Review
Medicine, General & Internal
Neha Jain, Upendra Nagaich, Manisha Pandey, Dinesh Kumar Chellappan, Kamal Dua
Summary: In the era of medical revolution, genomic testing plays a crucial role in the development of predictive, preventive, and personalized medicine. Techniques such as predictive screening and facial analysis can enhance diagnostic accuracy and treatment effectiveness. The application of genomic technologies, artificial intelligence, and machine learning is essential for the advancement of preventive and personalized medicine. Biomarkers are important for diagnosis, prognosis, and selection of personalized medicine. This review highlights the applicability and challenges of predictive diagnostics in the development of personalized medicine.
Article
Computer Science, Interdisciplinary Applications
Jared D. Huling, Menggang Yu
Summary: Numerous statistical methods have been proposed for subgroup identification, but many lack corresponding R packages. Recently, a unified software framework has been developed to provide estimation and evaluation of treatment effects within subgroups. This personalized package offers a variety of methods, including flexible machine learning tools, and allows for efficiency improvements in estimation.
JOURNAL OF STATISTICAL SOFTWARE
(2021)
Article
Health Care Sciences & Services
Umile Giuseppe Longo, Arianna Carnevale, Carlo Massaroni, Daniela Lo Presti, Alessandra Berton, Vincenzo Candela, Emiliano Schena, Vincenzo Denaro
Summary: The study proposes a personalized, predictive, participatory, precision, and preventive (P5) medicine model, highlighting the potential for tailoring diagnosis and therapy for patients with RC diseases by defining genetic predispositions and assessing lifestyle and environmental factors.
JOURNAL OF PERSONALIZED MEDICINE
(2021)
Article
Oncology
Sylvie Rodrigues-Ferreira, Clara Nahmias
Summary: Breast cancer is a common malignancy among women worldwide, and different treatment options have significantly improved patient outcomes. However, further research on predictive biomarkers is needed to combat treatment resistance and identify the most suitable patients for personalized treatment, aiming to avoid unnecessary overtreatment.
Review
Genetics & Heredity
Sarah G. Ayton, Martina Pavlicova, Carla Daniela Robles-Espinoza, Jose G. Tamez Pena, Victor Trevino
Summary: This study systematically assessed the ability of multiomics cancer subtyping methods to capture cancer prognosis and found that latent-variable subtyping methods better identify clinically prognostic cancer subtypes.
GENETICS IN MEDICINE
(2022)
Review
Oncology
Miao Su, Zhe Zhang, Li Zhou, Chao Han, Canhua Huang, Edouard C. Nice
Summary: Cancer is a major global public health issue and personalized/precision medicine plays a crucial role in its diagnosis and treatment. Research focuses on cancer biomarkers, the impact of the microbiome, the application of emerging omics technologies, and future prospects.
Article
Medicine, General & Internal
Si Chen, Nan Wang, Siqi Xiong, Xiaobo Xia
Summary: This bioinformatic study identifies NFKB1, IL18, KITLG, TLR9, FKBP2, and HDAC4 as key genes for POAG and GM regulation. Immune response modulated by macrophages plays an important role in POAG and may be potential targets for future predictive, preventive, and personalized diagnosis and treatment.
Article
Health Care Sciences & Services
Grace S. Shieh
Summary: This article reviews the applications of the synthetic lethal (SL) concept in translational cancer medicine over the past five years. It discusses the use of SL concept in drug combinations to overcome tumor resistance, the identification of prognostic and predictive biomarkers using synthetic lethality, the application of SL interactions in stratifying patients for targeted and immunotherapy, as well as the challenges and future directions in this field.
JOURNAL OF PERSONALIZED MEDICINE
(2022)
Article
Medicine, General & Internal
Xiaoni Meng, Biyan Wang, Xizhu Xu, Manshu Song, Haifeng Hou, Wei Wang, Youxin Wang
Summary: This study found that suboptimal health status (SHS) is associated with changes in immunoglobulin G (IgG) N-glycan profiles. IgG N-glycan profiles may serve as potential biomarkers for early detection of SHS. The logistic regression model including IgG N-glycans showed moderate performance in predicting SHS, providing an opportunity for SHS management and advanced treatment of non-communicable diseases (NCDs).
Review
Biochemistry & Molecular Biology
Manish Pratap Singh, Sandhya Rai, Ashutosh Pandey, Nand K. Singh, Sameer Srivastava
Summary: Molecular subtypes-based therapies provide a new potential framework for precise outcomes in clinical settings, as colorectal cancer is a highly heterogeneous malignancy with different pathological and genetic signatures in each subtype. Designing therapeutic stratification based on these features may lead to improved treatment outcomes.
Review
Health Care Sciences & Services
Francesca Cianci, Ivan Verduci
Summary: The loss of function and aberrant expression of ion channels and transporters, particularly the CLIC1 protein, play a crucial role in the development of diseases such as cancer and neurodegenerative diseases.
JOURNAL OF PERSONALIZED MEDICINE
(2021)
Review
Pharmacology & Pharmacy
Katja Goricar, Vita Dolzan, Metka Lenassi
Summary: EVs have emerged as a potential source of biomarkers in liquid biopsy, offering valuable insights for personalized cancer treatment in oncology.
FRONTIERS IN PHARMACOLOGY
(2021)
Article
Biochemistry & Molecular Biology
Bo Chen, Kang Xie, Jianzhong Zhang, Liting Yang, Hongshu Zhou, Liyang Zhang, Renjun Peng
Summary: Mitochondrial dysfunction and necroptosis play important roles in multiple cardiovascular diseases, and their implications in intracranial aneurysms (IAs) are unclear. This study aimed to explore their value in predictive, preventive, and personalized medicine for IAs. Key genes involved in mitochondrial dysfunction and necroptosis were identified, and their diagnostic value for IA was confirmed using machine learning. Single-cell sequencing analysis showed that mitochondrial dysfunction and necroptosis were up-regulated in monocytes/macrophages and vascular smooth muscle cells within IA lesions. Mitochondria-induced necroptosis may be a novel potential target for the diagnosis, prevention, and treatment of IA.
Review
Gastroenterology & Hepatology
Stephen L. Chan, Nathalie Wong, W. K. Jacky Lam, Ming Kuang
Summary: Advances in systemic treatment for hepatocellular carcinoma (HCC) have introduced more options but personalized treatment approaches are crucial due to differing patient responses. Biomarker data and tumor heterogeneity analysis have proven to be effective methods in guiding treatment decisions for HCC patients.
JOURNAL OF GASTROENTEROLOGY AND HEPATOLOGY
(2022)
Review
Biochemistry & Molecular Biology
Juntuo Zhou, Lijun Zhong
Summary: Metabolomics is a rapidly developing technique with promising applications in disease prediction and personalized medicine. Liquid chromatography-mass spectrometry is the most widely used analytical strategy, and recent studies have demonstrated its novel applications in predictive and personalized medicine, such as early diagnosis and prognostic evaluation. The application of metabolomics in COVID-19 research is also summarized.
FRONTIERS IN MOLECULAR BIOSCIENCES
(2022)
Article
Biology
Layla Parast, Tianxi Cai, Lu Tian
Summary: In studies requiring long-term and/or costly follow-up, surrogate markers are often used to evaluate the treatment effect. However, existing statistical methods often fail to account for the potential variation in the utility or strength of surrogate markers based on patient characteristics. This paper proposes an approach to measure surrogate strength as a function of baseline covariate and examines potential heterogeneity in surrogate marker utility.
Article
Biology
Jue Hou, Stephanie F. Chan, Xuan Wang, Tianxi Cai
Summary: This paper aims to develop an efficient risk prediction model for disease onset time by leveraging a small number of labels on the current status and a large number of unlabeled observations on imperfect proxies. By combining information from proxies and limited labels, the proposed semisupervised risk prediction method improves the accuracy of estimations. Simulation studies and a real case analysis demonstrate the good performance of the method in a finite sample.
Article
Mathematical & Computational Biology
Yuri Ahuja, Liang Liang, Doudou Zhou, Sicong Huang, Tianxi Cai
Summary: Leveraging large-scale electronic health record (EHR) data to estimate survival curves for clinical events is crucial for risk estimation and comparative effectiveness research. However, the lack of direct event time observations hinders the use of EHR data. In this article, the authors propose SCORNET, a method that utilizes current status labels and informative features to achieve consistent and efficient survival function estimation.
Article
Rheumatology
Sicong Huang, Tianrun Cai, Brittany N. Weber, Zeling He, Kumar P. Dahal, Chuan Hong, Jue Hou, Thany Seyok, Andrew Cagan, Marcelo F. DiCarli, Jacob Joseph, Seoyoung C. Kim, Daniel H. Solomon, Tianxi Cai, Katherine P. Liao
Summary: This study found that elevated inflammation early in the diagnosis of rheumatoid arthritis (RA) was associated with heart failure, specifically heart failure with preserved ejection fraction (HFpEF). The association was not observed with heart failure with reduced ejection fraction (HFrEF). This finding suggests a window of opportunity for prevention of HFpEF in RA patients.
ARTHRITIS CARE & RESEARCH
(2023)
Article
Biology
Xuan Wang, Layla Parast, Larry Han, Lu Tian, Tianxi Cai
Summary: Identifying effective surrogate markers is crucial for improving clinical trial efficiency. Replacing long-term outcomes with shorter-term and/or cheaper surrogate markers can shorten study duration and reduce costs. However, methods for combining multiple markers to construct a composite marker with improved surrogacy are limited.
Article
Immunology
Ashley Galloway, Yojin Park, Vidisha Tanukonda, Yuk-Lam Ho, Xuan-Mai T. Nguyen, Monika Maripuri, Andrew T. Dey, Hanna Gerlovin, Daniel Posner, Kristine E. Lynch, Tianxi Cai, Shiuh-Wen Luoh, Stacey Whitbourne, David R. Gagnon, Sumitra Muralidhar, Phillip S. Tsao, Juan P. Casas, J. Michael Gaziano, Peter W. F. Wilson, Adriana M. Hung, Kelly Cho
Summary: This retrospective cohort study examined the association between COVID-19 severity and long-term complications in 94,595 cases. The study found that COVID-19 severity was associated with an increased risk of long-term complications occurring 31-120 days postinfection. Most events occurred within the first 60 days postinfection and decreased after day 91, except for heart failure in severe patients and death in moderate patients, which peaked between days 91-120.
JOURNAL OF INFECTIOUS DISEASES
(2022)
Article
Gastroenterology & Hepatology
William Yuan, Jayson S. Marwaha, Shana T. Rakowsky, Nathan P. Palmer, Isaac S. Kohane, David T. Rubin, Gabriel A. Brat, Joseph D. Feuerstein
Summary: Population-scale analysis reveals patterns in prescribing trends for ulcerative colitis management, including earlier use of biologics, associations of step-up therapy with higher corticosteroid exposure, and association of combination therapy with positive patient outcomes.
INFLAMMATORY BOWEL DISEASES
(2023)
Article
Urology & Nephrology
James A. Diao, Gloria J. Wu, Jason K. Wang, Isaac S. Kohane, Herman A. Taylor, Hocine Tighiouart, Andrew S. Levey, Lesley A. Inker, Neil R. Powe, Arjun K. Manrai
Summary: The newly recommended 2021 CKD-EPI creatinine-based eGFR equation may result in substantial changes to recommended care for US patients of all racial and ethnic groups.
JOURNAL OF THE AMERICAN SOCIETY OF NEPHROLOGY
(2023)
Article
Statistics & Probability
Xinzhou Guo, Waverly Wei, Molei Liu, Tianxi Cai, Chong Wu, Jingshen Wang
Summary: In this study, a new data analysis pipeline and statistical methodology were introduced to address the limitations of previous research on the association between statin usage and the risk of developing type II diabetes. The study found that females with high genetic risk for diabetes are the most vulnerable subgroup for developing diabetes after taking statins.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Article
Health Care Sciences & Services
Roland A. Matsouaka, Yi Liu, Yunji Zhou
Summary: In this article, we propose methods to calculate the variance of the normalized, doubly robust average treatment effect of the treated and average treatment effect on the controls estimators and investigate their finite sample properties. We consider three sources of uncertainty when evaluating these estimators and their variances, and conduct an extensive simulation study to explore different methods.
STATISTICAL METHODS IN MEDICAL RESEARCH
(2023)
Article
Mathematical & Computational Biology
Layla Parast, Tianxi Cai, Lu Tian
Summary: The primary benefit of identifying a valid surrogate marker is the ability to use it in a future trial to test for a treatment effect with shorter follow-up time or less cost. However, previous work has shown potential heterogeneity in the utility of surrogate markers. Therefore, ignoring this heterogeneity may lead to inaccurate conclusions about the treatment effect. This article introduces a novel test that accounts for the heterogeneity in surrogate marker utility.
STATISTICS IN MEDICINE
(2023)
Article
Medicine, General & Internal
Byorn W. L. Tan, Bryce W. Q. Tan, Amelia L. M. Tan, Emily R. Schriver, Alba Gutierrez-Sacristan, Priyam Das, William Yuan, Meghan R. Hutch, Noelia Garcia Barrio, Miguel Pedrera Jimenez, Noor Abu-el-rub, Michele Morris, Bertrand Moal, Guillaume Verdy, Kelly Cho, Yuk-Lam Ho, Lav P. Patel, Arianna Dagliati, Antoine Neuraz, Jeffrey G. Klann, Andrew M. South, Shyam Visweswaran, David A. Hanauer, Sarah E. Maidlow, Mei Liu, Danielle L. Mowery, Ashley Batugo, Adeline Makoudjou, Patric Tippmann, Daniela Zoeller, Gabriel A. Brat, Yuan Luo, Paul Avillach, Riccardo Bellazzi, Luca Chiovato, Alberto Malovini, Valentina Tibollo, Malarkodi Jebathilagam Samayamuthu, Pablo Serrano Balazote, Zongqi Xia, Ne Hooi Will Loh, Lorenzo Chiudinelli, Clara-Lea Bonzel, Chuan Hong, Harrison G. Zhang, Griffin M. Weber, Isaac S. Kohane, Tianxi Cai, Gilbert S. Omenn, John H. Holmes, Kee Yuan Ngiam
Summary: This retrospective observational study analyzed data from 12,891 COVID-19 patients and found that age, severe COVID-19, severe acute kidney injury (AKI), and ischemic heart disease were associated with worse mortality outcomes. The severity of AKI was associated with poorer kidney function recovery, while the use of remdesivir was associated with better recovery. In patients without chronic kidney disease, age, male sex, severe AKI, and hypertension were associated with post-AKI kidney function impairment. COVID-19-associated AKI was also linked to higher mortality and worse long-term kidney function recovery.
Article
Computer Science, Information Systems
Jacqueline Honerlaw, Yuk-Lam Ho, Francesca Fontin, Jeffrey Gosian, Monika Maripuri, Michael Murray, Rahul Sangar, Ashley Galloway, Andrew J. Zimolzak, Stacey B. Whitbourne, Juan P. Casas, Rachel B. Ramoni, David R. Gagnon, Tianxi Cai, Katherine P. Liao, J. Michael Gaziano, Sumitra Muralidhar, Kelly Cho
Summary: The development of phenotypes using electronic health records is a resource-intensive process. The Department of Veterans Affairs (VA) has developed a standard for phenotype metadata collection, called CIPHER, which improves upon existing methods by capturing algorithm context and validation approach. This standard is applicable across healthcare systems and is currently used in the largest healthcare system in the United States.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Psychiatry
Yi-han Sheu, Jiehuan Sun, Hyunjoon Lee, Victor M. Castro, Yuval Barak-Corren, Eugene Song, Emily M. Madsen, William J. Gordon, Isaac S. Kohane, Susanne E. Churchill, Ben Y. Reis, Tianxi Cai, Jordan W. Smoller
Summary: This study finds that applying machine learning models to clinical data can outperform clinicians in suicide risk stratification. The researchers use a landmark model framework, aligned with clinical practice, and a large electronic health record database to predict suicide-related behaviors. The models developed using this approach achieve high predictive performance across different prediction windows and settings.
PSYCHIATRY RESEARCH
(2023)
Article
Biochemical Research Methods
Jun Wen, Xiang Zhang, Everett Rush, Vidul A. Panickan, Xingyu Li, Tianrun Cai, Doudou Zhou, Yuk-Lam Ho, Lauren Costa, Edmon Begoli, Chuan Hong, J. Michael Gaziano, Kelly Cho, Junwei Lu, Katherine P. Liao, Marinka Zitnik, Tianxi Cai
Summary: Predicting molecule-disease indications and side effects can be enhanced by comprehensively mining semantic dependencies between molecules, diseases, and diseases. This study introduces a Multi-Modal REpresentation Mapping Approach to Predicting molecular-disease relations (M2REMAP), which incorporates clinical semantics from electronic health records (EHR) of 12.6 million patients. M2REMAP achieves significant improvements in prediction performance on indications and side effects by combining multimodal molecule representation and disease semantic embedding.