Cancer Research Highlights
MicroRNAs Linked to Trastuzumab Resistance, Brain Metastases
Researchers have tied increased levels of a microRNA (miRNA) to resistance to the targeted therapy trastuzumab (Herceptin) in women with HER2-positive breast cancer. Another research team has discovered a “signature” of miRNAs in brain metastases in patients with melanoma—a signature that is also present in the primary tumor and could identify melanoma patients at increased risk of brain metastases. The results from both studies were presented yesterday at the AACR annual meeting.
In the first study, a team from the University of Texas M. D. Anderson Cancer Center showed that HER2-positive breast cancer cell lines with experimentally increased miR-21 levels were resistant to trastuzumab compared with cell lines with lower miR-21 levels, according to Sumaiyah K. Rehman, a graduate student in the lab of Dr. Dihua Yu, who presented the findings at a press briefing. Tumor samples from patients with HER2-positive tumors that were resistant to treatment with trastuzumab had elevated miR-21 levels, Rehman explained.
In both cell lines and mouse models of breast cancer, inhibition of miR-21 levels increased the sensitivity to trastuzumab. Lower levels of miR-21 were also associated with decreased expression of the tumor-suppressor gene PTEN, which previously has been linked to trastuzumab resistance. The findings suggest that expression of miR-21 and PTEN could be “biomarkers of who may or may not respond to trastuzumab treatment,” Rehman said.
In the second study, researchers from the New York University School of Medicine analyzed samples of metastatic tumors from a cohort of 59 patients with melanoma and identified a signature of 7 miRNAs present at different levels in metastases to the brain as compared with metastases to other sites in the body. They were able to validate this signature in tumor samples from a separate cohort of 36 patients, reported lead investigator Dr. Eva Hernando. A smaller signature involving some of these same RNAs was also seen in the primary tumors of these same patients.
Cell line and mouse model studies also showed that silencing one of the miRNAs in the signature, miR-199-3p, was associated with the ability of melanoma cells to adhere to endothelial cells and migrate to the brain.
Establishing such an miRNA signature “could allow us to identify [melanoma] patients at higher risk of brain metastases,” Dr. Hernando said, meaning that these patients could be identified for intensified surveillance and be enrolled in clinical trials. Some of the implicated miRNAs also “could be targets for novel therapeutic attacks,” she said.
Diabetes Drug Metformin Prevents Lung Tumors in Mice
The diabetes drug metformin helped prevent tumors in mice that were exposed to a cancer-causing agent found in tobacco, said researchers at the AACR annual meeting. Compared with untreated mice, those that received the drug had a 53 percent reduction in lung tumor burden after exposure to the carcinogen, which is called nicotine-derived nitrosamine. The animals were treated with an oral form of metformin for 13 weeks at drug levels that would be achievable in humans, the researchers said.
“Metformin is a very interesting drug for cancer prevention,” said lead researcher Dr. Phillip A. Dennis of NCI’s Center for Cancer Research, who presented the results. “We prevented over half of the lung tumors that would have occurred from exposure to the main carcinogen in tobacco, and that’s a real and important reduction.” Dr. Dennis’ group is planning a clinical trial to test the FDA-approved drug in people at highest risk of developing lung cancer, he said.
When the researchers administered metformin to mice by injection, the drug levels were higher and the anticancer effects were even stronger—the drug reduced lung tumors by 72 percent compared with no treatment. Although this method of administration is not feasible for cancer prevention in humans, analogs of metformin that are more potent may well be more effective than the drug itself, Dr. Dennis noted. With any chemopreventive agent, however, minimal toxicity is critical, and metformin was well tolerated in the mice. In fact, the livers of the treated mice not only showed no signs of toxicity, but they actually appeared healthier than those of untreated mice.
This may be the first study to show the potential for metformin in preventing lung cancer, but preclinical studies in mouse models of breast cancer have also yielded positive results. In addition, epidemiological studies of hundreds of thousands of patients with type 2 diabetes have shown that those who take metformin have a lower risk of many types of cancer. Together, these results provide a rationale for studying the role of this drug in cancer in a variety of clinical settings, the researchers noted.
Cancer Risk Tools May Need Regular Recalibration
A new study of a statistical tool for assessing a woman’s risk of breast cancer—commonly known as the Gail model—suggests that such tools may need to be updated regularly to account for underlying trends in the population that may influence results, researchers reported online in the Journal of Clinical Oncology on April 5.
This validation study of the Gail model found that the tool underestimated the number of invasive breast cancers among women in two large population-based cohort studies. This meant that a small percentage of women did not meet the recommended threshold for using chemoprevention drugs such as tamoxifen or raloxifene, even though the women would have qualified based on a recalibrated model. The cohort studies were the NIH-AARP Diet and Health Study and the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial.
The researchers, led by Dr. Ruth M. Pfeiffer of NCI’s Division of Cancer Epidemiology and Genetics, attributed the underestimation primarily to differences in the breast cancer rates that were used to develop the Gail model (SEER data covering 1983 to 1987) and population rates for the time periods covered by the cohort studies used in the validation (1995 through 2003 for NIH-AARP and 1993 through 2006 for PLCO).
Breast cancer rates in the United States rose steadily in the 1990s, leveled off in about 2000, and then dropped substantially starting in 2003, when the use of menopausal hormone therapy was linked to breast cancer risk. The performance of the model improved greatly after the model was updated with population rates of invasive breast cancer in SEER data for 1995 to 2003, the researchers noted. The Gail model was well calibrated when the validation was restricted to PLCO data for the years 2003 to 2006.
Many women and health care professionals use a publicly available version of the Gail model called the Breast Cancer Risk Assessment Tool (BCRAT) to inform decisions about breast cancer prevention. This tool, which takes into account a woman’s medical and reproductive history as well as her family history of the disease, is being updated using incidence data from the SEER program and will continue to be updated regularly in the future, the researchers said. As current SEER incidence rates are similar to the original rates included in the model, the updating will not substantially affect the current BCRAT outputs but may play a more important role in the future.
“Our study shows that to ensure that a risk model is continuously useful for clinical decisions, the model has to be updated on a regular basis,” said Dr. Pfeiffer. “The take-home message is that to do good risk prediction, we have to watch out for trends in the population over time that could affect the model.”
Profiling Airway Cells May Identify Smokers at Highest Risk of Cancer
Researchers have found that improper activation of a signaling pathway in the airway cells of some smokers and former smokers is an early step in the development of lung cancer. In addition, the researchers identified a compound that inhibited the activation of this pathway, called the PI3K pathway, in some individuals with precancerous lesions. A team led by Drs. Avrum Spira of the Boston University College of Medicine and Andrea Bild of the University of Utah reported their findings in the April 7 Science Translational Medicine.
In previous work, these researchers showed that profiling the activity of 80 genes in cells from a patient’s airway could help distinguish smokers with and without lung cancer. Although these airway cells, collected during bronchoscopy, may appear normal under the microscope, they may also show abnormal patterns of gene activity associated with lung cancer. Toxins in cigarette smoke can alter the behavior of genes in cells throughout the respiratory tract, and these changes can be used for diagnostic purposes, the researchers said.
After identifying the activation of the PI3K pathway, which plays a role in a number of cancers, the researchers asked whether a drug was available to counter this effect. In a study of a small number of smokers with bronchial dysplasia, they found that treatment with myoinositol—a compound naturally found in some foods that has shown anticancer effects in mice—caused the regression of these precancerous lesions in the majority of subjects. The activity of the PI3K pathway was decreased in subjects whose lesions regressed but not in those subjects whose lesions were unchanged.
“These results suggest that deregulation of the PI3K pathway in the bronchial airway epithelium of smokers is an early, measurable, and reversible event in the development of lung cancer and that genomic profiling of these relatively accessible airway cells may enable personalized approaches to chemoprevention and therapy,” the researchers concluded.
Larger studies are needed to confirm the role of the PI3K pathway in early lung cancer and to evaluate myoinositol in humans. NCI has an ongoing phase II clinical trial of myoinositol open to participants with bronchial dysplasia. The study will assess the compound’s ability to reverse precancerous changes in the lung, measure markers of lung cancer and lung cancer risk, and establish a safety profile.
Peptide Helps Chemotherapy Penetrate Deep into Tumors
Many anticancer drugs can only penetrate a few cells away from the blood vessels that carry them into a tumor, limiting their ability to kill cells that lie deeper. In a proof-of-concept study, researchers led by Dr. Kazuki Sugahara from the University of California, Santa Barbara, showed that a tumor-penetrating peptide called iRGD, injected along with several common chemotherapies, greatly increased the amount of drug that reached tumor cells in mice. Their results were reported April 8 in Science.
iRGD binds to proteins found in the lining of tumor blood vessels but not in blood vessels found in normal tissue. Upon binding, iRGD appears to trigger an active transport process that drives drugs into the tumor tissue from the blood stream.
The researchers injected iRGD along with chemotherapy drugs in mice bearing human tumor xenografts. The combination increased accumulation of nanoparticle-embedded paclitaxel in breast cancer cells 12-fold, of doxorubicin in prostate cancer cells 7-fold, and of trastuzumab in breast cancer cells 40-fold, compared with accumulation in tumor cells when the chemotherapy drugs were given alone. These increases were observed even though iRGD was not chemically attached to the drugs.
The addition of iRGD also increased drug potency. For example, 1 mg/kg of doxorubicin plus iRGD injected intravenously was as effective at eradicating tumors in mice as 3 mg/kg doxorubicin given without iRGD, and it did not increase toxicity.
“Coadministration of iRGD may be a [way] to enhance the efficacy of anticancer drugs while reducing their side effects, a primary goal of cancer therapy research,” concluded the authors. They caution that iRGD has not yet been tested for safety or efficacy in people.
Also in the Journals: Treating Childhood Cancers—How to Build on Success
A new analysis of childhood cancer statistics estimates that 38,000 cancer-related deaths in children were averted in the United States between 1975 and 2006, largely thanks to new drugs, improved treatment strategies, and past investments in clinical trials. But clinical research on these diseases is at a crossroads, the study authors cautioned in the April 19 Journal of Clinical Oncology. The era of steady improvements in outcomes by optimizing the delivery of conventional therapies is coming to a close, they wrote, while the era of targeted therapeutics is just beginning.
“Our progress has slowed, and now we need a deeper understanding of the biology of childhood cancers and the genes and pathways that drive them,” said lead author Dr. Malcolm Smith of NCI’s Cancer Therapy Evaluation Program in a statement. “With this understanding, we can identify susceptibilities that can be therapeutically exploited so that we continue moving forward toward the goal of curative therapy for every child diagnosed with cancer.”
The promise of this strategy has been demonstrated by early results such as those from the Childhood Cancer TARGET Initiative, an NCI-supported project that brings together investigators with expertise on childhood cancers and genome analysis, the authors noted.