FAQs: Dr. Damaj

Question: Should we know the dates for the Laws and amendments on p54? Also what did you want us to know from the Herbal products chart on p.171? indication? dose?

Answer: You do not need to know the dates for the laws and amendments; you only need to know their legal significance.  The dates are included only to give you an appreciation of the historical development of drug regulation in the U.S.

For the herbal products chart, you need only know the indication(s) for each product (i.e. you do NOT need to know botanical nomenclature or dosage -- knowledge of principal active constituent is good to have, but I won't ask that.)

Question: You said something in class about the numbers of patients that you would have to treat to initiate adverse reactions. I wrote something down about needing to expose 15-20 times the number of people who provoke the event. I don't really understand how to apply this or what the importance of this information is.

Answer: This issue is both confusing and difficult to verbalize, so I will attempt to verbalize it and will then give an example that I hope will demonstrate the application.

In order to achieve sufficient statistical power to associate an adverse effect with a drug as the causative agent (i.e., so that "I can be 95% certain that the drug is a cause of this adverse effect"), you need to have exposed to the drug approximately 15 to 20 times the number of people in which the adverse effect would be expected to occur 1 time. For example, if the adverse event is precipitated in 1 of 100 people, you would need to expose 15 to 20 times 100 (1500 to 2000) people to obtain the statistical association. If it occurs in 1 of 1000 exposed people, you would need to expose 15,000 to 20,000 to obtain the association. Another way of saying it is that you would need to expose enough people so that the effect would be expected to be observed in15 to 20 people in order for the association to be obtained. (Actually, the numbers will also vary depending on how frequently the event occurs spontaneously --i.e. not related to drug -- relative to the frequency due to drug exposure; so it is not quite as simple as I have described it, but our discussion makes the points that establishing the association requires large numbers of subjects and is not always easy to do, and the estimate of number of subjects needed is good enough for most purposes.)

Question: What's the relationship between lean body mass and aminoglycosides? 

Answer: Aminoglycoside antibiotics are just an example (probably the most frequent example) of a class of drugs that may have dosage adjustments made on the basis of lean body mass.  That is because the aminoglycosides are highly water-soluble and distribute primarily to lean tissues (do not enter fatty tissues well), so lean body mass serves better as a predictor of plasma and lean tissue levels of drug relative to dosage than total body weight serves as a predictor.

Question: Why is it that you need 15-20 times the number of people to recreate an adverse effect during clinical testing?

Answer: It is not that you need that number of people to recreate the adverse effect -- you need that number of people (15 to 20 times the number of people in which the drug-induced effect occurs one time, or the number of people in which you should observe the effect 15 to 20 times) in order to establish a statistically significant relationship between the drug and the effect. That number of people must be exposed in order to achieve the desired "statistical power" so that you can say that you are "95% certain that the effect is drug-related."

Question: Did you say that "ED50 is an expression of potency of the drug" in class?  I thought that by definition ED50 is the dose at which 50 percent of the subjects achieved the desired effect.  Isn't that more along the line of efficacy?    Also, if an clinician says drug A is better at treating heartburn than drug B, is that a statement of efficacy?  Does he/she means that it takes less of a drug A to achieve the same desired effect as drug B?  How is that different from saying that it takes less drug A in producing the desired effect in X number of patients compared to drug B?

Answer: Yes, I did say that ED50 is an expression of potency of the drug.  Your definition is correct (for a quantal response), but that is an indicator of potency (How much drug does it take to produce that effect?) rather than of efficacy (How much effect can that drug produce?).  For evaluation of efficacy, one must examine graded responses rather than quantal responses.

    Yes, I would interpret that as a statement of efficacy, that "drug A is more effective ("better") than drug B" in terms of producing a greater magnitude of effect (although in could be "better" in the sense that it produces an equal effect but with lesser adverse effects).  I would not draw a conclusion from that statement about relative amounts of either drug required to produce the desired effect (i.e. potency.)  If I understand your last question correctly, I don't think that it is any different from the previous question in the sense that both are dealing with the issue of potency.  When the issue is clinical superiority, or which drug is "better", that is usually concerned with efficacy.

Question: Could you explain again the difference between the therapeutic index and the certain safety factor.  When is one used vs. the other?  Why create different measures of safety?

Answer: The therapeutic index is defined as the ratio LD50/ED50, or the ratio of the dosage required to produce death in 50% of the tested population to the dosage required to produce the designated (desired) effect in 50% of the tested population.  It is the more commonly used measure.  However, it only provides relative median values, and does not provide information concerning extreme values.  Concern over the potentially disastrous overlap of extreme values led to development of the measure of the certain safety factor, defined as the ratio LD1/ED99, or the ratio of the dosage required to produce death in 1% of the population (i.e. practically no deaths) to the dosage required to produce the desired effect in 99% of the population (i.e. effective in practically everyone).  So the certain safety factor (if it is greater than 1) tells you that you can produce your desired effect in practically everyone without causing deaths, but it is unfortunately a value that is not readily available for most drugs.