Two critical factors – prior exposure to chemotherapy and a glycolytic metabolism – appear to degrade the potential of T cells to become chimeric antigen receptor–T cells.

Chemotherapy, especially with cyclophosphamide and doxorubicin, seems particularly toxic to T cells, damaging the mitochondria and decreasing the cells’ spare respiratory capacity – a measure of mitochondrial health, David Barrett, MD, said during a press briefing held in advance of the annual meeting of the American Association for Cancer Research.

Cells that relied primarily on glucose for fuel were much weaker and less able to withstand the chimeric antigen receptor (CAR) transformation and expansion process. Both of these characteristics were more common in cells from patients with solid tumors than in cells from patients with leukemia, said Dr. Barrett of the Children’s Hospital of Philadelphia.

These new findings may help explain why children with acute lymphoblastic leukemia (ALL) tend to respond so vigorously to CAR T treatment, and why T cells from patients with solid tumors simply don’t grow, or die soon after patient infusion, he said in an interview. They also suggest a benefit of harvesting T cells before any chemotherapy, a procedure Dr. Barrett and his colleagues have advocated.

“Based on these data we have altered our practice for T-cell therapy in high-risk leukemia patients. If we have a patient who may have a poor prognosis, we try to collect the cells early and store them before proceeding, because we know chemotherapy will progressively degrade them.”

There still is no successful CAR T-cell protocol for solid tumors, but Dr. Barrett said these findings eventually may help such patients, particularly if more advanced experiments in manipulating the cells’ metabolism prove successful.

He and his colleagues investigated why T cells from some patients result in a poor clinical product that either fails manufacture or does not proliferate in the patient. They examined T cells from 157 pediatric patients with a variety of cancers, including ALL, non-Hodgkin lymphoma, neuroblastoma, osteosarcoma, rhabdomyosarcoma, Wilms tumor, Hodgkin disease, chronic myelogenous leukemia, and Ewing sarcoma. The team obtained cells at diagnosis and after each cycle of chemotherapy.

They examined how well the cells grew in the transformation and expansion process. A “pass” was considered a fivefold expansion in response to CD3/CD28 exposure for 7 days. Normal donor cells typically expand 20- to 30-fold in this time.

Only T cells taken from ALL and Wilms tumor patients before chemotherapy achieved a pass, Dr. Barrett said. Most of the ALL expansions (80%) and half of the Wilms tumor expansions passed. “We noted very poor CAR T-cell potential in all the other tumor types – less than a 30% pass. We noted a decline in potential with cumulative chemotherapy in all cases, though this was particularly significant in children less than 3 years old.”

The team also used RNA profiling to look at the cells’ metabolic pathways. Dr. Barrett noted that T cells are highly metabolically adaptable, capable of using several different fuel types and switching from one to another. Glucose and fatty acids are frequent fuels. Most of the cells from patients with solid tumors exhibited a glycolytic metabolism, while cells from patients with ALL and Wilms tumor appeared to rely more on fatty acids.

“One is not inherently worse than the other,” he said. “But glycolysis appears to be a bad thing when we’re trying to turn them into CAR T cells. Those T cells were too exhausted to do anything.”

However, Dr. Barrett encouraged the cells to switch fuels by treating them in vitro with palmitic acid, the most common fatty acid in plants and animals.

“We were growing the cells in a media containing sugar, fatty acids, and amino acids,” he explained. “We just started overloading them with palmitic acid, which has a natural transporter on the T-cell surface, so it already had a good pathway to get into the cell. It helped restore some of the performance of these T cells in some assays, although it wasn’t a complete reversal. But it was encouraging that something as simple as providing an alternate fuel was enough to get some positive effect. Whether or not we would also have to block glucose use to get it to really work is something we continue to study.”

T cells that had been exposed to chemotherapy also did poorly. Cyclophosphamide and doxorubicin seemed particularly toxic. Cells with exposure to these two agents had severely depleted CAR T cell potential with very poor spare respiratory capacity. This is a marker of mitochondrial injury, Dr. Barrett said. “That wasn’t a huge surprise. We already knew that cyclophosphamide is very toxic to T cells.”

But the finding did suggest the simple intervention of harvesting T cells before chemotherapy, which is what Dr. Barrett and his colleagues now do in their high-risk ALL patients. Whether or not this would improve response in patients with solid tumors is still unknown.

He had no financial disclosures. This study was supported by the AACR, the Doris Duke Charitable Foundation Clinical Science Development Award, the Jeffrey Pride Foundation Research Award, and the St. Baldrick’s Foundation Scholar Award.

msullivan@mdedge.com

SOURCE: Barrett DM et al. AACR 2018, Abstract 1631.

Chemotherapy, metabolic pathway may affect CAR T-cell potential