Participants in a large colorectal cancer prevention study had an increased risk of serious cardiovascular events - cardiovascular death, heart attack, stroke, or heart failure - if they took the arthritis drug celecoxib (Celebrex) daily for an average of almost 3 years, according to an analysis released online by the New England Journal of Medicine on February 15. Celecoxib is one of several compounds that preferentially block one of two cycloxygenase (COX) enzymes that are produced in response to inflammation and by precancerous tissues. It was approved by the FDA for the treatment of osteoarthritis and adult rheumatoid arthritis in December 1998.
The participants in the Adenoma Prevention with Celecoxib (APC) Trial taking 200 mg of celecoxib twice a day had more than 2 times the risk of cardiovascular events, and those taking 400 mg of celecoxib twice a day had more than 3 times the risk of cardiovascular events compared with those taking a placebo twice daily. These results led to the December 2004 suspension of the drug within the trial, which was cosponsored by NCI and Pfizer, Inc., celecoxib's manufacturer. The APC Trial included more than 2,000 people with a history of precancerous colon polyps. It began in late 1999 and is scheduled to be completed this spring.
The effectiveness of celecoxib in preventing the recurrence of colon adenomas in APC participants is being analyzed. "The ability of celecoxib, or another agent that inhibits COX-2, to prevent colorectal cancer is an important question that remains to be answered," said Dr. Ernie Hawk, director of NCI's Office of Centers, Training, and Resources and project officer on the APC Trial. "The cardiovascular events seen in the trial were serious, but the total number of events was relatively small. The potential benefit of celecoxib to prevent cancer or to relieve pain must be weighed against this risk."
Using X-ray fluorescence spectroscopy, NCI researchers have found an inverse relationship between tissue zinc concentration and subsequent risk of developing esophageal squamous cell carcinoma. This study, appearing in the February 16 Journal of the National Cancer Institute, provides some of the first human evidence of an association between zinc deficiency and esophageal cancer, and also demonstrates a novel approach to studying mineral concentrations in a variety of tissues.
Dr. Christian Abnet of NCI's Cancer Prevention Studies Branch and colleagues obtained esophageal tissue samples from a cohort in Linzhou, China. They measured zinc, copper, iron, nickel, and sulfur levels in samples from 60 subjects who developed esophageal squamous cell carcinoma and from 72 subjects who did not develop the disease. The team detected these levels of key elements by bombarding the samples with high-intensity X-rays, causing each element to glow with its characteristic energy signature.
"This technique has many advantages," said Dr. Abnet. "You can apply it to most elements, you only need a tiny tissue sample, and it doesn't damage the tissue, so you can make multiple measurements on one sample."
The average tissue zinc concentration was significantly lower in subjects who developed esophageal cancer than in control subjects. Subjects in the highest quartile of zinc concentration had a 5-fold lower risk of developing esophageal cancer than those in the lowest quartile. Overall, 90 percent of subjects in the highest quartile were alive and cancer-free after 16 years, compared with 65 percent of the subjects in the lowest quartile. There were no consistent associations with cancer risk for any of the other elements studied.
A gene that helps guide prenatal brain development, orthodenticle homologue 2 (OTX2), is normally shut off after babies are born. But in the case of medulloblastoma, the most frequent pediatric malignant brain tumor, sometimes there are too many copies of OTX2 and the copies have been turned back on, according to two studies published in the February 1 Cancer Research.
In the first study, researchers at Johns Hopkins University School of Medicine cut DNA from medulloblastoma tumors into pieces and measured the quantity of each section to identify genomic duplications and deletions. Two portions of DNA, both of which came from chromosome 14, appeared more frequently than expected; only one gene in these sequences, however, had high transcription levels: OTX2. The researchers analyzed copies of OTX2 in 42 other medulloblastoma samples and found the gene was amplified in 19 percent of these, with anywhere from 8 to 56 copies present. By comparison, OTX2 was expressed only rarely in samples from normal adult tissue and from other tumor types.
The second study, led by researchers at Duke University Medical Center and collaborators at Johns Hopkins and the University of Utah Medical Center, also identified multiple copies - as many as 10-fold - of OTX2 in medulloblastomas, with high transcription levels in 93 percent of cases. They then tested the effect of a known OTX2 blocker, all-trans retinoic acid (ATRA), and found that it decreased gene expression and increased apoptosis in the tumors. They concluded, "Our studies of ATRA in medulloblastoma, in conjunction with the studies of others, lay the conceptual framework for clinical trials of retinoids in the treatment of a commonly lethal pediatric brain tumor."
The body's internal or circadian clock may influence chemotherapy toxicity through its effect on immune system B cells, researchers reported last week in a Proceedings of the National Academy of Sciences early online release. It has been long known that the time of day when chemotherapy is delivered can influence toxicity in animal studies. Researchers from the Cleveland Clinic and Northwestern University tested the chemotherapy agent cyclophosphamide (CY) in mice with mutations in genes that control circadian rhythms: the Clock, Bmal1, and Cryptochrome (Cry) genes.
According to the researchers, led by Dr. Marina Antoch, a cell biologist at the Cleveland Clinic Foundation, and Dr. Joseph Takahashi, a Howard Hughes Medical Institute (HHMI) investigator at Northwestern, the Clock and Bmal1 genes are major activators of the molecular circadian clock, and mutations in these genes slow or stop the clock at the low point of its cycle. However, defects in Cry genes, which are activated by Clock and Bmal1, lock up the clock at its most active point. Treatment with CY, the team found, was significantly toxic to mice with Clock or Bmal1 mutations at any time of day. Mice that lacked the Cry genes, on the other hand, were more resistant to toxicity, again, regardless of the time of day.
The cause of these time-of-treatment differences has traditionally been thought to be due to circadian differences in drug metabolism. What researchers found, however, was that severe reductions in B cells (immune system cells targeted by CY) were more pronounced in mice with Clock mutations. "Thus, this paper gives us specific mechanistic insight into the role of circadian rhythms in sensitivity to [chemotherapy] drugs," Dr. Takahashi said in an HHMI news release. "This is not some vague metabolic difference between night and day. This is a tangible difference in the immune system that influences sensitivity."
Neoadjuvant systemic therapy - delivery of hormone or chemotherapy before surgery or radiation - has been recognized as a promising option for nonmetastatic breast cancer patients because it controls local disease progression and can minimize surgery. But, as researchers from the University of Ioannina School of Medicine in Greece have shown, timing of treatment makes no difference in terms of mortality, disease progression, or disease recurrence, and treatment beforehand may actually have adverse consequences. Their findings were published in the February 2 Journal of the National Cancer Institute.
For their analysis, the researchers reviewed 9 randomized trials with collective data from 3,861 patients in which neoadjuvant therapy was compared with adjuvant therapy. In addition to showing that there was no difference between the two arms in terms of survival, disease progression, and recurrence, the data showed a 22-percent higher risk for local disease recurrence in women who received neoadjuvant therapy, likely because those who showed a positive response forewent surgery. "Consequently," the authors wrote, "despite gross clinical response, the tumor bed may not be free of malignant cells, and tumor cell foci may be present in the majority of patients, increasing the risk for subsequent loco-regional disease recurrence." They concluded that, for patients who show a positive response to neoadjuvant therapy, surgery should not be replaced by radiation.
In an editorial, researchers at the Johns Hopkins Kimmel Cancer Center and Fox Chase Cancer Center agreed with this conclusion, and added, "We should strive to move beyond this limited understanding by formulating a uniform definition of a pathologic complete response and delineation of its molecular and imaging predictors."