News of the WeekClinical Research

Gene Therapy a Suspect in Leukemia-like Disease

See allHide authors and affiliations

Science  04 Oct 2002:
Vol. 298, Issue 5591, pp. 34-35
DOI: 10.1126/science.298.5591.34

A French gene-therapy team that was hailed in 2000 for its breakthrough in curing children of a lethal immune deficiency reported a serious adverse event this week. One of 10 children they treated has developed a blood disorder resembling leukemia. Concerned that the therapy might have caused the problem, researchers Alain Fischer and Marina Cavazzana-Calvo of the Necker Hospital in Paris have halted the trial and urged others who use similar methods to hold off until the risks are assessed. At press time, French regulatory officials were preparing a public advisory.

Fischer says he and his group recognized the importance of the case “exactly 1 month ago” but decided to study it and explain it to their patients before going public. The French team quickly sent advisory letters to investigators in charge of similar gene therapy trials using gene transfer “vectors” made from a retrovirus called the mouse Moloney leukemia virus. When the warning reached the U.S. National Institutes of Health (NIH) in Bethesda, Maryland, a clinical group immediately cancelled a six- patient trial due to begin in September.

This French trial was designed to identify children with a type of severe combined immunodeficiency (SCID) caused by a mutation on the X chromosome and to treat them early (Science, 28 April 2000, p. 669). So far, the team has treated nine infants and one teenager. All faced the prospect of lethal infections or harsh therapy such as bone marrow transplantation, which itself often has fatal consequences. Gene therapy offered a way out; in most cases it restored the immune system without toxicity.

During a routine check of their fourth patient last spring, however, the French researchers noted that the child had a high number of γδ T cells in his blood. The import didn't hit home until late August, Fischer says, when the T cell count climbed “very high”—to 200,000 cells per microliter. Other symptoms also appeared, including mild anemia, and the child was hospitalized.

Molecular studies revealed that the T cells were monoclonal: All had come from a single cell. Furthermore, Fischer explains, all the cells contained the same DNA signature, a sequence reflecting the site where the retrovirus vector had integrated itself into the host's genome. “Unfortunately,” Fischer says, that site is in the coding region of a gene on chromosome 11 that's “aberrantly expressed” in a form of childhood acute lymphoblastic leukemia.

Fischer believes that the vector triggered “an insertional mutagenesis event”—splicing itself into a dangerous gene and stepping up its production. “Everyone was aware” of a theoretical risk that retrovirus vectors might do this, he says, but the risk seemed very small. The phenomenon did not turn up in animal experiments or in other clinical data.

Although gene therapy probably contributed to the patient's T cell response, Fischer says, other factors probably played a role, too. For example, the child might have been predisposed to disease, as other members of his family have had childhood cancers. And an infection might have been important as well; the child got chickenpox shortly before his T cell count spun out of control. But at the moment, Fischer acknowledges, it's not clear whether this was a “very unlucky” random event or a sign that the risk of using retrovirus vectors has been “underestimated” in the past.

Researchers at the Necker Hospital are collaborating with Christof von Kalle of the Institute for Molecular Medicine in Freiburg, Germany, to try to create a map of all known human DNA integration sites for this retrovirus vector. They hope this will enable them to estimate the risks better. Meanwhile, the child is receiving chemotherapy.

Few gene-therapy researchers were available to comment on the case at press time. But Jennifer Puck, a leader of the planned SCID therapy trial at NIH, knows of four other groups that are using or were planning to use similar gene-therapy techniques. At the moment, she says, “we don't know whether the risks [of insertional mutagenesis] are one in 80 or one in 10 million.”

U.S. regulatory officials declined to comment on the case. But NIH's Recombinant DNA Advisory Committee is reported to be preparing a broad review of the case at its next meeting, scheduled tentatively for 4 to 6 December.

View Abstract

Navigate This Article