Biomedical Text Research Articles

An extensive benchmark study on biomedical text generation and mining with ChatGPT.

In recent years, the development of natural language process (NLP) technologies and deep learning hardware has led to significant improvement in large language models (LLMs). The ChatGPT, the state-of-the-art LLM built on GPT-3.5 and GPT-4, shows excellent capabilities in general language understanding and reasoning. Researchers also tested the GPTs on a variety of NLP-related tasks and benchmarks and got excellent results. With exciting performance on daily chat, researchers began to explore the capacity of ChatGPT on expertise that requires professional education for human and we are interested in the biomedical domain. To evaluate the performance of ChatGPT on biomedical-related tasks, this article presents a comprehensive benchmark study on the use of ChatGPT for biomedical corpus, including article abstracts, clinical trials description, biomedical questions, and so on. Typical NLP tasks like named entity recognization, relation extraction, sentence similarity, question and answering, and document classification are included. Overall, ChatGPT got a BLURB score of 58.50 while the state-of-the-art model had a score of 84.30. Through a series of experiments, we demonstrated the effectiveness and versatility of ChatGPT in biomedical text understanding, reasoning and generation, and the limitation of ChatGPT build on GPT-3.5. All the datasets are available from BLURB benchmark https://microsoft.github.io/BLURB/index.html. The prompts are described in the article.

Healthcare: A Growing Role for Large Language Models and Generative AI

Abstract: Large language models and generative artificial intelligence (GAI) have recently demonstrated significant promise for revolutionizing a range of industries, including healthcare. The paper investigates how these cutting-edge AI developments are transforming healthcare applications. We focus on how big language models, like GPT-3 (generative pretrained transformer), Visual ChatGPT and generative AI, such Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), can be applied to solve important problems in the healthcare sector. Medical text analysis is one of the main uses of massive language models in the visual ChatGPT healthcare industry. These models have impressive natural language processing abilities that make it possible to effectively extract important information from electronic health records (EHRs), biomedical text data from large biobanks, scholarly articles and patient notes. The Biomedical Transformer Model represents a ground-breaking development in natural language processing for the biomedical field, exhibiting outstanding performance in comprehending and producing textual data. It opens up new avenues for biomedical research, diagnosis, and personalized therapy when combined with Multimodal Biomedical AI, which makes use of numerous data sources, including pictures, genomes, and clinical records. On the other side, generative AI has made great progress in medical picture analysis such as MRI scans and X-rays. The outstanding performance of GANs in medical picture synthesis and augmentation has helped to increase the precision and accuracy of diagnosis. Due to the issues with small and uneven medical datasets, VAEs have proven crucial in producing realistic medical images for training and research reasons. In addition to describing the various generative AI tools used in healthcare, this paper also provides an overview of multimodal medical LLMs and the biomedical transformer LLMs in the healthcare industry. Large language models and generative AI have great potential, but ethical issues and data privacy are still major problems in healthcare applications. Further, we investigate the potential role of multimodal medical LLMs as the foundation for novel assistive technologies in the fields of professional medicine, medical research, and consumer applications in the healthcare industry

SPBERE: Boosting span-based pipeline biomedical entity and relation extraction via entity information

Triplet extraction is one of the fundamental tasks in biomedical text mining. Compared with traditional pipeline approaches, joint methods can alleviate the error propagation problem from entity recognition to relation classification. However, existing methods face challenges in detecting overlapping entities and overlapping relations, which are ubiquitous in biomedical texts. In this work, we propose a novel pipeline method of end-to-end biomedical triplet extraction. In particular, a span-based detection strategy is used to detect the overlapping triplets by enumerating possible candidate spans and entity pairs. The strategy is further used to capture different contextualized representations via an entity model and a relation model, respectively. Furthermore, to enhance interrelation between spans, entity information from the output of the entity model is used to construct the input for the relation model without utilizing any external knowledge. Our approach is evaluated on the drug–drug interaction (DDI) and chemical-protein interaction (CHEMPROT) datasets, exhibiting improvement of the absolute F1-score in relation extraction by 3.5%–3.7% compared prior work. The experimental results highlight the importance of overlapping triplet detection using the span-based approach, acquisition of various contextualized representations via different in-domain pre-trained language models, and early fusion of entity information in the relation model.

Transformers for extracting breast cancer information from Spanish clinical narratives

The wide adoption of electronic health records (EHRs) offers immense potential as a source of support for clinical research. However, previous studies focused on extracting only a limited set of medical concepts to support information extraction in the cancer domain for the Spanish language. Building on the success of deep learning for processing natural language texts, this paper proposes a transformer-based approach to extract named entities from breast cancer clinical notes written in Spanish and compares several language models. To facilitate this approach, a schema for annotating clinical notes with breast cancer concepts is presented, and a corpus for breast cancer is developed. Results indicate that both BERT-based and RoBERTa-based language models demonstrate competitive performance in clinical Named Entity Recognition (NER). Specifically, BETO and multilingual BERT achieve F-scores of 93.71% and 94.63%, respectively. Additionally, RoBERTa Biomedical attains an F-score of 95.01%, while RoBERTa BNE achieves an F-score of 94.54%. The findings suggest that transformers can feasibly extract information in the clinical domain in the Spanish language, with the use of models trained on biomedical texts contributing to enhanced results. The proposed approach takes advantage of transfer learning techniques by fine-tuning language models to automatically represent text features and avoiding the time-consuming feature engineering process.

Drug-Drug Interaction Extraction from Biomedical Text Using Relation BioBERT with BLSTM

In the context of pharmaceuticals, drug-drug interactions (DDIs) occur when two or more drugs interact, potentially altering the intended effects of the drugs and resulting in adverse patient health outcomes. Therefore, it is essential to identify and comprehend these interactions. In recent years, an increasing number of novel compounds have been discovered, resulting in the discovery of numerous new DDIs. There is a need for effective methods to extract and analyze DDIs, as the majority of this information is still predominantly located in biomedical articles and sources. Despite the development of various techniques, accurately predicting DDIs remains a significant challenge. This paper proposes a novel solution to this problem by leveraging the power of Relation BioBERT (R-BioBERT) to detect and classify DDIs and the Bidirectional Long Short-Term Memory (BLSTM) to improve the accuracy of predictions. In addition to determining whether two drugs interact, the proposed method also identifies the specific types of interactions between them. Results show that the use of BLSTM leads to significantly higher F-scores compared to our baseline model, as demonstrated on three well-known DDI extraction datasets that includes SemEval 2013, TAC 2018, and TAC 2019.

RelCurator: a text mining-based curation system for extracting gene-phenotype relationships specific to neurodegenerative disorders.

The identification of gene-phenotype relationships is important in medical genetics as it serves as a basis for precision medicine. However, most of the gene-phenotype relationship data are buried in the biomedical literature in textual form. We propose RelCurator, a curation system that extracts sentences including both gene and phenotype entities related to specific disease categories from PubMed articles, provides rich additional information such as entity taggings, and predictions of gene-phenotype relationships. We targeted neurodegenerative disorders and developed a deep learning model using Bidirectional Gated Recurrent Unit (BiGRU) networks and BioWordVec word embeddings for predicting gene-phenotype relationships from biomedical texts. The prediction model is trained with more than 130,000 labeled PubMed sentences including gene and phenotype entities, which are related to or unrelated to neurodegenerative disorders. We compared the performance of our deep learning model with those of Bidirectional Encoder Representations from Transformers (BERT), Support Vector Machine (SVM), and simple Recurrent Neural Network (simple RNN) models. Our model performed better with an F1-score of 0.96. Furthermore, the evaluation done using a few curation cases in the real scenario showed the effectiveness of our work. Therefore, we conclude that RelCurator can identify not only new causative genes, but also new genes associated with neurodegenerative disorders' phenotype. RelCurator is a user-friendly method for accessing deep learning-based supporting information and a concise web interface to assist curators while browsing the PubMed articles. Our curation process represents an important and broadly applicable improvement to the state of the art for the curation of gene-phenotype relationships.

Biomolecular Event Extraction using Natural Language Processing

Biomedical research and discoveries are communicated through scholarly publications and this literature is voluminous, rich in scientific text and growing exponentially by the day. Biomedical journals publish nearly three thousand research articles daily, making literature search a challenging proposition for researchers. Biomolecular events involve genes, proteins, metabolites, and enzymes that provide invaluable insights into biological processes and explain the physiological functional mechanisms. Text mining (TM) or extraction of such events automatically from big data is the only quick and viable solution to gather any useful information. Such events extracted from biological literature have a broad range of applications like database curation, ontology construction, semantic web search and interactive systems. However, automatic extraction has its challenges on account of ambiguity and the diverse nature of natural language and associated linguistic occurrences like speculations, negations etc., which commonly exist in biomedical texts and lead to erroneous elucidation. In the last decade, many strategies have been proposed in this field, using different paradigms like Biomedical natural language processing (BioNLP), machine learning and deep learning. Also, new parallel computing architectures like graphical processing units (GPU) have emerged as possible candidates to accelerate the event extraction pipeline. This paper reviews and provides a summarization of the key approaches in complex biomolecular big data event extraction tasks and recommends a balanced architecture in terms of accuracy, speed, computational cost, and memory usage towards developing a robust GPU-accelerated BioNLP system.

S1000: a better taxonomic name corpus for biomedical information extraction.

The recognition of mentions of species names in text is a critically important task for biomedical text mining. While deep learning-based methods have made great advances in many named entity recognition tasks, results for species name recognition remain poor. We hypothesize that this is primarily due to the lack of appropriate corpora. We introduce the S1000 corpus, a comprehensive manual re-annotation and extension of the S800 corpus. We demonstrate that S1000 makes highly accurate recognition of species names possible (F-score =93.1%), both for deep learning and dictionary-based methods. All resources introduced in this study are available under open licenses from https://jensenlab.org/resources/s1000/. The webpage contains links to a Zenodo project and three GitHub repositories associated with the study.

Transformer versus traditional natural language processing: how much data is enough for automated radiology report classification?

Current state-of-the-art natural language processing (NLP) techniques use transformer deep-learning architectures, which depend on large training datasets. We hypothesized that traditional NLP techniques may outperform transformers for smaller radiology report datasets. We compared the performance of BioBERT, a deep-learning-based transformer model pre-trained on biomedical text, and three traditional machine-learning models (gradient boosted tree, random forest, and logistic regression) on seven classification tasks given free-text radiology reports. Tasks included detection of appendicitis, diverticulitis, bowel obstruction, and enteritis/colitis on abdomen/pelvis CT reports, ischemic infarct on brain CT/MRI reports, and medial and lateral meniscus tears on knee MRI reports (7,204 total annotated reports). The performance of NLP models on held-out test sets was compared after training using the full training set, and 2.5%, 10%, 25%, 50%, and 75% random subsets of the training data. In all tested classification tasks, BioBERT performed poorly at smaller training sample sizes compared to non-deep-learning NLP models. Specifically, BioBERT required training on approximately 1,000 reports to perform similarly or better than non-deep-learning models. At around 1,250 to 1,500 training samples, the testing performance for all models began to plateau, where additional training data yielded minimal performance gain. With larger sample sizes, transformer NLP models achieved superior performance in radiology report binary classification tasks. However, with smaller sizes (<1000) and more imbalanced training data, traditional NLP techniques performed better. Our benchmarks can help guide clinical NLP researchers in selecting machine-learning models according to their dataset characteristics.

SnorkelPlus: A Novel Approach for Identifying Relationships Among Biomedical Entities Within Abstracts

AbstractIdentifying relationships between biomedical entities from unstructured biomedical text is a challenging task. SnorkelPlus has been proposed to provide the flexibility to extract these biomedical relations without any human effort. Our proposed model, SnorkelPlus, is aimed at finding connections between gene and disease entities. We achieved three objectives: (i) extract only gene and disease articles from NCBI’s, PubMed or PubMed central database, (ii) define reusable label functions and (iii) ensure label function accuracy using generative and discriminative models. We utilized deep learning methods to achieve label training data and achieved an AUROC of 85.60% for the generated gene and disease corpus from PubMed articles. Snorkel achieved an AUPR of 45.73%, which is +2.3% higher than the baseline model. We created a gene–disease relation database using SnorkelPlus from approximately 29 million scientific abstracts without involving annotated training datasets. Furthermore, we demonstrated the generalizability of our proposed application on abstracts of PubMed articles enriched with different gene and disease relations. In the future, we plan to design a graphical database using Neo4j.

AIONER: all-in-one scheme-based biomedical named entity recognition using deep learning.

Biomedical named entity recognition (BioNER) seeks to automatically recognize biomedical entities in natural language text, serving as a necessary foundation for downstream text mining tasks and applications such as information extraction and question answering. Manually labeling training data for the BioNER task is costly, however, due to the significant domain expertise required for accurate annotation. The resulting data scarcity causes current BioNER approaches to be prone to overfitting, to suffer from limited generalizability, and to address a single entity type at a time (e.g. gene or disease). We therefore propose a novel all-in-one (AIO) scheme that uses external data from existing annotated resources to enhance the accuracy and stability of BioNER models. We further present AIONER, a general-purpose BioNER tool based on cutting-edge deep learning and our AIO schema. We evaluate AIONER on 14 BioNER benchmark tasks and show that AIONER is effective, robust, and compares favorably to other state-of-the-art approaches such as multi-task learning. We further demonstrate the practical utility of AIONER in three independent tasks to recognize entity types not previously seen in training data, as well as the advantages of AIONER over existing methods for processing biomedical text at a large scale (e.g. the entire PubMed data). The source code, trained models and data for AIONER are freely available at https://github.com/ncbi/AIONER.

Few-Shot Learning for Clinical Natural Language Processing Using Siamese Neural Networks: Algorithm Development and Validation Study

Background Natural language processing (NLP) has become an emerging technology in health care that leverages a large amount of free-text data in electronic health records to improve patient care, support clinical decisions, and facilitate clinical and translational science research. Recently, deep learning has achieved state-of-the-art performance in many clinical NLP tasks. However, training deep learning models often requires large, annotated data sets, which are normally not publicly available and can be time-consuming to build in clinical domains. Working with smaller annotated data sets is typical in clinical NLP; therefore, ensuring that deep learning models perform well is crucial for real-world clinical NLP applications. A widely adopted approach is fine-tuning existing pretrained language models, but these attempts fall short when the training data set contains only a few annotated samples. Few-shot learning (FSL) has recently been investigated to tackle this problem. Siamese neural network (SNN) has been widely used as an FSL approach in computer vision but has not been studied well in NLP. Furthermore, the literature on its applications in clinical domains is scarce. Objective The aim of our study is to propose and evaluate SNN-based approaches for few-shot clinical NLP tasks. Methods We propose 2 SNN-based FSL approaches, including pretrained SNN and SNN with second-order embeddings. We evaluate the proposed approaches on the clinical sentence classification task. We experiment with 3 few-shot settings, including 4-shot, 8-shot, and 16-shot learning. The clinical NLP task is benchmarked using the following 4 pretrained language models: bidirectional encoder representations from transformers (BERT), BERT for biomedical text mining (BioBERT), BioBERT trained on clinical notes (BioClinicalBERT), and generative pretrained transformer 2 (GPT-2). We also present a performance comparison between SNN-based approaches and the prompt-based GPT-2 approach. Results In 4-shot sentence classification tasks, GPT-2 had the highest precision (0.63), but its recall (0.38) and F score (0.42) were lower than those of BioBERT-based pretrained SNN (0.45 and 0.46, respectively). In both 8-shot and 16-shot settings, SNN-based approaches outperformed GPT-2 in all 3 metrics of precision, recall, and F score. Conclusions The experimental results verified the effectiveness of the proposed SNN approaches for few-shot clinical NLP tasks.

AJE: Attention Mechanism for Entity-relation Joint Extraction

Joint extraction of entities and relations is an significant issue of information extraction, which is very helpful for many downstream tasks, including knowledge base construction, question answering, and biomedical text diagnosis[1], etc. The common approach of existing models is to extract the subject and the relation first, then compute the subject and the relation to obtain the object, and finally, the triplet is judged. However, such an approach cannot efficiently handle of information extraction, and the results are not very good for Subject-Object Overlap (SOO) case. In this paper, a joint entity-relation extraction method AJE is proposed based on dot-product attention mechanism. The method first maps subject, object and relationship into three matrices of Q, K and V. After that, the attention weighting is achieved on these three matrices and the output vector is used to determine whether the triple is correct or not. The F1-score is used in experiments to show that the proposed model is more efficient than the current existing ones. It also has better results in handling other cases such as SOO, multi-triple problem, etc.

A Review on Electronic Health Record Text-Mining for Biomedical Name Entity Recognition in Healthcare Domain.

Biomedical-named entity recognition (bNER) is critical in biomedical informatics. It identifies biomedical entities with special meanings, such as people, places, and organizations, as predefined semantic types in electronic health records (EHR). bNER is essential for discovering novel knowledge using computational methods and Information Technology. Early bNER systems were configured manually to include domain-specific features and rules. However, these systems were limited in handling the complexity of the biomedical text. Recent advances in deep learning (DL) have led to the development of more powerful bNER systems. DL-based bNER systems can learn the patterns of biomedical text automatically, making them more robust and efficient than traditional rule-based systems. This paper reviews the healthcare domain of bNER, using DL techniques and artificial intelligence in clinical records, for mining treatment prediction. bNER-based tools are categorized systematically and represent the distribution of input, context, and tag (encoder/decoder). Furthermore, to create a labeled dataset for our machine learning sentiment analyzer to analyze the sentiment of a set of tweets, we used a manual coding approach and the multi-task learning method to bias the training signals with domain knowledge inductively. To conclude, we discuss the challenges facing bNER systems and future directions in the healthcare field.

PubExN: An Automated PubMed Bulk Article Extractor with Affiliation Normalization Package.

Biomedical article extraction is the preliminary step for every biomedical application. These applications are helpful in finding the gene, disease, chemical, drugs, protein entities. Finding entities relation such as gene-gene entities, drug-disease interaction, and chemical protein relation the PubExN can be helpful for these types of biomedical applications. In most cases, domain experts do this extraction process on their own. Human interference makes this process time-consuming and there is a high probability, that documents can be missed during the extraction process. To get rid of these complicated processes a python package is introduced to automate the process of bulk extraction from the PubMed database. The extraction process covers all the citation information with the associated abstract. The batch approach is used to extract the bulk extraction. The motivation for the development of PubExN was to provide flexibility for the extraction process of biomedical article's text data from NCBI's PubMed database. Basically, NCBI's PubMed database article contains the article id or can say PubMed-id (PMID), the title of the article, abstract, authors information, etc. This package will benefit many biomedical texts mining research including biomedical named entity recognition, biomedical relation extraction, literature discovery, knowledgebase creation, and various biomedical Natural Language Processing (NLP) tasks. In addition, it could be used in the author name disambiguation problems and new drug discoveries. This package will help save time and extra effort for the extraction and normalization process of PubMed articles.

Fine-grained biomedical knowledge negation detection via contrastive learning

Negation is a common and complex phenomenon in natural language, which will reverse the truth value of knowledge extracted from sentences and make knowledge validation indispensable. Negation detection at the granularity of knowledge triples is still a challenging and minimally explored task, especially in the biomedical domain where multiple knowledge triples frequently exist in one sentence and the same triple can express quite distinct meanings under different contexts. In this paper, we illustrate and define the label conflict problem in biomedical negation detection and propose CL-GSAN, a novel Contrastive Learning-based Gated Structured Attention Network for identifying the negated knowledge triples from the biomedical text. CL-GSAN jointly performs fine-grained knowledge-context modeling within a sentence via a gating mechanism and learns the target triple’s context representations by considering the semantic divergence of claims via contrastive learning. Experimental results demonstrate that CL-GSAN obtains promising performance on three biomedical corpora, including the Chinese Medical Knowledge Negation (CMKN) corpus constructed by us to facilitate fine-grained, i.e., triple-level, knowledge negation detection.

DDI-MuG: Multi-aspect graphs for drug-drug interaction extraction.

Drug-drug interaction (DDI) may lead to adverse reactions in patients, thus it is important to extract such knowledge from biomedical texts. However, previously proposed approaches typically focus on capturing sentence-aspect information while ignoring valuable knowledge concerning the whole corpus. In this paper, we propose a Multi-aspect Graph-based DDI extraction model, named DDI-MuG. We first employ a bio-specific pre-trained language model to obtain the token contextualized representations. Then we use two graphs to get syntactic information from input instance and word co-occurrence information within the entire corpus, respectively. Finally, we combine the representations of drug entities and verb tokens for the final classification. To validate the effectiveness of the proposed model, we perform extensive experiments on two widely used DDI extraction dataset, DDIExtraction-2013 and TAC 2018. It is encouraging to see that our model outperforms all twelve state-of-the-art models. In contrast to the majority of earlier models that rely on the black-box approach, our model enables visualization of crucial words and their interrelationships by utilizing edge information from two graphs. To the best of our knowledge, this is the first model that explores multi-aspect graphs to the DDI extraction task, and we hope it can establish a foundation for more robust multi-aspect works in the future.

Web Interface of NER and RE with BERT for Biomedical Text Mining

The BioBERT Named Entity Recognition (NER) model is a high-performance model designed to identify both known and unknown entities. It surpasses previous NER models utilized by text-mining tools, such as tmTool and ezTag, in effectively discovering novel entities. In previous studies, the Biomedical Entity Recognition and Multi-Type Normalization Tool (BERN) employed this model to identify words that represent specific names, discern the type of the word, and implement it on a web page to offer NER service. However, we aimed to offer a web service that includes Relation Extraction (RE), a task determining the relation between entity pairs within a sentence. First, just like BERN, we fine-tuned the BioBERT NER model within the biomedical domain to recognize new entities. We identified two categories: diseases and genes/proteins. Additionally, we fine-tuned the BioBERT RE model to determine the presence or absence of a relation between the identified gene–disease entity pairs. The NER and RE results are displayed on a web page using the Django web framework. NER results are presented in distinct colors, and RE results are visualized as graphs in NetworkX and Cytoscape, allowing users to interact with the graphs.

Multi-task learning for few-shot biomedical relation extraction

Artificial intelligence (AI) has advanced rapidly, but it has limited impact on biomedical text understanding due to a lack of annotated datasets (a.k.a. few-shot learning). Multi-task learning, which uses data from multiple datasets and tasks with related syntax and semantics, has potential to address this issue. However, the effectiveness of this approach heavily relies on the quality of the available data and its transferability between tasks. In this paper, we propose a framework, built upon a state-of-the-art multi-task method (i.e. MT-DNN), that leverages different publicly available biomedical datasets to enhance relation extraction performance. Our model employs a transformer-based architecture with shared encoding layers across multiple tasks, and task-specific classification layers to generate task-specific representations. To further improve performance, we utilize a knowledge distillation technique. In our experiments, we assess the impact of incorporating biomedical datasets in a multi-task learning setting and demonstrate that it consistently outperforms state-of-the-art few-shot learning methods in cases of limited data. This results in significant improvement across most datasets and few-shot scenarios, particularly in terms of recall scores.

Abstract 5365: Extracting novel knowledge from scientific literature to build a web portal for cancer researchers to keep up with the latest scientific discoveries

Abstract Millions of new papers are published in biomedical sciences every year. In many disciplines, it has become impossible to read all the published new papers to learn what is happening in the frontier of a particular area. This gap is widening as the publishing speed has been accelerating in recent years. To address this challenge, one can convert unstructured text data into a structured form, which can then support highly accurate information retrieval, information integration, and automated knowledge discovery. A plausible approach for such a task is to use named entity recognition (NER) and relation extraction (RE) methods to identify important biological entities and extract their relations to construct knowledge graphs (KGs). The LitCoin Natural Language Processing (NLP) Challenge was recently organized by NCATS of NIH and NASA to spur innovation by rewarding the most creative and high-impact uses of biomedical text to create KGs. In addition to entities and relations, the manually annotated LitCoin dataset also contains the annotations of relations being new discoveries or background knowledge. Our team participated in the challenge and ranked first place. The novelty prediction model of our pipeline has achieved an F1 score of 0.90. We have applied our model to all the PubMed abstracts published previously and the newly published ones to extract the novel discoveries in each article. A web portal has been created to allow scientists to view the latest discoveries in cancer research. The web portal is updated daily with versatile visualization tools for cancer researchers to quickly grasp the latest discoveries in a particular area. It also offers powerful functions to explore the existing literature and make sophisticated inferences about causal and indirect relationships. Citation Format: Feng Pan, Yuan Zhang, Xin Sui, Donghu Sun, Menghan Chung, Jinfeng Zhang. Extracting novel knowledge from scientific literature to build a web portal for cancer researchers to keep up with the latest scientific discoveries. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5365.