By David N. Leff

On May 10, 2001, the FDA granted fast-track approval to Novartis Pharmaceuticals Corp. for its anti-leukemia drug trademarked Gleevec (formerly STI-571). The event was hailed by print and television media as presaging a new generation of cancer chemotherapy agents.

FDA¿s authorization of Gleevec covered its use as oral therapy for treating patients with chronic myeloid leukemia (CML) in the accelerated phase of blast crisis, or in chronic phase after failure of interferon-alpha treatment. The agency acted on the basis of a multicenter Phase I clinical trial.

Now, via a fast-track electronic release, Science reports in its issue dated June 21, 2001: ¿Clinical resistance to STI-571 cancer therapy caused by BCR-ABL gene mutation or amplification.¿ Its senior author is clinical oncologist and molecular biologist Charles Sawyers, at the University of California, Los Angeles. He was a member of the panel that performed the Phase I trials of CML patients in 1999.

¿There are about 5,000 new cases a year of CML in the U.S. ¿ 10,000, if you include Europe,¿ Sawyers told BioWorld Today. ¿Average age at onset is 60.

¿CML,¿ he said, ¿is a pluripotent hematopoietic stem cell disorder characterized by a translocation from chromosome 9 to chromosome 22 ¿ the Philadelphia chromosome.¿ The resulting oncogenic BCR-ABL fusion gene encodes a cellular protein with tyrosine kinase activity.

¿The field would have thought,¿ Sawyers noted, ¿that at that advanced stage in a cancer¿s life, it wouldn¿t rely so much on a single oncogene. Our finding says that in fact, the cell does everything it can to maintain the activation of that initial BCR-ABL.

¿In chronic myeloid leukemia,¿ he added, ¿the oncogene is an enzyme, a tyrosine kinase, that goes by the name BCR-ABL. We¿ve known for years that BCR-ABL initiates the first hit in the development of this leukemia. We¿ve also known that when patients are in the end stage of the disease, which goes by the term blast crisis,¿ that there are numerous additional genetic hits that the cancerous cell has accumulated.¿

Cancer Cell Scams Outwit Gleevec

¿The point is,¿ he went on, ¿that the cancer cell now has five or 10 things that are used to drive its malignant growth. And when we treat with a drug directed against only one of them ¿ Gleevec against ABL ¿ when the cell becomes resistant, it does so by turning on BCR-ABL again, and gets around the drug by that oncogenic reaction.¿

Sawyers pointed out that, ¿Gleevec was designed to inhibit or block the BCR-ABL oncogene. So when the drug stops working, there are two possibilities: One, it is still blocking BCR-ABL, but the cancer cell doesn¿t care because it has other oncogene signals in which to grow. And that¿s what I think most people expected to find.

¿But possibility two,¿ he went on, ¿is that something about BCR-ABL changes itself in order to get around the drug ¿ so it can keep doing what it wants to do despite Gleevec being there. And that¿s exactly what happens.¿

At that point, the cell pulled off two tricks:

¿The first,¿ Sawyers related, ¿was that the gene amplified ¿ multiplied itself ¿ so there was more of the enzyme, and therefore it could outcompete Gleevec. And the second way, it could actually mutate itself by making an alteration in the amino acid sequence at the place where the drug binds to the enzyme, to block it. It did it in such a way that the enzyme now remains perfectly active, but the drug can¿t inhibit it any more.¿

As reported in Science, Sawyers described the clinical analysis of patients who had relapsed from early stage chronic CML to the late-stage blast crisis: ¿We examined the status of BCR-ABL in nine blast-crisis patients who had developed resistance to Gleevec. When they started with the drug, when they responded to it, and then when they relapsed. These individuals sometimes went into remission while on the drug ¿ which rarely happens with older treatments ¿ but 80 percent of them relapsed in less than a year.

¿We found that in all cases, resistance occurred because the enzymatic activity encoded by the original oncogene was switched back on.

¿We don¿t know what triggered that drug resistance,¿ Sawyers said. ¿We do know that we have seen it only in patients who had an advanced stage of CML. One characteristic of that stage,¿ he noted, ¿is that the cells are genetically unstable ¿ are able to change, unlike normal cells. The combination of a genetically unstable cell and Gleevec ¿ a cell that just wants to survive ¿ has found a way to get around the drug. Remissions in such patients have usually lasted only two to six months, despite continued Gleevec therapy.

¿Contrasting with that, in patients who were at very early stage CML ¿ the chronic phase ¿ we haven¿t really seen resistance there. And hopefully we won¿t. If there¿s a scientific explanation for why the difference,¿ he surmised, ¿it would probably be genetic instability.¿

New Drug Cocktails Against Oncogene

To counteract this complex, resistance-linked relapsing condition, Sawyers and others in the field are resorting to several strategies. ¿First of all,¿ he noted, ¿Gleevec is now being combined with chemotherapy to treat advanced stages of CML. That is, chemotherapeutic agents targeted and tailor-made to inhibit a specific cancer enzyme, and poison cell growth. That chemotherapy can work. Gleevec may work; we think the combination will work better. And we can do that today with drugs that we have available. So that¿s being tested now.

¿Based on this finding,¿ Sawyers observed, ¿it should be possible to use yet another targeted strategy. In the same way that Gleevec was identified and tested, a different therapeutic inhibitor of the mutant oncogene should be able to work against the same oncogenic target ¿ BCR-ABL. And what would presumably make sense is to use them in combination, so the cell cannot get around the drug¿s activity. It will try to circumvent the first drug, and there will be the second drug sitting there ¿ just waiting.

¿We and others are working toward this concept now,¿ Sawyers said, ¿but we¿re just at the beginning.¿