Studies Show Gene Expression Activity in Leukemia Cells
Two studies, both in the August 5 New England Journal of Medicine, offer new insights to some of the underlying mechanisms of acute lymphoblastic leukemia (ALL), a leading form of cancer among children. The first study, by researchers at St. Jude Children's Research Hospital in Tennessee; Erasmus University - Sophia Children's Hospital in Rotterdam, the Netherlands; and the COALL cooperative group in Germany; identified sets of genes that are differentially expressed in leukemia cells resistant to each of four antileukemia drugs tested, and also showed that the pattern of expression of these genes was related to overall patient outcomes. The second study, by researchers at the National Cancer Institute (NCI), found that the loss of a key protein (Smad3) is specific to one form of childhood leukemia, but not to other pediatric and adult leukemias.
Dr. William Evans of St. Jude and colleagues compared total gene expression of leukemia cells taken from 173 ALL patients and grouped the samples that exhibited resistance to one of four drugs: prednisolone, vincristine, asparaginase, or daunorubicin. They found 124 genes and 28 cDNA (potential genes) that are differently expressed in drug-resistant cells. "Interestingly, only three of those genes had previously been associated with drug resistance," said Dr. Evans, "so there are unexpected cellular mechanisms at work." There was also little gene overlap among the drugs, and no gene was present on all four lists. The expression profiles for drug resistance were similar across different ALL subtypes, indicating that the mechanism of disease resistance is independent of the molecular cause of the leukemia.
The exact level of drug resistance, whether moderate or high, also independently related to how well the patients generally responded to the drugs: The higher the resistance score, the greater the risk of relapse during treatment. After 10 years, more than 90 percent of the patients with drug-sensitive leukemia were symptomatically disease free, while only 60 percent of the patients with the most resistant cells were disease free. A second clinical trial of 98 patients, which used the same medications according to a different treatment protocol at a different institution, produced similar results.
While this study revealed some commonality among different leukemias, the discovery of Smad3 loss in T-cell ALL highlighted the fact that similar diseases can be quite different on the molecular level. NCI researchers, led by Dr. John Letterio, looked for the Smad3 protein, a key component of normal blood cells, in samples of leukemia cells collected from patients with one of several different forms of leukemia. Smad3 protein was absent in all 10 samples of childhood T-cell ALL, but present in specimens collected from patients with other leukemia subtypes, which included other forms of childhood leukemia (such as B-cell ALL) and adult T-cell leukemias (such as Sézary syndrome).
The researchers were intrigued by the biology behind Smad3's absence. The leukemia cells produced normal levels of Smad3 mRNA - the "instructions" that cells use to make protein - indicating that the Smad3 gene was turned on. Furthermore, the sequence of the Smad3 gene in patient samples was identical to the Smad3 gene found in healthy T-cells, so a genetic mutation was not the culprit. "We don't yet know the mechanisms behind this loss of Smad3 protein," said Dr. Letterio, "but two possibilities may be that protein synthesis is being blocked or that the protein is made but rapidly degraded."
Smad3 loss by itself does not lead to leukemia in mice, but spontaneous T-cell leukemia does develop when Smad3 expression is disrupted in mice lacking the gene for p27Kip1, another regulator of T-cell proliferation. The NCI researchers are now looking for other genetic and epigenetic alterations that might act along with Smad3 in the onset of T-cell ALL.