Optimizing dose-finding in early oncology trials by utilizing information from multiple measures and simulations

The goal of this research is to improve the process of determining the best dose of a new cancer treatment during early clinical trials. When developing cancer treatments, it is critical to find a dose that provides the most benefit while minimizing harmful side effects. Early-phase clinical trials often involve a small number of patients and try to minimize number of people treated with doses that are too high or too low. This can make it difficult to gather enough useful information about how the drug works.

To address this, our research will use a method called Monte Carlo simulation. Monte Carlo simulation is a statistical technique that creates synthetic data by predicting possible outcomes based on the information already collected from patients. By generating a range of possible results, this method can help researchers better understand the dose-response relationships both from the toxicity perspective (side effects), and efficacy perspective (how well the treatment works).

One of the main challenges is using all the information available about each patient at different dose levels measured by different variables. Our approach introduces a novel technique to combine various types of patient data more effectively and provides a structured way to evaluate the safety and effectiveness. The method also aligns with the recent guidance from the Food and Drug Administration (FDA) on dose optimization for cancer treatments.