← Back to Home

Connecting Cellular Responses to AI: Flagship Combines Three Platforms to Form Etcelera

Etcelera hopes to identify therapies worth developing earlier by studying cellular changes following genetic and drug interventions. Blood diseases will be among the first tests, but vast datasets and models cannot yet replace answers from human trials.

By SURL BioNews

A molecule's ability to change a cell does not mean it can cure a disease. Distinguishing mechanisms that truly drive disease progression from accompanying signals before entering human trials is a challenge in drug development. On October 29, Flagship Pioneering announced the formation of Etcelera, bringing together the platforms and AI capabilities of Cellarity, Etiome, and Empress Therapeutics in an effort to move that assessment earlier, into drug discovery. Ad Rawcliffe serves as CEO of the new company, with Alec Reynolds as chairman.

The three platforms each offer a different perspective: Cellarity looks for clues to disease in overall cellular states; Etiome tracks how diseases progress over time to help determine when to intervene; and Empress provides large-scale experiments and multiomics tools to clarify the targets and mechanisms of candidate molecules. Following the integration, Etcelera will connect transcriptomic data reflecting gene activity, chemoproteomics analyzing interactions between compounds and proteins, and data on cellular responses to interventions.

Intervention data are key to understanding this strategy. Researchers can inhibit a gene or add a compound, then measure how cells change. Compared with simply comparing healthy and diseased cells, these experiments offer a better opportunity to ask: which component must be altered to bring about a shift in a disease-related state? Rawcliffe told Fierce Biotech and BioPharma Dive that the company hopes to train models on these responses to predict the effects of new interventions and identify earlier which drug candidates might fail in human trials.

Flagship said the platform includes a foundation model trained on data from 150 million cells, along with more than 30 proprietary AI predictive models. However, data scale is a starting point for the platform and cannot be directly translated into a clinical success rate. The launch announcement provided no prospective validation results for the integrated platform's predictions of efficacy in humans, nor did it disclose accuracy compared with existing methods. Whether the models can bridge the gap between cell experiments and patients' conditions remains to be tested in subsequent research.

Etcelera will initially focus on blood, metabolic, and immune diseases, including myelofibrosis, essential thrombocythemia, and a jointly developed program for metabolic dysfunction-associated steatohepatitis (MASH). Rawcliffe told Chemical & Engineering News that clinical trials are expected to begin in 2028. According to BioPharma Dive, the company will take over Cellarity's MASH collaboration with Novo Nordisk, while another alliance with an unnamed large pharmaceutical company remains under discussion. No new funding round was announced at launch. Rawcliffe said existing funds from the predecessor companies would help support ongoing operations.

The integration also draws a clear clinical boundary: CLY-124, Cellarity's oral drug candidate for sickle cell disease that has already entered human trials, will not be included in Etcelera. In a separate announcement, Cellarity said the program would be advanced by an independent company retaining the Cellarity name, still led by John H. Leaman and equipped with a dedicated team and resources. The drug's clinical progress therefore cannot be directly regarded as validation of Etcelera's integrated platform.

Etcelera must next demonstrate whether these complementary tools can improve drug selection decisions: choosing more suitable targets, intervention timing, and candidate molecules, and having experimental results support the models' predictions. The initial blood disease programs will provide concrete tests. Until human efficacy and safety data emerge, “predicting how drugs act in patients” remains a research and development goal rather than an established capability.

References

  1. Fierce Biotech
  2. Flagship Pioneering
  3. Cellarity
  4. Chemical & Engineering News