Clozapine: Controversy Erupts Over Brand/Generic Equivalence
Clozapine: Controversy Erupts Over Brand/Generic Equivalence
Study Contradicts AB Rating, But One Patient May Hold Key
In a brewing war similar to recent "narrow therapeutic index" arguments over warfarin, Novartis and Zenith Goldline are at odds over the interchangeability of brand-name and generic formulations of clozapine. At the heart of the issue are two studies, one conducted by Zenith and used by FDA to give the generic product an "AB" rating, and the other a Novartis-sponsored assessment that shows pharmacokinetic differences between the products. In arguments that make "dimpled chads" sound like a refreshing change of pace, the two companies are throwing statistics and standards around in confusing, contradictory presentations.
The controversy gained national attention when the Wall Street Journal reported on October 24 a study by Larry Ereshefsky, PharmD, professor of pharmacy and psychiatry and director of the Psychiatric Pharmacy Program at the University of Texas College of Pharmacy. Ereshefsky's findings suggest that the Zenith product might not be bioequivalent. Based on the study, the Texas Department of Mental Health has advised hospitals to use caution when switching patients, and FDA is reviewing the data for possible action -- including a requirement for Zenith to conduct further studies. Novartis -- sponsor of Ereshefsky's study -- is using his data to discourage switches to the less expensive generic rival.
Zenith Goldline, its parent company IVAX, and the Generic Pharmaceutical Association have attacked Ereshefsky's work in press releases, letters to FDA, and news reports. "The medical community is sophisticated," said IVAX President and Vice Chairman Neil Flanzraich in a prepared statement, "and we are confident that the misleading study sponsored by our competitor Novartis...will have no significant effect on our sales of clozapine." But just in case, the company has launched a lawsuit against Novartis.
"I did this study because I wanted to know if one drug is different from the other [in terms of] of public policy, mental health, and patient protection," Ereshefsky said. "I don't really care which drug is better. There have been enough reports of sporadic worsening when patients are switched to the generic to say maybe there is something there."
Ereshefsky took 21 patients with severe schizophrenia who had been stable on Clozaril for at least 3 months and randomized them to receive either generic clozapine or Clozaril for 2 weeks. Pharmacokinetic and pharmacodynamic parameters were assessed at the end of that time. Patients were then treated with the alternate drug for an additional two weeks, and parameters were again assessed. The patients' other medications were not discontinued during the study, nor were patients blinded to the switch or observed between testing days.
Ereshefsky found no statistically significant difference in the extent of absorption between the two drugs, represented in Table 1 as AUC (area under the curve). However, he did find a statistically significant difference in maximum concentration (Cmax), a function of absorption rate and extent. "Clozaril is absorbed faster," Ereshefsky said, explaining the data. "That's where we see the difference. Patients on bigger doses of Clozaril had a spike in their blood level that was not apparent in patients on the generic. Higher peaks in Clozaril levels resulted in increased pulse, higher core temperatures, and slightly reduced cognitive measures."
"If I had a new patient, I might prefer the generic over Clozaril because the higher peak blood levels might not be good. That's one way to look at the data. But we did not," Ereshefsky emphasized, "do a bioequivalence study. The study was not powered to prove or disprove bioequivalence." Ereshefsky had screened 33 patients, randomized 26 patients into the study, and obtained complete results from 21 patients. He had requested funding for a larger number of patients but was denied by Novartis.
Ereshefsky's results will be published shortly in a supplement to the Journal of Clinical Psychiatry. Zenith Goldline was the only company that had a generic clozapine on the market when the study was conducted, but Mylan has since had its own formulation approved.
Ereshefsky suspects the difference in absorption rates is related to a difference in how the drugs are made. "The ratio of binders and excipients to active drugs, the pressure used to make the tablet, the content of the tablet [lipophilicity] -- there are a variety of pharmaceutics issues there," he explained.
However, professor Art Kibbe, PhD, Chair of the Wilkes University's Department of Pharmaceutical Sciences, thinks the rate difference may have more to do with one of the study subjects.
Kibbe has conducted many bioequivalence studies during his career. He has testified before Congress on generic drug approvals and was appointed by former FDA Commissioner James Benson to a task force that troubleshot the approval process and recommended improvements in the early 1990s.
"Patient 114 is a clear outlier," Kibbe said. "His peak concentration for the generic is about 650 ng/mL and his peak for Clozaril is 1170 ng/mL, which means his Cmax is more than 1.5 standard deviations away from other patients'. It's so dramatically different that it skews all the other data." (See Figure A).
FIGURE
Kibbe believes the numbers could be skewed because patient 114 was an alcoholic, which calls adherence into question. Given the "crossover" study design, it is possible patient 114 faithfully took Clozaril for 2 weeks, but did not take the generic alternative.
Kibbe suspects that if patient 114 were taken out of the analysis, the statistical difference in Cmax -- and the difference in rate of absorption that it suggests -- would disappear. "As it is now, the Clozaril Cmax is about 900 ng/mL and the Cmax for clozapine is 800 ng/mL. If subject 114 were removed, the Clozaril Cmax curve would drop and the clozapine Cmax curve would rise, perhaps even overlap." (See Figure B)
FIGURE
Zenith's original bioequivalency study -- the data that helped win the AB rating -- supports the argument. The blood level curves are almost superimposable. (See Figure C).
FIGURE
A 12.5 mg test dose was used to approve the generic because clozapine has linear pharmacokinetics -- the excretion rate is the same at all doses. It therefore was not necessary to replicate the 300 to 400 mg dosages used by patients in actual practice, dosages that would have put healthy test subjects at risk for the drug's hematologic and neurotoxic side effects.
"The study makes a big deal out of a little difference in Cmax that could disappear if one patient were excluded," Kibbe concluded. "The [Zenith] data clearly indicate that the generic should have been approved. Should we use Clozaril or generic clozapine? Generic clozapine is fine."
"With only 21 subjects, [patient 114] could affect our numbers," Ereshefsky acknowledged, "but there was no reason to exclude this patient from the analysis. The patient was drug and alcohol free throughout the study."
Ereshefsky also notes that patient 114 decompensated while taking the generic product and started drinking after the study ended. This, he argues, could indicate that the generic, with its slower absorption rate, did not adequately treat the patient's schizophrenia.
But the effectiveness of clozapine -- both brand and generic -- fluctuates over the course of therapy, and critics are quick to note that this could also explain the decline, as could the fact that the patient knew he had been taken off Clozaril -- a drug he had done well on and most likely trusted. Patient 114 was the only one to get worse while taking the generic.
Approximately 26,000 of the 75,000 patients who take clozapine use the generic, IVAX noted in a recent press release, and it claims that 21,000 of those patients were switched without incident from Clozaril.
FDA has subpoenaed Ereshefsky's data, and his study has been questioned in the press and on pharmacy listservs. He'll likely be called to testify as IVAX and Novartis wage battle in court.
Not surprisingly, Ereshefsky takes a grim view of the controversy surrounding his study. "As far as I am concerned, Novartis and IVAX are both interested in marketing and sales. Any time you've got hundreds of millions of dollars at stake, the science goes in the toilet and you end up with a bunch of innuendo."
In a year of "chad," truer words may never have been spoken.
Introduction
Study Contradicts AB Rating, But One Patient May Hold Key
In a brewing war similar to recent "narrow therapeutic index" arguments over warfarin, Novartis and Zenith Goldline are at odds over the interchangeability of brand-name and generic formulations of clozapine. At the heart of the issue are two studies, one conducted by Zenith and used by FDA to give the generic product an "AB" rating, and the other a Novartis-sponsored assessment that shows pharmacokinetic differences between the products. In arguments that make "dimpled chads" sound like a refreshing change of pace, the two companies are throwing statistics and standards around in confusing, contradictory presentations.
Texas Study Stirs Scrutiny
The controversy gained national attention when the Wall Street Journal reported on October 24 a study by Larry Ereshefsky, PharmD, professor of pharmacy and psychiatry and director of the Psychiatric Pharmacy Program at the University of Texas College of Pharmacy. Ereshefsky's findings suggest that the Zenith product might not be bioequivalent. Based on the study, the Texas Department of Mental Health has advised hospitals to use caution when switching patients, and FDA is reviewing the data for possible action -- including a requirement for Zenith to conduct further studies. Novartis -- sponsor of Ereshefsky's study -- is using his data to discourage switches to the less expensive generic rival.
Zenith Goldline, its parent company IVAX, and the Generic Pharmaceutical Association have attacked Ereshefsky's work in press releases, letters to FDA, and news reports. "The medical community is sophisticated," said IVAX President and Vice Chairman Neil Flanzraich in a prepared statement, "and we are confident that the misleading study sponsored by our competitor Novartis...will have no significant effect on our sales of clozapine." But just in case, the company has launched a lawsuit against Novartis.
"I did this study because I wanted to know if one drug is different from the other [in terms of] of public policy, mental health, and patient protection," Ereshefsky said. "I don't really care which drug is better. There have been enough reports of sporadic worsening when patients are switched to the generic to say maybe there is something there."
Not a Bioequivalence Study
Ereshefsky took 21 patients with severe schizophrenia who had been stable on Clozaril for at least 3 months and randomized them to receive either generic clozapine or Clozaril for 2 weeks. Pharmacokinetic and pharmacodynamic parameters were assessed at the end of that time. Patients were then treated with the alternate drug for an additional two weeks, and parameters were again assessed. The patients' other medications were not discontinued during the study, nor were patients blinded to the switch or observed between testing days.
Ereshefsky found no statistically significant difference in the extent of absorption between the two drugs, represented in Table 1 as AUC (area under the curve). However, he did find a statistically significant difference in maximum concentration (Cmax), a function of absorption rate and extent. "Clozaril is absorbed faster," Ereshefsky said, explaining the data. "That's where we see the difference. Patients on bigger doses of Clozaril had a spike in their blood level that was not apparent in patients on the generic. Higher peaks in Clozaril levels resulted in increased pulse, higher core temperatures, and slightly reduced cognitive measures."
"If I had a new patient, I might prefer the generic over Clozaril because the higher peak blood levels might not be good. That's one way to look at the data. But we did not," Ereshefsky emphasized, "do a bioequivalence study. The study was not powered to prove or disprove bioequivalence." Ereshefsky had screened 33 patients, randomized 26 patients into the study, and obtained complete results from 21 patients. He had requested funding for a larger number of patients but was denied by Novartis.
Ereshefsky's results will be published shortly in a supplement to the Journal of Clinical Psychiatry. Zenith Goldline was the only company that had a generic clozapine on the market when the study was conducted, but Mylan has since had its own formulation approved.
Why the Difference?
Ereshefsky suspects the difference in absorption rates is related to a difference in how the drugs are made. "The ratio of binders and excipients to active drugs, the pressure used to make the tablet, the content of the tablet [lipophilicity] -- there are a variety of pharmaceutics issues there," he explained.
However, professor Art Kibbe, PhD, Chair of the Wilkes University's Department of Pharmaceutical Sciences, thinks the rate difference may have more to do with one of the study subjects.
Kibbe has conducted many bioequivalence studies during his career. He has testified before Congress on generic drug approvals and was appointed by former FDA Commissioner James Benson to a task force that troubleshot the approval process and recommended improvements in the early 1990s.
"Patient 114 is a clear outlier," Kibbe said. "His peak concentration for the generic is about 650 ng/mL and his peak for Clozaril is 1170 ng/mL, which means his Cmax is more than 1.5 standard deviations away from other patients'. It's so dramatically different that it skews all the other data." (See Figure A).
FIGURE
Kibbe believes the numbers could be skewed because patient 114 was an alcoholic, which calls adherence into question. Given the "crossover" study design, it is possible patient 114 faithfully took Clozaril for 2 weeks, but did not take the generic alternative.
Closing the Gap
Kibbe suspects that if patient 114 were taken out of the analysis, the statistical difference in Cmax -- and the difference in rate of absorption that it suggests -- would disappear. "As it is now, the Clozaril Cmax is about 900 ng/mL and the Cmax for clozapine is 800 ng/mL. If subject 114 were removed, the Clozaril Cmax curve would drop and the clozapine Cmax curve would rise, perhaps even overlap." (See Figure B)
FIGURE
Zenith's original bioequivalency study -- the data that helped win the AB rating -- supports the argument. The blood level curves are almost superimposable. (See Figure C).
FIGURE
A 12.5 mg test dose was used to approve the generic because clozapine has linear pharmacokinetics -- the excretion rate is the same at all doses. It therefore was not necessary to replicate the 300 to 400 mg dosages used by patients in actual practice, dosages that would have put healthy test subjects at risk for the drug's hematologic and neurotoxic side effects.
"The study makes a big deal out of a little difference in Cmax that could disappear if one patient were excluded," Kibbe concluded. "The [Zenith] data clearly indicate that the generic should have been approved. Should we use Clozaril or generic clozapine? Generic clozapine is fine."
In Defense of Patient 114
"With only 21 subjects, [patient 114] could affect our numbers," Ereshefsky acknowledged, "but there was no reason to exclude this patient from the analysis. The patient was drug and alcohol free throughout the study."
Ereshefsky also notes that patient 114 decompensated while taking the generic product and started drinking after the study ended. This, he argues, could indicate that the generic, with its slower absorption rate, did not adequately treat the patient's schizophrenia.
But the effectiveness of clozapine -- both brand and generic -- fluctuates over the course of therapy, and critics are quick to note that this could also explain the decline, as could the fact that the patient knew he had been taken off Clozaril -- a drug he had done well on and most likely trusted. Patient 114 was the only one to get worse while taking the generic.
Approximately 26,000 of the 75,000 patients who take clozapine use the generic, IVAX noted in a recent press release, and it claims that 21,000 of those patients were switched without incident from Clozaril.
Lessons From the Frontline
FDA has subpoenaed Ereshefsky's data, and his study has been questioned in the press and on pharmacy listservs. He'll likely be called to testify as IVAX and Novartis wage battle in court.
Not surprisingly, Ereshefsky takes a grim view of the controversy surrounding his study. "As far as I am concerned, Novartis and IVAX are both interested in marketing and sales. Any time you've got hundreds of millions of dollars at stake, the science goes in the toilet and you end up with a bunch of innuendo."
In a year of "chad," truer words may never have been spoken.
Source...