By zeroing in on a notorious class of enzymes associated with cancer, Reverie Labs puts the power of cloud computing to its full potential
Based in Cambridge, Reverie Labs is applying the increasingly powerful discipline of machine learning to biology. This kinase-targeting, cancer-fighting biotech company has no lab facility of its own. But that didn’t stop them from raising $6.6 million a couple years ago in a virtual drug discovery deal that included investors like First Round Capital, Neo, Wireframe Ventures, and LifeForce Capital.
Now, Reverie Lab’s virtual approach has gotten the attention of the pharma industry. It recently announced a multi-target collaboration agreement with pharma giants Roche and Genentech, just three short years after Reverie started with a few laptops in a New York City apartment.
The startup’s meteoric rise marks the arrival of the next generation in precision medicine.
“We’re extremely excited to be working with some of the best investors in the business,” says Kallenbach. “It’s incredible to have folks in our corner who have expertise in building scalable tech organizations, as well as others who have deep expertise on the healthcare side.”
How it all started
Kallenbach and Gupta (Reverie CEO and CTO, respectively) met on their first day of college at Harvard. After hitting it off and studying together, the two developed their ideas and eventually participated in Y Combinator’s Winter 2018 batch. They focused intensely on developing their machine learning algorithms to improve the ability to find potential drug candidates. Reverie Labs estimates it takes $100 million and two years for big pharma companies to choose a drug candidate to go to clinical trials.
“Computation isn’t driving decision-making at big pharma,” says Kallenbach, “and we knew that we needed new computational tools for smarter decisions in drug discovery.”
This approach has proven fruitful. Reverie’s interdisciplinary team of engineers, chemists, and modelers has used the company’s unique technology platform to generate, select, and predict millions of chemical structures for new precision therapies. This drug development pipeline has yielded candidates that can penetrate the blood-brain barrier to target kinase-based brain cancers, one of the most devastating and notoriously difficult forms of cancer to treat. The scalability of Reverie’s infrastructure means that cures for brain cancer and other deadly diseases are now within reach.
Zeroing in on kinases
In 2001, the FDA approved a drug that would revolutionize the landscape of cancer treatment. Gleevec, also known as imatinib, transformed chronic myelogenous leukemia (CML) from a death sentence to a conquerable form of cancer.
Conventional chemotherapy kills fast-growing cells like cancer, but it also kills fast-growing healthy cells with devastating side effects. In contrast, Gleevec is more discriminate, destroying only cancerous leukemia cells while leaving healthy cells unharmed. This life-saving innovation marked a breakthrough in precision medicine and established a new class of drugs called targeted therapies.
Gleevec treats CML by targeting a protein called BCR-ABL. BCR-ABL is found only on cancerous blood cells, and belongs to a class of enzymes known as kinases. Because Gleevec targets this specific kinase, it cannot treat cancers caused by mutations in any of the other 500 kinases in the human genome.
While effective drugs targeting other kinases do exist, developing more highly specific kinase inhibitors—especially at higher speeds and at lower costs—remains an open and urgent problem.
Reverie Labs saw this opportunity and decided to focus on kinases. Machine learning algorithms have advanced significantly, but they can still only solve drug discovery problems within specific domains. As a class of molecules, kinases offer deep homology and structural similarities that narrow the search space in a way that’s ideal for current machine learning methods.
Reverie is an early example of a new kind of pharmaceutical company—virtual pharma—that is using cloud computing as a killer app to revolutionize drug development, all while eschewing the high costs of traditional pharma.
Follow me on Twitter at @johncumbers and @synbiobeta. Subscribe to my weekly newsletters in synthetic biology. Thank you to Aishani Aatresh for additional research and reporting in this article. I’m the founder of SynBioBeta, and some of the companies that I write about are sponsors of the SynBioBeta conference and weekly digest. Here’s the full list of SynBioBeta sponsors.0