In a quiet lab in Los Angeles, a tiny cluster of cells no wider than a poppy seed is changing how we fight cancer. Grown from a patient’s own tumor and printed layer by layer using 3D bioprinting, this miniature organoid is being bombarded with dozens of drug combinations—each move captured in real time by a high-speed camera and interpreted by artificial intelligence. At the UCLA Health Jonsson Comprehensive Cancer Center, Dr. Michael Teitell and his team have built a revolutionary platform that merges biology, engineering, and machine learning to test cancer therapies faster and more precisely than ever before. This isn’t science fiction—it’s a new frontier in personalized medicine, now detailed in Nature Protocols.
Cancer treatment has long relied on one-size-fits-all approaches, but tumors vary wildly between patients, even within the same cancer type. Traditional models, like cell lines grown on flat dishes, often fail to mimic the complexity of real tumors. Organoids—miniature, three-dimensional versions of tumors grown in the lab—offer a far more accurate model. But until now, scaling them for rapid drug testing has been a major hurdle. The UCLA team’s breakthrough lies in combining extrusion bioprinting with label-free quantitative phase imaging and AI-driven analysis, creating a system that can produce, monitor, and analyze thousands of organoids simultaneously—without damaging them.
The platform prints patient-derived tumor cells into 3D structures embedded in extracellular matrix gels, arranged in multiwell plates for high-throughput screening. Using high-speed imaging, it tracks changes in biomass and growth over time—up to 1,000 organoids per hour—without dyes or destructive assays. AI then steps in: deep learning algorithms segment images, while machine learning models track individual organoid responses to therapy, revealing patterns invisible to the human eye. In tests, the system successfully monitored responses in both established cancer cell lines and a patient-derived glioblastoma sample, identifying subtle differences in drug sensitivity and spotting rare, resistant cell populations that could lead to relapse.
"Instead of asking whether a drug works on average for a large number of tumor cells, we can now determine which specific organoids respond and which do not, and ultimately, have an approach to determine the underlying reasons for unique response profiles," said Dr. Teitell. This level of precision could transform how oncologists choose treatments, allowing them to test therapies on a patient’s own cells before a single drug is administered. For cancers with limited options—like pancreatic or brain tumors—this could mean the difference between life and death.
While still in the research phase, the platform is paving the way for clinical trials where treatment decisions are guided by a patient’s living tumor twin. In a world where cancer claims nearly 10 million lives a year, this fusion of bioprinting and AI offers a powerful new hope: that the right drug can finally find the right patient, faster.