National Cancer Institute NCI Cancer Bulletin: A Trusted Source for Cancer Research News
June 28, 2011 • Volume 8 / Number 13

Cancer Research Highlights

Common Gene Mutation Found in Hairy Cell Leukemia

A gene mutation that occurs in melanoma and other cancers has now been detected in hairy cell leukemia (HCL), a rare cancer of the blood. The mutation, called V600E, occurs in the BRAF gene and appears to contribute to the disease, researchers reported in the June 16 New England Journal of Medicine

All 48 patients with HCL in the study had the mutation. Patients with HCL produce large numbers of abnormal B lymphocytes, a type of white blood cell. These abnormal cells appear “hairy” under a microscope and give the disease its name.

Dr. Enrico Tiacci of the University of Perugia, Italy, and his colleagues began by sequencing the complete exome (that is, all protein-coding regions of the genome) of both tumor and normal cells of a patient with HCL. This revealed five variants that were found specifically in tumor cells but not in normal cells, one of which was the V600E mutation in BRAF.

Little is known about the other four affected genes, but BRAF is often mutated in cancer, and the V600E mutation occurs frequently. The researchers then found this mutation in 47 other patients with HCL; it was, however, absent from 195 additional patients with other B-cell lymphomas or leukemias who were tested.

The presence of this mutation could serve as a diagnostic tool to help doctors distinguish HCL from cancers that have similar features, the study authors wrote. Further research, they noted, is needed to determine the frequency of the V600E mutation in a larger patient group and to clarify whether mutations in other genes may cooperate with BRAF mutations in the development of HCL.

Drugs that target the effects of BRAF mutations are in development, such as vemurafenib (PLX-4032). When the authors treated cells from HCL patients with the BRAF inhibitor PLX-4720, they observed biochemical changes that indicated a reduction in the activity of a growth-promoting pathway. These inhibitors could be tested in patients whose disease does not respond to initial treatment with other agents or who relapse, the authors said.

The study findings are both “unexpected” and “very interesting,” said Dr. Robert Kreitman of NCI’s Center for Cancer Research. “The [V600E] mutation is a potential target for treatment and diagnosis, and it may also play a role in why normal lymphocytes become hairy cells,” added Dr. Kreitman, who leads clinical trials in HCL. (See “Treating Multiply Relapsed or Refractory Hairy Cell Leukemia” and “Eliminating Hairy Cell Leukemia Minimal Residual Disease.”)

New Cancer Vaccine Shows Promise against Tumors in Mice

A series of intravenous (IV) vaccinations with a virus carrying fragments of DNA coding for normal human prostate tissue surface antigens cured cancer in about 80 percent of mice with established prostate tumors, new study findings showed.

The study, published online June 19 in Nature Medicine, was led by Dr. Richard Vile of the Mayo Clinic in Rochester, MN, and Dr. Alan Melcher of the Cancer Research UK Clinical Centre in Leeds. If the new approach can be translated from mice to humans, such vaccines could “be readily constructed for off-the-shelf use and [could] be easily delivered” intravenously, the study authors wrote.

The researchers created the cancer vaccine by inserting an extensive “library” of DNA fragments from normal human prostate cells into a mutated form of a virus called vesicular stomatitis virus (VSV). Each VSV particle carries a different DNA fragment, and each fragment holds the code for making a particular human prostate antigen, a piece of protein capable of triggering an immune response.

When the researchers injected the virus particles intravenously into mice bearing prostate tumors, the DNA fragments, known as complementary DNAs (cDNAs) that coded for surface antigens, were translated in the mice into a wide range of human prostate antigens.

Bolstered by VSV itself, which produces a strong immune response, the antigens triggered an immune reaction that selectively targeted prostate tumor cells while sparing healthy prostate tissue and other normal mouse tissues. Injecting the vaccine directly into prostate tumors was not nearly as effective as intravenous injection and produced signs of an autoimmune response in the mice.

“By expressing all of these proteins in highly immunogenic viruses, we increased their visibility to the immune system,” Dr. Vile said in a news release. “The immune system now thinks it is being invaded by the viruses, which are expressing cancer-related antigens that should be eliminated.”

The researchers found that tumors in mice were cured more effectively with nine IV injections given every other day than with a series of three or six IV injections. Tumor cells that survived an initial round of three IV vaccinations “were readily treated by second-line virus-based immunotherapy” using DNA fragments that coded for antigens on the surfaces of resistant tumor cells, the study authors wrote.

Dr. James Gulley of NCI’s Center for Cancer Research, who studies human cancer vaccines but was not involved in the current work, said that a possible strength of the new approach “is the large number of potential antigens that can be targeted,” although he said that this might also make it harder to monitor the immune response to a similar vaccine in human studies. Dr. Gulley also noted the challenges inherent in taking promising vaccine approaches from mice into the clinic.

The method described by Dr. Vile and his colleagues is just one approach being explored in the search for effective cancer vaccines in humans, Dr. Gulley added. “I don’t think it really matters in the end what initiates an immunologically relevant killing of tumor cells,” he said. “The key is to get the immune system trained to kill the tumor cells, and it may be that multiple vaccine approaches will demonstrate similar effectiveness.”

Researchers Scrutinize Genetic Variants Linked to Melanoma

Researchers have identified five genetic variants that appear to be the most promising markers of susceptibility for melanoma from a set of 745 variants that have been linked to the disease in published studies. The results of this meta-analysis, which included 145 genetic association studies in melanoma, appeared online June 21 in the Journal of the National Cancer Institute.

Each of the five variants showed a statistically significant association with melanoma at the genome-wide level and had “strong epidemiological credibility,” Dr. Alexander Stratigos of the University of Athens Medical School in Greece and his colleagues found.

To conduct the analysis, the researchers collected and catalogued all genetic association studies published in the field of melanoma, including results from recent genome-wide association studies. Working with colleagues around the world, they then conducted a systematic meta-analysis for each genetic variant identified. For each variant that showed a statistically significant association with melanoma in the meta-analysis, the authors also assessed the epidemiologic evidence supporting such an association.

The strongest associations with melanoma included variants within genes that play an important role in determining the color of skin, hair, and eyes. This finding may reflect the fact that these variants had been tested in the original candidate gene studies precisely because of their effects on pigmentation, a well-known risk factor of melanoma, the study authors noted.

Nonetheless, the confirmation of some of these variants in genome-wide association studies
“strongly supports their contribution to melanoma susceptibility,” Dr. Stratigos wrote in an e-mail. “Whether their effect on melanoma risk is exclusively through their influence on pigmentation, or also partly independent of that, cannot be assessed with our systematic meta-analysis approach,” he added.

The results provide “an integrated perspective of the accumulated evidence of genetic associations” in melanoma, the authors concluded. Their evidence is available online at the MelGene Database, which provides “a regularly updated field synopsis of genetic association studies performed in melanoma,” according to the site.

Barrett Esophagus May Progress to Cancer Less Often than Thought

The risk of Barrett esophagus (BE) developing into esophageal cancer may be lower than previously thought, according to one of the largest studies of its kind. The study was conducted in Ireland, where the clinical parameters for diagnosing BE are different than they are in the United States. As a result, the implications of these results for care in the United States are unclear. The study was published online June 16 in the Journal of the National Cancer Institute.

Esophageal cancer incidence has been rising dramatically in the United States, with an increase of 300 to 400 percent in the last 30 years. BE is a strong risk factor for the development of esophageal cancer. Current clinical guidelines advise that people who have BE without evidence of significant cellular aberrations, known as dysplasia, undergo an endoscopy every 3 to 5 years, with more frequent endoscopies in people whose BE has more significant dysplasia.

In this study, researchers followed more than 8,500 BE patients who were listed in the Northern Ireland Barrett Esophagus Registry. After a mean follow-up of 7 years, 79 participants had been diagnosed with esophageal cancer, 16 with cancer of the gastric cardia (the junction between the esophagus and stomach), and 36 with high-grade dysplasia, meaning that the esophageal tissue had severe precancerous changes. The combined incidence of all three outcomes was 0.22 percent per year, lower than what has been seen in previous studies. Consistent with previous studies, cancer risk was higher among men and in patients over 50 years of age.

In addition to showing a lower risk of progression than previously seen, the study “makes it possible to more accurately determine risk in subgroups and to estimate changes in risk of progression over time,” wrote the study’s lead investigator, Dr. Shivaram Bhat of Queen’s University Belfast, and his colleagues.

However, only 46 percent of the BE patients in the registry had changes in the esophageal lining (identified during an endoscopy) known as specialized intestinal metaplasia (SIM), a condition which is used to diagnose BE in the United States. The remaining patients had been diagnosed with BE based on the presence of less severe changes to the esophageal lining, determined via a biopsy. The combined incidence of progression from BE to esophageal cancer, gastric cardia cancer, or high-grade dysplasia in patients with SIM was 0.38 percent per year. Patients without SIM at their first biopsy had a much lower incidence of progression, just 0.07 percent per year.

Diagnosing BE in patients without SIM “could meaningfully bias the results,” wrote Dr. Douglas Corley of the Kaiser Permanente Northern California Division of Research in an accompanying editorial. In the patients with SIM, he noted, the incidence of progression to cancer was only slightly lower than prior estimates. “The current study…provides general support for many paradigms underlying current thoughts about the management of Barrett esophagus,” explained Dr. Corley. Further studies will be necessary to determine “whether surveillance or treatment actually decrease cancer deaths and the development of better techniques for risk stratification.”

The conflicting diagnostic criteria for BE “makes it difficult to compare these results with other studies,” agreed Dr. Asad Umar of NCI’s Division of Cancer Prevention. The findings, he continued, “need to be analyzed with caution, as the practice of endoscopic and pathologic evaluations varies greatly among practices and research groups.”