Study Shows Phase I Cancer Findings Under Reported
Phase I Cancer
Phase I cancer studies, trials that are conducted to determine the safety and maximum dose of a new agent, are under-reported in peer-reviewed journals - a trend that could ultimately delay scientific progress and negatively affect patient care, say researchers at The University of Texas M. D. Anderson Cancer Center in a new study in Cancer.
Over the last decade, greater understanding of cancer at the molecular and cellular levels has resulted in a scientific discovery explosion, translating into the development of numerous potential anticancer agents. An excellent indication of this progress, says Luis Camacho, M.D., assistant professor of medicine in M. D. Anderson's Phase I Clinical Trials Program, is the 10-fold increase in the number of Investigational New Drug (IND) applications for oncology agents filed with the Food and Drug Administration - from 100 compounds in 1980 to over 1,000 in 1998.
"With all this new knowledge, the need to share information is paramount, now more than ever before," says Camacho, the study's senior author. "We, as clinicians and researchers, have a tremendous responsibility to not only investigate and discover new agents, but also to disseminate our discoveries - good and bad - to the medical community at large, to ensure the safety and well-being of our patients."
Given this acceleration of discovery, scientific meetings have become inundated with study submissions, reports Camacho. Yet, often these results are preliminary and subject to change after final analysis. Therefore, publishing in peer-reviewed scientific journals remains the "gold-standard" for sharing medical information, he says.
To understand the rate at which Phase I trials were being published, Camacho and his team analyzed the 275 Phase I abstracts accepted by the American Society of Clinical Oncology (ASCO) for presentation at its annual meeting in 1997, and determined whether those findings had been published over the next seven and a half years. Through repeated MEDLINE searches and emailed questionnaires, researchers discovered that by February 2001, two-thirds (67 percent) of the Phase I abstracts had been published in peer-reviewed medical journals.
Researchers found that the Phase I ASCO studies of novel (agents not approved by the FDA) and non-novel (studies including at least one FDA-approved agent) drugs were published at an equal rate. Research that was financially supported by the pharmaceutical industry was published at the same rate of those studies not industry-supported. There was no difference in the rate of oral and poster presentations published; however, oral presentations were published more quickly than those presented as posters.
Publishing positive or negative Phase I trial results is critically important in developing new, or modifying existing cancer therapies, Camacho says.
"Obviously, if a Phase I agent proves too toxic, we need to ensure that information is shared within the cancer medical community, so as to not put patients in harms way," Camacho says. "Of course, if Phase I studies are promising, publishing can encourage further investigation of these potential therapeutic agents."
In unique cases, Camacho says, positive Phase I results can have immediate impact on clinical care.
"Phase I studies are not necessarily first-in-human trials - they also can be combinations of already approved drugs," he adds. "If those combinations prove relatively non-toxic and show some effect, that combination can potentially be moved forward to Phase II and, in selected cases, used to treat patients, almost immediately."
The median time to publication - 3.4 years - also is very concerning, Camacho says. "Releasing data after such a lengthy interval may not allow researchers to build on their scientific experience, thereby significantly delaying further investigation with encouraging agents."
From the questionnaire sent to investigators, Camacho and his team learned that lack of time and investigator relocation were the major obstacles of publication. These reasons were unexpected, as researchers expected to learn that negative results, lack of interest in pursuing projects to completion, or the stringent peer-review process would be among the reasons why investigators had not published.
Given the investigators reasons for not publishing, Camacho says that academic institutions should provide academicians more time to concentrate on publishing efforts. In return, clinicians must be committed to reporting Phase I toxicities in peer-review literature.
The study was funded, in part, by a Clinical Cancer Research Award from the Cancer and Leukemia Group B (CALGB), a cooperative cancer working group.