IE 11 is not supported. For an optimal experience visit our site on another browser.

When medical reality doesn't match the spin

Presentation of the results from a recent breast cancer study demonstrates the National Cancer Institute's eagerness to proclaim success against the disease no matter what the data show. By NBC's Robert Bazell.
/ Source: NBC News

We’re making great progress against cancer, right?

The National Cancer Institute desperately wants you to believe that, even if it means spinning the scientific facts to fit the perception.

The NCI is the federal government’s lead organization in financing cancer research, with an annual budget of about $5 billion. Presentation of the results from a recent breast cancer study demonstrates the NCI’s eagerness to proclaim success no matter what the data show.

Called the STAR trial, the study enrolled 20,000 healthy post-menopausal women for five years to compare two drugs, tamoxifen and raloxifene (sold under the trade name Evista), to determine which is better at reducing the risk for breast cancer. Scientists and journalists label such studies “prevention” trials. Unfortunately, nothing prevents breast cancer. Some drugs shift the odds, and the goal here was to determine which one is best.

Reading a a press release issued by the NCI in April and listening to a telephone briefing, there was no doubt. Both drugs affected breast cancer incidence equally, but raloxifene has fewer side effects, according to the researchers.

Bingo! A great result for women’s health. Raloxifene, as headlines and news broadcasts across the country proclaimed, was the clear winner.

Since the drug is already on the market to protect against bone loss why shouldn’t eligible women, estimated to number 30 million, take it?

Ambiguous results
The problem? The detailed results of the study — appearing in this week’s issue of the Journal of the American Medical Association — differ substantially from the interpretation pushed by the press release. Both drugs have side effects and the study offers little clarity about which one is best. Most troubling for raloxifene, it brings a higher incidence of a breast condition called DCIS (or ductal carcinoma in situ). 

Oncologists debate whether this is truly breast cancer or a pre-cancerous condition. But tens of thousands of women a year require surgery (often mastectomy) and radiation to treat DCIS so anyone would think twice before taking a drug that might increase the risk. 

This equivocal result is especially troubling for the NCI because of the history of the effort to find a breast cancer “prevention” pill. In 1998 with great fanfare the NCI released results of a trial that compared tamoxifen to a placebo. Tamoxifen clearly reduced the risk of breast cancer. A government press release called it “an historic step toward more effective prevention of breast cancer.” 

But women and their doctors did not buy into the concept. Tamoxifen increases the risk for blood clots and uterine cancer and few healthy women wanted to take a pill every day. 

So the NCI really wanted to prove that raloxifene could be an acceptable alternative. But taxpayers spent $88 million (and raloxifene manufacturer Eli Lilly put in another $30 Million) for the ambiguous result.

Traditionally, researchers present results at scientific meetings or in peer-reviewed journals where other experts can see all the data in context. But the practice of announcing the results by press release — whether by biotechnology company or the NCI — is becoming ever more common.

Impress the public before the messy facts get in the way. 

Painting a rosy picture
NCI scientists have offered several reasons for the press release. They said it was necessary for “public health” and to inform the women in the trial.  But it is hard to see how either public health or the volunteers benefit from incomplete confusing information released in a hurry.

The NCI has faced a huge problem since 1971 when President Richard Nixon ordered the "war on cancer.” Deaths from cancer — not only total deaths — but the death rate adjusted for the age of the population — have been climbing. Only last year for the first time did the death rate drop by a mere 1 percent. 

To secure funding, NCI is always painting a rosy picture of progress. The optimism in the face of reality reached its apex in 2005 when Dr. Andrew von Eshenbachappointed by President Bush to head the NCI — declared a goal to "eliminate suffering and death due to cancer by 2015." Many scientists derided the statement.

Dr. von Eshenbach quit the NCI to head the Food and Drug Administration on June 10.

In the face of the enormous health burden of cancer, NCI should be spending even more to elucidate both the basic science and the clinical applications — including studies like the STAR trial. But a government agency with a charter to improve the public’s health does not serve it by trying to put a positive spin on every result.