and performance assessment in surgical procedures (see ). The most frequent environments were simulation training (36%, 20/56) [-,-,,,,,] and box trainers (23%, 13/56) [40–43,46-50,52,66,70-71], followed by surgical video analysis (16%, 9/56) [,,-,,,,] and robotic systems using the da Vinci platform (11%, 6/56) [,,,,,]. Less frequent configurations included training stations (4%, 2/56) [,] and in-vivo settings (4%, 2/56) [,], with single-study setups for case logs [], motion data [], medical images [], and a slave controller [] (each 2%, 1/56). Across these settings, studies reported the use of automated skill assessment, formative feedback, and adaptive progression, supported by video, kinematic, and performance-metric streams.

Over time, setup diversity increased, peaking in 2023 (see ). Simulation training and box trainers were consistently present, while surgical video and da Vinci deployments clustered in 2021-2023. These patterns mirror data availability and standardization in risk-managed environments, where AI can be trained and evaluated reliably.

‎

Discussion

Principal Findings

This section discusses the study’s implications and contributions to the field. The review maps and analyzes current applications of AI in surgical training, assessment, and evaluation, identifying the most common surgical procedures, AI techniques, training setups, and highlighting gaps and opportunities for future research. The results show that AI is most frequently reported in data-rich, risk-mitigated environments, notably simulation training and box-trainer setups, and that ML (unspecified) and DL (unspecified) approaches dominate model choices.

Within these settings, many studies report models that leverage synchronized inputs, for example, kinematics, video, and other performance metrics, to classify technical skill using consistent criteria, to characterize learning trajectories across repeated attempts, and to localize performance-limiting behaviors at the level of gestures, steps, or procedural phases. When embedded in iterative practice, these capabilities may enable individualized training pathways that adjust task parameters and feedback density to a trainee’s evolving competence, with the potential to shorten time to proficiency and to reduce instructor workload. These implications are consistent with the results, in which simulation training accounted for 36% (20/56) and box trainer setups for 23% (13/56) of the included studies.

Findings in Relation to the Research Questions

Regarding the first research question aimed at identifying the specific surgical procedures where AI algorithms are most frequently applied in surgical training, AI use concentrates on MIS skills [-], neurosurgery [-], and laparoscopy [-]. Rather than simple frequency, the common thread across these areas is structured, high-signal data capture and well-specified tasks. Endoscopic and robotic workflows generate synchronized video, robotic kinematics, and simulator logs, which enable reproducible labels such as phase boundaries, gesture events, and Objective Structured Assessment of Technical Skills–aligned rubrics. This ecosystem lowers barriers to annotation and validation, thereby accelerating method development. Beyond these clusters, activity in ophthalmology [-], open surgery [,], robot-assisted surgery [-], and single-study specialties including radiology [], urology [], and vascular surgery [] signals a widening scope. However, these domains often face less standardized capture or a more variable field-of-view, which complicates model training and external validation. The overall distribution, therefore, appears to reflect data tractability and curricular formalization more than inherent differences in educational need.

The second research question investigated which AI techniques have been used in surgical training and evaluation. Studies use ML (unspecified) [,,,,,,,,,-] and DL (unspecified) [,,,,,] as broad families, with task-appropriate specializations such as CNNs for video [,,,,,] and SVMs for lower-dimensional kinematics or hand-crafted features [,,,,]. NNs [,,,] support competency modeling when feature engineering is feasible, and CNN+LSTM hybrids [,] target temporal dynamics for suturing and task segmentation. DNNs are explicitly mentioned in [,]. Single-study categories (fuzzy systems [,], regression analysis [], Markov chains [], tutoring system (unspecified) [], Bayesian network [], transformers [], and SVM+RF []) illustrate exploratory breadth rather than established consensus. Consistent with coding, CNN+LSTM is treated as a distinct class and not double-counted under CNNs. No single approach emerges as universally optimal; instead, methods align with task structure (classification vs sequence prediction), signal characteristics (video and kinematics), and assessment granularity (summative scores versus frame- or gesture-level feedback).

The third research question investigated how AI techniques are being used to assess and improve surgical training. Across setups, a common pattern is the move from retrospective, manual scoring to prospective, automated analytics that are both standardized and timely. In simulation training, synchronized streams enable immediate feedback and progression gating, which supports deliberate practice cycles grounded in objective metrics. This aligns with the preponderance of simulation studies in the dataset and the consistent application of ML and DL to transform kinematics and video into competency-linked outputs. In box trainers, models quantify motion economy, tool path quality, and task efficiency, enabling skill stratification and targeted coaching [-,-,,,,]. In robotic systems on the da Vinci platform, studies demonstrate automated assessment, uncertainty-aware feedback, and domain adaptation for cross-site or cross-task transfer [,,,,,]. In surgical video pipelines, investigators focus on procedural understanding, ergonomics, and fine-grained performance analytics [,,-,,,,]. The unifying mechanism across these contexts is measurement at scale that reduces feedback latency, increases consistency, and enables adaptive progression rules without displacing instructor oversight.

Finally, the last research question investigated the way in which AI applications in surgical training affect the learning curve of surgical residents and fellows. Multiple studies report outcomes consistent with accelerated learning and improved technical performance under AI-enabled training. This includes predictive modeling of progression [], metric selection and learning-curve characterization in simulation [], a randomized comparison of feedback modalities [], competency-based training backed by neural models [], continuous monitoring of bimanual expertise with deep models [], and competency estimation in laparoscopic training []. Evidence from robotic contexts shows that automated assessment can structure practice with short feedback loops []. That said, effect sizes remain difficult to aggregate due to heterogeneous study designs, small sample sizes, nonstandard outcome measures, and limited external validation. The most defensible interpretation is that personalized, data-driven feedback and objective, repeated measurement are plausible mechanisms for the observed gains, with further multicenter validation needed to establish generalizability and durability.

The findings suggest that current AI deployment in surgical training follows data availability and standardization, that ML/DL with video and kinematics are dominant because they best match that data, and that automated, timely feedback is the primary lever through which AI influences performance and learning. Where capture is less standardized or external validation is sparse, adoption tends to lag. This synthesis directly motivates the recommendations presented later in the Discussion section on common benchmarks, transparent reporting, and SDG 4–aligned scalability.

Comparison With Previous Work

Systematic literature reviews in surgical training found in the literature have focused on specific training methods (eg, simulation-based training) or on specific types of surgery (eg, plastic surgery and orthopedic surgery) rather than providing a cross-specialty map of AI methods for training, assessment, and evaluation. Reviews focused on simulation-based training within specific domains underscore this pattern. Lawaetz et al [] examined simulation-based training and assessment in open vascular surgery, cataloguing common methods and commenting on effectiveness within that context. Abelleyra Lastoria et al [] surveyed simulation-based tools in plastic surgery and concluded that the validity of many approaches requires further investigation. Woodward et al [] reached a similar conclusion in orthopedic surgery, noting concerns about the construct validity and methodological rigor of simulation studies. Reviews centered on robotic-assisted surgery also reflect divergent emphases: Rahimi et al [] provided a descriptive overview of training modalities and assessment practices, whereas Boal et al [] explicitly scrutinized AI methods for technical skills in robotic surgery and highlighted that both manual and automated assessment tools are often insufficiently validated.

Closer to the scope of the present scoping review, several analyses have examined automation and AI across surgical training tasks. Levin et al [] identified families of automated technical skill assessment methods, including computer vision, motion tracking, ML and DL, and performance classification, but did not synthesize evidence on educational effectiveness. Lam et al [] focused specifically on ML methods and reported accuracy rates that generally exceeded 80 percent across included studies, offering a performance-oriented view rather than a training-context analysis. Pedrett et al [] emphasized the central role of video-derived motion and robotic kinematic data as inputs to AI models for technical skill assessment in minimally invasive surgery, reinforcing the importance of structured, high-signal data streams.

Findings from the present review are consistent with these previous observations in several respects. First, the centrality of simulation and other risk-managed environments recurs across literature, reflecting where ground truth is tractable and measurement can be standardized. Second, many reviews identify validation gaps, noting that reported metrics, dataset partitions, and labeling practices vary widely, which complicates comparison across sites and inhibits external generalizability [-]. Third, there is broad agreement that AI-assisted assessment is advancing rapidly in robotic and minimally invasive settings; yet, many frameworks remain descriptive or single-center, and their educational impact is not consistently established with robust designs [-].

At the same time, this review differs from earlier work in several ways. The scope extends across specialties and across training setups, linking procedures, techniques, and use cases in a single comparative framework. Rather than isolating a single algorithm family or specialty, the analysis connects the dominant AI techniques to the data modalities they exploit and to the assessment functions they serve. This mapping clarifies why ML and DL approaches, particularly CNN-based and hybrid temporal models, are prevalent where high-quality video and kinematics are available, and why adoption is slower where capture is less standardized. In addition, the review integrates signals relevant to learning curves, highlighting studies that associate AI-enabled feedback with improvements in proficiency trajectories, while also acknowledging heterogeneity and the need for external validation. By taking this comparative perspective, the review identifies shared deficiencies that cut across specialties, including nonstandard outcome measures, limited transparency in algorithmic reporting, and sparse multicenter testing, and points toward future work on benchmarks, interoperable data schemas, and scalable deployment aligned with SDG 4.

Whereas previous reviews have been primarily domain-specific or method-specific, this scoping review offers a cross-specialty synthesis that links where AI is used, which techniques are used, and how they are used to support training and assessment. This perspective complements existing literature by emphasizing comparability across contexts, illuminating mechanisms by which AI influences learning, and articulating the methodological steps needed to translate promising prototypes into reproducible, generalizable, and educationally meaningful tools.

Strengths and Limitations

This scoping review offers a broad, cross-specialty perspective on the application of AI in surgical training, assessment, and evaluation. It maps procedures, techniques, and training setups within a single comparative framework, which supports interpretation across contexts rather than within a single specialty. The review adheres to PRISMA-ScR guidance, applies explicit inclusion and exclusion criteria, and uses transparent counting rules that assign each study a primary AI technique and a primary setup to avoid double-counting. Results are presented as both narrative synthesis and structured summaries. The Discussion integrates an SDG 4 perspective, offering concrete implementation considerations related to access, scalability, and equity. Together, these elements provide a panoramic view of where AI is currently deployed, why certain methods dominate in specific data environments, and how these choices influence assessment and feedback in practice.

Several constraints should be considered. First, the search was limited to English-language publications and to the period ending March 18, 2024, which may omit relevant work outside this window. Second, many articles describe methods only at a general label level (AI, ML, and DL) without specifying architectures or training details, which limits interpretability and reproducibility. Third, the evidence base is concentrated in simulation, box-trainer, and video-centric settings, which may not fully capture transfer to live clinical performance, patient outcomes, or longer-term retention. Fourth, external validation is limited, as relatively few studies report multicenter testing, performance under domain shift, subgroup analyses, or calibration, which constrains confidence in portability.

To address these limitations, educational outcomes should also be mapped to recognized competency frameworks and reported with standardized metrics that enable replication and meta-synthesis. When multisetup or multi-technique pipelines are used, authors should specify proportional attribution. Reporting on access, resource requirements, and cost per trainee hour will support the deployment and equity assessment of SDG 4. Multicenter collaborations that release shared benchmarks and interoperable datasets will be necessary to improve reproducibility and to allow fair comparisons across techniques and settings.

Future Work Recommendations

This scoping review identified current applications of AI in surgical education and highlighted priority areas for further work. As summarized in and visualized in , a large proportion of studies focus on simulation training [-,-,,,,,], representing 36% (20/56) of the included articles. This concentration reflects the suitability of simulation for controlled data capture and iterative practice. Building on this foundation, AI can enhance simulation-based training with realistic, adaptive, and personalized learning experiences [,], while also enabling standardized and rapid feedback that supports deliberate practice.

Advances in computer vision are particularly significant where high-quality video and kinematic data are accessible, which aligns with the prevalence of simulation and box-trainer studies in the included literature. In these regulated, risk-mitigated environments, AI systems can produce timely and structured feedback linked to defined competency frameworks, including economy of motion, bimanual coordination, camera control, tissue handling, and ergonomics, thereby facilitating deliberate practice. Although natural language processing technologies are less represented in the current review, their growing maturity suggests near-term opportunities to integrate narrative guidance, rubric-based feedback, and reflective prompts alongside quantitative metrics, provided such outputs are aligned with curricular objectives and are appropriately validated.

Future efforts should pursue 5 complementary directions.

First, strengthen external validity. Studies should include multi-institution cohorts, predefined external test sets, and reporting of performance under domain shift, including different camera views, instruments, and case difficulty. Where feasible, researchers should evaluate the transfer from simulation or bench-top tasks to higher-fidelity or clinical settings with clearly specified outcome measures and follow-up intervals.

Second, standardize educational outcomes. Investigators should map AI outputs to recognized competency frameworks and report validity, reliability, learning curve parameters, and time to competency with consistent definitions. Agreement on core outcome sets will enable comparison across techniques and facilitate meta-synthesis.

Third, expand the breadth and transparency of data. New work should prioritize multimodal capture that combines video, kinematics, tool telemetry, where appropriate, eye tracking or physiological signals. Public or data-sharing consortia should release interoperable schemas, labeling protocols, and benchmark tasks that are specific to procedures and skill elements. Clear descriptions of models and training and validation splits will improve reproducibility.

Fourth, improve usability, equity, and scalability in alignment with SDG 4. Models should operate on standard hardware, interoperate with existing simulators and video platforms, and function reliably in low-bandwidth or offline environments. Reporting of access, installation steps, resource needs, and cost per trainee hour will support adoption in diverse settings. Interfaces should disclose uncertainty, make feedback interpretable, and integrate into educator workflows without adding undue burden.

Fifth, broaden methodological scope responsibly. There is an opportunity to study natural language technologies for rubric-based guidance, structured debriefs, and reflective prompts, provided outputs are aligned with curricular objectives and validated for educational use. Prospective trials that compare feedback modalities and density, and that measure downstream retention and transfer, will clarify how AI should be integrated pedagogically.

Together, these directions could move the field from promising prototypes toward reproducible, generalizable, and educationally meaningful tools that improve surgeon training while supporting equitable access to high-quality education.

Conclusions

This scoping review maps current applications of AI in surgical training, assessment, and evaluation across procedures, techniques, and training setups. From 1400 records, 56 studies met the inclusion criteria, with activity concentrated in minimally invasive surgery, neurosurgery, and laparoscopy. AI is most frequently deployed in data-rich, risk-mitigated environments, particularly simulation training and box trainers, where synchronized video and kinematic streams support objective measurement and timely feedback. Technique choices reflect these data conditions, with ML (unspecified) and DL (unspecified) methods predominating and task-specific variants, such as CNNs and hybrid temporal models, applied to video-centric problems.

Across settings, studies describe automated skill assessment, structured formative feedback, and adaptive progression, with several reporting improvements consistent with accelerated learning curves. At the same time, heterogeneity in study design, small samples, nonstandard outcome measures, and limited external validation constrain strong inferences about effect sizes and generalizability. The evidence, therefore, supports cautious optimism that AI-enabled feedback can enhance skill acquisition, while underscoring the need for more rigorous evaluation.

Future work should prioritize precise reporting of models and datasets, multicenter validation, and standardized educational outcomes linked to recognized competency frameworks. Interoperable data schemes, shared benchmarks, and transparent methods will be essential to enable comparison across sites and techniques. Attention to scalability, access, and usability will support alignment with SDG 4, ensuring that benefits extend beyond well-resourced centers. With these elements in place, AI has the potential to deliver reproducible, equitable, and educationally meaningful gains in surgical training.

Sue Harley

Sue Harley

SAN FRANCISCO (AP) — Meta has prevailed over an existential challenge to its business that could have forced the tech giant to spin off Instagram and WhatsApp after a judge ruled that the company does not hold a monopoly in social networking.

U.S. District Judge James Boasberg issued his ruling Tuesday after the historic antitrust trial wrapped up in late May. His decision follows two separate rulings that branded Google an illegal monopoly in both search and online advertising, dealing yet another regulatory blow to the tech industry that for years enjoyed nearly unbridled growth.

READ MORE: Mark Zuckerberg takes the stand in historic antitrust trial that could force breakup of Meta

The FTC “continues to insist that Meta competes with the same old rivals it has for the last decade, that the company holds a monopoly among that small set, and that it maintained that monopoly through anticompetitive acquisitions,” Boasberg wrote in his ruling. “Whether or not Meta enjoyed monopoly power in the past, though, the agency must show that it continues to hold such power now. The Court’s verdict today determines that the FTC has not done so.”

Samsung Recognized for Commitment to Employee Growth, Well-being, and Innovation

 

Samsung Electronics Co., Ltd. is proud to announce it’s been recognized as Canada’s Top 100 Employers for 2026, by Mediacorp Canada Inc. This recognition underscores Samsung’s dedication to fostering a forward-thinking workplace that prioritizes the growth, well-being, and success of its employees.

 

“We are incredibly proud of this recognition, which is a true testament to the dedication and hard work of our team,” said Brian Shin, Samsung Electronics Canada Inc., President & CEO. “We are committed to building a resilient and forward-looking organization that empowers our employees to thrive both professionally and personally.”

 

Samsung has strengthened it focus on talent investments, ensuring employees receive the support and skills necessary to drive innovation and success in the years ahead. From comprehensive health and wellness programs to training and development opportunities, Samsung continues to strive towards high engagement and satisfaction.

Pecans, America’s only native major nut, have a storied history in the United States. Today, American trees produce hundreds of million of pounds of pecans – 80% of the world’s pecan crop. Most of that crop stays here. Pecans are used to produce pecan milk, butter and oil, but many of the nuts end up in pecan pies.

Throughout history, pecans have been overlooked, poached, cultivated and improved. As they have spread throughout the United States, they have been eaten raw and in recipes. Pecans have grown more popular over the decades, and you will probably encounter them in some form this holiday season.

I’m an extension specialist in Oklahoma, a state consistently ranked fifth in pecan production, behind Georgia, New Mexico, Arizona and Texas. I’ll admit that I am not a fan of the taste of pecans, which leaves more for the squirrels, crows and enthusiastic pecan lovers.

The spread of pecans

The pecan is a nut related to the hickory. Actually, though we call them nuts, pecans are actually a type of fruit called a drupe. Drupes have pits, like the peach and cherry.

The pecan nuts that look like little brown footballs are actually the seed that starts inside the pecan fruit – until the fruit ripens and splits open to release the pecan. They are usually the size of your thumb, and you may need a nutcracker to open them. You can eat them raw or as part of a cooked dish.

The pecan derives its name from the Algonquin “pakani,” which means “a nut too hard to crack by hand.” Rich in fat and easy to transport, pecans traveled with Native Americans throughout what is now the southern United States. They were used for food, medicine and trade as early as 8,000 years ago.

Pecans are native to the southern United States, and while they had previously spread along travel and trade routes, the first documented purposeful planting of a pecan tree was in New York in 1722. Three years later, George Washington’s estate, Mount Vernon, had some planted pecans. Washington loved pecans, and Revolutionary War soldiers said he was constantly eating them.

Meanwhile, no one needed to plant pecans in the South, since they naturally grew along riverbanks and in groves. Pecan trees are alternate bearing: They will have a very large crop one year, followed by one or two very small crops. But because they naturally produced a harvest with no input from farmers, people did not need to actively cultivate them. Locals would harvest nuts for themselves but otherwise ignored the self-sufficient trees.

It wasn’t until the late 1800s that people in the pecan’s native range realized the pecan’s potential worth for income and trade. Harvesting pecans became competitive, and young boys would climb onto precarious tree branches. One girl was lifted by a hot air balloon so she could beat on the upper branches of trees and let them fall to collectors below. Pecan poaching was a problem in natural groves on private property.

Pecan cultivation begins

Even with so obvious a demand, cultivated orchards in the South were still rare into the 1900s. Pecan trees don’t produce nuts for several years after planting, so their future quality is unknown.

To guarantee quality nuts, farmers began using a technique called grafting; they’d join branches from quality trees to another pecan tree’s trunk. The first attempt at grafting pecans was in 1822, but the attempts weren’t very successful.

Grafting pecans became popular after an enslaved man named Antoine who lived on a Louisiana plantation successfully produced large pecans with tender shells by grafting, around 1846. His pecans became the first widely available improved pecan variety.

The variety was named Centennial because it was introduced to the public 30 years later at the Philadelphia Centennial Expedition in 1876, alongside the telephone, Heinz ketchup and the right arm of the Statue of Liberty.

This technique also sped up the production process. To keep pecan quality up and produce consistent annual harvests, today’s pecan growers shake the trees while the nuts are still growing, until about half of the pecans fall off. This reduces the number of nuts so that the tree can put more energy into fewer pecans, which leads to better quality. Shaking also evens out the yield, so that the alternate-bearing characteristic doesn’t create a boom-bust cycle.

US pecan consumption

The French brought praline dessert with them when they immigrated to Louisiana in the early 1700s. A praline is a flat, creamy candy made with nuts, sugar, butter and cream. Their original recipe used almonds, but at the time, the only nut available in America was the pecan, so pecan pralines were born.

During the Civil War and world wars, Americans consumed pecans in large quantities because they were a protein-packed alternative when meat was expensive and scarce. One ounce of pecans has the same amount of protein as 2 ounces of meat.

After the wars, pecan demand declined, resulting in millions of excess pounds at harvest. One effort to increase demand was a national pecan recipe contest in 1924. Over 21,000 submissions came from over 5,000 cooks, with 800 of them published in a book.

Pecan consumption went up with the inclusion of pecans in commercially prepared foods and the start of the mail-order industry in the 1870s, as pecans can be shipped and stored at room temperature. That characteristic also put them on some Apollo missions. Small amounts of pecans contain many vitamins and minerals. They became commonplace in cereals, which touted their health benefits.

In 1938, the federal government published the pamphlet Nuts and How to Use Them, which touted pecans’ nutritional value and came with recipes. Food writers suggested using pecans as shortening because they are composed mostly of fat.

The government even put a price ceiling on pecans to encourage consumption, but consumers weren’t buying them. The government ended up buying the surplus pecans and integrating them into the National School Lunch Program.

While you are sitting around the Thanksgiving table this year, you can discuss one of the biggest controversies in the pecan industry: Are they PEE-cans or puh-KAHNS?

Housing bosses at a Surrey council have said they need another £107,000 and more staff to fix deep-rooted problems in the service.

In September, the Regulator of Social Housing (RSH) gave Tandridge District Council a rating of C4 – meaning there were very serious failings and potential for government intervention.

Of the extra money, £87,000 would be spent on salaries for extra staff to help the department and £20,000 on “service costs”, according to the Local Democracy Reporting Service.

Head of housing at the authority, Jane Rochelle, said: “We’re working at a tremendous pace and I’m putting my whole team under pressure.”

She added: “I don’t intend to take my foot off the gas this side of Christmas at least.”

The council had already carved out £420,000 from the housing revenue accounts operating surplus to kickstart the housing improvement plan before the inspection results came in.

The scale of the work under way was outlined at the council’s housing committee on 11 November.

Housing officers are trying to catch up with national standards right across the service, from rewriting policies to overhauling IT systems and carrying out thousands of overdue tenancy audits.

Housing leaders have said they are focusing on what the RSH calls the “big six” safety areas – things like gas, electric, fire, asbestos and water safety.

One of the main issues is a backlog of about 2,000 tenancy audits, which are basic checks that confirm who lives in each property, identify vulnerabilities and pick up risks like fuel poverty or damp.

Savills is currently inspecting every council home, according to a council report, and said of the 710 properties it had already surveyed – about 30% of the stock – most windows and doors would need replacing “sooner rather than later”.

It also said many homes would need insulation upgrades, and many boilers would require associated pipework and radiators to be replaced.

But officers found kitchens and bathrooms to be generally in a fair condition, and said the stock overall is not in a poor condition, but would be “hungry for investment” in the next decade.

The new housing boss said she was “fairly comfortable” with the results and hoped there would not be any more nasty surprises.

Tandridge’s improvement plan will continue into 2026/27 with progress reported back to both the regulator and councillors.

Visma, one of Europe’s biggest software companies, has approached a leading City grandee to become its chair if it goes ahead with a blockbuster €20bn listing in London next spring.

Sir Ron Kalifa, former boss of payments group Worldpay and a director of the Bank of England, is considered the leading candidate for the potential role after a round of interviews in recent weeks, the Guardian understands.

However, sources close to the process cautioned that London is not yet certain to land the sought-after listing of the Norwegian company, which has been backed by UK-based private equity firm Hg Capital for almost two decades.

Stockholm has emerged as a rival because Visma is better known in Scandinavian markets, and because the Swedish bourse last month hosted the successful €13.7bn flotation, or initial public offering (IPO), of security services group Verisure, whose shares rose 25% on debut.

The absence of similar-sized IPOs in recent years is seen as one risk to listing in London. An offer to Kalifa to join Visma’s board may depend on London being chosen. Current executive chair, Øystein Moan, who was Visma’s chief executive from 1997 to 2020, could yet stay in that role.

The Oslo-based company is currently running “early look” meetings with major fund managers to gauge likely demand for the shares and investors’ preference for listing venue and governance. The company is being advised by investment banks Goldman Sachs, Morgan Stanley and UBS.

London was reported in the summer to have beaten Amsterdam to attract the listing in what was seen as a coup – Visma would be the biggest London listing for years. An eleventh-hour loss to Stockholm would be regarded as a heavy blow. “There are different routes this could go down and nothing is yet decided,” said one source close to the process.

The potential recruitment of Kalifa may become important to London’s case. Kalifa was chief executive of Worldpay for more than 10 years and later vice chair. He led the group through a period of rapid growth during which it was bought out of Royal Bank of Scotland after the financial crash and later re-emerged as a standalone FTSE 100 company. Worldpay was bought by US rival Vantiv for £9.3bn in 2017.

A London listing would also represent a personal win for Kalifa. He wrote a high-profile report for government in 2021 into how to boost the UK’s financial technology sector by attracting investment and new companies – including how to encourage them to choose London.

skip past newsletter promotion

after newsletter promotion

While Visma is not a pure technology company, it sees the development of AI products to “simplify and automate complex processes” as critical to its business in the next few years.

The group makes accounting, payroll and HR software products for 2.2 million customers. It has 17,500 employees and describes itself as the leading provider outside North America of “mission-critical business software”.

–   EPKINLY plus rituximab and lenalidomide (EPKINLY + R²) is now the first and only bispecific antibody combination therapy available for patients with relapsed or refractory follicular lymphoma after at least one line of systemic therapy
–   In the Phase 3 EPCORE^® FL-1 trial, EPKINLY + R² demonstrated significantly superior progression-free survival and overall response rates compared to standard of care R² with approximately 3 out of 4 patients achieving a complete response
–   Approval marks third indication for EPKINLY and first-ever FDA approval for a bispecific combination therapy in lymphoma

NORTH CHICAGO, Ill., Nov. 18, 2025 /PRNewswire/ — AbbVie (NYSE: ABBV) today announced that EPKINLY^® (epcoritamab-bysp), a T-cell engaging bispecific antibody administered subcutaneously, in combination with rituximab and lenalidomide (EPKINLY + R²) is approved by the U.S. Food and Drug Administration (FDA) for the treatment of adult patients with relapsed or refractory (R/R) follicular lymphoma (FL). This approval of EPKINLY is based on results from the pivotal Phase 3 EPCORE^® FL-1 study that evaluated fixed duration EPKINLY + R² compared to standard of care R² and demonstrates the potential of this combination therapy to reshape FL treatment and to reach patients earlier in their treatment.ⁱ 

“Recurrent follicular lymphoma can be an incurable, complex and persistent disease, creating a clear need for additional treatments that can change its course earlier in the treatment journey,” said Lorenzo Falchi, M.D., lymphoma specialist, department of medicine, Memorial Sloan Kettering Cancer Center. “The results shown with EPKINLY + R² in the EPCORE FL-1 study are incredibly meaningful, demonstrating durable responses compared to patients treated with R² alone. These data, delivered by a regimen that’s chemotherapy-free and can be administered in the outpatient setting, suggest that EPKINLY + R² could potentially become a new standard of care.”

FL is typically an indolent (slow-growing) form of non-Hodgkin lymphoma (NHL) that arises from B-lymphocytes and impacts approximately 15,000 new patients per year in the U.S.^ii,iii The disease is considered incurable with current available therapies.^iv Patients with FL often relapse, and in some cases, the disease can transform into a more aggressive form of NHL called diffuse large B-cell lymphoma (DLBCL).^v

The Phase 3 EPCORE FL-1 study included a broad range of patients, including those with indolent to aggressive disease. In the study, EPKINLY + R² reduced the risk of disease progression or death by 79% (HR 0.21, 95% CI: 0.13% – 0.33%, p<0.0001) compared to standard of care R² alone. In the dual primary endpoint of overall response rate (ORR), 89% of patients treated with EPKINLY + R² responded to treatment (n=216/243, 95% CI: 84% – 93%; p<0.0001) compared to 74% of patients treated with R² (n=181/245, 95% CI: 68%-79%). The median for dual primary endpoint of progression-free survival (PFS), was not reached (NR) among patients treated with EPKINLY + R² (95% CI: 21.9 months – NR) compared to 11.2 months for patients treated with R² (95% CI: 10.5 months – NR). Among patients who were treated with EPKINLY + R², 74% achieved a complete response (CR) (n=181/243, 95% CI: 69% – 80%, p<0.0001) compared to 43% of patients treated with R² (n=106/245, 95% CI: 37% – 50%).ⁱ

The safety profile of EPKINLY + R² in the EPCORE FL-1 study was generally consistent with the known safety profiles of the individual regimens (epcoritamab and R²).  The most common (≥ 20%) adverse reactions in patients who received EPKINLY + R² were rash, upper respiratory tract infections, fatigue, injection site reactions, constipation, diarrhea, cytokine release syndrome (CRS), pneumonia, COVID-19 and fever. The most common Grade 3 to 4 laboratory abnormalities (≥ 10%) were decreased neutrophil count, lymphocyte count, and platelets. CRS occurred in 24% of patients at the recommended 3 step-up dosage schedule, and was primarily low grade (19% Grade 1, 5% Grade 2). A single event of immune effector cell-associated neurotoxicity syndrome (ICANS) occurred in one patient, grade 1 (0.8%). The prescribing information has a Boxed Warning for serious or life-threatening CRS and ICANS. Warnings and precautions include infections, cytopenias, and embryo-fetal toxicity. Please see additional Important Safety Information below.

“Today’s milestone marks meaningful progress for people living with follicular lymphoma. With a bispecific-based therapy that can be administered in a variety of medical settings, patients have the possibility of accessing this treatment at sites of care closer to where they live,” said Meghan Gutierrez, chief executive officer, Lymphoma Research Foundation.

EPKINLY + R² was previously granted Breakthrough Therapy Designation (BTD) by the FDA for the treatment of R/R FL. This designation is an FDA process designed to expedite the development and review of drugs that are intended to treat a serious condition, and preliminary clinical evidence indicates that the drug may demonstrate substantial improvement over available therapy on a clinically significant endpoint(s).

“With this approval, EPKINLY is now the first bispecific antibody available for patients with follicular lymphoma in the second-line plus setting. New options are needed to improve outcomes for patients with relapsed or refractory disease,” said Daejin Abidoye, MD, vice president, therapeutic area head, oncology, solid tumor and hematology, AbbVie.

In June 2024, EPKINLY monotherapy was granted accelerated approval by the FDA for the treatment of R/R FL following two or more lines of systemic therapy. With the results of the confirmatory Phase 3 EPCORE FL-1 study, the FDA has also converted this accelerated approval to a full approval. Both companies will pursue additional international regulatory approvals for the R/R FL indication.

Data from the Phase 3 EPCORE FL-1 study will be presented at the Annual Meeting and Exposition of the American Society of Hematology (ASH) in December 2025.

About the EPCORE^® FL-1 Trial 
EPCORE FL-1 (NCT05409066) is a Phase 3 open-label randomized interventional trial to evaluate the safety and efficacy of epcoritamab plus rituximab and lenalidomide (R²) versus R² alone in patients with relapsed/refractory (R/R) follicular lymphoma (FL). Patients were randomized to receive EPKINLY in combination R² (n=243) or R² alone (n=245). Patients received EPKINLY in 28-day cycles for a total of 12 cycles or until disease progression or unacceptable toxicity, whichever occurred first. Efficacy was established based on the dual primary endpoints of progression free survival (PFS) and overall response rate (ORR) determined by Lugano 2014 criteria as assessed by Independent Review Committee (IRC). Additional efficacy outcome measures include complete response (CR) and duration of response (DOR).

EPKINLY® (epcoritamab-bysp) U.S. INDICATIONS AND IMPORTANT SAFETY INFORMATION

What is EPKINLY?
EPKINLY is a prescription medicine used to treat adults with:

• certain types of diffuse large B-cell lymphoma (DLBCL) or high-grade B-cell lymphoma that has come back (relapsed) or that did not respond (refractory), after 2 or more treatments.
  • EPKINLY for the treatment of DLBCL is approved based on patient response data. Studies are ongoing to confirm the clinical benefit of EPKINLY.
• follicular lymphoma (FL) that has come back or that did not respond to previous treatment, together with lenalidomide and rituximab
• follicular lymphoma (FL) that has come back or that did not respond after receiving 2 or more treatments.

It is not known if EPKINLY is safe and effective in children.

Important Warnings—EPKINLY can cause serious side effects, including:

• Cytokine release syndrome (CRS), which is common during treatment with EPKINLY and can be serious or lead to death. To help reduce your risk of CRS, you will receive EPKINLY on a step-up dosing schedule (when you receive 2 or 3 smaller step-up doses of EPKINLY before your first full dose during your first cycle of treatment), and you may also receive other medicines before and for 3 days after receiving EPKINLY. If your dose of EPKINLY is delayed for any reason, you may need to repeat the step-up dosing schedule.
• Neurologic problems that can be serious, and can be life-threatening, and lead to death. Neurologic problems may happen days or weeks after you receive EPKINLY.

People with DLBCL or high-grade B-cell lymphoma should be hospitalized for 24 hours after receiving their first full dose of EPKINLY on Day 15 of Cycle 1 due to the risk of CRS and neurologic problems.

People with follicular lymphoma (FL) may need to be hospitalized after receiving their first full dose of EPKINLY on Day 22 of Cycle 1 due to the risk of CRS.

Tell your healthcare provider or get medical help right away if you develop a fever of 100.4°F (38°C) or higher; dizziness or lightheadedness; trouble breathing; chills; fast heartbeat; feeling anxious; headache; confusion; shaking (tremors); problems with balance and movement, such as trouble walking; trouble speaking or writing; confusion and disorientation; drowsiness, tiredness or lack of energy; muscle weakness; seizures; or memory loss. These may be symptoms of CRS or neurologic problems. If you have any symptoms that impair consciousness, do not drive or use heavy machinery or do other dangerous activities until your symptoms go away.

EPKINLY can cause other serious side effects, including:

• Infections that may lead to death. Your healthcare provider will check you for signs and symptoms of infection before and during treatment and treat you as needed if you develop an infection. You should receive medicines from your healthcare provider before you start treatment to help prevent infection. Tell your healthcare provider right away if you develop any symptoms of infection during treatment, including fever of 100.4°F (38°C) or higher, cough, chest pain, tiredness, shortness of breath, painful rash, sore throat, pain during urination, feeling weak or generally unwell, or confusion.
• Low blood cell counts, which can be serious or severe. Your healthcare provider will check your blood cell counts during treatment. EPKINLY may cause low blood cell counts, including low white blood cells (neutropenia and lymphopenia), which can increase your risk for infection; low red blood cells (anemia), which can cause tiredness and shortness of breath; and low platelets (thrombocytopenia), which can cause bruising or bleeding problems.

Your healthcare provider will monitor you for symptoms of CRS, neurologic problems, infections, and low blood cell counts during treatment with EPKINLY. Your healthcare provider may temporarily stop or completely stop treatment with EPKINLY if you develop certain side effects.

Before you receive EPKINLY, tell your healthcare provider about all your medical conditions, including if you have an infection, are pregnant or plan to become pregnant, or are breastfeeding or plan to breastfeed. If you receive EPKINLY while pregnant, it may harm your unborn baby. If you are a female who can become pregnant, your healthcare provider should do a pregnancy test before you start treatment with EPKINLY and you should use effective birth control (contraception) during treatment and for 4 months after your last dose of EPKINLY. Tell your healthcare provider if you become pregnant or think that you may be pregnant during treatment with EPKINLY. Do not breastfeed during treatment with EPKINLY and for 4 months after your last dose of EPKINLY.

The most common side effects of EPKINLY when used alone in DLBCL or high-grade B-cell lymphoma or FL include CRS, injection site reactions, tiredness, muscle and bone pain, fever, diarrhea, COVID-19, rash, and stomach-area (abdominal) pain. The most common severe abnormal laboratory test results with EPKINLY when used alone include decreased white blood cells, decreased red blood cells, and decreased platelets.

The most common side effects of EPKINLY when used together with lenalidomide and rituximab in FL include rash, upper respiratory tract infections, tiredness, injection site reactions, constipation, diarrhea, CRS, pneumonia, COVID-19, and fever. The most common severe abnormal laboratory test results with EPKINLY when used together with lenalidomide and rituximab include decreased white blood cells and decreased platelets.

These are not all of the possible side effects of EPKINLY. Call your doctor for medical advice about side effects.

You are encouraged to report side effects to the FDA at (800) FDA-1088 or www.fda.gov/medwatch or to Genmab US, Inc. at 1-855-4GENMAB (1-855-443-6622).

Please see Medication Guide, including Important Warnings.

About EPKINLY^® (epcoritamab-bysp)
EPKINLY^® (epcoritamab-bysp) is an IgG1-bispecific antibody created using Genmab’s proprietary DuoBody^® technology and administered subcutaneously. Genmab’s DuoBody-CD3 technology is designed to direct cytotoxic T cells selectively to elicit an immune response toward target cell types. Epcoritamab is designed to simultaneously bind to CD3 on T cells and CD20 on B cells and induces T-cell-mediated killing of CD20+ cells.^vi 

Epcoritamab (approved under the brand name EPKINLY in countries including the U.S. and Japan, and as TEPKINLY^® in the European Union) has received regulatory approval in certain lymphoma indications in more than 65 countries. Epcoritamab is being co-developed by Genmab and AbbVie as part of the companies’ oncology collaboration. The companies will share commercialization responsibilities in the U.S. and Japan, with AbbVie responsible for further global commercialization. Both companies will pursue additional international regulatory approvals for the R/R FL indication and additional approvals for the R/R DLBCL indication. 

Genmab and AbbVie continue to evaluate the use of epcoritamab as a monotherapy, and in combination, across lines of therapy in a range of hematologic malignancies. This includes four additional ongoing Phase 3, open-label, randomized trials including a trial evaluating epcoritamab as a monotherapy in patients with R/R DLBCL compared to investigators choice immunochemotherapy (NCT04628494), a trial evaluating epcoritamab in combination with R-CHOP in adult patients with newly diagnosed DLBCL (NCT05578976), a trial evaluating epcoritamab in combination with rituximab and lenalidomide (R²) compared to chemoimmunotherapy in patients with previously untreated FL (NCT06191744), and a trial evaluating epcoritamab in combination with lenalidomide compared to chemotherapy infusion in patients with R/R DLBCL (NCT06508658). The safety and efficacy of epcoritamab has not been established for these investigational uses. Please visit www.clinicaltrials.gov for more information.

About AbbVie in Oncology
AbbVie is committed to elevating standards of care and bringing transformative therapies to patients worldwide living with difficult-to-treat cancers. We are advancing a dynamic pipeline of investigational therapies across a range of cancer types in both blood cancers and solid tumors. We are focusing on creating targeted medicines that either impede the reproduction of cancer cells or enable their elimination. We achieve this through various, targeted treatment modalities and biology interventions, including small molecule therapeutics, antibody-drug conjugates (ADCs), immuno-oncology-based therapeutics, multispecific antibody and novel CAR-T platforms. Our dedicated and experienced team joins forces with innovative partners to accelerate the delivery of potential breakthrough medicines.

Today, our expansive oncology portfolio comprises approved and investigational treatments for a wide range of blood cancers and solid tumors. We are evaluating more than 35 investigational medicines in multiple clinical trials across some of the world’s most widespread and debilitating cancers. As we work to have a remarkable impact on people’s lives, we are committed to exploring solutions to help patients obtain access to our cancer medicines. For more information, please visit http://www.abbvie.com/oncology.

About AbbVie
AbbVie’s mission is to discover and deliver innovative medicines and solutions that solve serious health issues today and address the medical challenges of tomorrow. We strive to have a remarkable impact on people’s lives across several key therapeutic areas including immunology, oncology, neuroscience and eye care – and products and services in our Allergan Aesthetics portfolio. For more information about AbbVie, please visit us at www.abbvie.com. Follow @abbvie on LinkedIn, Facebook, Instagram, X (formerly Twitter), and YouTube.

AbbVie Forward-Looking Statements 
Some statements in this news release are, or may be considered, forward-looking statements for purposes of the Private Securities Litigation Reform Act of 1995. The words “believe,” “expect,” “anticipate,” “project” and similar expressions and uses of future or conditional verbs, generally identify forward-looking statements. AbbVie cautions that these forward-looking statements are subject to risks and uncertainties that may cause actual results to differ materially from those expressed or implied in the forward-looking statements. Such risks and uncertainties include, but are not limited to, challenges to intellectual property, competition from other products, difficulties inherent in the research and development process, adverse litigation or government action, changes to laws and regulations applicable to our industry, the impact of global macroeconomic factors, such as economic downturns or uncertainty, international conflict, trade disputes and tariffs, and other uncertainties and risks associated with global business operations. Additional information about the economic, competitive, governmental, technological and other factors that may affect AbbVie’s operations is set forth in Item 1A, “Risk Factors,” of AbbVie’s 2024 Annual Report on Form 10-K, which has been filed with the Securities and Exchange Commission, as updated by its Quarterly Reports on Form 10-Q and in other documents that AbbVie subsequently files with the Securities and Exchange Commission that update, supplement or supersede such information. AbbVie undertakes no obligation, and specifically declines, to release publicly any revisions to forward-looking statements as a result of subsequent events or developments, except as required by law.

Contacts:

ⁱ EPKINLY (epcoritamab-bysp) [package insert]. Copenhagen, Denmark: Genmab, 2025.
ⁱⁱ Lymphoma Research Foundation official website. https://lymphoma.org/aboutlymphoma/nhl/fl/. Accessed November 2025.
ⁱⁱⁱ Leukemia & Lymphoma Society. https://www.lls.org/research/follicular-lymphoma-fl. Accessed November 2025.
^iv Ghione P, Palomba ML, Ghesquieres H, et al. Treatment patterns and outcomes in relapsed/refractory follicular lymphoma: results from the international SCHOLAR-5 study. Haematologica. 2023;108(3):822-832. doi: 10.3324/haematol.2022.281421.
^v Al-Tourah AJ, Gill KK, Chhanabhai M, et al. Population-based analysis of incidence and outcome of transformed non-Hodgkin’s lymphoma. J Clin Oncol. 2008 Nov 10;26(32):5165-9. doi: 10.1200/JCO.2008.16.0283. Epub 2008 Oct 6. PMID: 18838711.
^vi Engelberts PJ, et al. DuoBody-CD3xCD20 Induces Potent T-Cell-Mediated Killing of Malignant B Cells in Preclinical Models and Provides Opportunities for Subcutaneous Dosing. EBioMedicine. 2020;52:102625. doi: 10.1016/j.ebiom.2019.102625.

 

SOURCE AbbVie