Prostate cancer researcher Dr. John Lee garnered a National Institutes of Health New Innovator Award to fund his breakthrough approach that promises to help make scientific tumor models better capture the complex biology of tumors that arise naturally. Part of the NIH’s High Risk, High Reward Research program, New Innovator Awards support exceptionally creative early-career investigators carrying out innovative, high-impact projects.
“The goal was to close the gap [between natural tumors and scientific models] and make tumor models look and behave more like human cancers,” said Lee, who is a faculty member at Fred Hutchinson Cancer Research Center.
Researchers learn more about cancer using a variety of methods, from cancer cells growing in lab dishes, to mice that have been genetically engineered to try to reproduce some characteristics of human tumors.
Lee developed his new approach by building a more complex model of prostate cancer. The grant, $1.5 million over five years, will allow his team to create a set of prostate cancer models that better mimic existing subsets of human tumors and probe different steps governing prostate cancer progression. They hope this will help improve scientists’ understanding of the genetic factors shaping prostate cancer’s initiation, growth and progression. Lee also plans to expand the approach to model other types of cancer.
Lee hopes that eventually he and other scientists will be able to use insights from this approach to capitalize on the promise of precision oncology, and better match patients to effective treatments based on their tumors’ unique genetic makeup.
The problem with tumor models
“Cancer is very complex at the genetic level,” Lee said. “No two cancers are the same, in that they harbor different genetic abnormalities.”
Individual tumors can also have hundreds or thousands of mutations — a phenomenon that current tumor models don’t capture, he said.
Though genetic engineering offers scientists a lot of insight into how specific genes may be contributing to cancer initiation and progression, researchers can only modify two to three genes in any given model.
“It means we’re missing a whole heck of a lot of biology,” Lee noted.
The lack of genetic complexity in these models means that the information researchers glean about key molecular drivers of cancer may not always hold up when they transition from the lab to clinical trials in people.
“So we asked, is there a high-throughput way that we can figure out how all these diverse genetic alterations operate together in cancer?” he said.
A high-throughput approach to creating more complex, more accurate tumor models
Lee and his team decided to start with organoids, a way of modeling the three-dimensional structure and cellular interactions found in tissue. Rather than growing two-dimensionally across a relatively solid, nutrient dense medium, organoids grow in a more viscous medium that allows the cells to grow in all dimensions and form structures that better represent the 3D contacts found between cells in tissues. Organoids capture more of the natural biology of tissue than individual cells in a Petri dish, but are quicker and easier to produce than mouse models.
But there was a drawback: The usual methods that scientists use to disrupt genes (which rely on viruses called lentiviruses) don’t work efficiently in organoids. Lee’s breakthrough innovation turned out to be surprisingly simple: They mixed the lentivirus into the biological matrix that the organoids grow in, ensuring that many virus copies made their way into each cell in the growing organoid.
“That was a pretty significant breakthrough,” Lee said.
It allowed them to create organoids in which each cell has a different set of many genetic abnormalities. Next, Lee’s team grows the genetically manipulated organoids in mice.
“From the tumors that emerge, we can actually analyze cells at the single-cell level, and figure out exactly which lentiviruses and genetic alterations got into those cells,” Lee said.
Computational analyses of individual tumor cells then reveal which combinations of mutations work together to promote tumor growth and survival, and which don’t.
“The New Innovator Award will allow us to push forward to develop more models and improve our understanding of the biology of prostate cancer,” Lee said.
His ultimate goal is to create several laboratory models of prostate cancer whose genetic mutations reflect different subtypes of prostate cancer that occur in humans, which can have different prognoses and respond differently to treatment. Lee also plans to apply his approach to understanding different aspects of prostate cancer progression. Additionally, he is expanding into other types of cancer, including bladder cancer and, working with Hutch colleague Dr. Nina Salama, gastric cancer.
“Once you get a wide enough swath of models, you can treat them across the board and figure out which prostate cancers are sensitive to treatment, which are resistant, and determine the genetic factors responsible for that,” Lee said. “I think that approach will be really useful to specify the activity of treatment to different subsets of prostate cancer.”
