Avatars Provide Unique Window Into the Crucial Pathways Connecting Each Patient’s Cancer Cells and Genetic Mutations

by Tom O’Neill

Ben Mizukawa, MD and Jim Mulloy, PhD.

Right: Ben Mizukawa, MD, and Jim Mulloy, PhD, both of the Cancer and Blood Diseases Institute, are the principal investigators on the avatar project.

Cancer researchers here have more than 100 reasons for seeing a new horizon in personalized treatment for children with high-risk and relapsed cancer.

Each of those reasons is reflected in a mouse “avatar,” carefully engrafted with the cancer cells from children whose leukemia cells have been sequenced.

These mice serve as living test platforms, allowing scientists at Cincinnati Children’s to add a new level of precision for determining which treatments are most likely to work against cancers carrying particular genetic mutations.

Since launching the initial research phase of the avatar program in 2015, the team has developed lines of mouse avatars that mimic more than 100 children’s cancers, primarily from patients who developed hematologic cancers such as acute myeloid leukemia (AML) and acute lymphoid leukemia (ALL).

“When you’re down to the level of the genetic mutations that drive the formation of these leukemias, you see that each patient is unique,” says Ben Mizukawa, MD. “If a cluster of avatars share a common mutation or pathway that can be dysregulated, you can test drugs against those specific targets and learn to predict which therapies are likely to succeed in clinical trials.”

He and Jim Mulloy, PhD, both of the Cancer and Blood Diseases Institute (CBDI), are the principal investigators on the avatar project, which involves a group of about 25 collaborators, including clinicians, genomics technicians and experts in biomedical informatics.



A. been diagnosed but not yet begun initial intense rounds of chemotherapy, or,

B. have refractory disease that does not respond to chemotherapy, or,

C. have relapsed disease after being in remission for some time

After studying the cells’ molecular make-up, gene expression and proteins, the team grows the cells in the avatar and analyzes the effectiveness of different treatment options. 


The program is rooted in the innovative goals of John Perentesis, MD, Co-Executive Director of CDBI.

“John had the foresight years ago to routinely send for genetic sequence testing on cancer patients,” Mizukawa says. “So we have hundreds of patients where we can see the specific gene mutation, and match drug options to it. Traditionally, there was just an empiric approach, very trial-and-error.”

Avatars reflect a step beyond the standard method of evaluating human cancer cell lines in culture. Although growing cells in a petri dish can be informative, the process produces limited numbers of cell lines. Such cells also tend to adapt to the culture conditions, acquiring additional mutations to grow, which makes them less like their source over time.

Interestingly, cancer progresses faster in mice than it does in humans, allowing the researchers to test multiple different treatments in an avatar in less time than it would take to test a single treatment in humans.

“Typically, we use drugs that are targeting cells that are growing and dividing quickly,” Mizukawa says. “They’re not tailored to any particular biology of the cancer cell. With some leukemias, that has worked. Yet others do not respond, or the patient relapses and the tumor becomes harder to treat. There is now more emphasis on predicting who is going to respond well and who needs new approaches.”


Engraftment of as few as 10,000 cells can expand to millions in a mouse, which can then be engrafted into a second-generation of mice that are also genetic mimics of that patient.

“And we’re interested to know,” Mulloy says, “whether the sub clone that grows out from the minimal residual disease samples is actually the relapse clone that shows up months to years later in the patient.”

If the answer is yes, deep sequencing of the clone may answer why it is chemo-resistant, and whether specific mutations might be targets of new therapies. Deep sequencing acquires more information than standard sequencing, and lessens the chance that rare mutations are missed.

For now, data from the avatars are too new to begin guiding patient therapies. Researchers still face numerous obstacles. Engraftment of the cancer cells fails 25 percent of the time. As a heterogeneous disease, cancer presents something of a moving target.

Scientists envision that one day avatars will help them identify molecular subgroups that show exceptional responses — or just as importantly, drug resistance and toxicity.

The team is also still studying why engraftment works better with aggressive blood cancers than non-aggressive types like myelodysplastic syndromes.

They’ve also yet to engraft “solid tumors” such as brain tumors, neuroblastoma and sarcoma as part of their program. “That,” Mulloy says, “is on the horizon.”