Skip to main content

A treatment that uses immune system T-cells, combined with an immune-boosting drug packaged in an injectable gel, was found to preserve the vision of mice implanted with tissue from a human eye cancer known as retinoblastoma. The cancer is treatable in early stages but can still lead to the loss of vision in about 5% of cases.

The research findings from scientists at the University of North Carolina Lineberger Comprehensive Cancer Center were published in Nature Cancer on Oct. 12, 2020.

Retinoblastoma is primarily diagnosed in infants and young children. It is considered rare, with approximately 200-300 children diagnosed with the cancer each year in the U.S. Current treatments for retinoblastoma use cold, heat, chemotherapy, lasers or radiation but vision loss still occurs, so the UNC researchers wanted to search for methods that could preserve vision.

A man with short black hair wearing a light blue collared shirt with a dark blue patterned tie.
UNC Lineberger’s Zongchao Han, MD, PhD.

“Based on our mouse study and the existence of an active cell immunotherapy program at UNC Lineberger, along with infrastructure for generation of CAR-Ts for clinical use, we feel confident that our efforts could be translated into a phase I clinical study in people,” said Zongchao Han, MD, PhD, an associate professor in the UNC School of Medicine and UNC Eshelman School of Pharmacy and a UNC Lineberger member.

The researchers used an incremental process to determine the best method for treatment of retinoblastoma. First, the researchers turned to chimeric antigen receptor-T (CAR-T) cell therapy, a type of immunotherapy where T-cells that comprise the immune system are modified in the laboratory to express chimeric antigen receptors, CARs, that target surface proteins on cancer cells. In a lab test, they found that a molecule, GD2, is expressed in retinoblastoma but the possibility to target this molecule to safely eliminate the tumor in the eye was unknown.

Next, to test the safety and benefit of targeting GD2, the investigators injected the CAR-T that recognizes this molecule into the retina of mice implanted with retinoblastoma cancer cells and found the therapy delayed tumor development but did not eradicate the tumor. Then they combined the CAR-Ts with interleukin (IL)-15, a protein that can boost immune response, and found that 60% of mice were tumor-free for up to 70 days.

Finally, they injected a water-based gel containing the CAR-Ts and IL-15 into the retinas of the mice. The CAR-Ts and IL-15 retained an extended ability to attack the cancer cells, control tumor growth and prevent tumor recurrence. They corroborated the lack of tumor growth with several imaging exams of the retina.

A smiling woman with long brown hair wearing glasses, a black turtleneck with a silver necklace.
UNC Lineberger’s Barbara Savoldo, MD, PhD.

This gel-encapsulated therapy is currently being tested in clinical trials in children with neuroblastoma, an embryonal tumor that can progress rapidly and has some of the same genetic characteristics of retinoblastoma.

“We are always looking to improve the lives of children at Lineberger,” said Barbara Savoldo, MD, PhD, professor of pediatric Hematology/Oncology at UNC School of Medicine and UNC Lineberger member. “Therefore, we hope to look at the safety of gel injection in a clinical trial of retinoblastoma in children, and if that proves safe, we could move on to see if our methodology can reduce or eliminate these tumors.”

Authors and Disclosures

In addition to Han and Savoldo, the paper’s other authors are Kai Wang, PhD, Department of Ophthalmology, UNC School of Medicine; Yuhui Chen, PhD, UNC Lineberger; and Sarah Ahn, PhD, UNC Lineberger and Department of Microbiology and Immunology, UNC School of Medicine, who all contributed equally to the work; Min Zheng, Department of Ophthalmology, UNC School of Medicine; Elisa Landoni, UNC Lineberger; and Gianpietro Dotti, MD, UNC Lineberger and Department of Microbiology and Immunology, UNC School of Medicine.

Funding was received from NIH grants (R01EY026564 and 1R21CA226483-01A1), the Carolina Center of Cancer Nanotechnology Excellence, a NC TraCS Translational Research Grant, the Edward N. & Della L. Thome Memorial Foundation, the BrightFocus Foundation, and the University Cancer Research Fund.

The authors declare no competing interests.

Media contact: Bill Schaller,