Methods for randomized, blinded, controlled evaluation of putative disease interventions in multilaboratory, preclinical assessment networks

lab animal Article https://doi.org/10.1038/s41684-026-01683-z Methods for randomized, blinded, controlled evaluation of putative disease interventions in multilaboratory, preclinical assessment networks Check for updates Jessica Lamb    1 , Karisma Nagarkatti1, Marcio A. Diniz    2,26, Ryan Cabeen3, Monica Estrada3, Karen L. Crawford3, Andre Rogatko    2, Sungjin Kim2, Cenk Ayata    4,5, David C. Hess6, Mohammad Badruzzaman Khan6, Rakesh B. Patel    7, Mariia Kumskova7, Enrique C. Leira8,9,10, Anil K. Chauhan7, SPAN Consortium* & Patrick Lyden1,11 Science faces a reproducibility crisis, and public trust in science declines when large clinical trials, which had been qualified by promising preclinical studies, fail. While some clinical trial designs may have been inadequate, preclinical assessments of disease interventions might have lacked key elements of rigor such as treatment concealment, randomization, blinded outcomes, prespecified and adequate sample sizes, and models including comorbidities. Here, to demonstrate feasibility and practicality of enhanced rigor in preclinical assessment, we designed a six-laboratory network that implemented rigorous study elements, using acute ischemic stroke for demonstration. This network enrolled 2,615 rodents in 5 different models and implemented a multistage, multiarm statistical design that sequentially eliminated candidate interventions during interim analyses. The methods included centralized intervention packaging, randomization, data quality assessment and data archiving. Blinded analysis of 9,274 video-recorded behavioral tasks and 3,652 magnetic resonance images were evaluated. All tools and protocols are presented and could be adapted to preclinical assessment in other disease areas. Science faces skepticism from the lay public, and scientists have described problems with rigor, transparency and reproducibility. Many published findings, selected from high-impact journals, have failed replication outside of the original laboratories1–3. Many factors contribute to reproducibility issues in science: inadequate sample size and proper power analysis before initiating experiments, lack of control for repeated significance testing (‘P-hacking’), inadequate blinding of the investigators or insufficient or inappropriate controls, among other deficiencies1,4–7. 1 Many groups, including the National Academy of Science, have called on grant agencies and journals to enforce higher standards of rigor and experimental design to address these deficiencies. However, appropriate methods to implement greater scientific rigor may be lacking or insufficiently developed8. Here, we address one important type of scientific study: the use of preclinical animal disease models to assess the efficacy of proposed candidate interventions. Before launching pivotal clinical trials in patients, Department of Physiology and Neuroscience of the Zilkha Neurogenetic Institute of the Keck School of Medicine, Los Angeles, CA, USA. 2Biostatistics and Bioinformatics Research Center, Samuel Oschin Comprehensive Cancer Center, Cedars-Sinai Medical Center, Los Angeles, CA, USA. 3USC Stevens Neuroimaging and Informatics Institute, Los Angeles, CA, USA. 4Neurovascular Research Unit, Department of Neurology, Harvard Medical School, Massachusetts General Hospital, Boston, MA, USA. 5Department of Radiology, Harvard Medical School, Massachusetts General Hospital, Charlestown, MA, USA. 6Department of Neurology, Medical College of Georgia, Augusta University, Augusta, GA, USA. 7Department of Internal Medicine, Division of Hematology, Oncology and Blood and Marrow Transplantation, Carver College of Medicine, University of Iowa, Iowa City, IA, USA. 8Department of Neurology, Carver College of Medicine, University of Iowa, Iowa City, IA, USA. 9Department of Neurosurgery, Carver College of Medicine, University of Iowa, Iowa City, IA, USA. 10Department of Epidemiology, College of Public Health, University of Iowa, Iowa City, IA, USA. 11Department of Neurology of the Keck School of Medicine, Los Angeles, CA, USA. 26Present address: Department of Population Health Science and Policy, Icahn School of Medicine at Mount Sinai, New York, NY, USA. *A list of authors and their affiliations appear at the end of the paper. e-mail: Lab Animal Article https://doi.org/10.1038/s41684-026-01683-z a b d Publication requests MRI images Behavior videos SOP approval LONI Communication /website MGH Yale LONI IDA USC Stats Image analysis EAB SC decisions Protocol MRI acquisition UT-Houston c Primary outcome MCAo, NDS Corner Day –7 Corner, grid, hanging wire* Corner, grid, hanging wire*, NDS 7 28 NDS 0 1 2 5 Animal records EOS, tissue banking 29 30 Statistics raw Results REDCap forms by stage 21 MRI piled Com data Database e Timeline NDS, MRI Behavior scores Model choice Histology Augusta Stage-specific treatment Number of forms Univ. of Iowa Johns Hopkins NINDS 1 21 19 18 2 3 4 Stage Treatments as prescribed Fig. 1 | Description of the network. a, Geographical representation of the six laboratories marked with a yellow dot (Augusta, University of Iowa, Johns Hopkins, Mass General Hospital (MGH), Yale and UT-Houston) sending data to the CC at University of Southern California (USC), marked with a yellow star, where the data repositories are located, including IDA of LONI and Statistics. The External Advisory Board (EAB) provides feedback to the National Institute of Neurological Disorders and Stroke (NINDS) in Washington DC, which also advises the network. b, Graphical representation of the decisions the SC makes, including approving SOPs, communication and website development, stagespecific treatment protocols, the model choice, behavioral outcome measures and the experimental protocol, which includes when MRI images are collected as well as histology decisions and publication requests from within the network and outside the network. c, General experimental timeline through end of study (EOS): each animal starts with baseline corner testing performed seven days before MCAo surgery. NDS are collected on the day of surgery, day 1, day 2 and day 28. MRI is performed at day 2 and day 29. In addition to the baseline, corner testing is performed at day 7 and day 28, along with grid testing and hanging wire testing. *The hanging wire test was discontinued after stage 1. d, Pathways of data flow from the research laboratories to central storage and analysis. Data collected at the laboratories included animal records, MRI images and behavior videos, which are sent to either other laboratories, the database or LONI for analysis. Once compiled, all raw data are sent to statistics for results. e, Total number of data entry forms in the REDCap database for each stage (1–4) of the trial. Panels a–d created in BioRender; Lamb, J. https://biorender.com/sae4s8q (2026). many funders, sponsors and regulators require that therapeutic efficacy b (...truncated)