By David N. Leff
Until three or four decades ago, a diagnosis of leukemia spelled a speedy death sentence with no appeal. In 1847, a Scottish physician, John Hughes Bennett (1812-75) at the University of Edinburgh, gave the implacable disease its name ¿ leukemia.
Acute lymphoblastic leukemia (ALL) ¿ the commonest malignancy in children ¿ usually strikes its victims between 3 and 5 years of age. The cancerous cells move into the bone marrow and displace its white blood cells, the leukocytes. The blood-borne interlopers spread throughout the body, notably to liver, kidneys, spleen, lymph nodes, central nervous system and gonads.
By the mid-1960s, chemotherapy to treat this childhood hematopoietic dycrasia gradually improved, and outright cure became a realistic goal. Now, ALL gives 95 percent of chemo-treated kids a five-year lease on life. Two out of three who achieve this near-unanimous remission continue disease-free indefinitely, presumably cured.
The adult-onset version of the blood malignancy is acute myelogenous leukemia ¿ AML. Its salient risk factor is advancing age, and AML responds to fewer chemotherapy drugs than does ALL. Current remission induction rates run from 50 percent to 85 percent, with long-term disease-free survival at 20 percent to 40 percent. It¿s higher in younger AML patients, who are more likely to benefit from bone marrow transplantation.
Now ALL and AML are no longer alone in the traditional two-star constellation of leukemias. It has just been joined by MLL ¿ mixed-lineage leukemia. This novel third variant was ushered in by the December 2001 issue of Nature Genetics, under the title: ¿MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia.¿ Its senior author is cancer immunologist Stanley Korsmeyer at the Harvard-affiliated Dana-Farber Cancer Institute in Boston.
¿The important point of this paper is that we have thrown the dice, if you will,¿ Korsmeyer told BioWorld Today, ¿and looked at whole-genome profiling in acute lymphoblastic leukemias. We¿ve pulled out such a drastic difference in these diseases ¿ with translocations of this MLL mixed-leukemia gene ¿ that we think it ought to be distinctly separated, rather than thought of as solely an unfavorable portion of ALL.
¿The rationale is we took a look at that DNA-profiling diagram and were struck by 1,000 genes down and 200 up, comparing MLL to conventional ALL. So they¿re really quite different critters.¿
Newest Acute Leukemia ¿ Rare But Lethal¿
¿It tells us that the cell of origin of this leukemia is quite different from the rest,¿ Korsmeyer continued. ¿If they all responded to conventional therapy, you could argue that we¿re being kind of esoteric. On the other hand, this new MLL is a rare but lethal blood cancer that afflicts fewer than 100 babies a year in the U.S. in their first year. They tend to relapse very rapidly, and have a poor prognosis following conventional therapy. Their blood cells are the scene of a chromosomal accident. Portions of their fourth and 11th chromosomes have broken off and swapped places. This leads to the runaway growth of cancer.
¿So I think this puts the onus on us,¿ he went on, ¿to say Well, it¿s no wonder they don¿t respond to standard therapy, and we need to be looking for novel therapeutic approaches for this third subset.¿
¿And when we tried to identify the distinct essential genes that would separate, when we profiled MLL from ALL or AML, right at the top of that list was one essential minimal molecular target ¿ the flt3 tyrosine kinase gene. So that had to be validated and interrogated a bit, but I think this type of DNA-profiling study, in addition to subsetting leukemia ¿ the good, the bad and the ugly ¿ will also provide unanticipated molecular targets for testing that we couldn¿t have dreamed up if we had thousands of years of doing conventional experiments, one at a time.¿
Korsmeyer explained: ¿Flt 3 encodes a tyrosine kinase enzyme that is expressed in hematopoietic [blood-forming] cells. It¿s probably an early evolutionary survival factor for them, and has been seen also to be mutated in a handful of adult AMLs. So with the tyrosine kinase, the gene has an attractive little ATP-binding pocket there, of which there probably are small-molecule inhibitors that we¿re testing. The analogy, of course, that works so spectacularly well,¿ he pointed out, ¿is Novartis¿ drug Gleevac, back on another tyrosine kinase, as the paradigm that¿s been so successful and so nontoxic ¿ at least in treating CML ¿ chronic myelocytic leukemia.¿
To analyze MLL¿s gene expression, the co-authors turned to microarray chips ¿ commercially made by Affymetrix Inc. ¿ strewn with small stretches of DNA encoded by 12,600 genes. ¿They represent somewhere around a third or a fourth of the human genome,¿ Korsmeyer observed. ¿When you¿re looking for pattern recognition, even that genomic subset is adequate in terms of seeing RNA profiles. So one might guess that out of the 30,000 to 40,000 human genes, about 3,000 will be used to make RNA in any one given cell type, and we were after that signature.¿
On To Drug Discovery With Novartis
¿What we got,¿ he recounted, ¿was a higher expression of this flt3 tyrosine kinase in MLL compared to the other two leukemias ¿ ALL and AML. That doesn¿t necessarily mean that it¿s critical for the maintenance of a leukemia. It¿s just a guess that it might be a player. We then took a look at inhibitors of flt3, first in vitro with fresh leukemic cells from MLL patients vs. the other two to see if they¿re differentially susceptible. We found that about 1,000 genes were silent in the MLL cells, and 200 overactive, compared to those from patients with conventional ALL
¿Those findings were positive,¿ he recalled, ¿so we¿re now trying out these small molecules in mouse models. If everything works out we¿ll move into human clinical trials. So it¿s a sequential thing, a new idea to be tested.
¿Dana-Farber has a collaborative research agreement with Novartis AG in Basel, Switzerland,¿ Korsmeyer concluded, ¿in terms of a small-molecule inhibitor. It¿s a drug discovery program with their oncology group.¿