Many Group-Randomized Trials Lack Necessary Design Characteristics
Group-randomized trials in which participants are assigned to study arms in predetermined groups, instead of individually, infrequently met the unique statistical and design challenges required for this approach. These cancer trials may present misleading findings as a result.
Scientists increasingly use group-randomized trials to test interventions when it is impractical or impossible to randomize individual participants. But the quality of these trials has not been examined in the cancer field.
To determine whether group-randomized oncology trials are adequately designed and executed, David M. Murray, Ph.D., of The Ohio State University in Columbus and colleagues identified and examined 75 studies that used group-randomization and were published between 2002 and 2006 in the field of oncology.
The researchers found a substantial number of problems with the trials. For example, only 18 of the studies (24 percent) reported information about how they calculated the necessary sample size. Additionally, just 35 studies (45 percent) used only the analytical methods that Murray and colleagues considered appropriate for a group-randomized trial. Many of the trials relied on analyses that ignored the group randomization altogether.
“Given that 88 percent of these articles also reported statistically significant intervention effects, there is reason to be concerned that many of these reported effects may be type I errors. Continued reporting of spurious effects misleads investigators and policymakers and slows progress toward control and prevention of cancer,” the authors write.
In an accompanying editorial, Timothy R. Church, Ph.D., of the University of Minnesota School of Public Health in Minneapolis writes, “This disjuncture of health research and statistical practice has consequences that go far beyond academic niceties. The initiative to improve the health of all individuals depends heavily on the appropriate interpretation of well-designed studies, an activity that depends in turn heavily on the close marriage of researchers and statisticians.”
HPV-Based Screening More Accurate at Detecting Cervical Precancers than Cytology
Human papillomavirus (HPV) DNA-based tests are more accurate than cytology (Pap smears) in detecting cervical precancerous lesions, according to data from a randomized controlled trial.
HPV screening, using DNA-based technology, was known to uncover more infections and precancerous lesions than cytology. It was unclear, however, whether it would lead to treatment of patients whose immune systems would clear up the problem on its own and thus did not need to be treated at all.
Guglielmo Ronco, M.D., Ph.D., of CPO Piemonte in Torino, Italy, and colleagues randomly assigned nearly 50,000 women to receive either conventional cytology, followed by a more intensive examination called colposcopy if abnormalities were detected, or to HPV-based screening and referral to colposcopy if the concentration of HPV DNA was above a predetermined level.
Using the HPV test, researchers identified nearly twice as many premalignant lesions, called CIN2+, than they did with cytology. Among older women ages 35 to 60, testing for HPV was more likely than conventional cytology testing to identify women who had cervical lesions. In women between the ages of 25 and 34, the HPV test appeared to identify a large number of infections that resolved themselves, even if they had some abnormal cells at the time of initial detection. Based on these data, Ronco and colleagues propose that it might be most appropriate for younger women with a positive HPV test to be retested in 12 months, rather than to be immediately referred for colposcopy.
“It seems clear that an HPV DNA-based approach to primary screening is a very attractive option that should be actively developed and evaluated,” the authors write.
Contact: Guglielmo Ronco, email@example.com, +393357553701
Watchful Waiting May Be Proper Initial Treatment for Some HPV Infections
Watchful waiting may be an adequate initial treatment for many women infected with a carcinogenic type of human papillomavirus (HPV) if they have normal cervical cell cytology.
Some types of HPV cause cervical cancer. Persistent infections with these HPV types are more likely than transient infections to induce precancerous lesions. However, researchers have not fully quantified the risks of precancerous lesions associated with persistent versus transient infections.
To determine those risks, Ana Cecilia Rodríguez, M.D., of the National Cancer Institute in Bethesda, Md., and colleagues performed a study of 2,655 randomly selected sexually active women living in Costa Rica. All of the women underwent cytology-based cervical cancer screening (Pap smears) every 6 to 12 months as well as DNA-based HPV tests.
Of the study participants, 1,013 women had HPV infections at the start of the study and 599 of them were infected with a strain that can cause cervical cancer. (Some women were infected with two or more cancer-causing types, leading to 800 total infections in the analysis.) More than half of the infections with cancer-causing HPV types cleared within 6 months, and 67 percent cleared by 12 months. Only 4 percent of the infections with cancer-causing HPV types led to precancerous lesions, although that figure rose to 21 percent in women whose infections persisted beyond 12 months.
“These data suggest that, when possible, a patient with a normal cytology and initial positive HPV result should be managed with watchful waiting because a 12-month follow-up can safely exclude more than 50 percent of infections as transient,” the authors write.
Contact: National Cancer Institute press office, firstname.lastname@example.org, (301) 496 6641
Inhibition of Chemokine Receptor Activation May Improve Head and Neck Cancer Therapies
Squamous cell cancers of the head and neck that express a cell surface receptor, called CCR7, are less likely to respond to current therapies. Combining drugs that block this receptor with existing agents might improve patient outcomes.
Robert L. Ferris, M.D., Ph.D., of the University of Pittsburgh Cancer Institute and colleagues previously found that the chemokine receptor 7 (CCR7) is frequently active in metastatic squamous cell cancers of the head and neck.
In the current study, Ferris and colleagues used cancer cell lines and tumor cells isolated from primary and metastatic head and neck tumors to examine patterns of CCR7 activation. They also looked for expression of the ligands that stimulate the receptor.
Treatment of cells isolated from metastatic tumors with an anti-CCR7 antibody, which blocked activation of the receptor, reduced cell survival by 31 percent, compared with a 19.5 percent reduction in viability after treatment with a targeted therapy, called erlotinib, that is already being used in squamous cell head and neck cancer treatment. Combining the anti-CCR7 antibody and erlotinib, resulted in a 59.5 percent drop in cell survival. The researchers found that both the tumor environment and the tumor cells themselves can produce the ligand that stimulated CCR7 and those tumors that did so were more resistant to standard chemotherapy.
“The results of this study offer additional evidence that this chemokine receptor plays an important role in [squamous cell cancer of the head and neck] progression and in resistance to chemotherapies,” the authors write.
Contact: Robert Ferris, email@example.com, (412) 623-7771
Also in March 25 JNCI:
Note to Reporters:
We have started up an e-mail list to alert reporters when papers are available on the EurekAlert site. If you would be interested on being on this list, please let us know at firstname.lastname@example.org. The content will continue to be available through EurekAlert’s e-mail system and our EurekAlert page.
The Journal of the National Cancer Institute is published by Oxford University Press and is not affiliated with the National Cancer Institute. Attribution to the Journal of the National Cancer Institute is requested in all news coverage. Visit the Journal online at http://jnci.oxfordjournals.org/.
AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert! system.