Clinical Infectious Diseases Advance Access published November 21, 2013
1
Efficacy and quality of antibacterial generic products
cr ipt
approved for human use: a systematic review
Pierre Tattevina,b,*, Anne-Claude Crémieuxc,d, Christian Rabaude, Rémy Gauzitf a
Pontchaillou Univ. Hosp., Rennes, France
us
INSERM U835, Université Rennes 1, IFR140, F-35033, Rennes, France
c
EA 3647, Versailles Saint-Quentin Univ., Versailles, France Raymond Poincaré Univ. Hosp., Garches, France
e
Brabois Univ Hosp, Nancy, France
f
an
d
Hôtel- Dieu Univ Hosp, Assistance Publique-Hôpitaux de Paris (AP-HP), Paris, France Corresponding author. Address: Service des Maladies Infectieuses et de Réanimation
M
*
Médicale, CHU Pontchaillou, 2 rue Henri Le Guilloux, 35033 Rennes Cedex, France. Tel
pt ed
+33 299289564 ; fax +33 299282452, E-mail address: [email protected] Alternate corresponding author. Réanimation, Hôpital Hôtel-Dieu, Assistance PubliqueHôpitaux de Paris (AP-HP), 1 place du Parvis Notre Dame, 75004 Paris, France. E-mail
ce
[email protected]
KEY POINTS
Ac
A systematic literature review on antibacterial generics approved for use in humans analyzed 37 original articles with heterogeneous design, inconsistent findings, and discrepancies. Additional evidence are needed before considering a revision of the marketing authorization process for antibacterial generic products.
© The Author 2013. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: [email protected]
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
b
2 ABSTRACT
antibacterial generic products approved for use in humans.
cr ipt
Background Concerns have recently emerged about the efficacy and the quality of
Methods We searched Medline and Embase for original research articles on antibacterial generic products published in English or French before July 2013.
Results We selected 37 original research articles: 15 on beta-lactams, 10 on
us
published during 2008-2012. Study designs included analytical chemistry (n=9), in vitro
an
susceptibility studies (n=14), animal experiments (n=6, including 5 using the neutropenic mouse thigh infection model), and clinical studies in humans (n=15). Of the 37 studies,
M
14 (37.8%) suggested that some generic products may be inferior to the innovator in terms of purity (n=2), in vitro activity (n=3), in vivo efficacy in experimental models (n=4), clinical efficacy (n=2), taste (n=2), or compliance and acceptability in children
pt ed
(n=1). The majority of in vitro studies (78.6%) found no significant difference between generic products and the innovator. Most (5/6) in vivo studies suggesting a difference between generic products and the innovator were performed in an animal model that is not validated for the evaluation of the efficacy of antibacterial agents. The level of
ce
evidence was constantly low in clinical studies. Conclusions Published data on antibacterial generic products are limited and
Ac
heterogeneous, thus precluding any attempt to generalize the study results. This systematic review suggests that additional evidence would be needed before considering a revision of the marketing authorization process for antibacterial generic products.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
glycopeptides, and 12 on other antibacterial agents. The majority of articles (73.0%) were
3 INTRODUCTION Generic medicinal products are copies of patented drugs and can be marketed at low cost
cr ipt
following patent expiration of the brand leader product. As all pharmaceutical products, generic products must comply with standards of quality, efficacy, and reliability. The regulatory authorities of several countries, including the US Food and Drug
Administration (FDA) and the European Medicines Agency (EMA), and the World
us
under which generic medicinal products can be recognized as therapeutically equivalent
an
to their brand name counterpart (reviewed in [1]). Bioequivalence principles have been defined (i.e., 20% acceptance range (80–125%) for the 90% confidence interval of the ratio between test and reference least square means after log-transformation of the
M
pharmacokinetic parameters of interest, Cmax and area under curve (AUC)). The objectives of drug policies supporting the use of generic medicinal products are
pt ed
essentially economic: i) to decrease the cost of medicines as part of healthcare spending, in particular in developed countries (e.g., in France, it is estimated that €1.3 billion were saved by introducing such policy in 2008 [2]), and ii) to facilitate access to care in developing countries (e.g., expanded access to effective antiretroviral combinations for
ce
HIV-infected patients). However, these economic considerations should not occur at the price of lower quality of patient care.
Ac
The therapeutic equivalence of generic medicinal products approved for use in
humans has been challenged in various therapeutic areas, including neurology [3], endocrinology, and cardiovascular diseases [4]. The question was also raised for antibacterial agents, especially following publication of a study performed with vancomycin generic products suggesting that, despite similar pharmacokinetics
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Health Organization (WHO) have issued guidelines presenting the terms and conditions
4 parameters and in vitro antibacterial activities, some vancomycin generic products were less bactericidal than the innovator in vivo in a neutropenic mouse thigh infection model
cr ipt
[5], and could induce more resistant sub-populations [6]. Following publication of these results, a debate started worldwide, involving the scientific communities, drugs
regulatory agencies, and the general public [7]. Questions were raised about the current assumptions from the WHO, FDA, and EMA, that two products of parenteral use are
us
is of concern, as i) it would imply that patients may currently receive sub-optimal
an
antibacterial therapy depending on the product that they receive; ii) deep changes would be required in the approval process of antibacterial generic products, including the need
M
for more studies on efficacy, which would, in turn, translate into increased costs for approval. This systematic review of the literature was designed to objectively analyze
pt ed
published data regarding the efficacy and quality of antibacterial generic products.
METHODS
Search strategy and selection criteria
We searched Medline and Embase for articles in English or French published anytime
ce
before July 2013, using the following keywords in various combinations “generics”, “generic products”, “antibacterial”, “antibacterial agent”, “antibiotic”, “betalactam
Ac
agent”, “penicillin”, “cephalosporin”, “penem”, “fluoroquinolone”, “aminoglycoside”, “macrolide”, “cycline”, “glycopeptide”, and “vancomycin”.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
considered therapeutically equivalent if they are pharmaceutically equivalent. This issue
5 Data extraction Two independent researchers (PT and RG) reviewed all abstracts to identify articles that
cr ipt
required full-text review, with a final decision reached through consensus. All articles
were discussed with a third reviewer (ACC). For each reviewed article, we extracted data on the study setting, objectives, methods, and results, including details on the evaluated
generic products. We systematically searched for additional articles in the reference lists
RESULTS
us
an
articles available precluded any meta-analysis.
M
We identified 37 studies that met inclusion criteria: 15 focused on beta-lactam agents (Table 1), 10 on glycopeptides (Table 2), and 12 on other antibacterial agents (Table 3). The majority of articles (73.0%) were published during the last five years (2008-2012).
pt ed
Study designs included analytical chemistry (measurement of active pharmaceutical ingredients and/or impurities, n=9), in vitro susceptibility studies (e.g., determination of minimal inhibitory concentrations (MICs), minimal bactericidal concentration (MBCs), time-kill curves, population analysis, n=14), in vivo animal experiments (n=6, including
ce
five studies using the neutropenic mouse thigh infection model), and clinical studies in humans (i.e., pharmacokinetic studies, retrospective cohorts, studies on compliance
Ac
and/or taste, quasi-experimental study, case report, n=15). Studies originated from Asia (n=9), South America (n=8), Europe (n=7), North America (n=6) or covered more than one continent (n=7). Of the 37 studies, 14 (37.8%) suggested that some generic products may be inferior to the innovator in terms of purity (n=2), in vitro activity (n=3), in vivo
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
of all articles reviewed. The heterogeneity of study designs and the limited number of
6 efficacy in experimental models (n=4), clinical efficacy (n=2), taste (n=2), or compliance and acceptability in children (n=1).
cr ipt
Of the 15 studies on beta-lactam generic products, seven suggested that some
approved generic products may be inferior to the innovator. Rodriguez et al. found that, despite similar MICs and MBCs, the nine generic products of oxacillin that they
evaluated in the neutropenic mouse thigh Staphylococcus aureus infection model had
us
46 lots of piperacillin-tazobactam generic products manufactured in 17 countries and
an
found that their in vitro activity, evaluated by incremental MIC antimicrobial assay, was on average 10% lower than that of the innovator [9, 10]. Lambert et al. studied the
M
pharmaceutical qualities of 34 generic products of ceftriaxone approved for use, and found that quality standards were violated on 18 occasions [11]. Subsequently, in a mathematical model based on Monte Carlo simulations, Schito et al. suggested that most
pt ed
ceftriaxone generic products may not reach the required PK/PD parameters [12]. In a quasi-experimental study, Mastoraki et al. found that the incidence of post-cardiac surgery infections was significantly increased in patients who received prophylaxis with a generic product of cefuroxime, as compared to the innovator [13]. Lastly, Cohen et al.
ce
found that the acceptability and the compliance of amoxicillin-clavulanic acid generic products approved for oral use in France were lower than that of the innovator in
Ac
children, based on a questionnaire completed by their parents [14]. Of the 10 studies on glycopeptide generic products, three suggested that some
approved generic products may be inferior to the innovator. Rodriguez et al. published a case report where a liver transplant recipient with persistent bacteremia after 10 days of
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
lower Emax (maximum effect in log10 CFU/g) than the innovator [8]. Jones et al. evaluated
7 intravenous vancomycin generic product had sterile blood cultures 24 hours after a switch to vancomycin innovator [15]. Vesga et al. found that, despite similar MICs and MBCs,
cr ipt
three generic products of vancomycin imported from France, Argentina, and the US, had lower Emax than the innovator in the neutropenic mouse thigh S. aureus infection model [5]. The same team subsequently found, using the same model, that serial exposure to
generic vancomycin enriches resistant subpopulations, while exposure to the innovator
us
Of the 12 studies published to date on generic products of other antibacterial agent
an
classes, four suggested that some approved generic products may be inferior to the innovator. Zuluaga et al. found that, despite similar MICs and MBCs, 10 out of 20
M
generic products of gentamicin had lower Emax than the innovator in the neutropenic mouse thigh Escherichia coli infection model [16]. Nightingale et al. evaluated 65 generic products of clarithromycin from 18 countries and found that 9% did not contain
pt ed
between 95% and 105% of the clarithromycin content claimed in the label, and 19% exceeded the 3% limit for total impurities [17]. Two double-blind studies found that the taste of trimethoprim-sulfamethoxazole generic products was rated lower than the
ce
innovator [18, 19].
Regarding study design, almost half of the studies (n=17) in this systematic
review were performed in humans and studied pharmacokinetic parameters (n=7, mostly
Ac
bioequivalence studies), clinical efficacy (n=6), and tolerability or compliance in children (n=4). Nine studies focused on in vitro efficacy, comparing generic products with the innovator. Six studies were performed in animals, using the neutropenic mouse thigh infection model (n=5) or the rabbit endocarditis model (n=1). Five studies investigated
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
reduces resistant subpopulations [6].
8 the pharmaceutical qualities (purity, content, and potency) of antibacterial generic products. This heterogeneity in the outcomes measured further precluded any meta-
cr ipt
analysis, as pooling the studies focusing on major issues (effectiveness in Humans), relatively major issues (potency in vitro or in animal models), and relatively less
important - though still significant – issues (taste), would make little sense, and be of
an
DISCUSSION
This systematic review of studies that evaluated antibacterial generic products reveals
M
that the data available to date in the literature are limited and heterogeneous. This precludes any attempt to generalize their findings. Indeed, even for the classes of antibacterial agents who received the most attention (i.e., beta-lactams and
pt ed
glycopeptides), published studies have not consistently demonstrated that some generic products approved for use in humans were inferior to the innovator. The level of evidence from the six clinical studies which evaluated the efficacy of antibacterial generic products was low. One case report and one quasi-experimental study suggested that generic
ce
products of, respectively, vancomycin and cefuroxime, were less potent than the innovator, while three retrospective cohort studies that enrolled a total of 1,597 patients
Ac
treated with meropenem or imipenem/cilastatin found no significant difference between the generic products and the innovator in terms of clinical outcome. One randomized, open-label, clinical trial, found that a generic product of clarithromycin was not significantly different to the innovator in terms of clinical or bacteriological efficacy, and
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
us
limited clinical relevance.
9 in terms of tolerability. In summary, none of the clinical studies who compared the efficacy of generics and innovators in Humans is sufficient to raise significant concern on
study suggested that generics may be sub-optimal.
cr ipt
the efficacy of generics, as only one isolated case report, and one quasi-experimental
Of note, the four studies that are often cited to document the sub-optimal efficacy
us
animal model that has not been validated for the evaluation of the efficacy of antibacterial agents, i.e. the neutropenic mouse thigh infection model. Only one study evaluated the
an
efficacy of generic products in an animal model with long track records for the evaluation of the bactericidal effect of antibacterial agents, i.e. the rabbit endocarditis model [20].
M
The latter study evaluated 6 vancomycin generic products approved for use in America and Europe, and did not confirm the results observed in the neutropenic mouse thigh
pt ed
infection model [5]. Discrepancies between animal models are not rare, and may be responsible for erroneous assumptions, as was the case with initial studies on the role of Panton-Valentine leukocidin in the pathogenesis of community-associated meticillinresistant Staphylococcus aureus (MRSA) [21]. Hence, further studies with the
ce
appropriate animal model(s) would be required to judge the comparative in vivo efficacy of generic products and the innovator. Given that most in vitro studies (11/14, 78.6%) found no significant difference between the generic products and the innovator in terms
Ac
of MICs, and MBCs, we would expect similar in vivo efficacy in an animal model with
impaired immunity, such as the neutropenic mouse. Some authors suggested that discrepancies between in vitro and in vivo efficacy could be related to excess impurities, a significant issue with antibacterial agents such as vancomycin [5]. However, this
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
of generic products as compared to the innovator [5, 6, 8, 16], were all performed in an
10 hypothesis has not been confirmed by two recent studies where the quality parameters of all parenteral vancomycin products tested surpassed the US Pharmacopeia acceptance
cr ipt
criteria, including for generic products that were sub-optimal in the neutropenic mouse model [22, 23]. Lastly, the finding by Rodriguez et al. that serial exposure to generic
vancomycin enriches resistant subpopulations, while exposure to the innovator reduces
resistant subpopulations, remains unexplained [6]. It must be outlined that this intriguing
us
of experiments in the model (i.e., isolates recovered from an experiment were
an
reinoculated to new groups of animals, and this process was repeated 12 times). Given the complexity of this study design, the results may not be relevant to the clinical use of
M
vancomycin in humans. In addition, these findings have not been confirmed in the rabbit model of MRSA endocarditis [24].
pt ed
The current controversy on the equivalence of generic medicinal products is of paramount importance, as their volume surpasses that of branded medicinal products, and is continuously increasing, accounting for two-thirds of the worldwide consumption of antimicrobial agents in 2010 (source, IMS Health [2]). Promotion of the use of generic
ce
products has the objective to decrease healthcare costs without compromising quality of care and patient safety. Hence, to be approved for use in humans, antibacterial generic products must comply with standards of quality, efficacy, and reliability, as dictated by
Ac
regulatory authorities and international organizations (e.g., FDA, EMA, and WHO). Although in some countries, the quality of available medicinal products may be inadequate in terms of content of active ingredient, this issue also applies to branded drugs, as most of these substandard drugs are counterfeit products [25]. These sobering
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
phenomenon was observed in the neutropenic mouse thigh infection model after 12 sets
11 facts are out of the scope of this systematic review on antibacterial generic products that received regulatory approval. Among other concerns raised by the expanded access to
cr ipt
generic products, Jensen et al. found a relationship between community consumption and the number of trade names of oral ciprofloxacin. In their study, the introduction of generic ciprofloxacin in Denmark was followed by a sharp increase in the total
consumption of ciprofloxacin (from 0.13 to 0.33 defined daily doses/1,000 inhabitants-
us
and others are reminders that drug policies supporting the use of antibacterial generic
an
products must ensure that the reduced price of antibacterial treatments does not translate into increased use [27, 28]. Otherwise, excess use of antibiotics and its ecological impact
of generics.
M
on bacterial resistance would invariably reduce the economical benefit brought by the use
pt ed
This systematic review on the efficacy and quality of antibacterial generic products approved for use in humans has limitations. Firstly, as in most literature reviews, our findings are sensitive to publication bias, implying that the studies analyzed may not be representative of all the studies performed on this topic. Indeed, ‘positive’
ce
studies (i.e., studies that found a significant difference between generic products and the innovator), are probably more likely to be submitted for publication and more likely to be accepted by editors. In this regard, our finding that 37.8% studies reported significant
Ac
differences between generic products and the innovator, is probably an overestimate. Second, the economical consequences of drug policies for increased use of generic products may also carry a risk of bias, in two opposite directions. While pharmaceutical companies who own an innovator product would probably encourage the publication of
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
days), while the frequency of ciprofloxacin resistance increased by 200% [26]. This study
12 studies suggesting that generic products are sub-optimal, regulatory authorities and international organizations would try their best to demonstrate that this is not the case.
cr ipt
Lastly, most of these studies were not adequately powered to demonstrate non-inferiority. Hence, the absence of any significant difference in the parameters evaluated cannot be
interpreted as evidence that the generics are equivalent to the innovators. Despite these limitations, our systematic review provides a global picture of all studies published to
us
Indeed, studies comparing the effectiveness of generics and innovators in Humans could
an
obviously not be pooled with studies comparing taste and acceptability of oral solutions in children, or with studies comparing potencies in vitro and in animal models. However,
M
we found no convincing data that antibacterial generic products approved by regulatory authorities would be sub-optimal, as compared to the innovator. This suggests that additional evidence would be needed before considering a revision of the marketing
pt ed
authorization process for injectable and oral antibacterial generic products.
ce
Conflicts of interest
PT has received grants from Astellas, Astra-Zeneca, Aventis, Bristol-Myers Squibb, Galderma, Gilead Sciences, Janssen-Cilag, MSD, Novartis, Pfizer, and ViiV-Healthcare
Ac
for consultancies, workshops or travel to meetings and accommodation. ACC has received grants from Janssen–Cilag, Novartis, AstraZeneca, Aventis and Heraeus for consultancies, workshops and travel to meetings and accommodation. CR has received grants from Gilead Sciences, Tibotec, Bristol-Myers-Squibb, MSD, Sanofi, AbbVie,
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
date. Their limited number and their heterogeneity did not allow for a meta-analysis.
13 Janssen–Cilag, Thermo Scientific Biomarkers, ViiV-Healthcare, and GSK-vaccin for consultancies, workshops or travel to meeting and accommodation. RG has received
cr ipt
grants from BioMérieux, MSD, Bayer, Janssen-Cilag, Novartis, Sanofi, Astra-Zeneca and Astellas for consultancies, workshops or travel to meeting and accommodation.
Acknowledgments
Ac
ce
us
pt ed
M
an
reading of the manuscript.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
We are indebted to Dominique Monnet for his insightful comments, and for critical
Table 1. Beta-lactam generic products: description of included studies Reference
Evaluated generic
Study design
Main findings
products
rip t
14
Comments
Potential conflict of interest
9 oxacillin generic
- in vitro: determination of API,
- 4/9 generic products had lower
the neutropenic
Astra-Zeneca,
al. [8], 2010
products approved for
MIC, MBC
potency in vitro (API measured by
mouse thigh infection
Wyeth, Pfizer,
intravenous use,
- in vivo neutropenic mouse
microbiological assay)
model is not the gold
Allergan, Roche,
manufactured in
thigh infection model
- 9/9 generic products were
standard for the
GlaxoSmithKline,
Colombia
- comparison with the innovator
equivalent to the innovator in
evaluation of the
Bristol-Myers
(Bristol-Myers Squibb)
terms of MIC, MBC
potency of
Squibb, Merck
- 9/9 generic products had lower
antibacterial agents
Sharp and Dohme
none declared
M an u
sc
Rodriguez et
potency in vivo (Emax)
2 amoxicillin generic
- single-dose, randomized, three-
as compared with the innovator:
bioequivalence not
al. [29], 2009
products approved for
treatment, crossover, single-blind
- 90% CIs of AUC ratio were
demonstrated for
oral use (tablet
bioequivalence study in healthy
[0.8238-1.0502]Mérieu for generic
generic product A
formulations),
adult volunteers (n=24)
product A, and [0.8116-1.1007]
(inferior margin
available on the Italian
- comparison with the innovator
for generic product B
0.7921, i.e. slightly
- 90% CIs of Cmax ratio were
below 0.80)
market
ep t
ed
Del Tacca et
1 ampicillin/probenecid
- pharmacological and clinical
2010
generic product
study in healthy volunteers
developed in China
- comparison of the generic
Ac c
Wu et al. [30],
[0.7921-1.0134] for generic product A, and [0.8246-1.1199] for generic product B no significant differences
bioequivalence demonstrated
product and the innovator in
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
15
rip t
terms of PK properties, bioavailability, bioequivalence, and adverse events 26 samples of
in vitro study (incremental MIC
compared to the innovator, all but
data from this study
[10], 2008
piperacillin-tazobactam
antimicrobial assay)
one lot of generic product
were included in the
generic products from
demonstrated significantly
study by Moet et
Philippines (n=10 lots),
decreased activity, at -5 to -35%
al[9] (same team)
M an u
India (n=5), Greece
sc
Jones et al.
(average, -16%)
(n=3), China (n=2), Spain (n=2), Taiwan (n=2), Portugal (n=1) and Jordan (n=1) Moet et al.
46 lots of piperacillin-
in vitro study (incremental MIC
i) compared to the innovator, the
overall, the lots of the
[9], 2009
tazobactam generic
antimicrobial assay)
range of activity of the generic
innovator (Zosyn©)
products was -42% to +10%
averaged 9% to 10%
Silva et al.
multiple samples of
[31], 2010
piperacillin-
in vitro susceptibility tests
Ac c
tazobactam, and
ep t
countries)
ed
products (29 manufacturers, 17
Wyeth
(average, -16%)
greater activity per
ii) the range of activity between
vial, as compared
lots of the innovator was -19% to
with the lots of
+7% (average, -6%)
generic products
compared to the innovator, no significant differences with respect to potency, MICs, critical
meropenem generic
concentrations, and mutants
products purchased in
selection
different pharmacies in
Wyeth
Vitalis
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Colombia
rip t
16
Tschudin-
one piperacillin-
Sutter et al.
tazobactam generic
and median MIC between the
[32], 2011
product (Sandoz,
generic product and the innovator
Switzerland)
for all tested strains
no significant differences in mean
sc
in vitro susceptibility tests
34 ceftriaxone generic
pharmaceutical qualities of the
quality standards specified in the
[11], 2003
products
generic products compared with
European and US Pharmacopoeias
the innovator
Roche
M an u
Lambert et al
none declared
were violated for 18 generic products, including those for sterility (4 generic products) and impurities (5 generic products)
Schito et al.
34 ceftriaxone generic
mathematical modelling of
the innovator exceeded the
[12], 2005
products
PK/PD parameters based on fluid
required PK/PD parameters, while
concentrations of non-protein-
most generic products did not
ed
bound ceftriaxone and MonteCarlo simulations (pleural fluid, and plasma) 1 intravenous
al. [13], 2008
cefuroxime generic
quasi-experimental monocentric
product purchased in
national drug
clinical study, comparing the
infections higher in patients who
organization
generic product with the
received the generic product
identified gaps in the
innovator in the prophylaxis of
compared to the innovator (12.8%
chain of the
post-cardiac surgery infection
vs. 2.5%, p 48 hours with either
innovator in terms of clinical
the generic product or the
outcome (favourable, versus death)
Ac c
Angkasekwin
397 patients enrolled
ed
Tansuphasaw
none declared
600 patients enrolled
innovator
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
18
imipenem/cilastatin
retrospective monocentric cohort
no significant differences between
al. [36], 2010
generic product
study of hospitalized patients
the generic products and the
available in Thailand
treated for > 48 hours with either
innovator in terms of cure rates,
the generic product or the
superinfection rates, and mortality
innovator
(due to infection, and overall), but
600 patients enrolled
sc
rip t
Piyasirisilp et
a trend favouring the innovator for
M an u
all comparisons
API=active pharmaceutical ingredients. MIC=minimal inhibitory concentration. MBC=minimal bactericidal concentration. Cmax=maximal plasma concentration. Emax=bactericidal maximal effect. AUC=area under the curve. CI=confidence interval.
Ac c
ep t
ed
PK/PD=pharmacokinetic/pharmacodynamic.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Table 2. Glycopeptide generic products: description of included studies Reference
Evaluated
Study design
Main findings
generic products
rip t
19
5 vancomycin generic
- measurement of the active
- content per vial within the stipulated
al. [37], 2008
products approved in
component of vancomycin per
range for all drugs (i.e., −10; +15%)
Japan, imported from
vial
Slovenia, France,
- in vitro susceptibility tests
Hungary and Taipei
(MICs)
sc
Fujimura et
Comments
Potential conflict of interest none declared
M an u
- no significant differences with the innovator
- potency equivalent per vial Fujimura et
7 teicoplanin generic
al. [38], 2011
products purchased in
in vitro susceptibility tests
for 147 clinical isolates of MRSA,
none declared
MIC90 were similar (4 mg/L) for the
Japan
innovator and 5 generic products, while it was 8 mg/L for the 2
ed
remaining generic products
3 vancomycin generic
- in vitro: determination of
- in vitro: no difference between the
the neutropenic
al. [6], 2012
products purchased in
MIC, MBC, and population
generic products and the innovator
mouse thigh infection
Colombia
analysis profile
- in vivo: serial exposure to the
model is not the gold
- in vivo: serial passages of
generic product enriches resistant
standard for the
one clinical MRSA strain in
subpopulations, while exposure to the
evaluation of the
the neutropenic mouse thigh
innovator reduces resistant
efficacy of
infection model (12 cycles)
subpopulations
antibacterial agents
- in vitro studies: generic products
the neutropenic
Ac c
ep t
Rodriguez et
none declared
- comparison with the innovator (Eli Lilly)
3 vancomycin generic
- in vitro: determination of
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Vesga et al.
Astra-Zeneca,
20
products imported from
API, MIC, MBC, and time-kill
undistinguishable from the innovator
mouse thigh infection
Wyeth, Pfizer,
the US, France, and
curves
- in vivo studies: all generic products
model is not the gold
Allergan, Roche,
Argentina
- in vivo: neutropenic mouse
significantly inferior to the innovator
standard for the
GlaxoSmithKline,
thigh infection model
in terms of bactericidal effect (Emax)
evaluation of the
Bristol-Myers
efficacy of
Squibb, Merck
antibacterial agents
Sharp and Dohme
rip t
[5], 2010
sc
- comparison with the innovator (Eli Lilly) 1 vancomycin generic
al. [15], 2009
product purchased in
case report
- in a liver transplant recipient,
M an u
Rodriguez et
none declared
persistent bacteremia after 10 days of
the US (approved by
intravenous vancomycin generic
the FDA)
product
- blood cultures became sterile 24 hours after switch to the innovator
Diaz et al.
20 samples of
in vitro: determination of
no significant differences observed
[39], 2011
vancomycin generic
MIC, MBC, critical
between the generic products and the
products purchased
concentrations, and the
from the pharmacies in
production of spontaneous
different hospitals in
mutants
Conte et al.
the first vancomycin
[40], 1987
generic product
US
ed
innovator
in vitro: determination of
no significant differences observed
MIC, and MBC
between the generic product and the
Ac c
commercialized in the
ep t
Colombia
Vitalis
Lymphomed
innovator
Nambiar et al.
6 vancomycin generic
samples tested for purities,
the quality parameters of all generic
Study performed
[22], 2012
products approved for
content, and potency
products tested were above the US
by the US Food
Pharmacopeia acceptance criteria
and Drug
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
parenteral use in the
21
rip t
US
Administration (FDA)
samples of vancomycin
measurement of the active
the quality parameters of all generic
Study performed
al. [23], 2012
generic products
component of vancomcyin,
products tested were above the US
by the US Food
approved for parenteral
and the impurities, by two
Pharmacopeia acceptance criteria
and Drug
use in the US
independent laboratories
sc
Hadwiger et
M an u
(HPLC and ultra-HPLC)
Administration (FDA)
6 vancomycin generic
- in vitro: determination of
no significant differences observed
- the rabbit
Astra-Zeneca,
[20], 2012
products imported from
time-kill curves
between the generic products in terms
endocarditis model is
Astellas,
the US, France, Spain,
- in vivo: MRSA rabbit
of in vitro bactericidal activity (time-
the gold standard
Galderma, Pfizer,
and Switzerland
endocarditis model
kill curves), and in vivo efficacy
animal model for the
Roche, Novartis,
(rabbit endocarditis model)
evaluation of the
Sanofi-Aventis,
efficacy of
Merck Sharp and
antibacterial agents
Dohme,
- no comparison with
GlaxoSmithKline,
the innovator
Bristol-Myers
ep t
ed
Tattevin et al.
Squibb
API=active pharmaceutical ingredients. MIC=minimal inhibitory concentration. MBC=minimal bactericidal concentration.
Ac c
Emax=bactericidal maximal effect. MRSA=Meticillin-Resistant Staphylococcus aureus. HPLC= high-pressure liquid chromatography.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Table 3. Other antibacterial generic products: description of included studies Reference
Evaluated
Study design
Main findings
sc
generic products
rip t
22
Comments
Potential conflicts of interest
20 gentamicin generic
- in vitro study: MIC, MBC
- in vitro study: one generic product
the neutropenic
Astra-Zeneca,
[16], 2010
products approved for
- in vivo study: neutropenic
significantly inferior (MIC and MBC,
mouse thigh infection
Wyeth, Pfizer,
intravenous use,
mouse thigh infection model
respectively, 45.3 and 64 vs. 0.71 and
model is not the gold
Allergan, Roche,
0.79 mg/L, p
1
Efficacy and quality of antibacterial generic products
cr ipt
approved for human use: a systematic review
Pierre Tattevina,b,*, Anne-Claude Crémieuxc,d, Christian Rabaude, Rémy Gauzitf a
Pontchaillou Univ. Hosp., Rennes, France
us
INSERM U835, Université Rennes 1, IFR140, F-35033, Rennes, France
c
EA 3647, Versailles Saint-Quentin Univ., Versailles, France Raymond Poincaré Univ. Hosp., Garches, France
e
Brabois Univ Hosp, Nancy, France
f
an
d
Hôtel- Dieu Univ Hosp, Assistance Publique-Hôpitaux de Paris (AP-HP), Paris, France Corresponding author. Address: Service des Maladies Infectieuses et de Réanimation
M
*
Médicale, CHU Pontchaillou, 2 rue Henri Le Guilloux, 35033 Rennes Cedex, France. Tel
pt ed
+33 299289564 ; fax +33 299282452, E-mail address: [email protected] Alternate corresponding author. Réanimation, Hôpital Hôtel-Dieu, Assistance PubliqueHôpitaux de Paris (AP-HP), 1 place du Parvis Notre Dame, 75004 Paris, France. E-mail
ce
[email protected]
KEY POINTS
Ac
A systematic literature review on antibacterial generics approved for use in humans analyzed 37 original articles with heterogeneous design, inconsistent findings, and discrepancies. Additional evidence are needed before considering a revision of the marketing authorization process for antibacterial generic products.
© The Author 2013. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: [email protected]
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
b
2 ABSTRACT
antibacterial generic products approved for use in humans.
cr ipt
Background Concerns have recently emerged about the efficacy and the quality of
Methods We searched Medline and Embase for original research articles on antibacterial generic products published in English or French before July 2013.
Results We selected 37 original research articles: 15 on beta-lactams, 10 on
us
published during 2008-2012. Study designs included analytical chemistry (n=9), in vitro
an
susceptibility studies (n=14), animal experiments (n=6, including 5 using the neutropenic mouse thigh infection model), and clinical studies in humans (n=15). Of the 37 studies,
M
14 (37.8%) suggested that some generic products may be inferior to the innovator in terms of purity (n=2), in vitro activity (n=3), in vivo efficacy in experimental models (n=4), clinical efficacy (n=2), taste (n=2), or compliance and acceptability in children
pt ed
(n=1). The majority of in vitro studies (78.6%) found no significant difference between generic products and the innovator. Most (5/6) in vivo studies suggesting a difference between generic products and the innovator were performed in an animal model that is not validated for the evaluation of the efficacy of antibacterial agents. The level of
ce
evidence was constantly low in clinical studies. Conclusions Published data on antibacterial generic products are limited and
Ac
heterogeneous, thus precluding any attempt to generalize the study results. This systematic review suggests that additional evidence would be needed before considering a revision of the marketing authorization process for antibacterial generic products.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
glycopeptides, and 12 on other antibacterial agents. The majority of articles (73.0%) were
3 INTRODUCTION Generic medicinal products are copies of patented drugs and can be marketed at low cost
cr ipt
following patent expiration of the brand leader product. As all pharmaceutical products, generic products must comply with standards of quality, efficacy, and reliability. The regulatory authorities of several countries, including the US Food and Drug
Administration (FDA) and the European Medicines Agency (EMA), and the World
us
under which generic medicinal products can be recognized as therapeutically equivalent
an
to their brand name counterpart (reviewed in [1]). Bioequivalence principles have been defined (i.e., 20% acceptance range (80–125%) for the 90% confidence interval of the ratio between test and reference least square means after log-transformation of the
M
pharmacokinetic parameters of interest, Cmax and area under curve (AUC)). The objectives of drug policies supporting the use of generic medicinal products are
pt ed
essentially economic: i) to decrease the cost of medicines as part of healthcare spending, in particular in developed countries (e.g., in France, it is estimated that €1.3 billion were saved by introducing such policy in 2008 [2]), and ii) to facilitate access to care in developing countries (e.g., expanded access to effective antiretroviral combinations for
ce
HIV-infected patients). However, these economic considerations should not occur at the price of lower quality of patient care.
Ac
The therapeutic equivalence of generic medicinal products approved for use in
humans has been challenged in various therapeutic areas, including neurology [3], endocrinology, and cardiovascular diseases [4]. The question was also raised for antibacterial agents, especially following publication of a study performed with vancomycin generic products suggesting that, despite similar pharmacokinetics
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Health Organization (WHO) have issued guidelines presenting the terms and conditions
4 parameters and in vitro antibacterial activities, some vancomycin generic products were less bactericidal than the innovator in vivo in a neutropenic mouse thigh infection model
cr ipt
[5], and could induce more resistant sub-populations [6]. Following publication of these results, a debate started worldwide, involving the scientific communities, drugs
regulatory agencies, and the general public [7]. Questions were raised about the current assumptions from the WHO, FDA, and EMA, that two products of parenteral use are
us
is of concern, as i) it would imply that patients may currently receive sub-optimal
an
antibacterial therapy depending on the product that they receive; ii) deep changes would be required in the approval process of antibacterial generic products, including the need
M
for more studies on efficacy, which would, in turn, translate into increased costs for approval. This systematic review of the literature was designed to objectively analyze
pt ed
published data regarding the efficacy and quality of antibacterial generic products.
METHODS
Search strategy and selection criteria
We searched Medline and Embase for articles in English or French published anytime
ce
before July 2013, using the following keywords in various combinations “generics”, “generic products”, “antibacterial”, “antibacterial agent”, “antibiotic”, “betalactam
Ac
agent”, “penicillin”, “cephalosporin”, “penem”, “fluoroquinolone”, “aminoglycoside”, “macrolide”, “cycline”, “glycopeptide”, and “vancomycin”.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
considered therapeutically equivalent if they are pharmaceutically equivalent. This issue
5 Data extraction Two independent researchers (PT and RG) reviewed all abstracts to identify articles that
cr ipt
required full-text review, with a final decision reached through consensus. All articles
were discussed with a third reviewer (ACC). For each reviewed article, we extracted data on the study setting, objectives, methods, and results, including details on the evaluated
generic products. We systematically searched for additional articles in the reference lists
RESULTS
us
an
articles available precluded any meta-analysis.
M
We identified 37 studies that met inclusion criteria: 15 focused on beta-lactam agents (Table 1), 10 on glycopeptides (Table 2), and 12 on other antibacterial agents (Table 3). The majority of articles (73.0%) were published during the last five years (2008-2012).
pt ed
Study designs included analytical chemistry (measurement of active pharmaceutical ingredients and/or impurities, n=9), in vitro susceptibility studies (e.g., determination of minimal inhibitory concentrations (MICs), minimal bactericidal concentration (MBCs), time-kill curves, population analysis, n=14), in vivo animal experiments (n=6, including
ce
five studies using the neutropenic mouse thigh infection model), and clinical studies in humans (i.e., pharmacokinetic studies, retrospective cohorts, studies on compliance
Ac
and/or taste, quasi-experimental study, case report, n=15). Studies originated from Asia (n=9), South America (n=8), Europe (n=7), North America (n=6) or covered more than one continent (n=7). Of the 37 studies, 14 (37.8%) suggested that some generic products may be inferior to the innovator in terms of purity (n=2), in vitro activity (n=3), in vivo
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
of all articles reviewed. The heterogeneity of study designs and the limited number of
6 efficacy in experimental models (n=4), clinical efficacy (n=2), taste (n=2), or compliance and acceptability in children (n=1).
cr ipt
Of the 15 studies on beta-lactam generic products, seven suggested that some
approved generic products may be inferior to the innovator. Rodriguez et al. found that, despite similar MICs and MBCs, the nine generic products of oxacillin that they
evaluated in the neutropenic mouse thigh Staphylococcus aureus infection model had
us
46 lots of piperacillin-tazobactam generic products manufactured in 17 countries and
an
found that their in vitro activity, evaluated by incremental MIC antimicrobial assay, was on average 10% lower than that of the innovator [9, 10]. Lambert et al. studied the
M
pharmaceutical qualities of 34 generic products of ceftriaxone approved for use, and found that quality standards were violated on 18 occasions [11]. Subsequently, in a mathematical model based on Monte Carlo simulations, Schito et al. suggested that most
pt ed
ceftriaxone generic products may not reach the required PK/PD parameters [12]. In a quasi-experimental study, Mastoraki et al. found that the incidence of post-cardiac surgery infections was significantly increased in patients who received prophylaxis with a generic product of cefuroxime, as compared to the innovator [13]. Lastly, Cohen et al.
ce
found that the acceptability and the compliance of amoxicillin-clavulanic acid generic products approved for oral use in France were lower than that of the innovator in
Ac
children, based on a questionnaire completed by their parents [14]. Of the 10 studies on glycopeptide generic products, three suggested that some
approved generic products may be inferior to the innovator. Rodriguez et al. published a case report where a liver transplant recipient with persistent bacteremia after 10 days of
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
lower Emax (maximum effect in log10 CFU/g) than the innovator [8]. Jones et al. evaluated
7 intravenous vancomycin generic product had sterile blood cultures 24 hours after a switch to vancomycin innovator [15]. Vesga et al. found that, despite similar MICs and MBCs,
cr ipt
three generic products of vancomycin imported from France, Argentina, and the US, had lower Emax than the innovator in the neutropenic mouse thigh S. aureus infection model [5]. The same team subsequently found, using the same model, that serial exposure to
generic vancomycin enriches resistant subpopulations, while exposure to the innovator
us
Of the 12 studies published to date on generic products of other antibacterial agent
an
classes, four suggested that some approved generic products may be inferior to the innovator. Zuluaga et al. found that, despite similar MICs and MBCs, 10 out of 20
M
generic products of gentamicin had lower Emax than the innovator in the neutropenic mouse thigh Escherichia coli infection model [16]. Nightingale et al. evaluated 65 generic products of clarithromycin from 18 countries and found that 9% did not contain
pt ed
between 95% and 105% of the clarithromycin content claimed in the label, and 19% exceeded the 3% limit for total impurities [17]. Two double-blind studies found that the taste of trimethoprim-sulfamethoxazole generic products was rated lower than the
ce
innovator [18, 19].
Regarding study design, almost half of the studies (n=17) in this systematic
review were performed in humans and studied pharmacokinetic parameters (n=7, mostly
Ac
bioequivalence studies), clinical efficacy (n=6), and tolerability or compliance in children (n=4). Nine studies focused on in vitro efficacy, comparing generic products with the innovator. Six studies were performed in animals, using the neutropenic mouse thigh infection model (n=5) or the rabbit endocarditis model (n=1). Five studies investigated
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
reduces resistant subpopulations [6].
8 the pharmaceutical qualities (purity, content, and potency) of antibacterial generic products. This heterogeneity in the outcomes measured further precluded any meta-
cr ipt
analysis, as pooling the studies focusing on major issues (effectiveness in Humans), relatively major issues (potency in vitro or in animal models), and relatively less
important - though still significant – issues (taste), would make little sense, and be of
an
DISCUSSION
This systematic review of studies that evaluated antibacterial generic products reveals
M
that the data available to date in the literature are limited and heterogeneous. This precludes any attempt to generalize their findings. Indeed, even for the classes of antibacterial agents who received the most attention (i.e., beta-lactams and
pt ed
glycopeptides), published studies have not consistently demonstrated that some generic products approved for use in humans were inferior to the innovator. The level of evidence from the six clinical studies which evaluated the efficacy of antibacterial generic products was low. One case report and one quasi-experimental study suggested that generic
ce
products of, respectively, vancomycin and cefuroxime, were less potent than the innovator, while three retrospective cohort studies that enrolled a total of 1,597 patients
Ac
treated with meropenem or imipenem/cilastatin found no significant difference between the generic products and the innovator in terms of clinical outcome. One randomized, open-label, clinical trial, found that a generic product of clarithromycin was not significantly different to the innovator in terms of clinical or bacteriological efficacy, and
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
us
limited clinical relevance.
9 in terms of tolerability. In summary, none of the clinical studies who compared the efficacy of generics and innovators in Humans is sufficient to raise significant concern on
study suggested that generics may be sub-optimal.
cr ipt
the efficacy of generics, as only one isolated case report, and one quasi-experimental
Of note, the four studies that are often cited to document the sub-optimal efficacy
us
animal model that has not been validated for the evaluation of the efficacy of antibacterial agents, i.e. the neutropenic mouse thigh infection model. Only one study evaluated the
an
efficacy of generic products in an animal model with long track records for the evaluation of the bactericidal effect of antibacterial agents, i.e. the rabbit endocarditis model [20].
M
The latter study evaluated 6 vancomycin generic products approved for use in America and Europe, and did not confirm the results observed in the neutropenic mouse thigh
pt ed
infection model [5]. Discrepancies between animal models are not rare, and may be responsible for erroneous assumptions, as was the case with initial studies on the role of Panton-Valentine leukocidin in the pathogenesis of community-associated meticillinresistant Staphylococcus aureus (MRSA) [21]. Hence, further studies with the
ce
appropriate animal model(s) would be required to judge the comparative in vivo efficacy of generic products and the innovator. Given that most in vitro studies (11/14, 78.6%) found no significant difference between the generic products and the innovator in terms
Ac
of MICs, and MBCs, we would expect similar in vivo efficacy in an animal model with
impaired immunity, such as the neutropenic mouse. Some authors suggested that discrepancies between in vitro and in vivo efficacy could be related to excess impurities, a significant issue with antibacterial agents such as vancomycin [5]. However, this
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
of generic products as compared to the innovator [5, 6, 8, 16], were all performed in an
10 hypothesis has not been confirmed by two recent studies where the quality parameters of all parenteral vancomycin products tested surpassed the US Pharmacopeia acceptance
cr ipt
criteria, including for generic products that were sub-optimal in the neutropenic mouse model [22, 23]. Lastly, the finding by Rodriguez et al. that serial exposure to generic
vancomycin enriches resistant subpopulations, while exposure to the innovator reduces
resistant subpopulations, remains unexplained [6]. It must be outlined that this intriguing
us
of experiments in the model (i.e., isolates recovered from an experiment were
an
reinoculated to new groups of animals, and this process was repeated 12 times). Given the complexity of this study design, the results may not be relevant to the clinical use of
M
vancomycin in humans. In addition, these findings have not been confirmed in the rabbit model of MRSA endocarditis [24].
pt ed
The current controversy on the equivalence of generic medicinal products is of paramount importance, as their volume surpasses that of branded medicinal products, and is continuously increasing, accounting for two-thirds of the worldwide consumption of antimicrobial agents in 2010 (source, IMS Health [2]). Promotion of the use of generic
ce
products has the objective to decrease healthcare costs without compromising quality of care and patient safety. Hence, to be approved for use in humans, antibacterial generic products must comply with standards of quality, efficacy, and reliability, as dictated by
Ac
regulatory authorities and international organizations (e.g., FDA, EMA, and WHO). Although in some countries, the quality of available medicinal products may be inadequate in terms of content of active ingredient, this issue also applies to branded drugs, as most of these substandard drugs are counterfeit products [25]. These sobering
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
phenomenon was observed in the neutropenic mouse thigh infection model after 12 sets
11 facts are out of the scope of this systematic review on antibacterial generic products that received regulatory approval. Among other concerns raised by the expanded access to
cr ipt
generic products, Jensen et al. found a relationship between community consumption and the number of trade names of oral ciprofloxacin. In their study, the introduction of generic ciprofloxacin in Denmark was followed by a sharp increase in the total
consumption of ciprofloxacin (from 0.13 to 0.33 defined daily doses/1,000 inhabitants-
us
and others are reminders that drug policies supporting the use of antibacterial generic
an
products must ensure that the reduced price of antibacterial treatments does not translate into increased use [27, 28]. Otherwise, excess use of antibiotics and its ecological impact
of generics.
M
on bacterial resistance would invariably reduce the economical benefit brought by the use
pt ed
This systematic review on the efficacy and quality of antibacterial generic products approved for use in humans has limitations. Firstly, as in most literature reviews, our findings are sensitive to publication bias, implying that the studies analyzed may not be representative of all the studies performed on this topic. Indeed, ‘positive’
ce
studies (i.e., studies that found a significant difference between generic products and the innovator), are probably more likely to be submitted for publication and more likely to be accepted by editors. In this regard, our finding that 37.8% studies reported significant
Ac
differences between generic products and the innovator, is probably an overestimate. Second, the economical consequences of drug policies for increased use of generic products may also carry a risk of bias, in two opposite directions. While pharmaceutical companies who own an innovator product would probably encourage the publication of
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
days), while the frequency of ciprofloxacin resistance increased by 200% [26]. This study
12 studies suggesting that generic products are sub-optimal, regulatory authorities and international organizations would try their best to demonstrate that this is not the case.
cr ipt
Lastly, most of these studies were not adequately powered to demonstrate non-inferiority. Hence, the absence of any significant difference in the parameters evaluated cannot be
interpreted as evidence that the generics are equivalent to the innovators. Despite these limitations, our systematic review provides a global picture of all studies published to
us
Indeed, studies comparing the effectiveness of generics and innovators in Humans could
an
obviously not be pooled with studies comparing taste and acceptability of oral solutions in children, or with studies comparing potencies in vitro and in animal models. However,
M
we found no convincing data that antibacterial generic products approved by regulatory authorities would be sub-optimal, as compared to the innovator. This suggests that additional evidence would be needed before considering a revision of the marketing
pt ed
authorization process for injectable and oral antibacterial generic products.
ce
Conflicts of interest
PT has received grants from Astellas, Astra-Zeneca, Aventis, Bristol-Myers Squibb, Galderma, Gilead Sciences, Janssen-Cilag, MSD, Novartis, Pfizer, and ViiV-Healthcare
Ac
for consultancies, workshops or travel to meetings and accommodation. ACC has received grants from Janssen–Cilag, Novartis, AstraZeneca, Aventis and Heraeus for consultancies, workshops and travel to meetings and accommodation. CR has received grants from Gilead Sciences, Tibotec, Bristol-Myers-Squibb, MSD, Sanofi, AbbVie,
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
date. Their limited number and their heterogeneity did not allow for a meta-analysis.
13 Janssen–Cilag, Thermo Scientific Biomarkers, ViiV-Healthcare, and GSK-vaccin for consultancies, workshops or travel to meeting and accommodation. RG has received
cr ipt
grants from BioMérieux, MSD, Bayer, Janssen-Cilag, Novartis, Sanofi, Astra-Zeneca and Astellas for consultancies, workshops or travel to meeting and accommodation.
Acknowledgments
Ac
ce
us
pt ed
M
an
reading of the manuscript.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
We are indebted to Dominique Monnet for his insightful comments, and for critical
Table 1. Beta-lactam generic products: description of included studies Reference
Evaluated generic
Study design
Main findings
products
rip t
14
Comments
Potential conflict of interest
9 oxacillin generic
- in vitro: determination of API,
- 4/9 generic products had lower
the neutropenic
Astra-Zeneca,
al. [8], 2010
products approved for
MIC, MBC
potency in vitro (API measured by
mouse thigh infection
Wyeth, Pfizer,
intravenous use,
- in vivo neutropenic mouse
microbiological assay)
model is not the gold
Allergan, Roche,
manufactured in
thigh infection model
- 9/9 generic products were
standard for the
GlaxoSmithKline,
Colombia
- comparison with the innovator
equivalent to the innovator in
evaluation of the
Bristol-Myers
(Bristol-Myers Squibb)
terms of MIC, MBC
potency of
Squibb, Merck
- 9/9 generic products had lower
antibacterial agents
Sharp and Dohme
none declared
M an u
sc
Rodriguez et
potency in vivo (Emax)
2 amoxicillin generic
- single-dose, randomized, three-
as compared with the innovator:
bioequivalence not
al. [29], 2009
products approved for
treatment, crossover, single-blind
- 90% CIs of AUC ratio were
demonstrated for
oral use (tablet
bioequivalence study in healthy
[0.8238-1.0502]Mérieu for generic
generic product A
formulations),
adult volunteers (n=24)
product A, and [0.8116-1.1007]
(inferior margin
available on the Italian
- comparison with the innovator
for generic product B
0.7921, i.e. slightly
- 90% CIs of Cmax ratio were
below 0.80)
market
ep t
ed
Del Tacca et
1 ampicillin/probenecid
- pharmacological and clinical
2010
generic product
study in healthy volunteers
developed in China
- comparison of the generic
Ac c
Wu et al. [30],
[0.7921-1.0134] for generic product A, and [0.8246-1.1199] for generic product B no significant differences
bioequivalence demonstrated
product and the innovator in
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
15
rip t
terms of PK properties, bioavailability, bioequivalence, and adverse events 26 samples of
in vitro study (incremental MIC
compared to the innovator, all but
data from this study
[10], 2008
piperacillin-tazobactam
antimicrobial assay)
one lot of generic product
were included in the
generic products from
demonstrated significantly
study by Moet et
Philippines (n=10 lots),
decreased activity, at -5 to -35%
al[9] (same team)
M an u
India (n=5), Greece
sc
Jones et al.
(average, -16%)
(n=3), China (n=2), Spain (n=2), Taiwan (n=2), Portugal (n=1) and Jordan (n=1) Moet et al.
46 lots of piperacillin-
in vitro study (incremental MIC
i) compared to the innovator, the
overall, the lots of the
[9], 2009
tazobactam generic
antimicrobial assay)
range of activity of the generic
innovator (Zosyn©)
products was -42% to +10%
averaged 9% to 10%
Silva et al.
multiple samples of
[31], 2010
piperacillin-
in vitro susceptibility tests
Ac c
tazobactam, and
ep t
countries)
ed
products (29 manufacturers, 17
Wyeth
(average, -16%)
greater activity per
ii) the range of activity between
vial, as compared
lots of the innovator was -19% to
with the lots of
+7% (average, -6%)
generic products
compared to the innovator, no significant differences with respect to potency, MICs, critical
meropenem generic
concentrations, and mutants
products purchased in
selection
different pharmacies in
Wyeth
Vitalis
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Colombia
rip t
16
Tschudin-
one piperacillin-
Sutter et al.
tazobactam generic
and median MIC between the
[32], 2011
product (Sandoz,
generic product and the innovator
Switzerland)
for all tested strains
no significant differences in mean
sc
in vitro susceptibility tests
34 ceftriaxone generic
pharmaceutical qualities of the
quality standards specified in the
[11], 2003
products
generic products compared with
European and US Pharmacopoeias
the innovator
Roche
M an u
Lambert et al
none declared
were violated for 18 generic products, including those for sterility (4 generic products) and impurities (5 generic products)
Schito et al.
34 ceftriaxone generic
mathematical modelling of
the innovator exceeded the
[12], 2005
products
PK/PD parameters based on fluid
required PK/PD parameters, while
concentrations of non-protein-
most generic products did not
ed
bound ceftriaxone and MonteCarlo simulations (pleural fluid, and plasma) 1 intravenous
al. [13], 2008
cefuroxime generic
quasi-experimental monocentric
product purchased in
national drug
clinical study, comparing the
infections higher in patients who
organization
generic product with the
received the generic product
identified gaps in the
innovator in the prophylaxis of
compared to the innovator (12.8%
chain of the
post-cardiac surgery infection
vs. 2.5%, p 48 hours with either
innovator in terms of clinical
the generic product or the
outcome (favourable, versus death)
Ac c
Angkasekwin
397 patients enrolled
ed
Tansuphasaw
none declared
600 patients enrolled
innovator
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
18
imipenem/cilastatin
retrospective monocentric cohort
no significant differences between
al. [36], 2010
generic product
study of hospitalized patients
the generic products and the
available in Thailand
treated for > 48 hours with either
innovator in terms of cure rates,
the generic product or the
superinfection rates, and mortality
innovator
(due to infection, and overall), but
600 patients enrolled
sc
rip t
Piyasirisilp et
a trend favouring the innovator for
M an u
all comparisons
API=active pharmaceutical ingredients. MIC=minimal inhibitory concentration. MBC=minimal bactericidal concentration. Cmax=maximal plasma concentration. Emax=bactericidal maximal effect. AUC=area under the curve. CI=confidence interval.
Ac c
ep t
ed
PK/PD=pharmacokinetic/pharmacodynamic.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Table 2. Glycopeptide generic products: description of included studies Reference
Evaluated
Study design
Main findings
generic products
rip t
19
5 vancomycin generic
- measurement of the active
- content per vial within the stipulated
al. [37], 2008
products approved in
component of vancomycin per
range for all drugs (i.e., −10; +15%)
Japan, imported from
vial
Slovenia, France,
- in vitro susceptibility tests
Hungary and Taipei
(MICs)
sc
Fujimura et
Comments
Potential conflict of interest none declared
M an u
- no significant differences with the innovator
- potency equivalent per vial Fujimura et
7 teicoplanin generic
al. [38], 2011
products purchased in
in vitro susceptibility tests
for 147 clinical isolates of MRSA,
none declared
MIC90 were similar (4 mg/L) for the
Japan
innovator and 5 generic products, while it was 8 mg/L for the 2
ed
remaining generic products
3 vancomycin generic
- in vitro: determination of
- in vitro: no difference between the
the neutropenic
al. [6], 2012
products purchased in
MIC, MBC, and population
generic products and the innovator
mouse thigh infection
Colombia
analysis profile
- in vivo: serial exposure to the
model is not the gold
- in vivo: serial passages of
generic product enriches resistant
standard for the
one clinical MRSA strain in
subpopulations, while exposure to the
evaluation of the
the neutropenic mouse thigh
innovator reduces resistant
efficacy of
infection model (12 cycles)
subpopulations
antibacterial agents
- in vitro studies: generic products
the neutropenic
Ac c
ep t
Rodriguez et
none declared
- comparison with the innovator (Eli Lilly)
3 vancomycin generic
- in vitro: determination of
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Vesga et al.
Astra-Zeneca,
20
products imported from
API, MIC, MBC, and time-kill
undistinguishable from the innovator
mouse thigh infection
Wyeth, Pfizer,
the US, France, and
curves
- in vivo studies: all generic products
model is not the gold
Allergan, Roche,
Argentina
- in vivo: neutropenic mouse
significantly inferior to the innovator
standard for the
GlaxoSmithKline,
thigh infection model
in terms of bactericidal effect (Emax)
evaluation of the
Bristol-Myers
efficacy of
Squibb, Merck
antibacterial agents
Sharp and Dohme
rip t
[5], 2010
sc
- comparison with the innovator (Eli Lilly) 1 vancomycin generic
al. [15], 2009
product purchased in
case report
- in a liver transplant recipient,
M an u
Rodriguez et
none declared
persistent bacteremia after 10 days of
the US (approved by
intravenous vancomycin generic
the FDA)
product
- blood cultures became sterile 24 hours after switch to the innovator
Diaz et al.
20 samples of
in vitro: determination of
no significant differences observed
[39], 2011
vancomycin generic
MIC, MBC, critical
between the generic products and the
products purchased
concentrations, and the
from the pharmacies in
production of spontaneous
different hospitals in
mutants
Conte et al.
the first vancomycin
[40], 1987
generic product
US
ed
innovator
in vitro: determination of
no significant differences observed
MIC, and MBC
between the generic product and the
Ac c
commercialized in the
ep t
Colombia
Vitalis
Lymphomed
innovator
Nambiar et al.
6 vancomycin generic
samples tested for purities,
the quality parameters of all generic
Study performed
[22], 2012
products approved for
content, and potency
products tested were above the US
by the US Food
Pharmacopeia acceptance criteria
and Drug
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
parenteral use in the
21
rip t
US
Administration (FDA)
samples of vancomycin
measurement of the active
the quality parameters of all generic
Study performed
al. [23], 2012
generic products
component of vancomcyin,
products tested were above the US
by the US Food
approved for parenteral
and the impurities, by two
Pharmacopeia acceptance criteria
and Drug
use in the US
independent laboratories
sc
Hadwiger et
M an u
(HPLC and ultra-HPLC)
Administration (FDA)
6 vancomycin generic
- in vitro: determination of
no significant differences observed
- the rabbit
Astra-Zeneca,
[20], 2012
products imported from
time-kill curves
between the generic products in terms
endocarditis model is
Astellas,
the US, France, Spain,
- in vivo: MRSA rabbit
of in vitro bactericidal activity (time-
the gold standard
Galderma, Pfizer,
and Switzerland
endocarditis model
kill curves), and in vivo efficacy
animal model for the
Roche, Novartis,
(rabbit endocarditis model)
evaluation of the
Sanofi-Aventis,
efficacy of
Merck Sharp and
antibacterial agents
Dohme,
- no comparison with
GlaxoSmithKline,
the innovator
Bristol-Myers
ep t
ed
Tattevin et al.
Squibb
API=active pharmaceutical ingredients. MIC=minimal inhibitory concentration. MBC=minimal bactericidal concentration.
Ac c
Emax=bactericidal maximal effect. MRSA=Meticillin-Resistant Staphylococcus aureus. HPLC= high-pressure liquid chromatography.
Downloaded from http://cid.oxfordjournals.org/ at Belgorod State University on November 24, 2013
Table 3. Other antibacterial generic products: description of included studies Reference
Evaluated
Study design
Main findings
sc
generic products
rip t
22
Comments
Potential conflicts of interest
20 gentamicin generic
- in vitro study: MIC, MBC
- in vitro study: one generic product
the neutropenic
Astra-Zeneca,
[16], 2010
products approved for
- in vivo study: neutropenic
significantly inferior (MIC and MBC,
mouse thigh infection
Wyeth, Pfizer,
intravenous use,
mouse thigh infection model
respectively, 45.3 and 64 vs. 0.71 and
model is not the gold
Allergan, Roche,
0.79 mg/L, p
