scholarly journals A framework for validating AI in precision medicine: considerations from the European ITFoC consortium

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Rosy Tsopra ◽  
Xose Fernandez ◽  
Claudio Luchinat ◽  
Lilia Alberghina ◽  
Hans Lehrach ◽  
...  
Keyword(s):  
Treatment Response ◽  
Real World ◽  
Clinical Decision Making ◽  
Precision Oncology ◽  
Clinical Validation ◽  
Learning Approaches ◽  
Real World Data ◽  
Privacy And Security ◽  
The Future ◽  
Real World Datasets

Abstract Background Artificial intelligence (AI) has the potential to transform our healthcare systems significantly. New AI technologies based on machine learning approaches should play a key role in clinical decision-making in the future. However, their implementation in health care settings remains limited, mostly due to a lack of robust validation procedures. There is a need to develop reliable assessment frameworks for the clinical validation of AI. We present here an approach for assessing AI for predicting treatment response in triple-negative breast cancer (TNBC), using real-world data and molecular -omics data from clinical data warehouses and biobanks. Methods The European “ITFoC (Information Technology for the Future Of Cancer)” consortium designed a framework for the clinical validation of AI technologies for predicting treatment response in oncology. Results This framework is based on seven key steps specifying: (1) the intended use of AI, (2) the target population, (3) the timing of AI evaluation, (4) the datasets used for evaluation, (5) the procedures used for ensuring data safety (including data quality, privacy and security), (6) the metrics used for measuring performance, and (7) the procedures used to ensure that the AI is explainable. This framework forms the basis of a validation platform that we are building for the “ITFoC Challenge”. This community-wide competition will make it possible to assess and compare AI algorithms for predicting the response to TNBC treatments with external real-world datasets. Conclusions The predictive performance and safety of AI technologies must be assessed in a robust, unbiased and transparent manner before their implementation in healthcare settings. We believe that the consideration of the ITFoC consortium will contribute to the safe transfer and implementation of AI in clinical settings, in the context of precision oncology and personalized care.

scholarly journals Online Multitask Relative Similarity Learning

2017 ◽  
Author(s):  
Shuji Hao ◽  
Peilin Zhao ◽  
Yong Liu ◽  
Steven C. H. Hoi ◽  
Chunyan Miao
Keyword(s):  
Real World ◽  
Learning Algorithm ◽  
Learning Problems ◽  
Similarity Function ◽  
Learning Approaches ◽  
Similarity Learning ◽  
Real World Data ◽  
Real World Datasets ◽  
Online Learning Algorithm ◽  
Relative Similarity

Relative similarity learning~(RSL) aims to learn similarity functions from data with relative constraints. Most previous algorithms developed for RSL are batch-based learning approaches which suffer from poor scalability when dealing with real-world data arriving sequentially. These methods are often designed to learn a single similarity function for a specific task. Therefore, they may be sub-optimal to solve multiple task learning problems. To overcome these limitations, we propose a scalable RSL framework named OMTRSL (Online Multi-Task Relative Similarity Learning). Specifically, we first develop a simple yet effective online learning algorithm for multi-task relative similarity learning. Then, we also propose an active learning algorithm to save the labeling cost. The proposed algorithms not only enjoy theoretical guarantee, but also show high efficacy and efficiency in extensive experiments on real-world datasets.

scholarly journals PO2RDF: Representation of Real-world Data for Precision Oncology Using Resource Description Framework

2021 ◽  
Author(s):  
Yiqing ZHAO ◽  
Anastasios Dimou ◽  
Feichen Shen ◽  
Nansu Zong ◽  
Jaime I. Davila ◽  
...  
Keyword(s):  
Real World ◽  
Targeted Therapies ◽  
Resource Description Framework ◽  
Clinical Decision Making ◽  
Egfr Mutations ◽  
Precision Oncology ◽  
Real World Data ◽  
World Data ◽  
Description Framework ◽  
Resource Description

Abstract Background: Next-generation sequencing provides comprehensive information about individuals’ genetic makeup and is commonplace in precision oncology practice. Due to the heterogeneity of individual patient’s disease conditions and treatment journeys, not all targeted therapies were initiated despite actionable mutations. To better understand and support the clinical decision-making process in precision oncology, there is a need to examine real-world associations of patients’ genetic information and treatment choice.Methods: To fill the gap of insufficient use of real-world data (RWD) in electronic health records (EHRs), we generated a single Resource Description Framework (RDF) resource, called PO2RDF (precision oncology to RDF) by integrating information regarding gene, variant, disease, and drug from genetic reports and EHRs. Results: There are total 2,309,014 triples contained in the PO2RDF. Among them 32,815 triples are related to Gene, 34,695 triples are related to Variant, 8,787 triples are related to Disease, 26,154 triples are related to Drug. We performed one use case analysis to demonstrate the usability of the PO2RDF: we examined real-world associations between EGFR mutations and targeted therapies to confirm existing knowledge and detect off-label use. Conclusions: In conclusion, our work proposed to use RDF to organize and distribute clinical RWD that is otherwise inaccessible externally. Our work serves as a pilot study that will lead to new clinical applications and could ultimately stimulate progress in the field of precision oncology.

The master observational trial–a novel method to unify precision oncology data collection.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e19313-e19313
Author(s):  
Dane J. Dickson ◽  
Jennifer Maria Johnson ◽  
Raymond C. Bergan ◽  
Rebecca Owens ◽  
Vivek Subbiah ◽  
...  
Keyword(s):  
Real World ◽  
Unmet Needs ◽  
Data Availability ◽  
Precision Oncology ◽  
Real World Data ◽  
Observational Trial ◽  
World Data ◽  
Patient Consent ◽  
National Implementation ◽  
Real World Datasets

e19313 Background: The Master Observational Trial (MOT) was recently created as a new master protocol that hybridizes the power of master interventional trials with the richness of real-world data (Cell, 2020). The MOT can be described as a series of prospective observational studies that are tied together through a common protocol, infrastructure, and organization. The MOT has broad application in many disease states but is particularly powerful in oncology. We herein expand our prior work to describe key details regarding how the MOT concept can fill multiple unmet needs in oncology. Methods: Through published information, white papers, and expert opinions we identified key unmet needs of oncology stakeholders. We reviewed the publicly available information of structure, organization, and data availability of the five largest genomic-outcome real-world data efforts. Common concerns included variability and reliability of biomarkers, the scientific rigor in real-world data, data silos, patient consent, and duplicated or disparate activities. We then determined how a specific application of the MOT in oncology could answer stakeholder concerns, integrate with current efforts, and also how to provide a model that would be equally valuable to academic and community clinics. Results: We identified significant scientific challenges with many of the current oncology real-world datasets in answering key concerns of stakeholders. We developed the Master Registry of Oncology Outcomes Associated with Testing and Treatment (ROOT) as the first national implementation of an oncology-centric MOT. We modeled how ROOT could fill scientific gaps in current data efforts and integrate with interventional and real-world efforts and help answer key concerns of stakeholders. We also identified solutions that would allow community and academic groups to participate in the same effort. Conclusions: An oncology-centric MOT has the potential to improve the quality of RWD in oncology and advance precision oncology in ways that are not fully addressed by current retrospective efforts. Reference Dickson DJ, Johnson J, Owens R, Bergan R, Subbiah V, Kurzrock R. (2020). The Master Observational Trial: A New Class of Master Protocol to Advance Precision Medicine. Cell 180, 9-14. Clinical trial information: NCT04028479 .

scholarly journals POS0461 RESPONSE TO BIOLOGIC THERAPY IN RHEUMATOID ARTHRITIS: RETHINKING OUR CLASSIFICATION

2021 ◽  
Vol 80 (Suppl 1) ◽  
pp. 462.1-462
Author(s):  
E. Vallejo-Yagüe ◽  
S. Kandhasamy ◽  
E. Keystone ◽  
A. Finckh ◽  
R. Micheroli ◽  
...  
Keyword(s):  
Treatment Response ◽  
Real World ◽  
Observational Studies ◽  
Initial Response ◽  
Primary Response ◽  
Sustained Response ◽  
Secondary Response ◽  
Real World Data ◽  
World Data ◽  
Secondary Failure

Background:In rheumatoid arthritis (RA), primary failure with biologic treatment may be understood as lack of initial clinical response, while secondary failure would be loss of effectiveness after an initial response. Despite these clinical concepts, there is no unifying operational definition of primary and secondary non-response to RA treatment in observational studies using real-world data. On top of data-driven challenges, when conceptualizing secondary non-responders, it is unclear if the mechanism behind loss of effectiveness after a brief initial response is similar to loss of effectiveness after previous benefit sustained over time.Objectives:This viewpoint aims to motivate discussion on how to define primary and secondary non-response in observational studies. Ultimately, we aim to trigger expert committees to develop standard terminology for these concepts.Methods:We discuss different methodologies for defining primary and secondary non-response in observational studies. To do so, we shortly overview challenges characteristic of performing observational studies in real-world data, and subsequently, we conceptualize whether treatment response should be a dichotomous classification (Primary response/non-response; Secondary response/non-response), or whether one should consider three response categories (Primary response/non-response; Primary sustained/non-sustained response; Secondary response/non-response).Results:RA or biologic registries are a common data source for studying treatment response in real-world data. While registries include disease-specific variables to assess disease progression, missing data, loss of follow-up, and visits restricted to the year or mid-year visit may present a challenge. We believe there is a general agreement to assess primary response within the first 6 month of treatment. However, conceptualizing secondary non-response, one could wonder if a patient with brief initial response and immediate loss of it should belong to the same response category as a patient who relapses after a period of prior benefit that was sustained over time. Until this concern is clarified, we recommend considering a period of sustained response as a pre-requisite for secondary failure. This would result in the following three categories: a) Primary non-response: Lack of response within the first 6 months of treatment; b) Primary sustained response: Maintenance of a positive effectiveness outcome for at least the first 12 months since treatment start; c) Secondary non-response: Loss of effectiveness after achieved primary sustained response. Figure 1 illustrates this classification through a decision tree. Since the underlying mechanisms for treatment failure may differ among the above-mentioned categories, we recommend to use the three-category classification. However, since this may pose additional methodological challenges in real-world data, optionally, a dichotomous 12-month time-point may be used to assess secondary non-response (unfavourable outcome after 12-months) in comparison to primary non-response or non-sustained response (unfavourable outcome within the first 12-months). Similarly, to study primary response, the solely 6-month timepoint may be used.Conclusion:A unified operational definition of treatment response will minimize heterogeneity among observational studies and help improve the ability to draw cross-study comparisons, which we believe would be of particular interest when identifying predictors of treatment failure. Thus, we hope to open the room for discussion and encourage expert committees to work towards a common approach to assess treatment primary and secondary non-response in RA in observational studies.Disclosure of Interests:Enriqueta Vallejo-Yagüe: None declared, Sreemanjari Kandhasamy: None declared, Edward Keystone Speakers bureau: Amgen, AbbVie, F. Hoffmann-La Roche Inc., Janssen Inc., Merck, Novartis, Pfizer Pharmaceuticals, Sanofi Genzyme, Consultant of: AbbVie, Amgen, Bristol-Myers Squibb Company, Celltrion, Myriad Autoimmune, F. Hoffmann-La Roche Inc, Gilead, Janssen Inc, Lilly Pharmaceuticals, Merck, Pfizer Pharmaceuticals, Sandoz, Sanofi-Genzyme, Samsung Bioepsis, Grant/research support from: Amgen, Merck, Pfizer Pharmaceuticals, PuraPharm, Axel Finckh Speakers bureau: Pfizer, Eli-Lilly, Paid instructor for: Pfizer, Eli-Lilly, Consultant of: AbbVie, AB2Bio, BMS, Gilead, Pfizer, Viatris, Grant/research support from: Pfizer, BMS, Novartis, Raphael Micheroli Consultant of: Gilead, Eli-Lilly, Pfizer and Abbvie, Andrea Michelle Burden: None declared

scholarly journals In-depth phenotyping for clinical stratification of Gaucher disease

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Simona D’Amore ◽  
Kathleen Page ◽  
Aimée Donald ◽  
Khadijeh Taiyari ◽  
Brian Tom ◽  
...  
Keyword(s):  
Real World ◽  
Gaucher Disease ◽  
Orthopaedic Surgery ◽  
Clinical Decision Making ◽  
Burden Of Illness ◽  
Fragility Fractures ◽  
Neurological Manifestations ◽  
Real World Data ◽  
World Data ◽  
Advanced Therapies

Abstract Background The Gaucher Investigative Therapy Evaluation is a national clinical cohort of 250 patients aged 5–87 years with Gaucher disease in the United Kingdom—an ultra-rare genetic disorder. To inform clinical decision-making and improve pathophysiological understanding, we characterized the course of Gaucher disease and explored the influence of costly innovative medication and other interventions. Retrospective and prospective clinical, laboratory and radiological information including molecular analysis of the GBA1 gene and comprising > 2500 variables were collected systematically into a relational database with banking of collated biological samples in a central bioresource. Data for deep phenotyping and life-quality evaluation, including skeletal, visceral, haematological and neurological manifestations were recorded for a median of 17.3 years; the skeletal and neurological manifestations are the main focus of this study. Results At baseline, 223 of the 250 patients were classified as type 1 Gaucher disease. Skeletal manifestations occurred in most patients in the cohort (131 of 201 specifically reported bone pain). Symptomatic osteonecrosis and fragility fractures occurred respectively in 76 and 37 of all 250 patients and the first osseous events occurred significantly earlier in those with neuronopathic disease. Intensive phenotyping in a subgroup of 40 patients originally considered to have only systemic features, revealed neurological involvement in 18: two had Parkinson disease and 16 had clinical signs compatible with neuronopathic Gaucher disease—indicating a greater than expected prevalence of neurological features. Analysis of longitudinal real-world data enabled Gaucher disease to be stratified with respect to advanced therapies and splenectomy. Splenectomy was associated with an increased hazard of fragility fractures, in addition to osteonecrosis and orthopaedic surgery; there were marked gender differences in fracture risk over time since splenectomy. Skeletal disease was a heavy burden of illness, especially where access to specific therapy was delayed and in patients requiring orthopaedic surgery. Conclusion Gaucher disease has been explored using real-world data obtained in an era of therapeutic transformation. Introduction of advanced therapies and repeated longitudinal measures enabled this heterogeneous condition to be stratified into obvious clinical endotypes. The study reveals diverse and changing phenotypic manifestations with systemic, skeletal and neurological disease as inter-related sources of disability.

scholarly journals Ondansetron use is associated with lower COVID-19 mortality in a Real-World Data network-based analysis

2021 ◽  
Author(s):  
Gregory M Miller ◽  
Austin J Ellis ◽  
Rangaprasad Sarangarajan ◽  
Amay Parikh ◽  
Leonardo O Rodrigues ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Real World ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
Long Term Effects ◽  
Real World Data ◽  
World Data ◽  
Antiemetic Agent ◽  
Future Loss ◽  
Selection Operator

Objective: The COVID-19 pandemic generated a massive amount of clinical data, which potentially holds yet undiscovered answers related to COVID-19 morbidity, mortality, long term effects, and therapeutic solutions. The objective of this study was to generate insights on COVID-19 mortality-associated factors and identify potential new therapeutic options for COVID-19 patients by employing artificial intelligence analytics on real-world data. Materials and Methods: A Bayesian statistics-based artificial intelligence data analytics tool (bAIcis®) within Interrogative Biology® platform was used for network learning, inference causality and hypothesis generation to analyze 16,277 PCR positive patients from a database of 279,281 inpatients and outpatients tested for SARS-CoV-2 infection by antigen, antibody, or PCR methods during the first pandemic year in Central Florida. This approach generated causal networks that enabled unbiased identification of significant predictors of mortality for specific COVID-19 patient populations. These findings were validated by logistic regression, regression by least absolute shrinkage and selection operator, and bootstrapping. Results: We found that in the SARS-CoV-2 PCR positive patient cohort, early use of the antiemetic agent ondansetron was associated with increased survival in mechanically ventilated patients. Conclusions: The results demonstrate how real world COVID-19 focused data analysis using artificial intelligence can generate valid insights that could possibly support clinical decision-making and minimize the future loss of lives and resources.

Analyzing treatment patterns and time to the next treatment in chronic lymphocytic leukemia real-world data using automated temporal phenotyping.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. e19512-e19512
Author(s):  
Kyeryoung Lee ◽  
Zongzhi Liu ◽  
Meng Ma ◽  
Yun Mai ◽  
Christopher Gilman ◽  
...  
Keyword(s):  
Chronic Lymphocytic Leukemia ◽  
Real World ◽  
Clinical Decision Making ◽  
Single Agent ◽  
Treatment Patterns ◽  
Lymphocytic Leukemia ◽  
Time Interval ◽  
Real World Data ◽  
World Data ◽  
Therapeutic Class

e19512 Background: Targeted therapy is an important treatment for chronic lymphocytic leukemia (CLL). However, optimal strategies for deploying small molecule inhibitors or antibody therapies in the real world are not well understood, largely due to a lack of outcomes data. We implemented a novel temporal phenotyping algorithm pipeline to derive lines of therapy (LOT) and disease progression in CLL patients. Here, the CLL treatment pattern and time to the next treatment (TTNT) were analyzed in real-world data (RWD) using patient electronic health records. Methods: We identified a CLL cohort with LOT from the Mount Sinai Data Warehouse (2003-2020). Each LOT consisted of either a single agent or combinations defined by NCCN CLL guidelines. We developed a natural language processing (NLP)-based temporal phenotyping approach to automatically identify the number of lines and therapeutic regimens. The sequence of treatment and time interval for each patient were derived from the systematic treatment data. Time to event analysis and multivariate (i.e., age, gender, race, other treatment patterns) Cox proportional hazard (CoxPH) models were used to analyze the patterns and predictors of TTNT. Results: Four hundred eleven CLL patients received 1 to 7 LOTs. Ibrutinib was the predominant 1st LOT (40.8% of patients) followed by anti-CD20-based antibody therapies and chemotherapy in 30.6 and 19.2% of patients, respectively, followed by Acalabrutinib, Venetoclax, and Idelalisib in 3.4, 2.7, and 0.7% of patients, respectively (Table 1). The 2nd to 5th LOT showed the same or similar trends. We next analyzed the TTNT in the 1st line of each therapeutic class. Acalabrutinib resulted in a longer median TTNT than Ibrutinib. Both Acalabrutinib and Ibrutinib showed longer TTNT compared to Venetoclax (median TTNTs were 742 and 598 vs. 373 days: HR = 0.23, p=0.015 and HR = 0.48, p=0.03, respectively). In addition, patients with age equal to or older than 65 showed longer TNNT (HR=0.16, p=0.016). Conclusions: Our result shows the potential of RWD usage in clinical decision making as real-world evidence reported here is consistent with results derived from clinical trial data. Linking this study to genetic data and other covariates affecting treatment outcomes may provide additional insights into the optimal sequences of the targeted therapies in CLL. Table 1: Therapeutic class and patient numbers (%) in each line.[Table: see text]

The future of research in hematology: Integration of conventional studies with real-world data and artificial intelligence

Blood Reviews ◽  
2021 ◽  
pp. 100914
Author(s):  
Francesco Passamonti ◽  
Giovanni Corrao ◽  
Gastone Castellani ◽  
Barbara Mora ◽  
Giulia Maggioni ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Real World ◽  
Real World Data ◽  
World Data ◽  
The Future
Sign in / Sign up

Export Citation Format

Share Document