Reviews/Evaluations

 

Generic Substitution

 

Overview

The 1984 Drug Competition and Patent Term Restoration Act, also called the Hatch-Waxman Act, expedited approval of generics. Generic use climbed 19% that year, driven by economic concerns.1 Generics impact budgets and public policy. Improved access to generics saved $8-$10 billion at retail according to the Congressional Budget Office. 2

Manufacturers of generic drugs file an Abbreviated New Drug Application (ANDA) with the Food and Drug Administration (FDA). An ANDA requires generic manufacturers to certify they will not infringe upon a patent listed in the Orange Book.3 The first manufacturer to file is eligible for 180 days of generic exclusivity from the time of marketing.

Obstacles delay availability of generics. The Federal Trade Commission (FTC) found the 180 exclusivity periods and patent litigation may have prevented availability of generics, and took antitrust action against some companies.1 State attorney generals have participated in class action suits about collusion and anti-competitive practices concerning generics.4 The FDA is reviewing comments prompted by amendment of the Hatch-Waxman Act by the Medicare Prescription Drug, Improvement, and Modernization Act of 2003.5

Generic Quality

FDA states the same standards apply to generics and innovator drugs.2 The agency requires the same quality, purity, and stability for both. This includes identical active ingredient(s), strength, dosage form, and method of administration as the reference drug. Generics may differ in inert ingredients, such as excipients, binders, fillers, coloring agents, flavors, preservatives, release mechanisms, and scoring configuration. In some patients, these differences may induce allergies.6 Firms provide full descriptions of facilities to process, test, package, label, and control the drug; comply with federal Good Manufacturing Practices; and undergo FDA inspection. FDA must approve any major change or reformulation. An allowable variation in content can occur among lots and even within a single lot of both brand drugs and generic drugs.

The FDA states generics possess the same benefit/risk assessment for efficacy and safety. ANDAs do not require clinical trials.2 Labeling must be essentially the same. Manufacturers must report adverse effects. No evidence of more adverse drug reactions has been found with generic drugs. The focus of concern is bioequivalence.7,8

Bioequivalence

The Hatch-Waxman Act established bioequivalency as the basis for approving generics. Generics must deliver the same amount of active ingredient at the same rate and extent of absorption as the reference drug.2

In 1986 FDA established a methodology for an average bioequivalence test.9 Data from a single-dose, 2-way crossover bioavailability study of 24 to 36 healthy, normal volunteers are analyzed using a complex statistical model that allows evaluation of the least squares means of the bioavailability parameters and their standard errors. Results are used to construct a 90% confidence interval (CI) around the peak serum concentration (Cmax) and areas under the curve (AUC) of the generic falling within 80-125% of the reference drug. This measurement of bioequivalence is not the same as therapeutic equivalence for narrow therapeutic index (NTI) drugs and could have adverse clinical outcomes, according to Medical Letter.10

According to FDA, they have received no documented examples of failure of a generic due to bioequivalence determination. Products declared bioequivalent should not require additional clinical testing or monitoring.8,9

Substitutability of generics remains contentious for narrow therapeutic index (NTI) drugs. NTIs are defined in federal regulation (21 CFR 320.33(c)) as follows: 1. there is less than a 2-fold difference in median lethal dose (LD50) and median effective dose (ED50) values, or 2. there is less than a 2-fold difference in the minimum toxic concentrations and minimum effective concentrations in the blood, and 3. Safe and effective use of the drug products require careful titration and patient monitoring.

NTIs include levothyroxine and warfarin. In a letter to the National Boards of Pharmacy on 4-16-97, the FDA stated they believed drugs do not fall into discrete groups that would allow one to consider NTI drugs as clearly different from other drugs for purposes of therapeutic substitution. FDA claimed they had not received any data to support changes in relation to NTI drugs.11

Countering the FDA is a single-dose, open-label pharmacokinetic study of LT4 in 36 healthy euthyroid volunteers.12 Using three doses (400, 450, and 600 mcg) of Synthroid, the study compared FDA methodology with three modifications to correct for endogenous thyroxine levels to determine if these could distinguish dose differences of 12.5%, 25%, or 33%. None of the methods distinguished differences of 12.5%. Without baseline correction for endogenous source, differences of 25-33% were not detected. The three mathematical correction methods did distinguish differences of 25-33%. The study, sponsored by the manufacturer, was cited by the American Association of Clinical Endocrinologists, The Endocrine Society, and American Thyroid Association in a joint statement of concern on interchangeability of thyroxine products in 2004. These organizations object to FDA reliance on pharmacokinetic methods to establish therapeutic equivalence, to not accounting for endogenous contribution to thyroxine levels, and not using TSH levels in determinations of equivalence. They believe a 33% or 25% difference in thyroxine dose may have substantial clinical impact, especially in the elderly, cardiac disease, or pregnancy. Their best physician practice calls for patients to be maintained on the same formulation of levothyroxine, not switching from brand to generic, or among generic manufacturers.13

Warfarin

Warfarin generic substitutability has been extensively litigated.7 Generic warfarin, withdrawn in 1992, was reintroduced in 1997.7,14 There are many other variables that contribute to difficulties in stabilizing INR, such as dietary vitamin K, drug interactions, exercise, CHF, diarrhea, alcohol, and non-compliance.8,9

Seven citations for clinical studies comparing Coumadin to generic products were identified via Medline query. Two involved generics not available in the United States.15,16 One randomized, controlled, crossover, observer-blind trial was available in abstract only.17 It enrolled 55 atrial fibrillation patients at a VA hospital clinic, but studied only 39 that did not require warfarin dose changes or experience adverse events during the study period of 42 days.

The remaining four clinical studies are summarized in Table 1. Each concluded therapeutic equivalency with some caveats. Milligan et. al. used a statistical process control, common to quality improvement evaluations, to detect changes that were not random. There were 12 serious adverse events prior to switch and 3 after, suggesting the sample size was too small to detect differences in rare adverse events. Witt et al had a statistical difference in the primary endpoint, “Time in target INR range” but the clinical relevance of this difference was not borne out in number of INR tests per patient or dose changes per patient. However, 72% of patients did have a 10% change in INR control during the study period. Weibert et al did not detect differences in any endpoints, but the external validity of this study is minimal. Less than 10% of patients screened were enrolled and the focus of the study was primarily men who were on stable doses of warfarin, being treated for atrial fibrillation. Finally, Swenson et al compared patients who volunteered to switch to control patients and found no differences in INR control.

Table 1 Published Trials Comparing Coumadin with Generic Warfarin

Authors	Study Design	Population	Outcome Measures	Results
Milligan PE, et al7	Prospective; Observational (non-blinded, non-randomized) Statistical Process Control	n=182 (8 lost to follow-up)	INR Control = (# in INR Target Range/ # Total INR)/month	No detectable change
		No detectable change Mean Age = 75.1 Female = 43.4%	INR Frequency = # of INRs/patient-days	No detectable change
	Voluntarily switched: Coumadin to Barr warfarin Data collected 8 month prior and 10 month after switch	Indication: AF = 60.4% Valve = 13.2% TIA/stroke = 11% DVT/PE = 8.8% other = 6.6%	Rate of Dose Changes = # Dose Changes/patient-days	No detectable change
	BJC Health System & Washington University Physicians Network multidisciplinary anticoagulation clinic		Rate of thrombotic/hemorrhagic events needing emergency evaluation or hospital admit	No detectable change
Witt DM, et al14	Retrospective; uncontrolled	n=2299	Primary: Time in target INR range during study period	Pre = 65.9% + 28.5 Post = 63.3% + 28.7 p = 0.0002
	Mandated switch: pre 90days - Coumadin post 90days - Barr warfarin	Mean Age = 68.6; Female = 46.4%	Secondary: # of INR tests / patient	NS
	Kaiser, Colorado pharmacist-managed anticoagulation clinic	Indication: AF = 40.7% Valve = 12% TIA/stroke = 7.7% DVT/PE = 27% other = 12.6%	Mean Dose	NS
			Percentage of patients with >1 dose change	NS
			Mean Complication Rate/100 patient-years	NS
Weibert RT18	Prospective; randomized; single-blind; multi-center; cross-over design Coumadin vs. Apothecon warfarin 4 week run-in 4 weeks of each product post 4 weeks returned to Coumadin prn MD, RN or RPh managed anticoagulation clinics	1181 patients screened; n=113 (9 did not complete) Mean Age = 70 Female = 25% Stable AF patients only = same dose and min of 2 INRs between 1.8 - 3.2 in 4 weeks prior Excluded patients with risk factors: alcohol use; dietary fluctuations; liver disease, etc.	Dose change ³ 20%	Apothecon = 7 Coumadin = 7
			Patients out of INR range	< 1.8: Apothecon = 9 Coumadin = 9
				> 3.2: Apothecon = 10 Coumadin = 9
			Hemorrhagic/thrombolytic events	none for either product
Swenson CN, et al19	Prospective, observational; non-randomized; unblinded; unmatched control	n=210 105 control 105 conversion	INR either clinically important (>0.5) or statistically significant (p < 0.5) difference from baseline	Control: Pre INR = 2.6 + 0.4; Post INR = 2.8 + 0.5 Conversion: Pre INR = 2.6 + 0.4; Post INR = 2.7 + 0.5 NS
	Coumadin vs. Barr warfarin Data collected 4 weeks prior and 8 weeks after switch	³ 3mths Coumadin therapy prior and INR in target at baseline
	Group Health Cooperative of Puget Sound anticoagulation clinics	Baseline Demographics comparable: Mean Age: 76-78 Female: 49.5-50.5% AF: 64-70%

 

Levo-Thyroxine

Before 1962 levothyroxine, an unstable drug, lacked bioequivalence data. Between 1987 and 1994, fifty-eight Adverse Drug Reports (ADR) were filed with FDA on levothyroxine. Forty-seven related to sub potency, nine to super potency. These ADR were caused by switching products and also by inconsistencies in bioavailability between lots from the same manufacturer.9

Literature on levothyroxine trials with generics is sparse. The primary study is summarized in Table 2. Published in 1997, it concluded Synthroid, Levoxyl, and two generics to be bioequivalent and interchangeable in the majority of patients. Controversy swelled with publication delayed seven years by the study sponsor, who marketed Synthroid.9 These products have since been reformulated and thus the study’s current relevance is diminished.

Table 2 Published Trials Comparing Synthroid with Generic Levothyroxine

Authors	Study Design	Population	Results
Dong BJ et al20	Single-Blind, Random, 4-way crossover comparison of Synthroid, Levoxyl, & 2 generic l-thyroxine 6 mos.	22 women Hypothyroid, euthyroid 0.1-1.5 mg l-thyroxine	Total thyroxine, Total triiodothyronine Thyroxine Index No Significant Differences

In 2000, FDA issued the first NDA for oral levothyroxine.9 FDA currently rates Synthroid (Abbot), Levo-T (Alara distributed by Sandoz), and levothyroxine by Mylan as AB2, a rating that indicates changing from one product to another is unlikely to cause variation in thyroid function.21 New studies are needed using generics marketed since the NDA.

Recommendations

It is recommended that when patients are newly prescribed warfarin or levothyroxine that an AB rated generic be specified initially. Prescribers can indicate not to switch manufacturers even of generics. Switching Coumadin to an AB rated warfarin in stable patients could be considered as studies of this switch have successfully been repeated. For patients already taking Synthroid, switching to a generic lacks substantial data at this time.

References

1. US Federal Trade Commission, Generic Drug Entry Prior to Patent Expiration: An FTC Study, July, 2002, http://www.ftc.gov/opa/2002/07/genericdrugstudy.htm, accessed 11-14-04
2. US Food & Drug Administration, Equivalence of Generic Drugs, September, 1999, http://www.fda.gov/cder/consumerinfo/generic_equivalence.htm, accessed 10-21-04
3. A Primer: Generic Drugs, Patients and the Pharmaceutical Marketplace, National Institute for Health Care Management Research and Educational Foundation, June, 2002
4. Families USA, Collusion and Other Anticompetitive Practices: A Survey of Class Action Lawsuits Against Drug Manufacturers, January, 2003, http://www.familiesusa.org
5. US Food & Drug Administration, Guidance for Industry Listed Drugs, 30-Month Stays, and Approval of ANDAs and 505(b) (2) Applications Under Hatch-Waxman as Amended by the Medicare Prescription Drug Improvement, and Modernization Act of 2003, Questions and Answers, http://www.fda.gov/cder/guidance/6174dft.htm, accessed 11-5-04
6. US Food & Drug Administration, Approved Drug Products with Therapeutic Equivalence Evaluations, 24th Edition, 2-26-04, http://www.fda.gov/cder/ob/docs/preface/ecpreface.htm, accessed 10-21-04
7. Milligan P E et al, Substitution of Generic Warfarin for Coumadin in an HMO Setting, Ann Pharmacother 2003; 36:764-78
8. Noviasky, JA, Evidenced-based pharmacy versus opinion on generic product selection of warfarin, Am J Health-Syst Pharm 1999, 56: 2246-7
9. Henderson JD and Esham RH, Generic Substitution: Issues for Problematic Drugs, SMJ 2001, 94 (1); 16-21
10. Generic Levothyroxine, The Medical Letter, 46 (1192): 77-78, 9-27-04
11. US Food & Drug Administration, Therapeutic Equivalence of Generic Drugs Response to National Association of Boards of Pharmacy, http://www.fda.gov/cder/news/ntiletter.htm, accessed 10-21-04
12. Blakesley V et al, Are Bioequivalence Studies of Levothyroxine Sodium Formulations in Euthyroid Volunteers Reliable?, Thyroid, 2004, 14: 191-200
13. AACE, TES, and ATA, Joint Position Statement on the Use and Interchangeability of Thyroxine Products, 2004, http://www.aace.com/clin/guidelines/AACE, accessed 10-11-04
14. Witt DM et al, Generic Warfarin Substitution, Pharmacotherapy, 2003; 23 (3): 361-8
15. Halkin H et al, Increased warfarin doses and decreased international normalize ratio response after nationwide generic switching, Clin Pharmacol Ther 2003; 74: 215-21
16. Yacobi A et al, Who needs individual bioequivalence studies for narrow therapeutic index drugs: A case for warfarin, J Clin Pharmacol 2000 Aug; 40 (8): 826
17. Neutel, JM, Smith DHG, A randomized crossover study to compare the efficacy and tolerability of Barr warfarin sodium to the currently available Coumadn, Cardiovasc Rev Rep 1998; 19:49-59
18. Weibert RT et al, A Randomized, Crossover Comparison of Warfarin Products in the Treatment of Chronic Atrial Fibrillation, Ann Pharmacother 2000, 34: 981-8
19. Swenson CN and Fundak G, Observational cohort study of switching warfarin sodium products in a managed care organization, Am J Health-Syst Pharm 2000, 57:452-5
20. Dong BJ et al, Bioequivalence of Generic and Brand-name Levothyroxine Products in the Treatment of Hypothyroidism, JAMA, 1997; 277: 1205-1213
21. US Food & Drug Administration, thyroxine, warfarin, http://www.accessdata.fda.gov/scripts/cder/ob/docs/tempai.cfm, accessed 11-3-04