MORE Data, MORE Questions
MORE Data, MORE Questions
By Sarah L. Berga, MD
I remember when i used to eagerly turn back the cover of JAMA in anticipation of that week’s medical findings. I still have the anticipation, but the sense of potential enlightenment has been replaced by a more utilitarian goal, that of making sure I am up-to-date on the latest in confusing information. I want to make sure that I will be able to explain the most recent report to my patients trying to decide about whether to start or remain on hormonal therapy. Typically, patients hear a sound bite on the radio or read a small blurb in the newspaper. Naturally, if the findings are ambiguous, as they mostly are, the patients are worried and want to know what I think. I can’t blame them. I also want to know what I think. Thus, I am compelled to turn pages. Last week’s JAMA was a case-in-point. The lead article (Barrett-Connor E, et al. Raloxifene and cardiovascular events in osteoporotic women. JAMA. 2002;287:847-857) reported the effect of raloxifene use in the Multiple Outcomes of Raloxifene Evaluation (MORE) trial upon cardiovascular events. The abstract suggested that raloxifene conferred cardioprotection for those most at risk for CVD. I was surprised because I knew that cardiovascular events were not one of the end points of the trial in question. So I read more, hoping for some clarification. Here is what I learned.
The MORE trial was a double-blind, placebo-controlled trial of 7705 postmenopausal women with osteoporosis that was conducted from 1994 to 1999. The mean age of participants was 67 years. The rate of cardiovascular events and mortality was about half of that reported for women in the general US population, so the number of events for the entire cohort during the 4 years was small (272), ie, 3.5% of the cohort. Overall, there was no difference in cardiovascular events among the 3 groups, which were those treated with placebo, raloxifene 60 mg/d, and raloxifene 120 mg/d. But in a subset of 1035 women at increased risk of cardiovascular disease, there was a decreased risk of cardiovascular events after the first year of raloxifene use. However, this decrease was statistically significant only in the group treated with the 60-mg daily dose of raloxifene and not in the group treated with the 120-mg raloxifene dose. The number of events in this subset of 1035 "high-risk" women was tiny, only 55 coronary events and 41 cerebrovascular events over the entire 4 years. Only in the discussion do Barrett-Conner and colleagues point out that the reduction in risk of cardiovascular events did not translate into a reduction in risk of death from cardiovascular disease. Further, the analysis was by intention-to-treat and 26% of the participants in each arm dropped out, mostly due to adverse events, the nature of which are not disclosed in the manuscript. Barrett-Conner et al do note that the trial was not designed to look at cardiovascular events, so the data on death and events are from the safety end points, rather than from direct prospective ascertainment. The groups differed from the outset in terms of diabetes and use of lipid-lowering agents, both of which were higher in the raloxifene groups. Thus, one wonders if the raloxifene group might have received more intensive medical care unrelated to the study. There also tended to be fewer women with a prior history of myocardial infarction or having had a coronary artery bypass graph (CABG) in the 60-mg/d raloxifene group.
I must acknowledge that I started to read the article in my customary (skeptical) frame of reference. However, the introduction was succinct and made a good case for the hypothesis. I started to become hopeful that the results would be convincing. The description of the methods was thorough and detailed, lending the impression that the study design might be adequate for the query under examination. But as I read a little more, the troublesome details began to emerge. In the end, I felt betrayed. After wading through 10 pages of fine print, I concluded that the study was so problematic that the results could not be trusted. I also felt that the positive aspects of the study had been highlighted in the abstract while the negative or problematic details were buried in the back. Further, while estrogen was not one of the study arms, sprinkled throughout the discussion were references as to how the present results compared to results obtained in postmenopausal women using conventional estrogens. Although the data do not permit such conclusions, the implication was that raloxifene might be better than estrogen for cardioprotection.
Why am I harassing you with these details? In part, I aim to sound a cautionary note about the reliability of the results of a study likely to receive more attention than it deserves. I would also like to point out the obvious, if only to make it more so. The company that makes the drug in question funded the study. This is not optimal. The study was conducted by a group of clinical investigators at reputable academic institutions. This is good. But why are distinguished clinical investigators doing a study funded by a pharmaceutical house? Because there is almost no money available from other sources for clinical investigations of this type. When the federal government undertook a similar study, the Women’s Health Initiative, they too asked for and received free drugs, thus becoming unwitting handmaidens of the drug makers. It is not that the pharmaceutical industry is intrinsically evil; their representatives are trying to let us know the most favorable points about the products that they sell. If we are ever to get the kind of reliable information we need to make informed decisions about the array of available products, then we will have to find a way for adjudicators (clinical investigators) to have the resources required for conducting head-to-head comparisons free of commercial bias. v
Reader Questions
Question: Dear Editor, In a recent edition of OB/GYN Clinical Alert, Dr. Noller stated that there is no longer any indication for the use of 5-FU in the lower genital tract. In 1999 we published a randomized controlled trial of the use of 5-FU in HIV-positive women with significant intraepithelial neoplasia. Those women who received the active agent after ablative or excisional therapy had more and longer disease-free intervals than those who received the placebo (Obstet Gynecol. 1999;94:954-961). We believe that there is a place for the use of this agent in this select population. Can you comment? — Mitchell Maiman, MD, Chairman, Dept. of Ob-Gyn, Staten Island University Hosptial, NY
Kenneth Noller, MD responds: Dr. Maimon’s article did indeed demonstrate that 5-FU can be useful for HIV-positive women who have been treated for intraepithelial neoplasia. As I have previously indicated in these pages, a few women who have condylomatous colpitis as a result of immunosuppression (usually iatrogenic) also can benefit from careful use of small amounts of intravaginal 5-FU. As far as I know, these are the only examples of proven benefit from the use of this caustic agent in the lower genital tract. Unfortunately, I continue to see women who have been injured by this drug, the majority of whom have used it on the vulva.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.