Safe Harbor? Leukemia, Gene Transfer, and Lentiviral Vectors

A few further observations from the American Society of Gene Therapy Meeting...

A recurrent theme in this blog is the frequency with which novel research fields encounter safety problems that confound laboratory predictions. One presentation at the 2009 ASGT meeting brought this point home.

Recall my entry on May 12 discussing various refinements to retroviral gene transfer that are aimed at reducing risk of malignancy. Researchers have postulated that HIV-derived lentiviral vectors might not cause the same leukemia-inducing mutations as retroviruses, and RAC recently passed a favorable judgment on a lentiviral vector gene transfer protocol for X-SCID.

How confident can we be that lentiviruses will not trigger leukemias? Some indication is provided in a May 2009 review by John Rossi in Molecular Therapy. It concluded "overall, the results of these [safety] analyses [of lentiviral vectors] are highly encouraging..." but "clearly, more careful analyses... are warranted in appropriate animal models."

At ASGT, researchers from France reported preliminary results from a phase 1 trial testing lentiviral vectors in patients with beta-thalassemia. The study involved two patients. Though no malignancies have been detected, tests in one patient showed signs that some cells were repopulating the patients blood much faster than others (what researchers call "clonal dominance"). This is a worrisome signal, as it might indicate a premalignant state.

The lessons here are not that lentiviral vectors are unsafe (we don't know whether this will lead to a malignancy), or that such vectors shouldn't be used in human beings (we can't say anything yet about the risk-benefit balance). Instead, I think the lesson is: in novel research areas, be very wary of anyone who makes emphatic claims that their system provides safe harbor. Expect the unexpected. (photo credit: dark matter, 2005)