Every scientist doing a long-term study wonders whether his or her conclusions would be different if the experiment had lasted longer. And a brief communication published in the April 27 issue of Nature said that for the side effects of X-SCID gene therapy, the answer is yes.

But clinical researchers remain skeptical, with some alleging that the paper was rushed to publication. Those conducting X-SCID gene therapy trials at a British hospital called the paper "unhelpful," according to a report by the BBC. And Harry Malech, director of the laboratory on host defenses at the National Institutes of Allergy and Infectious Disease, concurred. Malech, who also is conducting X-SCID gene therapy trials, told BioWorld Today that a brief communication is "by its very nature a short report. But boy, it is a really short report."

In 2000, the successful treatment of babies suffering from x-linked severe combined immunodeficiency disorder (X-SCID) was hailed as gene therapy's first triumph. But subsequently, for some of those children, the cure turned out to be as bad as the disease: To date, three of 14 patients have developed leukemia after the therapy. One of the children died. While not scrapping gene therapy trials altogether, the FDA has recommended that only patients for whom conventional therapy has failed be treated with gene therapy. (See BioWorld Today, May 1, 2000; Jan. 16, 2003; and March 7, 2005.)

Why some of the children developed leukemia is not fully understood, but one apparent problem is that the corrective gene, which is randomly inserted into the host genome, was inserted in or near an oncogene. But according to the Nature paper, that cannot be the only problem. Researchers from the Salk Institute in La Jolla, Calif.; the National Center for Tumor Diseases and German Cancer Research Centers, both located in Heidelberg, Germany; and the University Hospital and Institute of Molecular Medicine and Cell Research, both in Freiburg, Germany, have published data suggesting that the corrective gene itself also can contribute to the development of leukemia.

In fact, in what first author Niels-Bjarne Woods called "a surprising finding," mice treated with a copy of the IL2RG gene actually started developing leukemia before the positive controls, which were treated with an outright oncogene. IL2RG-treated mice began developing lymphomas six months after gene therapy, while the first mice treated with the oncogene showed evidence of lymphomas a few weeks later.

Over the course of 18 months, a third of the mice treated with IL2RG, half of the mice treated with an oncogene and none of the mice treated with a control vector developed T-cell lymphomas.

The communication did not include any studies about the mechanism by which IL2RG caused cancer. But IL2RG codes for the gamma chain of the interleukin-2 receptor, which is involved in growth and differentiation of B and T cells. An obvious possibility is that the interleukin-2 receptor signaling is somehow altered when the gamma chain is delivered via gene therapy.

In clinical circles, the reaction to the paper has been decidedly mixed. Malech summed up his dilemma for BioWorld Today: "No one can take this paper and ignore it, because the authors may well be right in their conclusions. So it's very important. But neither do you have all the facts to decide what to do. I can't go to sleep at night and know whether it's right or not."

Some critics have noted that the mice were treated with high doses of IL2RG gene, making the results irrelevant to clinical trials. First author Woods agreed that the doses the scientists used were high. But he called the idea that they were definitely higher than what patients would be exposed to during a clinical trial "a bit risky."

Woods told BioWorld Today that the level of gene expression will differ from patient to patient and from cell to cell depending on the exact site that the therapeutic gene integrates. "So the level of expression in a patient could be very high."

Malech's problem with the paper, though, goes beyond any one finding or method; he said it was "just not reported very well" and called its publication "premature." He said that the there is a "desperate" need for larger-scale replication and more thorough experimentation to either confirm or refute the findings.

In the short term, Malech, who is conducting a clinical trial on gene therapy for older children with X-SCID who have responded poorly to conventional treatment, said that the paper has no practical effects on his work because he does not have any patients that are close to starting treatment at this time: "I am able to wait to see what the community and the regulatory agency decide about this."