Gemcitabine Extends Disease-Free Survival in Pancreatic Cancer

Results from the largest randomized clinical trial of chemotherapy after surgery for pancreatic cancer, published in the January 17 Journal of the American Medical Association, show that the drug gemcitabine can increase disease-free survival without excessive side effects.

Investigators from Germany and Austria enrolled 354 eligible patients into the trial. All patients underwent complete resection of their pancreatic cancer. Following surgery, 179 patients received 6 cycles of chemotherapy with gemcitabine, each cycle consisting of 3 weekly injections followed by a 1-week rest period; 175 patients were assigned to an observation-only control group.

Dose modifications were allowed if patients experienced high-grade side effects. Investigators followed all patients with regular physical exams, blood tests, imaging, and quality-of-life assessments until death. The primary endpoint of the trial was disease-free survival, but investigators also measured toxicity and overall survival.

Sixty-two percent of patients in the gemcitabine group received all six cycles of scheduled chemotherapy, and 87 percent received at least one full cycle. High-grade side effects occurred infrequently and their incidence did not increase during the course of chemotherapy. Patients in the gemcitabine group experienced significantly longer median disease-free survival than patients in the control group (13.4 months vs. 6.9 months), without a decrease in quality of life.

Investigators observed a trend toward improved overall survival with the administration of gemcitabine, but this was not statistically significant. However, explained the authors, in light of their estimated survival analysis, "It seems highly likely that the difference in overall survival between groups will become statistically significant with a longer follow-up."

Review Reveals Common Flaws in Microarray Gene-Expression Studies

A detailed review of 42 cancer studies that used microarray technology to identify gene-expression profiles that can predict such things as therapeutic response or survival revealed common flaws in the studies' statistical analyses, according to a new report. Published in the January 17 Journal of the National Cancer Institute (JNCI), the "critical review" of the 42 studies, all published in 2004, was conducted by Drs. Alain Dupuy, from the Hospital Saint-Louis in Paris, and Richard M. Simon, chief of NCI's Biometric Research Branch.

Although the volume of studies that use microarrays is increasing, the authors explained, questions have been raised about the validity of such studies' findings. The authors indicated that "microarray-based clinical investigations have generated both unrealistic hype and excessive skepticism."

Drs. Dupuy and Simon reviewed 90 studies published through the end of 2004 that related microarray-expression profiling to clinical outcome. They grouped the studies into three general categories: those that had an "outcome-related gene finding," which attempted to find specific genes expressed differently depending on patients' prognoses; those that focused on "class discovery" using statistical methods to group tumor specimens with similar gene-expression profiles; and those focused on "supervised prediction," which produce an algorithm that could predict clinical outcome on the basis of individual gene-expression profiles.

They then more closely reviewed 42 studies from this larger group published in 2004 and concluded that 50 percent contained at least one basic flaw. In 9 of the 23 outcome-related gene-finding studies, for example, there was "an unstated, unclear, or inadequate control for multiple testing." Other important flaws they found included overuse and/or inappropriate use of cluster analysis and reporting of biased estimates of prediction accuracy for supervised classifiers.

To help researchers avoid similar problems in future microarray data analyses, the paper includes guidelines for conducting these types of studies.

New Mechanisms Found for PTEN Protein

Tumor suppressor genes such as PTEN (phosphatase and tensin homolog) play a key role in controlling cell proliferation. Normal PTEN protein acts in a biochemical pathway that signals damaged cells to stop dividing and triggers them to self-destruct. Research teams from Memorial Sloan-Kettering Cancer Center and NCI's Center for Cancer Research (CCR) have uncovered ways that cancer cells interfere with this suppressor action. Their findings, published in the January 12 Cell, could eventually yield new clinical strategies.

In one study, Dr. Xuejun Jiang and colleagues identified a key regulator of PTEN protein, a ubiquitin ligase known as NEDD4-1. In a mouse model, they found that NEDD4-1 was highly expressed in tumor cells and involved in posttranslationally modifying the PTEN protein by adding ubiquitin. Though the PTEN gene was not mutated, the ubiquinated PTEN protein was largely destroyed and its ability to suppress tumors was lost, thus qualifying NEDD4-1 as a potential proto-oncogene.

In a second study, Dr. Pier Paolo Pandolfi and colleagues uncovered a novel role for the PTEN protein in the cell nucleus. In collaboration with Dr. Tom Misteli in CCR, they showed that normal PTEN protein is synthesized in the cytoplasm and modified by the NEDD4-1 ligase for entry into the nucleus, where it contributes to chromosome stability. A cancer mutation in the PTEN gene alters the protein, preventing it from entering the nucleus and acting as a tumor suppressor. "This is a beautiful example of basic research uncovering a novel cancer mechanism and pointing the way to entirely novel therapeutic strategies," said Dr. Misteli.

Calcium Offers Prolonged Protection From Colorectal Adenomas

Calcium supplements decrease the risk of colorectal adenomas 5 years after treatment ends, according to study results published in the January 17 JNCI.

Dr. John A. Baron of Dartmouth Medical School and colleagues followed participants of the Calcium Polyp Prevention Study, a double-blind, placebo-controlled, randomized trial of calcium supplements for subjects with a previous colorectal adenoma from 1988 to 1992. Researchers sent participants annual follow-up questionnaires that addressed medical events, including colonoscopies, as well as use of medication, vitamins, and dietary supplements. They obtained posttreatment information from 1999 to 2003 for 822 of the original 930 subjects.

Both the calcium treatment group and the placebo group had a similar number of colonoscopies, time to first and last colonoscopy, and use of calcium supplements after treatment ended. However, during the 5 years after treatment, those participants in the calcium treatment group had a significantly lower risk of all adenomas than those in the placebo group (31.5 percent vs. 43.2 percent).

The researchers wrote, "The persistence of the effect is a provocative finding, but one that is difficult to explain. It suggests that a 4-year course of calcium treatment alters the colorectal mucosa in such a way that it can resist the development of new adenomas that would otherwise become apparent several years later."