Raviv Pryluk’s Post

Very proud to share some of the work we've been busy with, led by Oshri Machluf, Tzviel Frostig, Tomer Milo, Gal Shoham Levin and Raviv Pryluk. Our preprint on Robust CATE Estimation Using Novel Ensemble Methods is now available on arXiv! Causal Machine Learning for CATE estimation is a new and promising approach for identifying heterogeneous treatment effects and promising subgroups in clinical trials. Originally developed in the econometrics domain, this multi-disciplinary approach combines Causal Inference, Machine Learning, and Statistics. In this paper, we compare existing methods across a wide range of scenarios specifically tuned for clinical trials and propose two novel methods with improved robustness. As part of this work, we developed a unique Causal Biological Model of the PD-L1 pathway, trained on real clinical data, and used it to create more realistic data for our benchmarking. This is yet another example of how integrating biology and machine learning can drive innovation. Excited to see the impact this will have on the future of clinical trials!

We're thrilled to share our latest research on Conditional Average Treatment Effects (CATE) estimation in clinical trials. CATE estimation provides crucial value of understanding treatment effect heterogeneity, which is inherent to designing patient-centered trials. Our extensive simulation study demonstrates the limitation of traditional methods and introduces novel ensemble methods, which present better results across diverse use cases, varying in complexity, sample size, and underlying mechanisms. These robust methods reveal the factors driving heterogeneity in the presence of multiple covariates and complex biological signals, across therapeutic areas and trial types. Understanding the factors driving the heterogeneity enables us to predict treatment effects and probability of success and extract actionable, clinically significant, and business-meaningful insights to maximize the success of the next trial. Oshri Machluf Raviv Pryluk Elad Berkman Tzviel Frostig Gal Shoham Levin Tomer Milo

𝗧𝗵𝗲 𝗴𝘂𝘆𝘀 𝗵𝗮𝘃𝗲 𝗴𝗮𝘁𝗵𝗲𝗿𝗲𝗱 𝘀𝗼𝗺𝗲 𝗽𝗿𝗶𝗻𝗰𝗶𝗽𝗹𝗲𝘀 𝗮𝗻𝗱 𝗯𝘂𝗶𝗹𝘁 𝗮 𝗹𝗼𝗴𝗶𝗰 𝗳𝗼𝗿 𝘀𝗲𝗹𝗲𝗰𝘁𝗶𝗻𝗴 𝗯𝗶𝗼𝗺𝗮𝗿𝗸𝗲𝗿𝘀 𝗳𝗼𝗿 𝗮 𝗰𝗲𝗿𝘁𝗮𝗶𝗻 𝗰𝗮𝗻𝗰𝗲𝗿 𝗻𝗼𝘀𝗼𝗹𝗼𝗴𝘆. If this logic is the basis for LIMS, it could be a compromise between two aspirations: to make medicine a craft with clear rules and the reluctance of the physician, with his art and experience, to follow outside laboratory guidelines. Creating shared rules for test prescribing can help in many controversial cases - https://lnkd.in/g-tWxcdF #ADLM #CanserNosology #LIMS #VivicaLIMS #LifeDataLab https://about.vivica.us https://lifedl.net

The energy distance or the potential energy of data is a statistical concept promoted by the great mathematician-statistician Gábor J. Székely with a great impact on the field of statistics. In our latest research we extend this statistical notion to the case of survival analysis and show that new hypothesis tests and goodness-of-fit methods are very effective for comparing different immunotherapy treatments in the field of oncology compared to other classical tests such as the log-rank test. Energy Distance and Kernel Mean Embeddings for Two-Sample Survival Testing with Applications in Immunotherapy Clinical Trials https://lnkd.in/dzNJRWqi #biostatistics #statisticalscience #survivalanalysis #personalizedmedicine CiTIUS (Centro Singular de Investigación en Tecnoloxías Intelixentes) Universidad de Santiago de Compostela

BioPorto’s novel NGAL (neutrophil gelatinase-associated lipocalin) biomarker, The NGAL Test, is part of a new study funded by the National Institutes of Health. The study, titled "Combining Biomarkers and Electronic Risk Scores to Predict AKI in Hospitalized Patients," is actively recruiting participants and holds promise for significantly advancing the field of Acute Kidney Injury (AKI) diagnostics. The study, accessible at https://lnkd.in/eucCxqFF, aims to validate the use of renal biomarkers in conjunction with an electronic health record AKI risk score to identify individuals at high risk for severe AKI. Principal investigators Jay Koyner and Matthew Churpek anticipate that this study will be a game-changer, potentially setting a new standard in the identification and management of patients at risk for severe AKI. #EHR #biomarkers Association for Diagnostics & Laboratory Medicine International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) University of Chicago University of Wisconsin-Madison BioPorto A/S

Join this CAP complimentary webinar to explore the complexities of biomarker testing in oncology. Discover case studies where different modalities yield conflicting results. We will delve into various testing methods, including immunohistochemistry (IHC), flow cytometry, fluorescence in situ hybridization (FISH), polymerase chain reaction (PCR), and next-generation sequencing (NGS), and their limitations to better address challenges in diagnosis and treatment decisions. #CAP #PrecisionMedicine

Chen, S., Liang, B. & Xu, J. Unveiling heterogeneity in MSCs: exploring marker-based strategies for defining MSC subpopulations. J Transl Med 22, 459 (2024). https://lnkd.in/dB_MJfke "Abstract Mesenchymal stem/stromal cells (MSCs) represent a heterogeneous cell population distributed throughout various tissues, demonstrating remarkable adaptability to microenvironmental cues and holding immense promise for disease treatment. However, the inherent diversity within MSCs often leads to variability in therapeutic outcomes, posing challenges for clinical applications. To address this heterogeneity, purification of MSC subpopulations through marker-based isolation has emerged as a promising approach to ensure consistent therapeutic efficacy. In this review, we discussed the reported markers of MSCs, encompassing those developed through candidate marker strategies and high-throughput approaches, with the aim of explore viable strategies for addressing the heterogeneity of MSCs and illuminate prospective research directions in this field."

Delving into the Future of Medicine: Next-Gen Antibody-Drug Conjugates Exciting strides are being made in the realm of medicine! 🚀 Researchers are exploring the potential of next-generation antibody-drug conjugates, aiming to revolutionize treatment modalities and enhance patient outcomes. Dive into the latest findings and discoveries in this groundbreaking field! 💉🔬 #MedicalInnovation #DrugDevelopment #ResearchAdvancements Read more: https://lnkd.in/dGCM_Sv2

Real-world evidence (RWE) can complement data from randomized controlled trials (RCTs) where patients are subject to strict eligibility criteria, and provide outcome data for patients who may be underrepresented in RCTs, for example, black patients, patients ≥75 years, patients with poor performance status and other illnesses that may be underrepresented in clinical trials Watch our peer-reviewed video publication: https://bit.ly/3QpDHTv in the Video Journal of Biomedicine of a Medical Oncologist discussing results from a large real-world retrospective study assessing overall survival in patients with HR+/HER2- metastatic #BreastCancer treated with 1L CDK4/6i + aromatase inhibitor (AI) vs AI alone in the US. Find the original peer reviewed paper here: https://bit.ly/3l9IlcR Find Pfizer Oncology’s Interactive tool to the original papers mentioned in the video here: https://bit.ly/3MtLD4W

Prof. Dr. Hilmar Bading at Heidelberg University has led a breakthrough in ALS research with FP802, a novel 'TwinF interface inhibitor' that targets damaging protein interactions involved in ALS. Demonstrating efficacy in ALS models by preserving motor neurons and enhancing motor functions, FP802 offers new hope for treatment. Developed in collaboration with FundaMental Pharma and supported by leading institutions, this research highlights a significant step towards clinical trials. Dr. Jing Yan's involvement underscores the project's collaborative effort to combat this challenging disease, promising a brighter future for ALS patients. #InnovativeTreatment #NeurodegenerativeDiseases #ALSResearch #FutureOfHealthcare #Healthcare 🔗 For more details, explore the full article: https://lnkd.in/ghBhrVRh