Generic vs Branded Psychiatric Meds: Is There a Difference?
Generic vs Branded Psychiatric Meds: Is There a Difference?
Medscape: In terms of bioequivalence, how much variability does the FDA allow for equivalent drugs?
Dr. Howland: The 2 parameters are the rate of absorption (Cmax) and the extent of absorption (AUC). According to the FDA guidelines, the Cmax and the AUC are determined for the test drug and for the innovator or brand-name drug. Because these parameters are determined for a group of persons taking the test drug and a group taking the brand-name drug, the average values for Cmax and AUC are calculated, along with their degree of variability. Variability is a measure of how much the actual data for Cmax and AUC vary from the average values. The ratio of Cmax for test drug over Cmax of brand-name drug is determined. Similarly, the ratio of AUC for test drug over brand-name drug is determined. A difference between test drug and brand-name drug of greater than 20% for Cmax or for AUC is considered significant by the FDA, so the lower limit of the Cmax and AUC ratios must be 80%. Because the data are log-transformed, the upper limit of these ratios must be 125% (that is, the reciprocal of the lower 80% limit, which is 1 divided by 80%).
Because Cmax and AUC are average values, the variability of the data around these averages is also calculated. Variability is an important concept. If you think about a roomful of athletes, some of whom are basketball players and others are jockeys and gymnasts, you can calculate the average height, but there's a lot of variability. According to FDA guidelines, a 90% confidence interval is used for these averages. In other words, the average values for Cmax and AUC are in the center of the confidence interval, and 90% of the data are contained within the interval.
What becomes confusing, if you think about the difference between 80% and 125%, is that a 45% difference seems very large. To take into account the inherent variability around the average values for Cmax and AUC, the FDA specifies that the ratios for Cmax and AUC, along with their 90% confidence intervals, must all fall within the 80%-125% window for a generic drug to be considered bioequivalent to a brand-name drug.
Because the averages and the 90% confidence intervals have to fall within the 80%-125% range, differences between generic drugs and brand-name drugs don't vary so extensively if they fulfill the FDA's definition of bioequivalence. And you have to satisfy those criteria for both the Cmax and the AUC.
[Editor's note: A summary of FDA definitions for bioequivalence methodology can be found here.]
Medscape: So, the FDA is looking at how much each drug varies within itself, in addition to how different they are from each other?
Dr. Howland: Right, because when you do test studies of the bioavailability of a brand-name drug, or any other medication, there's normal variance from one person to another. If you conduct a study and you have 20 people, there's going to be normal variability in terms of absorption. They collect data in all the participants, and they determine the average Cmax and also the AUC. But, to take into account the variability, and to make sure the variability of the proposed generic is not significantly different from that of the brand-name drug, the FDA wants to make sure that the variability still falls within that 80%-125% range. If any part of the variability falls outside that range, the drugs are not considered bioequivalent.
It's tough to describe these things without looking at graphs, or especially understanding confidence intervals and how the FDA defines the 80% and 125% boundaries, because that's what creates a lot of confusion in the minds of physicians and the public. They think, "Wow, that's a 45% variability. If I get a drug that's at the 81st percentile, that might be quite different from another," but that really isn't what's being measured.
Medscape: In addition to having established bioequivalence, will the generic drugs also carry the warnings and other FDA language associated with the branded drugs?
Dr. Howland: Because the FDA intends for generic drugs and brand-name drugs to be therapeutically equivalent -- having the same active ingredient and being bioequivalent -- they mandate that the generic drug labeling be similar to the previously approved labeling for a brand-name product. There was a recent US Supreme Court ruling (Pliva, Inc. v. Mensing) on a case involving the labeling of a generic product in terms of serious adverse effects. There is a product, metoclopramide (Reglan®), which is used for treating nausea, and it's a dopamine-blocking drug. There are many generic formulations of Reglan, and like other dopamine-blocking drugs, such as antipsychotics, it carries the risk for movement disorders, including tardive dyskinesia.
My understanding of this case is that a woman who had developed tardive dyskinesia while taking a generic form of Reglan sued, arguing that, consistent with state law, the label should have mentioned this possibility. The Supreme Court ruled that a generic manufacturer of an old drug can't necessarily be held liable for a safety change, because they're bound to maintain the label of the originator drug, according to the FDA guidelines.
Labeling for generics has to be very similar to brand-name products. The issue for some older drugs is that you can't even prescribe the brand-name version anymore because it's not manufactured. New knowledge about the drug may not be incorporated into a product label because there's no brand-name product manufacturer who's being mandated to change the label in accordance with FDA guidelines on new safety data. A person taking an old medication that's only available as a generic may be exposed to risks that the original manufacturer would have been obliged to report to the FDA.
How Much Variability Is Too Much?
Medscape: In terms of bioequivalence, how much variability does the FDA allow for equivalent drugs?
Dr. Howland: The 2 parameters are the rate of absorption (Cmax) and the extent of absorption (AUC). According to the FDA guidelines, the Cmax and the AUC are determined for the test drug and for the innovator or brand-name drug. Because these parameters are determined for a group of persons taking the test drug and a group taking the brand-name drug, the average values for Cmax and AUC are calculated, along with their degree of variability. Variability is a measure of how much the actual data for Cmax and AUC vary from the average values. The ratio of Cmax for test drug over Cmax of brand-name drug is determined. Similarly, the ratio of AUC for test drug over brand-name drug is determined. A difference between test drug and brand-name drug of greater than 20% for Cmax or for AUC is considered significant by the FDA, so the lower limit of the Cmax and AUC ratios must be 80%. Because the data are log-transformed, the upper limit of these ratios must be 125% (that is, the reciprocal of the lower 80% limit, which is 1 divided by 80%).
Because Cmax and AUC are average values, the variability of the data around these averages is also calculated. Variability is an important concept. If you think about a roomful of athletes, some of whom are basketball players and others are jockeys and gymnasts, you can calculate the average height, but there's a lot of variability. According to FDA guidelines, a 90% confidence interval is used for these averages. In other words, the average values for Cmax and AUC are in the center of the confidence interval, and 90% of the data are contained within the interval.
What becomes confusing, if you think about the difference between 80% and 125%, is that a 45% difference seems very large. To take into account the inherent variability around the average values for Cmax and AUC, the FDA specifies that the ratios for Cmax and AUC, along with their 90% confidence intervals, must all fall within the 80%-125% window for a generic drug to be considered bioequivalent to a brand-name drug.
Because the averages and the 90% confidence intervals have to fall within the 80%-125% range, differences between generic drugs and brand-name drugs don't vary so extensively if they fulfill the FDA's definition of bioequivalence. And you have to satisfy those criteria for both the Cmax and the AUC.
[Editor's note: A summary of FDA definitions for bioequivalence methodology can be found here.]
Medscape: So, the FDA is looking at how much each drug varies within itself, in addition to how different they are from each other?
Dr. Howland: Right, because when you do test studies of the bioavailability of a brand-name drug, or any other medication, there's normal variance from one person to another. If you conduct a study and you have 20 people, there's going to be normal variability in terms of absorption. They collect data in all the participants, and they determine the average Cmax and also the AUC. But, to take into account the variability, and to make sure the variability of the proposed generic is not significantly different from that of the brand-name drug, the FDA wants to make sure that the variability still falls within that 80%-125% range. If any part of the variability falls outside that range, the drugs are not considered bioequivalent.
It's tough to describe these things without looking at graphs, or especially understanding confidence intervals and how the FDA defines the 80% and 125% boundaries, because that's what creates a lot of confusion in the minds of physicians and the public. They think, "Wow, that's a 45% variability. If I get a drug that's at the 81st percentile, that might be quite different from another," but that really isn't what's being measured.
Medscape: In addition to having established bioequivalence, will the generic drugs also carry the warnings and other FDA language associated with the branded drugs?
Dr. Howland: Because the FDA intends for generic drugs and brand-name drugs to be therapeutically equivalent -- having the same active ingredient and being bioequivalent -- they mandate that the generic drug labeling be similar to the previously approved labeling for a brand-name product. There was a recent US Supreme Court ruling (Pliva, Inc. v. Mensing) on a case involving the labeling of a generic product in terms of serious adverse effects. There is a product, metoclopramide (Reglan®), which is used for treating nausea, and it's a dopamine-blocking drug. There are many generic formulations of Reglan, and like other dopamine-blocking drugs, such as antipsychotics, it carries the risk for movement disorders, including tardive dyskinesia.
My understanding of this case is that a woman who had developed tardive dyskinesia while taking a generic form of Reglan sued, arguing that, consistent with state law, the label should have mentioned this possibility. The Supreme Court ruled that a generic manufacturer of an old drug can't necessarily be held liable for a safety change, because they're bound to maintain the label of the originator drug, according to the FDA guidelines.
Labeling for generics has to be very similar to brand-name products. The issue for some older drugs is that you can't even prescribe the brand-name version anymore because it's not manufactured. New knowledge about the drug may not be incorporated into a product label because there's no brand-name product manufacturer who's being mandated to change the label in accordance with FDA guidelines on new safety data. A person taking an old medication that's only available as a generic may be exposed to risks that the original manufacturer would have been obliged to report to the FDA.
Source...