We predict novel synergistic combinations beyond the reach of foundation models and traditional machine learning, where there is no precedent to interpolate from. The case for combinations
Standard of care shrinks a tumour, then resistant cells regrow and take back over. We look for combinations that hold every cancer population down at once, driving toward near-zero residual disease instead of a temporary response.
Just because a drug worked in one disease does not mean it will work in a similar setting. Its effect depends on a complex and evolving web of factors that have to create a necessary and sufficient environment for that drug to have an additive or multiplying effect. We unpick that web of interactions to determine what will be additive and what will likely disappoint.
The day before the meeting we published a PASS or FAIL and a confidence for fourteen phase 3 read-outs on @bigpicturebio, before any of them reported.
We are building our own internal combinations, both based on existing drugs and completely novel approaches.
We also work with others to help identify the right target combination, the right existing drugs in a portfolio to pair together, the most effective cargo for a delivery vector, or to simply sense check if a combination trial will perform as hoped, across pharma, biotech and investors.
Our model runs in seconds, so we can often do the first sense check within hours for free, and then love to work on solving the challenges that often come up. If that sounds good, get in touch at the email below.