Team publications
Synapsis AI publications
Publications 6
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read moreAnalysis of a large language model-based system versus manual review in clinical data abstraction and deduction from real-world medical records of patients with melanoma for clinical trial eligibility assessment. (opens in a new tab)
On 1,125 eligibility questions across 50 Cleveland Clinic melanoma records, the Synapsis LLM reached 95.73% accuracy in 2.5 minutes. A melanoma research nurse scored 95.11% in 427 minutes and an oncology research nurse 88.09% in 540 minutes, both measured against physician consensus.
Read moreAutomated abstraction of colorectal cancer registry data using AI: Accuracy and implementation insights. (opens in a new tab)
Synapsis AI abstracted 33 colorectal cancer registry fields (60 questions per patient) from the EMR, mostly from unstructured text, and was compared with registrar-abstracted tumor registry data. Overall accuracy was 94.78%, reaching 100% for lymph node yield and positivity, TNM stage, MMR and MSI status, immunotherapy regimens and surgical procedures.
Read moreUsing a medically trained LLM-based end-to-end system in automating patient eligibility screening across a health system for a phase 3 study evaluating the safety of givinostat in patients with polycythemia vera. (opens in a new tab)
Starting from 4.7 million Cleveland Clinic records, Synapsis AI narrowed to 904 patients with polycythemia vera and completed full eligibility assessment within one week. It identified 22 eligible patients for the GIV-IN PV trial, all verified by research staff (100% positive predictive value), about seven times the yield of standard screening.
Read moreSelf-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection (opens in a new tab)
Extends LLM self-evaluation to multi-step tasks by comparing holistic scoring with step-by-step scoring on two benchmarks. Stepwise evaluation generally detects errors better, with up to a 15% relative gain in AUC-ROC.
Read more
Leadership
Eirini Schlosser
CEO and Founder
Publications 5
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreAutomated abstraction of colorectal cancer registry data using AI: Accuracy and implementation insights. (opens in a new tab)
Synapsis AI abstracted 33 colorectal cancer registry fields (60 questions per patient) from the EMR, mostly from unstructured text, and was compared with registrar-abstracted tumor registry data. Overall accuracy was 94.78%, reaching 100% for lymph node yield and positivity, TNM stage, MMR and MSI status, immunotherapy regimens and surgical procedures.
Read moreAnalysis of a large language model-based system versus manual review in clinical data abstraction and deduction from real-world medical records of patients with melanoma for clinical trial eligibility assessment. (opens in a new tab)
On 1,125 eligibility questions across 50 Cleveland Clinic melanoma records, the Synapsis LLM reached 95.73% accuracy in 2.5 minutes. A melanoma research nurse scored 95.11% in 427 minutes and an oncology research nurse 88.09% in 540 minutes, both measured against physician consensus.
Read moreUsing a medically trained LLM-based end-to-end system in automating patient eligibility screening across a health system for a phase 3 study evaluating the safety of givinostat in patients with polycythemia vera. (opens in a new tab)
Starting from 4.7 million Cleveland Clinic records, Synapsis AI narrowed to 904 patients with polycythemia vera and completed full eligibility assessment within one week. It identified 22 eligible patients for the GIV-IN PV trial, all verified by research staff (100% positive predictive value), about seven times the yield of standard screening.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read more
Weiqi Sun, PhD
Chief AI Officer
Publications 19
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreSelf-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection (opens in a new tab)
Extends LLM self-evaluation to multi-step tasks by comparing holistic scoring with step-by-step scoring on two benchmarks. Stepwise evaluation generally detects errors better, with up to a 15% relative gain in AUC-ROC.
Read moreAutomated abstraction of colorectal cancer registry data using AI: Accuracy and implementation insights. (opens in a new tab)
Synapsis AI abstracted 33 colorectal cancer registry fields (60 questions per patient) from the EMR, mostly from unstructured text, and was compared with registrar-abstracted tumor registry data. Overall accuracy was 94.78%, reaching 100% for lymph node yield and positivity, TNM stage, MMR and MSI status, immunotherapy regimens and surgical procedures.
Read moreAnalysis of a large language model-based system versus manual review in clinical data abstraction and deduction from real-world medical records of patients with melanoma for clinical trial eligibility assessment. (opens in a new tab)
On 1,125 eligibility questions across 50 Cleveland Clinic melanoma records, the Synapsis LLM reached 95.73% accuracy in 2.5 minutes. A melanoma research nurse scored 95.11% in 427 minutes and an oncology research nurse 88.09% in 540 minutes, both measured against physician consensus.
Read moreUsing a medically trained LLM-based end-to-end system in automating patient eligibility screening across a health system for a phase 3 study evaluating the safety of givinostat in patients with polycythemia vera. (opens in a new tab)
Starting from 4.7 million Cleveland Clinic records, Synapsis AI narrowed to 904 patients with polycythemia vera and completed full eligibility assessment within one week. It identified 22 eligible patients for the GIV-IN PV trial, all verified by research staff (100% positive predictive value), about seven times the yield of standard screening.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read moreCompositional Task-Oriented Parsing as Abstractive Question Answering (opens in a new tab)
Task-oriented parsing (TOP) aims to convert natural language into machine-readable representations of specific tasks, such as setting an alarm. A popular approach to TOP is to apply seq2seq models to generate linearized parse trees. A more recent line of work argues that pretrained seq2seq2 models are better at generating outputs that are themselves natural language, so they replace linearized parse trees with canonical natural-language paraphrases that can then be easily translated into parse trees, resulting in so-called naturalized parsers. In this work we continue to explore naturalized semantic parsing by presenting a general reduction of TOP to abstractive question answering that overcomes some limitations of canonical paraphrasing. Experimental results show that our QA-based technique outperforms state-of-the-art methods in full-data settings while achieving dramatic improvements in few-shot settings.
Read moreUnfreeze with Care: Space-Efficient Fine-Tuning of Semantic Parsing Models (opens in a new tab)
Semantic parsing is a key NLP task that maps natural language to structured meaning representations. As in many other NLP tasks, SOTA performance in semantic parsing is now attained by fine-tuning a large pretrained language model (PLM). While effective, this approach is inefficient in the presence of multiple downstream tasks, as a new set of values for all parameters of the PLM needs to be stored for each task separately. Recent work has explored methods for adapting PLMs to downstream tasks while keeping most (or all) of their parameters frozen. We examine two such promising techniques, prefix tuning and bias-term tuning, specifically on semantic parsing. We compare them against each other on two different semantic parsing datasets, and we also compare them against full and partial fine-tuning, both in few-shot and conventional data settings. While prefix tuning is shown to do poorly for semantic parsing tasks off the shelf, we modify it by adding special token embeddings, which results in very strong performance without compromising parameter savings.
Read more
Clinical Innovations
Olga Tasopoulou
VP, Clinical Innovations
Publications 6
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreAutomated abstraction of colorectal cancer registry data using AI: Accuracy and implementation insights. (opens in a new tab)
Synapsis AI abstracted 33 colorectal cancer registry fields (60 questions per patient) from the EMR, mostly from unstructured text, and was compared with registrar-abstracted tumor registry data. Overall accuracy was 94.78%, reaching 100% for lymph node yield and positivity, TNM stage, MMR and MSI status, immunotherapy regimens and surgical procedures.
Read moreAnalysis of a large language model-based system versus manual review in clinical data abstraction and deduction from real-world medical records of patients with melanoma for clinical trial eligibility assessment. (opens in a new tab)
On 1,125 eligibility questions across 50 Cleveland Clinic melanoma records, the Synapsis LLM reached 95.73% accuracy in 2.5 minutes. A melanoma research nurse scored 95.11% in 427 minutes and an oncology research nurse 88.09% in 540 minutes, both measured against physician consensus.
Read moreUsing a medically trained LLM-based end-to-end system in automating patient eligibility screening across a health system for a phase 3 study evaluating the safety of givinostat in patients with polycythemia vera. (opens in a new tab)
Starting from 4.7 million Cleveland Clinic records, Synapsis AI narrowed to 904 patients with polycythemia vera and completed full eligibility assessment within one week. It identified 22 eligible patients for the GIV-IN PV trial, all verified by research staff (100% positive predictive value), about seven times the yield of standard screening.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read moreEarly Biological Therapy in Operated Crohn's Disease Patients Is Associated With a Lower Rate of Endoscopic Recurrence and Improved Long-term Outcomes: A Single-center Experience. (opens in a new tab)
Among 141 Crohn's disease patients followed after surgery, those on biologics at their postoperative colonoscopy had far less endoscopic recurrence (45.2% vs 80.8%). None of them needed hospitalization or repeat surgery, compared with 12.1% of untreated patients.
Read more
Cassandra Kosmidou
Publications 6
Genomic copy number predicts esophageal cancer years before transformation. (opens in a new tab)
Shallow whole-genome sequencing of 777 biopsies from 88 patients in Barrett's esophagus surveillance showed that genomic copy number signals separate progressive from stable disease up to 10 years before cancer appears on histopathology. The result held in two independent cohorts of 76 and 248 patients.
Read moreComputational pathology aids derivation of microRNA biomarker signals from Cytosponge samples. (opens in a new tab)
An AI-based computational pathology tool measured columnar epithelium on Cytosponge slides from 117 people with reflux, so microRNA signals could be traced to their cell of origin. Two microRNAs detected gastric cells with AUCs of 0.97 and 0.95, giving a quality-control metric for large-scale Cytosponge testing.
Read moreKCNQ potassium channels modulate Wnt activity in gastro-oesophageal adenocarcinomas. (opens in a new tab)
An analysis of 897 patients with gastro-oesophageal adenocarcinoma found KCNQ potassium channel genes mutated in about 30%. The channels drive growth through WNT and MYC signaling, and inhibiting them slowed cancer cell proliferation in the lab.
Read moreIssues with the Specificity of Immunological Reagents for NLRP3: Implications for Age-related Macular Degeneration. (opens in a new tab)
Using validated antibodies, this study did not detect the NLRP3 inflammasome in human retinal pigment epithelium cells under several priming conditions, or in macular tissue from patients with age-related macular degeneration. It prompts a second look at earlier reports tying NLRP3 in these cells to the disease.
Read moreThicknessTool: automated ImageJ retinal layer thickness and profile in digital images. (opens in a new tab)
ThicknessTool is a free ImageJ plugin that measures retinal nuclear layer thickness automatically. It agreed with masked manual observers more closely than they agreed with each other, and picked up outer nuclear layer thinning in retinal detachment images that manual measurement did not show as significant.
Read more
Katerina Karathanasopoulou
Publications 6
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreAutomated abstraction of colorectal cancer registry data using AI: Accuracy and implementation insights. (opens in a new tab)
Synapsis AI abstracted 33 colorectal cancer registry fields (60 questions per patient) from the EMR, mostly from unstructured text, and was compared with registrar-abstracted tumor registry data. Overall accuracy was 94.78%, reaching 100% for lymph node yield and positivity, TNM stage, MMR and MSI status, immunotherapy regimens and surgical procedures.
Read moreAnalysis of a large language model-based system versus manual review in clinical data abstraction and deduction from real-world medical records of patients with melanoma for clinical trial eligibility assessment. (opens in a new tab)
On 1,125 eligibility questions across 50 Cleveland Clinic melanoma records, the Synapsis LLM reached 95.73% accuracy in 2.5 minutes. A melanoma research nurse scored 95.11% in 427 minutes and an oncology research nurse 88.09% in 540 minutes, both measured against physician consensus.
Read moreUsing a medically trained LLM-based end-to-end system in automating patient eligibility screening across a health system for a phase 3 study evaluating the safety of givinostat in patients with polycythemia vera. (opens in a new tab)
Starting from 4.7 million Cleveland Clinic records, Synapsis AI narrowed to 904 patients with polycythemia vera and completed full eligibility assessment within one week. It identified 22 eligible patients for the GIV-IN PV trial, all verified by research staff (100% positive predictive value), about seven times the yield of standard screening.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read moreBeta 1, Beta 2 and Beta 3 Adrenergic Receptor Gene Polymorphisms in a Southeastern European Population. (opens in a new tab)
Genotyped 1,021 volunteers for five beta-adrenergic receptor variants tied to cardiovascular and respiratory disease and to drug response. Allele frequencies resembled those of Southwestern Europeans and differed from Northwestern Europeans.
Read more
AI Research
Lisheng Fu
Publications 6
Learning Relatedness between Types with Prototypes for Relation Extraction (opens in a new tab)
Relation types in different datasets often overlap in meaning. This work represents each type with prototypical examples so that related types from another dataset can augment training, easing the shortage of labeled data.
Read moreA Case Study on Learning a Unified Encoder of Relations (opens in a new tab)
Trains one shared encoder across relation datasets with different schemas (ACE05 and ERE), regularized by adversarial training. The multi-task model improves results on both datasets.
Read moreDomain Adaptation for Relation Extraction with Domain Adversarial Neural Network (opens in a new tab)
Relation extractors lose accuracy outside the domain they were trained on. This unsupervised method needs labels only from the source domain and pairs a CNN relation classifier with a domain-adversarial classifier.
Read more
Liang Zhao
Publications 2
A Deep Learning-Based Clinical Classification System for the Differential Diagnosis of Hip Prosthesis Failures Using Radiographs: A Multicenter Study. (opens in a new tab)
Hip-Net, trained on 2,908 routine dual-view radiographs from 1,454 patients at three centers, distinguished five causes of hip replacement failure with 0.904 accuracy and an AUC of 0.937 in an external cohort. Its infection probability maps lined up with intraoperative and pathology findings.
Read moreMultimodal Deep Learning-Based Screening of Degenerative Temporomandibular Joint Disease Using 2D Radiography: A Cost-Effective and Low-Radiation Approach. (opens in a new tab)
A model that fuses panoramic radiographs, lateral cephalograms, age and sex screened 1,000 joints from 500 patients for degenerative TMJ disease, reaching an AUC of 0.898 against cone-beam CT. Using both image types outperformed panoramic radiographs alone.
Read more
Vaibhav Mavi
Sr. NLP Applied Scientist
Publications 6
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreSelf-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection (opens in a new tab)
Extends LLM self-evaluation to multi-step tasks by comparing holistic scoring with step-by-step scoring on two benchmarks. Stepwise evaluation generally detects errors better, with up to a 15% relative gain in AUC-ROC.
Read moreAnalysis of a large language model-based system versus manual review in clinical data abstraction and deduction from real-world medical records of patients with melanoma for clinical trial eligibility assessment. (opens in a new tab)
On 1,125 eligibility questions across 50 Cleveland Clinic melanoma records, the Synapsis LLM reached 95.73% accuracy in 2.5 minutes. A melanoma research nurse scored 95.11% in 427 minutes and an oncology research nurse 88.09% in 540 minutes, both measured against physician consensus.
Read moreUsing a medically trained LLM-based end-to-end system in automating patient eligibility screening across a health system for a phase 3 study evaluating the safety of givinostat in patients with polycythemia vera. (opens in a new tab)
Starting from 4.7 million Cleveland Clinic records, Synapsis AI narrowed to 904 patients with polycythemia vera and completed full eligibility assessment within one week. It identified 22 eligible patients for the GIV-IN PV trial, all verified by research staff (100% positive predictive value), about seven times the yield of standard screening.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read moreMulti-hop Question Answering (opens in a new tab)
A monograph on answering questions that require combining several pieces of information over multiple reasoning steps. It formalizes the task and organizes the field's datasets, models and evaluation methods, including work built on large language models.
Read more
Shubh Jaroria
NLP Applied Scientist
Publications 1
Self-Evaluating LLMs for Multi-Step Tasks: Stepwise Confidence Estimation for Failure Detection (opens in a new tab)
Extends LLM self-evaluation to multi-step tasks by comparing holistic scoring with step-by-step scoring on two benchmarks. Stepwise evaluation generally detects errors better, with up to a 15% relative gain in AUC-ROC.
Read more
Clinical Fellows
Chase Wehrle
Publications 146
Development and validation of a machine-learning model to reduce futile procurements in donations after circulatory death in liver transplantation in the USA: a multicentre study. (opens in a new tab)
A gradient-boosting model built on 2,221 donors from six US centers predicted whether a donor would progress to death within 30 minutes of extubation with an AUC of 0.83, and held that performance in prospective validation. Against surgeon predictions it cut the futile procurement rate from 0.195 to 0.078.
Read moreLong-Term Follow-Up in Liver Transplantation and Combined Cardiac Surgery: Validation and Refinement of the LT+CS Risk Score Using Advanced Machine Learning. (opens in a new tab)
In 46 patients who had a liver transplant and cardiac surgery in a single operation, machine-learning cross-validation refined the LT-CS risk score. Five-year survival was 87.7% in the low-risk group and 49.5% in the high-risk group.
Read moreCombining a deep learning model with clinical data better predicts hepatocellular carcinoma behavior following surgery. (opens in a new tab)
A deep learning model read H&E whole-slide images from 469 Cleveland Clinic patients treated by transplant or resection and predicted recurrence-free and disease-specific survival. Combining it with clinical variables beat either approach alone in every subgroup.
Read moreEnsuring the Needle Is in the Haystack: Enhancing Electronic Health Record Data Capture for Artificial Intelligence-Enabled Therapy Implementation. (opens in a new tab)
An editorial on improving what the electronic health record captures so that AI tools can support delivery of guideline-directed therapy in heart failure.
Read more
Harsh Patolia
Publications 23
Automating Chart Review Using an Artificial Intelligence-Enabled System for Assessing Transthyretin Amyloid Cardiomyopathy Trial Eligibility. (opens in a new tab)
Synapsis AI, embedded in Cleveland Clinic's EMR, pre-screened 1,476 patients in one week for the phase 3 DepleTTR-CM trial and flagged 46 potential matches. Clinician review confirmed 30, and traditional recruitment had missed 29 of them. Accuracy was 96.2% across 7,700 trial-specific questions, with a 99% negative predictive value.
Read moreEvaluating a medically trained artificial intelligence and natural language processing-enabled, large-language model in screening for transthyretin cardiac amyloidosis clinical trials. (opens in a new tab)
Synapsis AI reviewed 1,476 patients' EMRs in one week on 2 GPUs against 11 inclusion and 24 exclusion criteria. It returned 4 complete matches and 26 high partial matches, and all 30 were appropriate. Manual screening over the preceding 90 days had not identified 29 of the 30.
Read moreEnsuring the Needle Is in the Haystack: Enhancing Electronic Health Record Data Capture for Artificial Intelligence-Enabled Therapy Implementation. (opens in a new tab)
An editorial on improving what the electronic health record captures so that AI tools can support delivery of guideline-directed therapy in heart failure.
Read moreTissue Screening and Diagnosis of Transthyretin Amyloid Cardiomyopathy. (opens in a new tab)
Non-cardiac signs of ATTR amyloidosis such as carpal tunnel syndrome can precede the cardiac diagnosis by up to a decade. This review covers screening tissue taken during otherwise indicated procedures to catch ATTR-CM earlier, and weighs variable diagnostic yield, patient anxiety and treatment cost.
Read more
Advisors
Dimitrios Iliopoulos, PhD, MBA
Senior Scientific Advisor & Board Member
Publications 135
Results of the Seventh Scientific Workshop of ECCO: Precision medicine in IBD – what, why, and how. (opens in a new tab)
Many diseases that affect modern humans fall in the category of complex diseases, thus called because they result from a combination of multiple aetiological and pathogenic factors. Regardless of the organ or system affected, complex diseases present major challenges in diagnosis, classification, and management. Current forms of therapy are usually applied in an indiscriminate fashion based on clinical information, but even the most advanced drugs only benefit a limited number of patients and to a variable and unpredictable degree. This ‘one measure does not fit all’ situation has spurred the notion that therapy for complex disease should be tailored to individual patients or groups of patients, giving rise to the notion of ‘precision medicine’ [PM]. Inflammatory bowel disease [IBD] is a prototypical complex disease where the need for PM has become increasingly clear. This prompted the European Crohn’s and Colitis Organisation to focus the 7 th Scientific Workshop on this emerging theme. The articles in this special issue of the Journal address the various complementary aspects of PM in IBD, including what is PM; why it is needed and how it can be used; how PM can contribute to prediction and prevention of IBD; how IBD PM can aid in prognosis and improve response to therapy; and the challenges and future directions of PM in IBD. This first article of this series is structured on three simple concepts [what, why, and how] and addresses the definition of PM, discusses the rationale for the need of PM in IBD, and outlines the methodology required to implement PM in IBD in a correct and clinically meaningful way.
Read moreLysine methyltransferase 2D regulates pancreatic carcinogenesis through metabolic reprogramming. (opens in a new tab)
We define a new antitumorous function of the histone lysine (K)-specific methyltransferase 2D (KMT2D) in pancreatic cancer. KMT2D is transcriptionally repressed in human pancreatic tumours through DNA methylation. Clinically, lower levels of this methyltransferase associate with poor prognosis and significant weight alterations. RNAi-based genetic inactivation of KMT2D promotes tumour growth and results in loss of H3K4me3 mark. In addition, KMT2D inhibition increases aerobic glycolysis and alters the lipidomic profiles of pancreatic cancer cells. Further analysis of this phenomenon identified the glucose transporter SLC2A3 as a mediator of KMT2D-induced changes in cellular, metabolic and proliferative rates.
Read moreThe IBD interactome: an integrated view of aetiology, pathogenesis and therapy. (opens in a new tab)
Crohn’s disease and ulcerative colitis are prototypical complex diseases characterized by chronic and heterogeneous manifestations, induced by interacting environmental, genomic, microbial and immunological factors. These interactions result in an overwhelming complexity that cannot be tackled by studying the totality of each pathological component (an ‘-ome’) in isolation without consideration of the interaction among all relevant -omes that yield an overall ‘network effect’. The outcome of this effect is the ‘IBD interactome’, defined as a disease network in which dysregulation of individual -omes causes intestinal inflammation mediated by dysfunctional molecular modules. To define the IBD interactome, new concepts and tools are needed to implement a systems approach; an unbiased data-driven integration strategy that reveals key players of the system, pinpoints the central drivers of inflammation and enables development of targeted therapies. Powerful bioinformatics tools able to query and integrate multiple -omes are available, enabling the integration of genomic, epigenomic, transcriptomic, proteomic, metabolomic and microbiome information to build a comprehensive molecular map of IBD. This approach will enable identification of IBD molecular subtypes, correlations with clinical phenotypes and elucidation of the central hubs of the IBD interactome that will aid discovery of compounds that can specifically target the hubs that control the disease.
Read moreAssessment of Circulating MicroRNAs for the Diagnosis and Disease Activity Evaluation in Patients with Ulcerative Colitis by Using the Nanostring Technology. (opens in a new tab)
We have identified a signature of 12 circulating microRNAs that differentiate patients with UC from control subjects. Moreover, 6 of these microRNAs significantly correlated with UC disease activity. Importantly, a set of 4 microRNAs (hsa-miR-4454, hsa-miR-223-3p, hsa-miR-23a-3p, and hsa-miR-320e), which correlated with UC disease activity were found to have higher sensitivity and specificity values than C-reactive protein. Circulating microRNAs provide a novel diagnostic and prognostic marker for patients with UC. The use of an FDA-approved platform could accelerate the application of microRNA screening in a gastrointenstinal clinical setting. When used in combination with current diagnostic and disease activity assessment modalities, microRNAs could improve both IBD screening and care management.
Read moreMicroRNA-124 regulates STAT3 expression and is down-regulated in colon tissues of pediatric patients with ulcerative colitis. (opens in a new tab)
miR-124 appears to regulate the expression of STAT3. Reduced levels of miR-124 in colon tissues of children with active UC appear to increase expression and activity of STAT3, which could promote inflammation and the pathogenesis of UC in children. Thus, the activation of miR-124 could have a therapeutic potential in IBD patients.
Read more
Allan J. Pantuck, MD
Senior Medical Advisor
Publications 268
Adjuvant Sunitinib in High-Risk Renal-Cell Carcinoma after Nephrectomy. (opens in a new tab)
Among patients with locoregional clear-cell renal-cell carcinoma at high risk for tumor recurrence after nephrectomy, the median duration of disease-free survival was significantly longer in the sunitinib group than in the placebo group, at a cost of a higher rate of toxic events.
Read moreCarbonic anhydrase-IX score is a novel biomarker that predicts recurrence and survival for high-risk, nonmetastatic renal cell carcinoma: Data from the phase III ARISER clinical trial. (opens in a new tab)
The largest, multicenter, prospective analysis of patients with high-risk nonmetastatic ccRCC demonstrates the utility of CAIX score as a statistically significant prognostic biomarker for survival. It recommends that CAIX score be quantified for all patients with high-risk disease after nephrectomy.
Read moreClinical, molecular, and genetic correlates of lymphatic spread in clear cell renal cell carcinoma. (opens in a new tab)
Predictive model consisting of smoking history (p=0.040), T stage (p<0.0001), Fuhrman grade (p<0.0001), Eastern Cooperative Oncology Group performance status (p<0.0001), and microvascular invasion (p<0.0001) was independently associated with lymphatic spread. After adjustment with these clinical variables, low carbonic anhydrase IX (CAIX) (p=0.043) and high epithelial vascular endothelial growth factor receptor 2 (p=0.033) protein expression were associated with a higher risk of lymphatic spread, and loss of chromosome 3p (p<0.0001) with a lower risk.
Read moreSmoking negatively impacts renal cell carcinoma overall and cancer-specific survival. (opens in a new tab)
In patients with RCC, a history of smoking was associated with worse pathologic features and survival outcomes and with an increased risk of having mutated p53. Further investigation of the genetic and molecular mechanisms associated with decreased CSS in patients with RCC who have a history of smoking is indicated.
Read moreCarbonic anhydrase IX in bladder cancer: a diagnostic, prognostic, and therapeutic molecular marker. (opens in a new tab)
CAIX was expressed differentially in noninvasive versus invasive tumors, in low-grade versus high-grade bladder cancer, and in primary tumors versus metastases. The current results indicated that CAIX is a strong predictor of recurrence, progression, and overall survival of patients with bladder cancer; and the integration of CAIX expression into conventional prognostic models significantly improved their predictive accuracy. The data suggest a tripartite role of CAIX as a diagnostic, prognostic, and therapeutic molecular marker in bladder cancer.
Read more