Extensive scientific studies are conducted to assess the relative risks of various candidate modified risk tobacco products compared with those of smoking cigarettes. As the scientific community conducts such assessments for diverse products and in a variety of laboratory models, knowledge on toxicity is spread across numerous scientific articles. We believe that by fostering the consolidation of data and knowledge gained from studies assessing novel tobacco/nicotine delivery products on a community platform, new hypotheses may be generated, and the weight of evidence may be increased. Therefore, we have created and are further developing INTERVALS (www.intervals.science), an online platform supporting independent, third-party collaboration by proactively sharing detailed protocols, tools, and data from assessment studies. Data files are accompanied by relevant information to foster reproducible research and encourage data reanalysis. We will present a meta-analysis of in vitro toxicology assessment studies, including aerosol characterization, neutral red uptake assay, and mouse lymphoma assay, for various e-liquid and heat-notburn platforms compared with the 3R4F reference cigarette. These studies have been conducted by multiple organizations using different methods and models. The content of the separate publications has been curated and included in INTERVALS in an interoperable format so that a metaanalysis of results can be performed. The direct comparison of the platforms tested in separate studies with different study designs (e.g., different lists of chemicals quantified in the aerosols) makes it difficult to compare every single result across all individual studies. However, the overall result is consistent in that all of the studies included in this analysis demonstrate the reduction of harmful or potentially harmful chemicals and of toxicity assessed in vitro for the tested platforms compared with cigarettes. As the scientific community integrates more studies and datasets into INTERVALS, it will become easier to conduct such meta-analyses and review results obtained across institutions, models, and platforms.