Cancer Research Highlights

High Breast Density Does Not Predict Death among Breast Cancer Patients

High breast density is a strong risk factor for developing breast cancer, but it does not affect a breast cancer patient’s risk of death, according to a study recently published in the Journal of the National Cancer Institute.

Denser breasts have more glandular tissue (cells that produce milk during lactation) and supportive connective tissue than fatty tissue. Doctors use a scale called the Breast Imaging Reporting and Data System (BI-RADS) to classify breast density as observed on mammograms on a scale from 1 to 4, with 1 being the least dense and 4 the most dense.

To examine the relationship between breast density and risk of death from breast cancer, Dr. Gretchen Gierach and her colleagues from NCI’s Division of Cancer Epidemiology and ­­­Genetics and the NCI-sponsored Breast Cancer Surveillance Consortium (BCSC) examined medical records from more than 9,000 breast cancer patients collected by the BCSC.

After accounting for age, body mass, treatment, and other factors that could influence the risk of death, the researchers found that among women diagnosed with breast cancer, women with dense breasts were not more likely to die of the disease or of other causes than women with less-dense breasts during nearly 7 years of follow-up, on average.

Unexpectedly, the researchers observed that breast cancer patients with the least-dense breasts had an increased risk of death from breast cancer if they had large tumors or were obese. However, given that this result was based on relatively small numbers of women and has not been previously suggested by other studies, “these findings need to be replicated in larger studies,” said Dr. Gierach.

Obesity is a risk factor for death from breast cancer and is also inversely related to breast density. (That is, obese women are less likely to have dense breasts.) Therefore, obesity could affect associations between breast density and breast cancer death.

“We already know that obesity is a poor prognostic factor for breast cancer in general, but this particular analysis showed that the subgroup of women who were obese and had less-dense, fatty breasts were at greatest risk,” she explained. “Our hypothesis is that the fat content in the breast might be enhancing obesity-related mechanisms that heighten tumor aggressiveness in breast cancer. We are conducting studies to better understand the biology of breast density.”

This research was supported in part by grants from the National Institutes of Health (U01CA63740,U01CA86076, U01CA86082, U01CA63736, U01CA70013, U01CA69976, U01CA63731, U01CA70040, and HHSN261201100031C).

Flavored Cigar Use Common among Young Adult Cigar Smokers

Flavored cigar use is common among adults who smoke cigars, and particularly high among young adult cigar smokers, according to the first nationwide survey to assess adult use of these products. Flavored cigar smoking also varies by geographic region, the study found.

Dr. Brian King and colleagues from the Centers for Disease Control and Prevention (CDC) reported the findings online August 27 in Nicotine & Tobacco Research.

Using data from the 2009–2010 National Adult Tobacco Survey, which included about 119,000 landline and cell phone users, the researchers estimated that 6.6 percent of adults in the United States smoke cigars. Cigar smoking was especially high (15.5 percent) among 18 to 24 year olds. Approximately 43 percent of adults who smoke cigars use flavored cigars, they found. Among 18- to 24-year-old cigar smokers, 57 percent use flavored cigars.

Flavored cigar use was also more common among certain groups, including women and Hispanic cigar smokers, as well as cigar smokers with less education and lower incomes. The highest rates of flavored cigar use among cigar smokers were in North Dakota (71.6 percent) and New Mexico (69.0 percent), and the lowest rates were in New Hampshire (11.1 percent) and New Jersey (23.7 percent).

In 2009, the U.S. Food and Drug Administration (FDA) prohibited the use of certain flavors in cigarettes—such as vanilla, chocolate, cherry, and others that have a distinguishable taste or aroma—under authority granted by the Family Smoking Prevention and Tobacco Control Act. However, the FDA does not currently regulate cigars, and flavored cigars may still be manufactured and sold.

Flavorings, the study authors said, “mask the natural harshness and taste of tobacco.” Cigars contain many of the same toxic substances as cigarettes and smokeless tobacco and raise the risk of several cancers, heart disease, and chronic obstructive pulmonary disease.  

Given the high rate of use among cigar smokers, the authors concluded, “efforts to curb flavored cigar smoking have the potential to reduce the prevalence of overall cigar smoking among U.S. adults.”

Gene in Stem Cell Donors May Decrease Relapse Risk in Leukemia Patients

An analysis of patients who had undergone blood stem cell transplants for acute myelogenous leukemia (AML) suggests that certain genetic features of the donors are related to risk of AML relapse. The findings of this retrospective study show that stem cell transplants from genetically similar donors who carry a gene known as KIR2DS1, which can activate cancer-fighting natural killer (NK) cells, appear to be associated with a reduced risk of relapse in AML patients.

The study, by Drs. Katharine Hsu and Bo Dupont of Memorial Sloan-Kettering Cancer Center and their colleagues, appeared August 30 in the New England Journal of Medicine.

KIR genes code for members of a group of proteins called killer-cell immunoglobulin-like receptors, which are found on the surface of NK cells—white blood cells that can kill tumor cells.

Researchers also found that the reduced relapse rate associated with KIR2DS1 was restricted to stem cell donors with specific variants of the gene HLA-C. HLA genes code for cell-surface HLA proteins, which make up a person’s tissue type.

Hematopoietic stem cell transplants (HSCT) are given to patients who have AML to replace the diseased cells in the bone marrow with healthy stem cells that can develop into all types of blood and immune system cells to fight the leukemia. When selecting donors for HSCT, physicians match the HLA tissue type of donor and recipient to reduce the chances that the recipient’s new immune system will damage his or her other tissues and organs.

 “Studies over the last 15 to 20 years have pointed to the fact that the NK cells that emerge after transplant are quite potent in destroying any remaining leukemia cells, particularly in AML,” Dr. Hsu said.

“We know that NK cells are inhibited or activated by the various KIR receptors and their interactions with HLA molecules,” she continued. “The real goal is to figure out which KIR-HLA combinations are going to allow the donor-derived NK cell to have its maximal effects in terms of leukemia control.”

The researchers examined the outcomes of HSCT in 1,277 patients with AML and 427 patients with acute lymphoblastic leukemia (ALL) who had received transplants between 1989 and 2008 from unrelated donors. The donors and recipients were matched for at least 9 of 10 possible versions of the five HLA genes. The researchers used stored blood and DNA samples to test for various types of KIR genes in stem cell donors and for HLA genes in donors and recipients.

They found that patients with AML (but not those with ALL) who received transplants from donors carrying the KIR2DS1 gene had a significantly lower risk of relapse than patients with transplants from donors without KIR2DS1. But KIR2DS1 did not improve outcomes when the donor had two copies, or alleles, of the HLA-C2 gene. This last finding is consistent with studies showing that high levels of HLA-C2 protein reduce the activity of NK cells bearing KIR2DS1 receptors.

Dr. Hsu commented that future studies should examine the effects of other KIR-HLA interactions or combinations in HSCT. “As we do this, we can become more and more refined in our donor selection criteria, with the goal of selecting a donor who is going to give the greatest benefit in terms of reducing leukemic relapse and increasing survival.”

This research was supported in part by grants from the National Institutes of Health (U01 AI69197, KL2 RR024997, R01 HL088134, and P01 CA23766).

Sequencing "Outlier" Genome Suggests Some Patients May Benefit from Cancer Drug

Clinical trials that fail to identify drugs that benefit most patients may still yield valuable information by identifying subsets of patients who would benefit from the drugs. That is the conclusion of a study in which researchers used whole-genome sequencing to salvage a potentially beneficial drug that might have been discarded.

In a report published August 23 in Science, researchers at Memorial Sloan-Kettering Cancer Center describe how they sequenced the tumor genome of a 73-year-old woman with advanced bladder cancer who had a complete response to the drug everolimus (Afinitor) that has lasted for more than 2.5 years. The patient was part of a 45-patient, early-phase clinical trial in which treatment with everolimus failed to improve progression-free survival, the trial’s primary endpoint, among the patient population enrolled in the trial as a whole.

“The patient was a dramatic outlier in terms of her clinical response,” said the study’s senior author, Dr. David Solit.

After targeted sequencing of a few specific genes in the woman’s tumor “didn’t turn up anything revealing,” Dr. Solit continued, the researchers decided to sequence the woman’s entire tumor genome to see if they could discover a molecular basis for her strong response.

Among the many genomic alterations the researchers identified, inactivating mutations in two genes, TSC1 and NF2, stood out. Evidence from laboratory studies had suggested that loss-of-function mutations in TSC1 and NF2 might increase sensitivity to everolimus.

The researchers then analyzed the tumor DNA from 13 other patients enrolled in the trial and found TSC1-inactivating mutations in the tumors of three more patients, two of whom had measurable tumor shrinkage and lived longer without their cancer progressing than patients whose tumors lacked a TSC1 mutation. None of the 13 patients’ tumors had an NF2 mutation.

Sequencing the TSC1 and NF2 genes in tumor samples from 96 other patients with advanced bladder cancer revealed five more patients with TSC1 mutations, but no NF2 mutations.

Dr. Solit and his colleagues hope to launch a small clinical trial testing everolimus in patients with bladder cancer tumors that have mutations in TSC1 or a related gene, TSC2.

“We often see cases where just a few patients in an otherwise negative trial have significant benefit,” Dr. Solit said. “With the in-depth analysis provided by novel technologies like whole-genome sequencing, we can possibly salvage potentially beneficial drugs that were otherwise going to fail.”

Researchers Use Gene Deletions to Find Cancer Treatment Targets

Chromosomal damage that can transform healthy cells into cancer cells may also create weaknesses that can be exploited to kill the cancer cells, a new study suggests. The idea, called “collateral vulnerability,” could be used to identify new targets for drug therapy in multiple cancers, according to researchers from the Dana-Farber Cancer Institute and the University of Texas MD Anderson Cancer Center. The study was published August 16 in Nature.

Directly targeting genetic mutations that drive cancer with drugs is difficult, particularly in the case of mutations that delete tumor suppressor genes. Using data on the brain cancer glioblastoma multiforme (GBM) from The Cancer Genome Atlas (TCGA) initiative, the research team identified a number of “collateral” or “passenger” gene deletions that occurred during chromosomal damage that resulted in the loss of tumor suppressor genes.

The researchers next looked for passenger gene deletions that met two criteria: the deleted genes were involved in functions vital to cell survival, and the deleted genes were closely related to existing genes that perform similar functions. This loss of redundancy caused by passenger gene deletions can potentially be exploited to selectively kill tumor cells, the authors explained.

One gene that met these criteria is ENO1. ENO1 produces enolase 1, an enzyme that plays a central role in a process cells use to make energy. Human cells have a closely related gene (ENO2) that produces the enzyme enolase 2, which acts as a back-up for enolase 1 in brain tissue. Brain cells normally have a high level of enolase 1 activity and a small amount of enolase 2 activity. In some patients with GBM, however, the tumor cells lack enolase 1 activity because ENO1 was deleted when a tumor suppressor gene was deleted. This lack of enolase 1 activity could make these tumor cells more vulnerable to enolase inhibition.

This idea was tested using two targeting strategies. First, in GBM cell lines that lacked ENO1, the investigators showed that silencing ENO2 gene expression with a short hairpin RNA (a short RNA sequence that blocks the production of enolase 2 protein from ENO2 messenger RNA) sharply reduced cell growth, and tumors failed to form in mice injected with the treated cells.

The second approach involved a drug that targets the enolase 1 and enolase 2 proteins. Treatment of GBM cell lines lacking ENO1 with the drug caused the cancer cells to die because of the low overall enolase levels in these cells. But drug treatment had little effect on normal brain cells or GBM cells that had ENO1, since these cells have high levels of ENO1 gene expression and are, therefore, less sensitive to the drug.

The collateral vulnerability concept is similar in some respects to the idea of synthetic lethality, which uses genetic mutations in cancer-associated genes to identify other potential cellular vulnerabilities, explained the study’s co-lead author, Dr. Florian Muller of MD Anderson.

There are many more passenger gene deletions than tumor suppressor gene deletions, “and some of these passenger-deleted genes perform functions critical for cell survival,” Dr. Muller continued. “So, by expanding the concept to passenger genes, we vastly expand the possibility of finding these relationships, and, in the case of essential-redundant gene pairs like ENO1 and ENO2, we also provide a rational, knowledge-based method of drug-target discovery.”

The researchers are extending their work to other passenger gene deletions in GBM, Dr. Muller said.

This research was supported in part by the National Institutes of Health (CA95616-10 and CA009361).