National Cancer Institute NCI Cancer Bulletin: A Trusted Source for Cancer Research News
July 13, 2010 • Volume 7 / Number 14

Cancer Research Highlights

Adherence Low among Women Taking Adjuvant Hormone Therapy for Early-stage Breast Cancer

Also in the Journals: ASCO Recommends Using an Aromatase Inhibitor in Adjuvant Breast Cancer Therapy

Updated guidelines from the American Society of Clinical Oncology (ASCO) recommend prescribing an aromatase inhibitor (AI) to women with hormone receptor-positive breast cancer at some point during their treatment. Treatment with an AI can last as long as 5 years, the guidelines explain, and AIs can be used in several ways: as initial adjuvant therapy, following 2 to 3 years of treatment with tamoxifen, or after 5 years of tamoxifen. The updated guidelines were published online July 12 in the Journal of Clinical Oncology

“Our panel carefully reviewed the explosion of research that has emerged in the past 5 years on anti-estrogen drugs, and filled in gaps in our understanding of how best to use these newer treatments, and what the trade-offs and side effects of therapy would be,” said Dr. Harold Burstein, chair of the expert committee assembled by ASCO to update the guidelines, in a news release.

Findings from one of the largest studies of its kind to date indicate that many women, particularly younger women, with early-stage breast cancer stop taking adjuvant hormone therapy before completing the 5-year regimen. The results were published online June 28 in the Journal of Clinical Oncology (JCO). In addition, an as-yet-unpublished analysis from the same study presented last month at the American Society of Clinical Oncology (ASCO) annual meeting indicates that failure to complete the treatment is associated with worse survival.

The findings published in JCO found that by 4.5 years after they began taking tamoxifen or one of three aromatase inhibitors, 32 percent of the nearly 8,800 women in the study had stopped taking their treatment. And among those who continued treatment, only 72 percent adhered to their prescribed regimen (that is, took most of their medication as prescribed) for the entire 4.5-year study period.

The study included women who are part of the Kaiser Permanente of Northern California (KPNC) health system, an integrated managed care program. The research team identified subjects via the KPNC cancer registry and tracked adjuvant therapy use via prescriptions filled through KPNC’s pharmacy information management system. Discontinuation was defined as going 180 days without filling a prescription, and nonadherence was defined as taking less than 80 percent of a given prescription.

Overall, only 49 percent of women completed the treatment regimen as prescribed. While women younger than 40 and older than 75 were the most likely to be nonadherent, the 202 women younger than 40 had the worst track record; they were 50 percent more likely to discontinue treatment and 40 percent more likely to be nonadherent than the rest of the study population.

Other studies have pointed to problems with discontinuation and adherence in women receiving adjuvant hormonal therapy. Nevertheless, given its “proven track record of reducing breast cancer recurrence,” the study authors were surprised by the extent of nonadherence, particularly in younger women. “Perhaps we need to do a better job of making patients aware that, to get the full benefit, they need to take their medications on time and for the full duration,” said lead author Dr. Dawn Hershman from Columbia University Medical Center, in a news release.

In the analysis presented at the ASCO meeting, early discontinuation and lack of adherence were associated with increases in mortality risk of 36 percent and 40 percent, respectively. The research team from Columbia University, led by Dr. Alfred Neugut, is conducting a prospective cohort study to learn why women discontinue treatment or do not adhere to the prescribed regimen, Dr. Hershman explained in an e-mail.

Drug that Inhibits DNA-repair Enzyme Shrinks Some Breast and Ovarian Tumors

In two small phase II clinical trials, the drug olaparib, which blocks a DNA-repair enzyme called PARP, shrank tumors or stopped the progression of advanced breast and ovarian cancer in some women with inherited BRCA1 or BRCA2 gene mutations. Cancer cells with BRCA mutations are thought to be especially sensitive to PARP inhibition because they have a defect in another DNA repair pathway and therefore rely on the pathway that involves PARP. The results of both studies were published online July 6 in The Lancet.

In the breast cancer trial, Dr. Andrew Tutt of King’s College London School of Medicine and his colleagues enrolled 54 women with locally advanced or metastatic breast cancer and verified BRCA1 or BRCA2 mutations. All of the women had undergone at least one prior chemotherapy regimen. Half of the women received 400 mg of olaparib twice daily—the maximum tolerated dose—and the other half received 100 mg of olaparib twice daily, which preliminary studies showed is the minimum amount to have antitumor effects. Preliminary results were presented at the 2009 ASCO annual meeting.

Tumors shrank in 11 of the 27 women in the 400 mg group and 6 of the 27 women in the 100 mg group. The median duration of this response was 144 days in the former group and 141 days in the latter. Twelve women who participated in the study experienced temporary stabilization of their disease.

The women who received 400 mg had a median survival without disease progression of 5.7 months compared with 3.8 months in the 100 mg group. Forty-one percent of patients in the 400 mg group and 33 percent of patients in the 100 mg group experienced a high-grade side effect, including fatigue, nausea, and vomiting.

Similar results were seen in a parallel phase II trial of olaparib for recurrent ovarian cancer in women with BRCA1 or BRCA2 mutations. In that trial, led by Dr. M. William Audeh of Cedars-Sinai Medical Center in Los Angeles, 33 women received 400 mg of olaparib twice daily and 24 received 100 mg twice daily. Tumors shrank in 11 women in the 400 mg group (with a median response duration of 290 days) and 3 women in the 100 mg group (a median response duration of 269 days).
 
The authors of the breast cancer paper noted that women in both the breast and ovarian trials had received a median of three previous chemotherapy regimens, meaning their disease was highly resistant to treatment. Future trials should compare the efficacy and toxicity of olaparib with traditional cytotoxic chemotherapy, they said.

Colorectal Cancer Screening Rates Rise, Mammography Rates Dip Slightly in the U.S.

Between 2006 and 2008, the percentage of adults age 50 to 75 who had undergone screening for colorectal cancer with a method recommended by the U.S. Preventive Services Task Force rose from 51.9 percent to 62.9 percent. During the same time period, the percentage of women age 50 to 74 who had received a mammogram in the previous 2 years declined slightly, from 81.5 percent to 81.1 percent.

These findings come from the ongoing Behavioral Risk Factor Surveillance System (BRFSS), run by the CDC, and were published online July 9 in the Morbidity and Mortality Weekly Report.

Adults age 50 to 59 years, Hispanics, and persons with lower income, less than a high school education, and without health insurance were least likely to have been screened for colorectal cancer. Women age 50 to 59 years, women with less than a high school education, American Indians and Alaskan Natives, women without health insurance, and those with an annual income of less than $15,000 were least likely to have received a mammogram.

The report highlights the need for evidence-based interventions to increase community cancer screening rates, such as those recently recommended in the CDC’s Guide to Community Preventive Services, concluded the authors. They also stressed the importance of physicians recommending screening services to their patients. “The most common reason women give for not having a mammogram is that no one recommended the test,” they wrote. “Therefore, health care providers have the most important role in increasing the prevalence of up-to-date mammography among women in the United States.”

Length of Telomeres Linked to Risk for Some Cancers

In one of the first prospective studies of its kind, European researchers have found an association between the length of DNA-protein complexes, known as telomeres, that are at the ends of chromosomes and the risk of incident cancer and cancer mortality. Dr. Peter Willeit of the Innsbruck Medical University in Austria and his colleagues published their findings July 7 in the Journal of the American Medical Association.

Telomeres are biological markers for aging; each time a cell divides, telomeres lose a small amount of DNA and become shorter. These structures help protect the genome, and shortened or damaged telomeres may lead to the chromosome instability that is associated with cancer.

To prospectively test for associations between telomere length and cancer incidence or death, the researchers measured telomere length in white blood cells, or leukocytes, from 787 men and women in the Bruneck Study, a cross-sectional population-based survey of atherosclerosis carried out in the small town of Bruneck, Italy. This baseline testing was done in 1995, when participants were free of cancer. After 10 years of follow-up, 92 of 787 participants (11.7 percent) had developed cancer and 44 had died of the disease.

As had been seen in previous studies, individuals with shorter telomeres at baseline had higher risks of cancer. The risk was highest in the group with shortest telomeres and lowest in the group with longest telomeres. An association was also seen with cancer mortality overall.
 
The authors noted that the sample size was too small to precisely assess associations between telomere length and specific types of cancer. Future studies would need to include participants of other races and ethnicities, as well as geographic regions. These studies could also test telomere length in a variety of tissues rather than in the easily accessible leukocytes, the authors said.

Study Estimates Cancer Cases Due to Radiation Exposure in the Marshall Islands

NCI researchers have estimated the risk of radiation-related cancers in the Marshall Islands caused by fallout from U.S. nuclear weapons testing over a 12-year period in the 1940s and 1950s. They concluded that as many as 1.6 percent of all cancers among Marshall Islands residents alive between 1948 and 1970 may be attributable to radiation exposure, and they projected that 170 radiation-related cancers will occur among more than 25,000 Marshallese. This result is based on a sophisticated analysis of available information about the weapons testing, weather patterns, population demographics, and other factors.

The estimates were released as part of a series of eight papers published online July 7 in Health Physics that describe in detail the methods and findings from the research. The studies determined the deposition of the radioactive fallout from the 66 nuclear tests conducted from 1946 to 1958, of which 20 were associated with measureable fallout in the islands. They also analyzed the internal (e.g., via ingestion of contaminated food) and external (e.g., via irradiation from radioactive fallout) radiation doses experienced by island residents.

The findings will be of benefit to Congress and health officials in the Marshall Islands, wrote lead author Dr. Steven L. Simon of NCI’s Division of Cancer Epidemiology and Genetics (DCEG) and colleagues. The methods used to study the nuclear tests’ impact “are also illustrative of methods that may be useful in broader circumstances,” they continued, such as nuclear detonations “due to accidents or intentional actions in wartime or by terrorists.”

In 2004, members of Congress asked DCEG researchers to estimate cancer risk in this population using the data that were immediately available. Those initial “unrefined estimates” were not peer reviewed, the research team explained, and “were generally conservative and purposely intended to not underestimate the actual number of cancers that might occur.”

Of the 170 estimated excess cancer cases, all but 65 are likely to have already occurred, the researchers reported. The most common radiation-related cancer was thyroid cancer, followed by colon cancer and leukemia. Residents of the northeastern atolls, or islands, who were closest to the two primary testing sites (the Bikini and Enewetak atolls) experienced the highest exposures (hundreds to more than 2,000 mGy on average) and developed about 30 percent of the estimated excess cancer cases. Prior to the tests, Bikini and Enewetak residents were relocated. Radiation exposures were likely lowest (between 5 and 12 mGy on average) among residents from the southern part of the island chain, which is farthest away from the testing sites.

“For the first time, NCI scientists have completed a comprehensive dose and cancer risk assessment for Marshallese exposed to fallout from all nuclear weapons tested in their country,” said Dr. Simon. “It was a difficult technical achievement that will be of benefit to the radiation epidemiology and dosimetry communities.”