Testimony on Problems with Prescription-to-OTC Switch of Cholesterol-Lowering Drugs
July 13, 2000
Sidney M. Wolfe, M.D.
Director, Public Citizen's Health Research Group
Before FDA Endocrinologic/Metabolic and Non-Prescription Drugs Advisory Committees
I. Risk assessment/ effectiveness
Before even evaluating the risk implications of different cholesterol levels, a reliable test must be done which, in the context of an OTC cholesterol product, may well be a home cholesterol test. There are a number of questions about the accuracy of such tests, especially because of the varying skills of those people who are performing them, often on themselves---for the first time. At the end of this testimony is a verbatim description of the steps necessary to perform one such test. Other related steps include the validity of patients' self-assessment (without a new test being performed) of their own cholesterol levels and of other risk factors such as the presence of cardiovascular disease or diabetes. In the various studies submitted by Merck, there is abundant evidence that most people who showed up because of an interest in the studies were not eligible using the criteria of the studies. For example, in the randomized efficacy study (075), 63.9% of the 582 participants did not go through with the study, mainly because they failed to meet the eligibility criteria such as cholesterol out of range. In study 079, a storefront study to evaluate compliance, of 4878 patients screened because they were interested in the medication, only 460 or 9.4% eventually got the drug, many not qualifying because of their cholesterol levels, raising serious questions about patients' ability to self-select into using these OTC drugs. Of 1149 participants who self-reported their cholesterol to be between 190 and 250, 30.5% were not correct about these estimates based on subsequent testing. Similarly, in study 076, an open label study at pharmacies, 53% of the 1351 subjects who reported their cholesterol level to be between 200 and 240, the criteria in the Mevacor and Pravachol labeling, did not, in fact, have levels in that range. These serious problems in self-selection, which would not likely be caught in the real world where people screen themselves, often without a cholesterol test, and decide to use the drug, are only the beginning of a cascade of other serious problems.
Even if the patient self-selection worked and arrived at a group of patients who met the defined eligibility criteria, these very criteria are seriously flawed. There is no evidence that the majority of those who are "eligible" for the drug will have any clinical benefit from taking 10 mg daily of Mevacor. The primary prevention study known as AFCAPs/TexCAPS found an overall reduction in strokes and heart attacks but the dose of Mevacor employed was 20 to 40 mg per day, not 10 mg per day. But even if there were—which there is not-- evidence of primary prevention (reduction of strokes and heart attacks) at 10 mg, those in AFCAPS/TexCAPS who had HDL levels of 40 or greater (not even listed as a risk factor in the proposed label for these drugs) showed no evidence of a clinical benefit. This is extremely germane to the issue of the OTC switch because a large proportion of people in Merck's estimated population of potential OTC users based on NHANES---78% of 15.5 million potential users—have HDL levels of 40 mg or greater. They are thus in a group not found to have a significant clinical benefit in AFCAPS/TEXCAPS even at the doses of 20-40 mg per day used in that study.
II. Compliance/monitoring of dose and of ADR's
Prevention of cardiovascular disease must be a multi-pronged strategy to reduce risk. The use of heavily advertised statins, out of the context of medical consultation, may impair the development of an integrated long-term strategy for preventing strokes or heart attacks. Diet and exercise, critically important components, may be thought to be less important if the primary strategy seems to be a statin drug. The evidence of poor compliance, even in the short term (six months or less) with 25.6% to 31% of people dropping out by that time, bodes poorly for the long term compliance necessary for the drug to work in those for whom the use may be appropriate. Similar problems with compliance would be guaranteed as far as doing routine liver function monitoring as stated in the prescription versions of these drugs.
III. Safety issues such as drug interactions; literacy issues; and liver tests
According to FDA's analysis of the Merck Label Comprehension Study, only 66% of low literacy label readers (only 82% of high literacy) knew about the counter indicated lovastatin/ erythromycin interaction, whereby erythromycin inhibits the metabolism of lovastatin and dangerously high levels of lovastatin can accumulate. In two years alone (11/97-12/99), there were eight reports to the FDA of rhabdomyolysis (life threatening acute destruction of muscle tissue with significant risk of kidney damage) requiring hospitalization in patients because they were simultaneously taking lovastatin and erythromycin. If statins are switched to OTC, the number of such serious adverse reactions would skyrocket as they would with the many other prescription drugs with which lovastatin should not be used. With multiple physicians there is a significant problem with interactions between two independently prescribed prescription drugs. But the situation is made much more dangerous with OTC availability because: 1/ neither of the doctors may know about the addition of an OTC statin and 2/ even if the patient is going to one pharmacy which keeps prescription records to prevent a harmful interaction, OTC products—bought in a supermarket, 7-11 or even in the same pharmacy but without an interaction with the pharmacist, will not get into this data system.
IV. Benefit risk ratio
The benefit/risk ratio for the approval of a drug must clearly be greater than 1 to merit approval. If the benefit is 0, because of the lack of evidence of clinical benefit for the majority of those who self-select into using OTC statins, then no amount of risk, no matter how small, will make this ratio acceptable.
The ADVANCED CARE Cholesterol Test, now known as Cholestrak, was reviewed by an author from the University of Kentucky's School of Pharmacy: (http://www.pharm.uky.edu/hometest/cholesterol/CHO.html)
Patient Instructions (verbatim):
- Wash hands thoroughly with soap and water. Rinse and dry. To increase blood flow to the hands, warm them by rubbing together for 10 seconds and allow the arms to hand to side for 30 seconds.
- Using the lancet provided with the test kit, hold it against the outside of the "ring" or middle fingertip and press the trigger with the thumb. You may need to squeeze the finger to obtain enough blood to fill the well.
- Add blood to the well of the cassette until the black line cannot be seen. The well must be filled within five minutes of adding the first drop of blood.
- Wait at least two minutes, but not more than four minutes, after filling the well with blood, then pull the clear plastic tab on the right side of the cassette until the red line appears. This tab may need to be pulled hard. Tap the kit against a flat surface one or two times. The "OK" indicator should turn purple in about five minutes. This tells you that the test is working. About 10-12 minutes later, the "END" indicator will turn green which lets you know the test is finished. You may now read the results.
- After the "END" indicator turns green, find the tip of the purple color in the window of the cassette and record the number to the left of it. Using CHOLESTEROL RESULT CHART, match the cassette reading to the corresponding cholesterol level in units of mg/dl. The cassette reading must be recorded within 30 minutes of performing the test.
In the FAQ section: Can both indicators change color without giving me a correct reading?
Answer: Yes, both indicators may still work ("OK" indicator turns purple and "END" indicator turns green) even if you do not follow the directions exactly. A false result may occur if:
- The test well is not filled to the black line, or
- More than five minutes pass when filling the test well, or
- You do not wait a full 2 minutes before pulling out the plastic tab, or
- You do not pull the tab out completely to the red line, or
- You perform the test in direct sunlight