A large consortium of scientists has created a new, detailed map of the genetic mutations that have a hand in acute myeloid leukemia, a ravenous variety of blood cancer.
The work, done by the Cancer Genome Atlas Project and published in the New England Journal of Medicine on Wednesday, doesn't immediately suggest possible new avenues of treatment for the disease, but does lay the biology of AML barer than it’s ever been seen before.
Researchers peered closely at the genomes of 200 adults with AML. They took samples from both the patient’s primary tumor and from a skin biopsy – a way to pinpoint which genetic changes are likely related to leukemia, and not to the patient’s natural genetic variation. In addition to sequencing the subjects’ DNA, the authors sequenced RNA from the patients and analyzed patterns of DNA methylation – both ways of looking at how the expression of the genes encoded in the DNA are being regulated on their way to becoming proteins.
The results don't point to a single smoking gun; there doesn't appear to be a single gene that plays a primary role in causing this kind of leukemia. But that’s hardly surprising, according to coauthor Tim Graubert, an oncologist at the Washington University School of Medicine in St. Louis.
“Like every other type of cancer, AML is heterogeneous,” Graubert said in a phone interview. “It’s not one disease; it’s many.”
But in some ways, AML is strikingly different from other kinds of cancer. The cancerous cells in the leukemia patients had an average of 13 mutated genes; in breast or lung tumors, there’s usually several hundred genetic mutations associated with the diseased tissue. In addition, 23 mutated genes cropped up in a significant number of patients, and another 237 mutations were found in at least two patients.
“You can be reasonably confident that most commonly mutated genes [that drive AML] are now known,” Graubert said.
The results also put another nail in the coffin of a previous theory about AML’s biology. Some scientists had thought that most AML tumors might stem from some impairment of the tyrosine kinase signaling pathway, which can result in uncontrolled cell growth. But in the current crop of patients, 40 percent didn't have mutations in this pathway, so it’s clearly not a required step for all kinds of AML.
One new area of interest that the work points toward is genes involved in the cohesin complex, a group of proteins that helps shepherd chromosomes into the right place throughout cell division.
“An important point is that this study represents the beginning of the analysis of this data, not the end,” Graubert said.
Researchers working with the Cancer Genome Atlas Project are also publishing a big find in the journal Nature on Thursday concerning endometrial cancer – cancer that affects the lining of the uterus. The disease is the fourth most common source of malignant tumors among women in the U.S., and is expected to crop up in 49,500 new cases and cause 8,200 deaths this year.
The group analyzed DNA and RNA from the cancers of 373 women, and found that some endometrial tumors look genetically similar to certain kinds of ovarian cancer and basal-like breast cancer. That suggests drugs for those other cancers might prove useful in treating these types of endometrial cancer. The results also point toward four new subtypes of endometrial cancer, which could help steer patients toward more tailored treatments.
Currently, there are two recognized subtypes of endometrial cancer: serous, generally found in older women and harder to treat, even with intensive courses of chemotherapy; and endometrioid, generally associated with obesity and excess estrogen and usually treatable with surgery and radiation. But determining which type of cancer a woman has is a subjective process – the doctor examines the features of tumor cells and makes a judgment call.
Analyzing tumors based on genetic factors, such as the frequency of mutations in the cancer cells, could provide a more precise classification. Some mutations may also point toward treatments with existing therapies such as the breast cancer drug Herceptin.
“We are entering an era when tumors can be evaluated from a genomics standpoint, not just by looking at cancer cells under a microscope,” author Elaine Mardis, co-director of WUSTL’s Genome Institute, said in a statement. “This more comprehensive approach provides a clearer picture of the way particular endometrial cancers will behave and will be important to gynecological oncologists who treat this disease.”
SOURCE: The Cancer Genome Atlas Research Network. Ley et al. “Genomic and Epigenomic Landscapes of Adult De Novo Acute Myeloid Leukemia.” New England Journal of Medicine published online 1 May 2013; Levine et al. “Integrated genomic characterization of endometrial carcinoma.” Nature published 2 May 2013.