About

Vivek Charu is an Assistant Professor of Pathology and of Medicine (Quantitative Sciences) at Stanford University. He is a physician and a biostatistician with clinical expertise in the diagnosis of non-neoplastic kidney and liver disease, including transplantation. His research interests focus on the design of observational studies and clinical trials, the analysis of observational data, and causal inference. He completed his medical education at Johns Hopkins University School of Medicine in 2017, earning an MD and PhD in Biostatistics from Johns Hopkins University Bloomberg School of Public Health in 2017. He is board certified by the American Board of Pathology in Anatomic Pathology (2020) and completed his pathology residency at Stanford University in 2020.

Research topics

• Computer Science
• Medicine
• Genetics
• Biology
• Computational biology
• Economics
• Pathology
• Economic growth
• Statistics
• Internal medicine
• Database
• Mathematics
• Cell biology
• Gastroenterology
• Environmental health

Selected publications

• 1445 Kidney Biopsy in Pregnancy Is Safe and Directly Impacts Patient Management

  Laboratory Investigation · 2026-03-01

  article
  PublisherDOI

• Donor-Recipient Age Mismatch and Long-Term Graft Outcomes After Adolescent Liver Transplant

  JAMA Network Open · 2026-01-07

  articleOpen access

  Importance: Donor-recipient age mismatch is an established risk factor in adult liver transplants (LTs), yet its effect in adolescents, who require long-term graft durability, has not been fully characterized. Despite pediatric prioritization, some adolescent donor livers are allocated to adults, limiting access to age-matched grafts for adolescents. Objective: To assess whether a donor-recipient age difference of 10 or more years is associated with inferior graft survival in LTs among adolescents and to estimate the benefits of broader geographic sharing of adolescent donor livers. Design, Setting, and Participants: This retrospective, registry-based case-control study used data from the Organ Procurement & Transplantation Network database, a nationwide US transplant registry. Participants were adolescents aged 12 to 17 years who received liver-only grafts from donation after brain death between March 1, 2002, and December 31, 2024, with follow-up until April 4, 2025. Propensity score matching (1:1) was performed on graft type and size mismatch, donor sex, donor-recipient sex mismatch, transplant center volume, and recipient variables. Exposure: Donor-recipient age difference of 10 or more years (age-mismatched graft) vs less than 10 years (age-matched graft). Main Outcomes and Measures: The primary outcome was 10-year graft survival. The secondary outcome was 10-year overall survival. Waiting time to an age-matched graft offer under alternative donor-sharing radii (1500 nautical miles [NM], 1000 NM, or no limit vs 500 NM) were also estimated. Results: Among 2020 adolescents receiving LTs (median age, 15.0 [IQR, 13.0-16.0] years; 1081 [53.5%] female), 612 (30.3%) received age-mismatched grafts (median donor age, 36.0 [IQR, 29.0-45.0] years) and 1408 (69.7%) received age-matched grafts (median donor age, 16.0 [IQR, 13.0-17.0] years). The age-mismatched group had a higher proportion of recipient candidates in the intensive care unit at transplant (287 [46.9%] vs 250 [17.8%]; P < .001). After propensity score matching (n = 526 per group), 10-year graft survival was 61.5% in the age-mismatched group and 74.2% in the age-matched group (P < .001), with consistent results across recipients' pretransplant hospitalization status. A simulation estimated that expanding the adolescent allocation radius to 1000 NM would allow 90% of adolescent candidates to receive age-matched offers within 15 days, compared with 44 days under the current 500-NM limit. Conclusions and Relevance: In this case-control study of a US national cohort of adolescents receiving LT, donor-recipient age mismatch of 10 or more years was associated with inferior graft survival. Broader allocation of adolescent donors may improve access to age-matched grafts and long-term outcomes.

  PublisherDOI

• Unmasking primary aldosteronism after kidney transplantation: a case series

  Journal of Human Hypertension · 2026-04-25

  article
  PublisherDOI

• Designing genome editing experiments with EditABLE

  Genome biology · 2026-05-11

  articleOpen access

  While many computational tools exist for designing CRISPR-Cas experiments, there is a need for a centralized resource that combines individual tools to predict the most efficient genome editing strategy for a given application. To fill this gap, we develop EditABLE (EditABLE-app.stanford.edu), an online resource that provides optimal CRISPR editors and guide RNAs based on user provided sequence data with functionalities for base editing, prime editing, and integrase-mediated editing. We demonstrate the utility of EditABLE by applying it to one of the most common monogenic disorders, autosomal dominant polycystic kidney disease (ADPKD), identifying specific editing tools across the ADPKD mutation landscape.

  PublisherDOI

• Clinical and histologic predictors of non-diabetic kidney disease in patients with diabetes mellitus

  Kidney International · 2026-04-01

  articleOpen access

  INTRODUCTION: The number of kidney biopsies performed in patients with diabetes has increased rapidly over the last two decades. However, the overall value of the biopsy has been questioned because any coexisting non-diabetic kidney disease (NDKD) is not identified in many patients. Here, we quantify the frequency of identifying NDKD and examine clinical indications, demographic factors, and histologic parameters that increase the odds of finding NDKD and the impact on developing end stage kidney disease. METHODS: A retrospective analysis of clinical and pathologic parameters of 49,075 biopsied patients with diabetes with and without diabetic nephropathy (DN) from 2001-2024 was performed. Data from the United States Renal Data Service were examined to determine the impact of a NDKD on disease progression to end stage kidney disease. RESULTS: NDKD was found in 58.8% of patients with diabetes who underwent kidney biopsy, including 35.9% without concurrent DN and 22.9% as a second diagnosis in DN. Acute kidney injury and acute nephritic syndrome had greater odds of a NDKD diagnosis in patients with DN. The youngest (under 30 years) and oldest (60 years and older) patients had a higher prevalence of NDKD. Higher chronicity on biopsy was associated with a lower prevalence of NDKD diagnosis. Patients with NDKD were 2.56-fold less likely to develop end stage kidney disease compared to patients with DN alone. CONCLUSIONS: This is the largest analysis examining prevalence of a NDKD in patients with diabetes and the impact of biopsy indication on finding a second diagnosis in biopsy-proven DN. The objective is to provide nephrologists with guidance in when to perform a biopsy based on the odds of finding a NDKD related to the patient's clinical indication. Given the high prevalence of NDKD, our study shows that kidney biopsy remains a critical tool in the care of diabetic patients.

  PublisherOA PDFDOI

• Learning Interpretable Point-Based Clinical Risk Scores via Direct Optimization

  arXiv (Cornell University) · 2026-05-18

  preprintOpen access

  Many clinical risk scores are deployed as additive rules with nonnegative integer points assigned to relevant binary predictive features. These integer weights not only make the score easier to use in practice but also promote sparsity in the resulting prediction model. Such risk scores are often derived by first fitting a regression model and then rounding the estimated coefficients to the nearest integer after appropriate scaling. This approach is computationally fast but does not guarantee optimality of the resulting score. Alternatively, one may search over all possible integer weights to directly optimize a value function by posing the problem as an integer programming task. However, the associated computational burden can be substantial, especially when the value function is nonconcave or even discontinuous. In this paper, we develop new machine learning algorithms that employ a flexible greedy optimization strategy to learn such additive scoring directly under explicit and sensible optimality objectives. We apply the proposed method to a large electronic health record (EHR) cohort in Epic Cosmos to construct an integer-weighted comorbidity score for measuring the risk of post-discharge mortality. We also conduct a simulation study to examine the finite-sample operating characteristics.

  PublisherDOI

• Learning Interpretable Point-Based Clinical Risk Scores via Direct Optimization

  ArXiv.org · 2026-05-18

  articleOpen access

  Many clinical risk scores are deployed as additive rules with nonnegative integer points assigned to relevant binary predictive features. These integer weights not only make the score easier to use in practice but also promote sparsity in the resulting prediction model. Such risk scores are often derived by first fitting a regression model and then rounding the estimated coefficients to the nearest integer after appropriate scaling. This approach is computationally fast but does not guarantee optimality of the resulting score. Alternatively, one may search over all possible integer weights to directly optimize a value function by posing the problem as an integer programming task. However, the associated computational burden can be substantial, especially when the value function is nonconcave or even discontinuous. In this paper, we develop new machine learning algorithms that employ a flexible greedy optimization strategy to learn such additive scoring directly under explicit and sensible optimality objectives. We apply the proposed method to a large electronic health record (EHR) cohort in Epic Cosmos to construct an integer-weighted comorbidity score for measuring the risk of post-discharge mortality. We also conduct a simulation study to examine the finite-sample operating characteristics.

  PublisherOA PDF

• Heterogenous Treatment Effects of Intensive vs. Standard Blood Pressure Control on eGFR Slope in Patients with CKD

  Journal of the American Society of Nephrology · 2025-10-01

  articleSenior author
  PublisherDOI

• Complexities in Variant Analysis, Classification, and Interpretation in Kidney Disease–Related Genes

  Seminars in Nephrology · 2025-08-06 · 1 citations

  reviewOpen accessSenior author

  Advancements in chronic kidney disease (CKD) genetic research and next-generation sequencing have improved CKD diagnosis and personalized treatment. Broad gene panel testing or whole exome/genome sequencing has greatly improved understanding of the genetic etiology of kidney disease but has also increased the complexity of interpretation. Standardized variant classification guidelines help, but challenges remain due to subjective evidence and limited functional and phenotypic data. Careful consideration of genetic and clinical evidence, along with collaboration between clinicians, genetics experts, and laboratories, is essential for accurate interpretation and patient care. This article examines nephrology genetic testing, focusing on the complexities of variant analysis, classification, and interpretation. Variant classification in monogenic kidney diseases is crucial for accurate diagnosis and patient management. We outline the classification methods highlighting several variant examples using the ACMG/AMP framework and quantitative approaches for pathogenicity assessment. We highlight challenges in integrating genetic findings into nephrology and emphasize the clinical impact of accurate genetic diagnoses for precision medicine in CKD.

  PublisherDOI

• Heterogeneity and Scoring Reproducibility of Folate Receptor 1 Immunohistochemistry in High-grade Serous Carcinoma

  International Journal of Gynecological Pathology · 2025-10-06 · 2 citations

  articleCorresponding

  Mirvetuximab soravtansine (MIRV) is an antibody-drug conjugate approved for the treatment of adult patients with folate receptor 1 (FRα; FOLR1) positive, platinum-resistant epithelial ovarian, fallopian tube, or primary peritoneal cancer, who have received one to three prior systemic treatment regimens. Per the FDA approval, FOLR1 positivity is defined as ≥75% of viable tumor cells showing moderate (2+) or strong (3+) membranous immunostaining ("PS2+"). Given this disease's high recurrence rate and relatively limited therapeutic options, there is utility in exploring consistency in FOLR1 reporting. Tubo-ovarian high-grade serous carcinoma (HGSC) samples from our institution's archives were included in tissue microarrays (n=806), whole tissue sections (n=51), or cell blocks (n=30) and evaluated using the Ventana FOLR1 (FOLR1-2.1) RxDx Assay. FOLR1 staining was heterogeneous across different anatomic sites (average FOLR1 PS2+ was 50.2 from adnexal sites compared with 47.4 from omental sites, P =0.015). Similarly, heterogeneity was noted in pre- versus post- neoadjuvant chemotherapy specimens (on average, FOLR1 PS2+ score increased by 17.7 from pre- to post- therapy, P =0.0089). Lastly, specimen type may also influence FOLR1 staining (average abdominal fluid FOLR1 PS2+ score was 25.5 and average surgical FOLR1 PS2+ score was 56.9, P =0.000034). Agreement among 9 readers was initially substantial, with a Fleiss kappa of 0.661 (95% CI: 0.636-0.685). For the subset of cases with the worst agreement initially, a training session with reference cases improved interobserver agreement. Our study highlights several factors contributing to heterogeneity in FOLR1 reporting. Future studies are needed to better understand the impact of FOLR1 heterogeneity on patient response to therapy.

  PublisherDOI

Frequent coauthors

• Lone Simonsen

  Fogarty International Center

  66 shared

• Cécile Viboud

  58 shared

• Bryan T. Grenfell

  Princeton Public Schools

  30 shared

• Megan L. Troxell

  University of Washington

  27 shared

• Julia R. Gog

  University of Cambridge

  24 shared

• Manjula Kurella Tamura

  Stanford University

  18 shared

• Neeraja Kambham

  18 shared

• Brooke E. Howitt

  Cancer Institute (WIA)

  16 shared

Labs

• Vice Provost for Student AffairsPI