Sorafenib Increases Risk of High Blood Pressure

A meta-analysis published online January 24 in Lancet Oncology reports that patients receiving the standard clinical dose of sorafenib (Nexavar), an anticancer drug that targets the growth of tumor blood vessels, or angiogenesis, have a significantly increased incidence of hypertension.

Previous studies have found an increased risk of hypertension with the use of other targeted drugs and antiangiogenesis agents, including sunitinib (Sutent). To see if sorafenib also increases this risk, and subsequent risk of heart attacks and other serious cardiac events, the authors combined clinical trials data from three published papers and six meeting abstracts that included data on hypertension for patients assigned a starting dose of 400 mg of sorafenib twice daily, which is the current starting dosage approved by the FDA.

From these 9 studies, 4,599 patients were available for analysis; 3,567 received sorafenib and the other 1,032 served as controls. Data on all grades of hypertension were available from 3,363 patients enrolled in 7 trials. The overall incidence of all-grade hypertension was 23.4 percent in patients receiving sorafenib. Patients taking sorafenib were approximately six times as likely to develop any grade of hypertension as the control patients.

Data on high-grade hypertension, which often requires treatment with multiple antihypertensive drugs or discontinuation of sorafenib, were available from all 3,567 patients in the 9 trials who received the drug. The overall incidence of high-grade hypertension was 5.7 percent for these patients.

"The hypertensive and cardiovascular side-effects of sorafenib need thorough postmarketing surveillance and reporting, and future studies will be needed to identify the mechanism and appropriate treatment of sorafenib-induced hypertension," conclude the authors. In particular, they explain, further studies need to identify the classes of antihypertensive medications that can be safely given in tandem with sorafenib to minimize potential drug-drug interactions and possible interference with the efficacy of sorafenib.

Clinical Outcomes in Colon Cancer Linked to microRNA Gene

Colon tumors that produced high expression levels of a microRNA gene called miR-21 were associated with poor survival and therapeutic outcome in two patient populations, one in the U.S. and the other in China, according to a study in the January 30 Journal of the American Medical Association.

Like other microRNA genes, miR-21 produces short strands of RNA that control multiple genes, including some involved in cancer. The findings support a growing view that this gene, which shows increased activity in more than a dozen major cancers, may be a useful biological marker and therapeutic target.

"We are interested in this gene in part because our results may have implications for other cancers," said Dr. Curtis Harris of NCI's Center for Cancer Research (CCR), who led the study. "And a therapy directed against miR-21 may be effective against a variety of cancers." He cautioned that the results are preliminary and potential therapies may be years away.

The researchers first profiled microRNA gene activity in colon tumors and matched normal tissues from 84 patients from Maryland, and they validated the results in 113 patients from Hong Kong. The most striking finding was the association between high miR-21 activity in tumors and poor survival and therapeutic outcome in patients receiving adjuvant chemotherapy based on 5-fluorouracil. The associations were consistent even though slightly different techniques were used to assess microRNA activity in each group.

The results also indicate that microRNA gene activity is systematically altered in colon tumors compared to normal tissues. "There are systematic changes in microRNA expression during the cancer process," said first author Dr. Aaron Schetter of NCI's CCR.

Protein in Breast Tumors Does Not Predict Chemotherapy Benefit

High levels of the protein Ki-67 in breast tumors did not predict which women in two clinical trials would benefit from chemotherapy added to endocrine (antiestrogen) therapy. Some studies have suggested that the protein, which is associated with cell proliferation, might be used to identify women with early-stage disease who may benefit from adjuvant chemotherapy after surgery and endocrine therapy.

To test this idea, Dr. Giuseppe Viale of the University of Milan and his colleagues in the International Breast Cancer Study Group retrospectively assessed Ki-67 expression in tumors from patients involved in two randomized trials - each of which compared endocrine therapy given after chemotherapy with endocrine therapy alone. The analysis included only women with early-stage, lymph node-negative disease.

Although high levels of Ki-67 expression did not predict a chemotherapy benefit, they were prognostic of poor survival. "Other biomarkers will be required to define which patients with endocrine-responsive, node-negative early breast cancer would benefit from the addition of adjuvant chemotherapy to endocrine therapy," the researchers conclude in the February 6 Journal of the National Cancer Institute (JNCI).

The study supports the long-held position of the American Society of Clinical Oncology Tumor Marker Expert Panel that measurements of the cell cycle, such as Ki-67 expression, should not be used to make decisions about chemotherapy.

In an accompanying editorial, Dr. Matthew Ellis of Washington University in St. Louis says that biomarker studies with negative results that can exclude a positive interaction are important to publish because "clinging to a long-favored but incorrect hypothesis in the face of negative evidence impedes scientific and clinical progress."

Survey Highlights Threat of Global Tobacco-Related Mortality

Results of a new World Health Organization (WHO)-led survey of youths ages 13 to 15 suggest that the already alarmingly high number of tobacco-related deaths estimated to occur annually by 2020 may be too low, officials from the Centers for Disease Control and Prevention (CDC) and colleagues report.

By 2020, global tobacco-related deaths are estimated to double from 5 million a year to 10 million. Approximately 70 percent of these deaths will occur in developing countries.

But the results of the most recent Global Youth Tobacco Survey (GYTS) - an international, school-based survey conducted in 140 WHO member nations and other sites - suggest the situation may be worse than previously thought, wrote Dr. Charles W. Warren and colleagues in the January 25 Morbidity and Mortality Weekly Report.

Among the findings are a higher prevalence of young female smokers compared with adult females, the "high susceptibility" to smoking in the next year reported by never-smokers, and elevated levels of exposure to secondhand smoke.

Other key findings included the similar prevalence of cigarette and other tobacco product use and the high percentage of respondents who had recently bought tobacco products at a store without being refused service. Seventy percent of respondents who smoke expressed an interest in quitting, while 80 percent of all respondents supported a public ban on smoking.

"The findings in this report suggest that interventions that decrease tobacco use among youth (e.g., increasing excise taxes, media campaigns, school programs in conjunction with community interventions, and community interventions that decrease minors' access to tobacco) must be broad-based, focused on boys and girls, and have components directed toward prevention and cessation," the authors wrote.

Expanding the Search for Cancer Drug Targets

Two studies in the February 1 Science describe a genome-wide strategy for discovering "essential" genes in cancer cells that are toxic to the cells when silenced. These genes may lack the hallmarks of cancer genes, such as mutations or increased activity; however, they are essential for cell survival and proliferation, making them potential drug targets.

The strategy, described in the first study, exploits a natural defense mechanism in cells that blocks invading viruses by degrading viral RNA. Researchers now use this mechanism, known as RNA interference, as a way to target and silence genes within the cell. But technical constraints have prevented researchers from doing this type of screen on a genome-wide scale.

During the last 6 years, Drs. Stephen J. Elledge of Harvard Medical School, Gregory Hannon of Cold Spring Harbor Laboratory, and their colleagues have been working to change that. Among other advances, the researchers developed libraries of short hairpin RNA molecules that can silence every gene in the human and mouse genomes. They also created a system for using molecular tags, or bar codes, to determine which genes were toxic to cancer cells when deleted from the cells.

As reported in the second study in Science, the researchers screened several thousand genes in colon and breast tumor cells, as well as in normal cells - using some 10,000 short hairpin RNAs. They identified a few dozen genes that, when silenced, impaired the growth of cancer cells but not normal cells. The researchers believe the results are preliminary but promising.

The strategy could complement The Cancer Genome Atlas and other large-scale efforts to catalog genetic differences between tumor cells and normal cells, they continued. Specifically, the screens could reveal potential therapeutic targets that might otherwise be missed.

"I've always felt that the focus on oncogene addiction in developing cancer drugs was missing a lot of other cancer drug targets," says Dr. Elledge, referring to the phenomenon where tumors become dependent on certain cancer-causing genes. "Cancer cells seem to depend on a lot of genes that are not oncogenes, but they are important nonetheless. We call this non-oncogene addiction."

His team is preparing to apply the strategy against the entire human genome. At the same time, they hope other researchers will begin to apply the tools and the methods to many different types of cancers. Now that the resources have been developed, carrying out the experiments is relatively cost effective, the researchers note.

Study Finds No Link Between Hormones, Prostate Cancer

An analysis of the original data from 18 prospective studies indicates that prostate cancer risk is not influenced by levels of certain circulating sex hormones in the blood, reports an international research team.

Published early online January 29 in JNCI, the analysis pooled data from studies that included 3,886 men with prostate cancer and 6,438 controls. Contrary to what has long been suspected but never definitively established, no association was found between prostate cancer risk - regardless of stage or grade - and levels of various androgens as well as estradiol, a form of estrogen.

Further, reported Dr. Andrew Roddam of the University of Oxford and colleagues from the Endogenous Hormones and Prostate Cancer Collaborative Group, "there was no evidence of interactions between concentrations of any of the hormones considered and risk of prostate cancer." There was, however, a slight reduction in the risk of low-grade, localized cancer associated with circulating levels of sex hormone-binding globulins (SHBGs).

"The factors that promote localized or nonaggressive prostate cancer may be entirely different than those that promote fast-growing, high-grade prostate cancers," wrote Dr. Paul Godley and colleagues from the University of North Carolina at Chapel Hill, in an accompanying editorial. "Because most of the cancers included in these studies were localized and low-grade, pooling data from many studies allowed a sample size that was large enough to perform subanalyses of advanced and high-grade cancers, and the power to investigate interactions among hormones and [SHBGs]."

The study results suggest that "perhaps it is time for research efforts to focus instead on developing more sophisticated hypotheses and novel study designs," the editorial concluded, which can be facilitated by NIH and NCI programs already in place, including the NIH Roadmap and biospecimen storage systems.