Antibacterial drug discovery in the 21st century

K. Bush

Johnson & Johnson Pharmaceutical Research & Development, L.L.C., NJ, USA

ABSTRACT
Antibacterial research over the past 50 years has been focused on meeting medical needs caused by
infectious, life-threatening pathogens. In spite of the introduction of a variety of antibacterial agents in
multiple unrelated drug classes, resistance continues to emerge. The pharmaceutical industry must
respond to these clinical challenges by bringing forward a stream of new agents with antibacterial
activity against resistant bacteria. Although the projected growth of the anti-infective area may not be as
large as for some therapeutic areas, development advantages for these agents include their higher
predictability for success, well-defined biomarkers, shorter clinical trials, and shorter duration of
therapy leading to fewer long-term safety concerns. Anti-infectives are still attractive commercially,
representing the third largest therapeutic area in terms of worldwide sales of $45 bn, with growth
predicted at least through 2010, particularly for the hospital-related products. Finally, companies that
conduct anti-infective research demonstrate their social responsibility by developing agents to treat
patients with acute and potentially fatal illnesses.
1 Keywords Antibacterial agents, antibiotics, drug discovery, research, anti-infective, resistance

Clin Microbiol Infect 2004; 10 (Suppl. 4): 10–17

Although many of the former corporate leaders

INTRODUCTION
Antibacterial drug discovery research, accom-
panied by clinical development, has historically
been conducted by large pharmaceutical com-
panies. Although the earliest antibiotics were
first identified in academic laboratories such as
those of Alexander Fleming (penicillin) [1] and
Selman Waksman (streptomycin) [2], pharma-
ceutical companies were responsible for success-
ful strain optimisation, compound scale-up,
formulation and clinical development activities
that allowed anti-infective drug research to gain
prominence as a viable area for corporate
investment. After the successful commercialisa-
tion of penicillin following the Second World
War, companies like Abbott, Beecham, Bristol,
Glaxo, Lederle, Lilly, Merck, Pfizer, Roche,
Schering and Squibb became leaders in antibi-
otic development and maintained active antibac-
terial research organisations for decades.
Corresponding author and reprint requests: Karen Bush,
Johnson & Johnson Pharmaceutical Research & Development,
LLC 1000 Route 202 Raritan, NJ 08869, USA
E-mail: kbush@prdus.jnj.com
have recently scaled back their efforts in this area,
antibacterial drug discovery remains an import-
ant opportunity for at least some of the pharma-
ceutical sector. In this article, four major reasons
to support the continuation of antibacterial drug
discovery are provided: medical needs based on
the continued increase in resistant pathogens,
development advantages, commercial viability,
and social factors.
MEDICAL NEED
Resistance
Soon after each antibacterial agent entered into
clinical practice, resistance was reported in at
least one bacterial pathogen (Table 1). Because the
first antibiotics, excluding the synthetic sulfa
drugs, were all identified or derived from natural
products, resistance determinants had already
accumulated in the environments from which
these agents originated. It was only a short period
of time before selection pressures allowed these
environmental resistance determinants to become
incorporated into the pathogenic bacteria that
were being treated with the new antibiotics.

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases
 Bush Antibacterial drug discovery in the 21st century 11

Table 1. Reported resistance to new agents following approval for clinical use

Agent for which Resistance Mechanism of

resistance was observed FDA approval reported resistance

Penicillin G 1943 1940 Penicillinase production [3]

Streptomycin 1947 1947 (Mutation in ribosomal protein S12)a [4]
Tetracycline 1952 1952 (Efflux)a [4]
Penicillin and tetracycline 1943 and 1952 1976–80 Plasmid-encoded broad spectrum
(Neisseria gonorrhea b-lactamases and tetracycline efflux pump [5]
and enterobacteriaceae)
Methicillin (and 1960 1961 MecA (penicillin-binding protein 2a) [6]
eventually all b-lactams
in S. aureus)
Nalidixic acid 1964 1966 Topoisomerase mutations [7]
Gentamicin 1967 1969 Aminoglycoside modifying enzyme [8]
Cefotaxime 1981 1981 AmpC b-lactamase [9]
1983 Extended-spectrum b-lactamases (ESBLs) [10]
Linezolid 2000 1999 23S RNA mutation [11,12]
a
Assumed to be resistance mechanism. Not identified at time of report.

Selection of resistant strains occurred so the plasmid-mediated resistances, whereby a

quickly for some bacteria-antibiotic combinations
that clinical utility of the antibiotic was severely
diminished within a 5-year time span. The first
documented example of the rapid selection of a
resistant population was the increase in penicillin
resistance from £8% to almost 60% in Staphylo-
coccus aureus from 1945 to 1949 (Fig. 1). In many
cases, chromosomal mutations led to class resi-
stances to other closely related agents, allowing
for the emergence of clonal strains when subjec-
ted to antibiotic pressures. Examples of chromo-
somal mutations leading to class resistances
include topoisomerase mutations selected by nal-
idixic acid and later fluroroquinolones, and selec-
tion of derepressed AmpC b-lactamases by any of
a number of second and third generation cephalo-
sporins. Of perhaps even graver consequence are
100
Penicillinase-producing
resistant strain selected by one antibiotic can
result in resistance to other antibiotic classes due
to the presence of multiple resistance determi-
nants in the same operon. Most importantly, these
plasmids can be transferred among species,
resulting in multifactorial resistance in organisms
such as Escherichia coli, Klebsiella pneumoniae and
Pseudomonas aeruginosa.
Following the identification of resistance to a
class or to a specific agent, the pharmaceutical
industry has responded with new generations or
new classes of drugs (Table 2). Examples of these
were the development of the ‘penicillinase-stable
penicillins’ to counteract Gram-positive b-lacta-
mases, second generation aminoglycosides to
avoid streptomycin resistance, third generation
cephalosporins to provide improved activity
against Gram-negative pathogens, the ketolide
telithromycin to avoid macrolide resistance in the
pneumococci, and linezolid in a novel class of

synthetic oxazolidinones with no cross-resistance

75
to any known antibiotic class. However, in all
S. aureus

cases, resistance has emerged to the new agents,

thus repeating the cycle (Table 1).
%

50
At this time in the history of the antibiotic
world, one might imagine that all resistance
25 mechanisms could be overcome by at least some
agent. Indeed, some have stated that virtually all
0 infections can be treated by at least a combination
1940 1950 1960 1970 1980 1990
of effective drugs, so it is not necessary to develop
Year new agents. Unfortunately, that is not the case, as
Fig. 1. Rapid increase in penicillinase production in S. au- evidenced by the multidrug-resistant enteric bac-
reus [13,14]. teria, and the pan-resistant pseudomonads that

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
12 Clinical Microbiology and Infection, Volume 10 Supplement 4, 2004

Table 2. Medical needs and drugs developed to meet them

Drugs developed to counteract specific resistance

Resistance
causing medical need Approved drugs Drugs currently in development

Penicillinase producing Penicillinase-stable penicillins: oxacillin, BAL5788, RWJ-54428

S. aureus methicillin, cloxacillin, dicloxacillin
First- and second-generation cephalosporins:
cefaclor, cephradine, cephalexin, cephalothin
Tetracycline resistance Minocycline, doxycycline Tigecycline
Gentamicin resistance Tobramycin, amikacin None
Nalidixic acid Norfloxacin, ciprofloxacin, ofloxacin, Garenoxacin
levofloxacin, gatifloxacin, moxifloxacin,
gemifloxacin
TEM-1 b-lactamase in Third- and fourth-generation cephalosporins: No specified drugs
Neisseria gonorrhea followed cefoperazone, cefotaxime, ceftazidime,
by increase in plasmid-mediated ceftriaxone, cefepime
penicillin resistance in Monobactams: aztreonam
Gram-negative rods Carbapenems: imipenem
b-Lactamase inhibitors: clavulanic acid,
sulbactam, tazobactam
b-Lactam resistance caused by Quinupristin-dalfopristin, linezolid, Oritavancin, dalbavancin, ramoplanin,
production of MecA (penicillin- daptomycin tigecycline, iclaprim, BAL5788,
binding protein 2a) in S. aureus RWJ-54428
Extended-spectrum b-lactamases Carbapenems: meropenem, ertapenem Tigecycline, doripenem
(ESBLs) in Enterobacteriaceae
Vancomycin resistance Quinupristin-dalfopristin, linezolid, Oritavancin, ramoplanin, tigecycline
in enterococci daptomycin
Macrolide-resistant streptococci Telithromycin No specified drugs

are currently being treated with the toxic mem-
brane-disruptive polymyxins [15].
With the approval of the three most recent
antibacterial agents, linezolid in 2000, daptomycin
in 2003 and telithromycin in 2002–04, three new
classes of agents have been introduced into the
marketplace. However, as might be expected,
resistance has already been reported for all three
agents, thus providing an opportunity for addi-
tional agents in these classes to overcome the new
resistances identified.
Unmet needs and resistance
Why, or when, does a research organisation make
the decision to develop a new drug to counteract
resistance? A major factor must be unmet medical
need. This need is closely tied to the rapidity and
proliferation of resistance development, and the
number of other drugs that may be used for
the same therapeutic indication. One can cite
the recent flurry of activity to develop
drugs to treat multidrug resistant (MDR) Gram-
positive bacteria, including MRSA (methicillin-
resistant Staphylococcus aureus) and VRE
(vancomycin-resistant enterococci). With the
increasing threats from less resistant but more
virulent community-acquired MRSA strains [16],
it should be expected that orally active agents
other than b-lactams will be of high interest. As
indicated by Livermore in this supplement [17],
other medical needs may be met with new agents
to treat infections caused by MDR pathogens such
as, Pseudomonas aeruginosa, Stenotrophomonas
maltophilia, Acinetobacter spp., Enterococcus faecium,
MDR enterobacteriaceae and MDR Mycobacterium
tuberculosis. All but MDR M. tuberculosis are
usually associated with hospital infections, thus
indicating that drugs in the hospital market will
continue to provide opportunities to meet in-
creased medical needs.
Emerging diseases
As we try to predict the future needs in infectious
diseases, there are always going to be unpredict-
able factors such as the appearance of new
diseases, or the newly recognised association of
established diseases with infectious agents.
Examples of previously unrecognised infectious

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
 Bush
diseases are Legionnaire’s disease that was found
to be associated with Legionella pneumophila [18],
and multiple viral diseases such as ebola, AIDS,
hepatitis C and SARS. Diseases described before
they were known to be caused by infectious
agents include cat scratch disease caused by
Bartonella henselae [19], and gastric ulcers caused
by Helicobacter pylori [20]. In addition, diseases
that were once thought to be under control by
existing therapies may re-emerge, such as cholera,
dengue and pertussis.
As with the case of H. pylori and ulcers, there is
a great deal of curiosity in the infectious disease
community as to possible anti-infective links with
other chronic diseases such as cardiovascular or
neurological disorders. Clinical studies that have
studied the effect of azithromycin therapy on the
treatment of atherosclerosis have been equivocal,
in part, because it has not been convincingly
determined whether the Chlamydia pneumoniae in
arterial plaque is the causative agent or present as
an innocent bystander in a favourable environ-
mental niche [21,22]. However, if a causal associ-
ation is proven, a large commercial opportunity
awaits the company that can present data from a
positive trial with a new antichlamydial agent.
Likewise, hints of associations between C. pneu-
moniae and Alzheimer’s disease or asthma [23,24]
suggest that there may be additional chronic
illnesses that may ultimately benefit from anti-
bacterial therapy. If these opportunities are devel-
oped in the future, this will provide another set of
pressures to increase antibacterial resistance, and
will provide incentives for companies to identify
new organism-specific agents.
Finally, there is now the threat of epidemic
outbreaks of resistant organisms in attacks of
bioterrorism. Although the mail attacks of anthrax
in the USA were fortunately conducted with a
highly susceptible Bacillus anthracis strain in 2001
[25], it is now well established that antibiotic-
resistant strains can be readily selected in vitro [26],
thus leading to the possibility of more resistant
strains appearing in the future. Because inhala-
tional anthrax is most frequently a fatal disease
due to rapid toxin production, opportunities are
available to approach this disease state more
creatively, with the development of drugs to attack
the toxin itself, in addition to the use of classical
antibacterial agents to eradicate the bacteria [27].
And, development of safe and effective vaccines is
still an option that is frequently ignored.

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
Antibacterial drug discovery in the 21st century 13
Dosing advantages
In addition to a medical need to treat a resistant
infection, there can be less obvious medical
advantages offered by new agents. Frequency of
dosing is one area in which new drugs have made
an important difference. An early example was
the replacement of ampicillin by amoxicillin
because the dosing regimen changed from four
times a day to three times a day. Ceftriaxone with
once-a-day dosing became a more highly used
drug than cefotaxime dosed three times a day, in
spite of the fact that cefotaxime was introduced to
the market first and had an interchangeable
spectrum of activity with ceftriaxone. Also,
levofloxacin with once-a-day dosing replaced
twice-a-day ciprofloxacin as the leading hospital
fluoroquinolone. These changes in dosing regi-
mens can result in higher patient compliance,
especially in the outpatient population where
once-a-day dosing has become the standard of
care. In the hospital setting, reduced dosing
frequencies can lead to lower medical costs due
to the resources involved in delivering parenteral
agents. There also are advantages offered by
providing both oral and parenteral agents so that
hospitalised patients can be transferred from
intravenous (i.v.) therapy to oral therapy on the
same drug, and subsequently released from the
hospital earlier. These attributes are shared by
many of the fluoroquinolones and by linezolid
where both oral and i.v. forms are available.
Safety improvements
Improvement in safety can also provide an oppor-
tunity for development of a second- or third-in-
class agent. Although some adverse events are
inherently associated with specific antibiotic clas-
ses, the degree to which these manifest can vary
greatly. Nephrotoxicity with cephalosporins can be
a major problem, with drugs like cephaloridine
removed from the market because of their renal
toxicity. However, the plethora of ‘cephawhat-
chamacallums’ in the 1980s [28] can be attributed to
the fact that the newer cephalosporins exhibited
good safety profiles with minimal nephrotoxicity
when normal doses were given. Fluoroquinolones
generally exhibit binding to the hERG ion channel
as a class, but the IC50 values for binding in the
patch clamp assay can vary by as much as two
orders of magnitude, from 18 lM for sparfloxacin to
14 Clinical Microbiology and Infection, Volume 10 Supplement 4, 2004

1420 lM for ofloxacin [29]. Thus, there may be room Table 3. Average months in each development phase
to improve safety profiles and drug–drug interac- according to therapeutic area [34]
tions for currently existing antibacterial agents Average number of months Phase Phase Phase
such as linezolid with its reversible monoamine Therapeutic Area I II II NDA
oxidase inhibition [30] and telithromycin with its
reversible cytochrome P450 interactions [31]. Anti-infectives 11 14 36 24
Cardiovascular 17 35 39 30
NSAID 21 23 55 39
DEVELOPMENT FACTORS Neuropharmacological 14 24 41 43

Attractive development opportunities

well-established in vitro testing to show that drugs
Development of anti-infective agents offers cer-
can get to their targets and have the desired
tain attractions not seen with some other thera-
therapeutic effect, i.e., killing of the micro-organ-
peutic areas. Based on historical data through the
ism. Second, good predictive in vivo efficacy mod-
end of 1999, the probability of an investigational
els have been established for a variety of infections
anti-infective agent entering the market was
caused by specific target organisms, leading to
higher than for agents in other therapeutic areas,
reassurance that clinical trials have a high probab-
as shown in Fig. 2 [32]. This advantage is
ility of success. Third, there are clearly defined
especially notable when success was measured
biomarkers to allow for monitoring of patients in
from the first human dose (Phase 1), first patient
clinical trials. Fourth, the use of pharmacodynam-
dose (Phase 2) or first pivotal dose (Phase 3) when
ics to predict efficacy, optimal doses, and the
anti-infective agents were compared with cardio-
potential for resistance selection has become a
vascular, oncology or central nervous system
sophisticated art in the anti-infective area. And
(CNS) drugs [33].
fifth, safety issues, especially for hospital-admin-
Other historical data indicate that FDA appro-
istered drugs, tend to be less than for drugs that are
val times were faster for anti-infective agents
prescribed for chronic illnesses where a patient will
compared with drugs in the other therapeutic
take a drug for the rest of his life. Finally, for
areas. As seen in Table 3, during each develop-
community drugs with decreased dosing regimens
ment phase, anti-infectives had the shortest resi-
of 3–5 days, long-term safety and follow-up stud-
dency time compared with cardiovascular,
ies lasting for years are generally not required,
NSAIDs and CNS agents [34]. In addition, the
thereby leading to shorter clinical trials.
capitalised clinical costs per approved NCE were
lowest for anti-infective drugs compared with the
other therapeutic areas [34]. Development role for large companies
Several possible reasons may be cited for the
Large pharmaceutical companies must continue to
development advantages for anti-infective drugs.
play a major role in the discovery and development
First, the drug targets are clearly delineated, with
of new anti-infective agents. Companies with
active antibacterial research and development
% of NCEs approved by January 1, 2000 programmes currently have the resources and
Number of NCEs with INDs filed from 1981 to 1992
talent to be able to take a drug from the conception
30
stage through to full FDA approval. The basic
20 research scientists in these groups have established
10
resources in terms of financial support and facilit-
ies, and have the experience and commitment to
0 follow a drug from identification through to study-
e
y

tic
r
S

al

ing clinical isolates for mechanisms of resistance.

la
ce
r

in

ive
N

in
to

he
cu

cr
C

an

st
ra

ct
st
do
as

te
C
pi

fe

Development scientists have expertise to run the

ae
ov

in
En
es

in
tro

an

ti-
di
R

An
ar

c/
as

toxicology and drug metabolism studies, to scale-

si
C

ge
al

up compounds under GMP conditions for clinical

An

Fig. 2. Success rates according to therapeutic class for self-

supplies, and to formulate the agents in pharma-
originated NCEs that had INDs first filed from 1981 to 1992 cologically acceptable forms. Clinical scientists
[32]. have experience in running clinical trials, with

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
 Bush Antibacterial drug discovery in the 21st century 15

appropriate contacts and clinical sites in place for Table 4. Worldwide pharmaceutical sales of agents in
future studies. And, regulatory groups have various therapeutic areas including prescription and over-
the-counter sales
experience with the various regulatory agencies
so that proper documentation and records are 2002 Total
maintained and filed throughout the development Sales Sales Population
process. It is imperative that companies with these Therapeutic Area ($bn)a (%)a sharesb
sets of experienced investigators maintain a critical Cardiovascular 67 16.8 23.2
mass before this expertise is filtered away into CNS 64 16.0 17.9
other therapeutic areas. Anti-infectives 45 11.2 12.8
Because many of the same functions are in Gastrointestinal 35 8.7 2.1
Respiratory 33 8.2 6.0
place for all therapeutic areas, a large company
Endocrinology ⁄ 32 8.0 7.0
can capitalise on the knowledge from other metabolic
groups. Shared resources such as scale-up chem- Oncology ⁄ immunology 27 6.7 12.4
istry, formulations, toxicology, metabolism and Rheumatology ⁄ 23 5.7 NDc
regulatory groups can work on compounds from musculoskeletal
Haematology 22 5.5 ND
any therapeutic area, so dedicated personnel are Miscellaneous 53 13.2 18.3
not lying in wait for the next compound to enter Total 401 (100) (100)
development. Also, because the return on invest- a
IMS Health World Review.
ment may be perceived to be lower for anti- b
Data through 2000 [37].
infective drugs [35], the ability to balance a c
No data provided.
portfolio with drugs from multiple therapeutic
classes means that large companies can afford to
stay in less lucrative areas of research such as to introduce a billion-dollar drug every year to be
antibacterial development. profitable. If a mix of parenteral agents is sold by
a single company, sales exceeding $1 bn can be
achieved by using the same hospital sales force to
COMMERCIAL VIABILITY market three drugs with sales of $400 m or less.
Over the next 5 years growth in the critical care
Worldwide sales antibacterial market is projected to increase con-
In addition to the scientific and developmental sistently, based on the increase in antibiotic sales
opportunities in the anti-infectives area, it is still from 1999 to 2002 [38]. This growth can be directly
necessary to show that there is a commercial related to the increase in serious infections caused
rationale for staying in this area. When worldwide by problem Gram-negative pathogens that are
pharmaceutical sales are broken down by thera- invading the intensive care units of hospitals. It is
peutic area, anti-infective agents rank third, interesting to note that sales appear to grow for
behind cardiovascular and CNS drugs (Table 4) quinolones, carbapenems and new agents, but are
[36,37]. The anti-infective market with US $45 bn
in sales is further divided into antibacterial,
Table 5. Worldwide anti-infective market sales and pro-
antifungal and antiviral areas (Table 5), with jected growth [36]
antibacterial sales providing almost two-thirds of
the total revenue [36]. Although other areas of Sales Annual
anti-infective drugs have been projected to grow Anti-infective in $bn 4 years growth
area (% Total) CAGRa in sales ($bn)
faster than antibacterial agents for the next few
years, all sectors show a positive growth rate, with Antibacterials 27.9 (62%) +6% 1.7
an overall growth of 10% for all anti-infectives. HIV antivirals 5.4 (12%) +22% 1.2
Interestingly, when the increase in sales is Non-HIV antivirals 2.7 (6%) +22% 0.6
Antifungals 3.2 (7%) +10% 0.3
computed in terms of actual dollars, the largest Sera, vaccines and 5.9 (13%) +14% 0.8
absolute increase in growth is projected for the immunoglobulins
antibacterial group (Table 5). This total represents Total 45 (100%) +10% (4.5)b
an increase of $1.7 bn per year. Although com- a
Compounded annual growth rate based on data from
panies like to boast about the number of billion- 1998 to 2002.
b
dollar drugs in their portfolio, one does not need Values do not add up to $4.5 bn due to rounding.

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
16 Clinical Microbiology and Infection, Volume 10 Supplement 4, 2004
projected to decrease or remain steady for the
penicillins and cephalosporins. This may reflect the
increase in generic agents available in those areas,
or the increasing resistance to these agents among
Gram-negative pathogens.
CORPORATE RESPONSIBILITY
Social commitment
Every pharmaceutical company has established a
position of social and corporate responsibility to
meet the medical needs of its customers. A survey
of corporate websites has identified statements
of corporate responsibility for virtually every
pharmaceutical company in the Western world.
These statements, although unique to a specific
company, all include many similar expressions
such as ‘responsibility’, ‘commitment’, ‘commu-
nity’, ‘values’ and ‘vision’. However, this corpor-
ate responsibility extends beyond internet
webpages. Examples of this social commitment
include the corporate donations or price-lowering
for ivermectin to treat river blindness (Merck), for
miconazole for AIDS-affected individuals (Tibo-
tec), and for anti-HIV drugs (GSK) to African
populations that cannot afford to pay full price for
these agents. Large companies such as AstraZen-
eca and Johnson & Johnson have demonstrated a
social commitment to less profitable research by
working on drugs to treat multidrug resistant
tuberculosis.
All companies exemplify a concern for the
welfare of their customers who are constantly
looking for new drugs and new treatments to
improve their quality of life. However, in the area
of anti-infective research, the goal is perhaps one
of even greater commitment. The agents that have
been developed in the past, and that are being
developed for future needs, are not drugs that
improve the ‘quality of life’, but are agents that
frequently treat acute, life-threatening diseases.
CONCLUSIONS
We need to remember that the antibiotic era
began only about 60 years ago at a time when a
strep throat could mean a death sentence.
Although we have been prolific in the past by
developing new antibiotics rapidly, those days
are behind us, as so clearly outlined by a number
of recent reports, including those from Projan and

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
Shlaes [39] and from Powers [40] in this supple-
ment. With the increasing multidrug resistance
that threatens our entire armamentarium of anti-
biotics, it is imperative that new agents continue
to be developed. And, perhaps more importantly,
it is critical that large pharmaceutical companies
continue to fund basic research to identify the
new agents that will be able to circumvent newly
emerging mechanisms. There are sound scientific,
medical, developmental and commercial reasons
for companies to remain in the anti-infective
therapeutic area. Although we will always live
in a world coinhabited by antibiotic-resistant
bacteria, we have a social commitment to devise
new approaches to contain them before it is too
late.
REFERENCES
1. Diggins FW. The discovery of penicillin: so many get it
wrong. Biologist 2000; 47: 115–9.
2. Schatz A, Bugie E, Waksman SA. Selman Abraham
Waksman, PhD, 22 July 1888–16 August 1973. Strepto-
mycin reported. Ann Intern Med 1973, 79: 678.
3. Abraham EP, Chain E. An enzyme from bacteria able to
destroy penicillin. Nature 1940; 146: 837.
4. Davies JE. Origins, acquisition and dissemination of anti-
biotic resistance determinants. Ciba Foundation Symp 1997;
207: 15–27; discussion 27–35.
5. Perine PL, Schalla W, Siegel MS, et al. Evidence for two
distinct types of penicillinase-producing Neisseria gonor-
rhoeae. Lancet 1977; 2: 993–7.
6. Hiramatsu K, Cui L, Kuroda M, Ito T. The emergence and
evolution of methicillin-resistant Staphylococcus aureus.
Trends Microbiol 2001; 9: 486–93.
7. 1993. Peterson LR. Quinolone resistance in clinical prac-
tice: occurrence and importance. In: Quinolone Antimi-
crobial Agents, 2nd Edn. (eds Hooper, DC, Wolfson, JS
3 American Society for Microbiology): 119–37.
8. Lacey RW, Mitchell AA. Gentamicin-resistant Staphylo-
coccus aureus. Lancet 1969; 2: 1425–6.
9. Then RL, Angehrn P. Trapping of nonhydrolyzable
cephalosporins by cephalosporinases in Enterobacter cloacae
and Pseudomonas aeruginosa as a possible resistance
mechanism. Antimicrob Agents Chemother 1982; 21: 711–7.
10. Knothe H, Shah P, Krcmery V, Antal M, Mitsuhashi S.
Transferable resistance to cefotaxime, cefoxitin, cefa-
mandole and cefuroxime in clinical isolates of Klebsiella
pneumoniae and Serratia marcescens. Infection 1983; 11:
315–7.
11. Clemett D, Markham A, Linezolid. Drugs 2000; 59: 815–27;
discussion 828.
12. Bozdogan B, Appelbaum PC. Oxazolidinones. activity,
mode of action, and mechanism of resistance. Int J Anti-
microb Agents 2004; 23: 113–9.
13. Bush K. The evoution of b-lactamases. In: Antibiotic
Resistance: Origins, Evolution, Selection and Spread. (eds
Chadwick, DJ, Goode, J. ). John Wiley & Sons 1997, 152–66.
Ciba Foundation Symposium, Vol. 207.
 Bush
14. Medeiros AA. Evolution and dissemination of b-lacta-
mases accelerated by generations of b-lactam antibiotics.
Clinic Infect Dis 1997; 24: S19–45.
15. Evans ME, Feola DJ, Rapp RP. Polymyxin B sulfate and
colistin: old antibiotics for emerging multiresistant
Gram-negative bacteria. Ann Pharmacother 1999; 33: 960–
7.
16. Liassine N, Auckenthaler R, Descombes MC, Bes M,
Vandenesch F, Etienne J. Community-acquired methicil-
lin-resistant Staphylococcus aureus isolated in Switzerland
contains the Panton-Valentine leukocidin or exfoliative
toxin genes. J Clin Microbiol 2004; 42: 825–8.
17. Livermore DM. The need for new antibiotics. Clin Micro-
biol Infect 2004; 10(Suppl. 4): 1–9.
18. Porschen RK. Legionnaires’ disease. more on staining and
isolation of Legionella pneumophila. N Engl J Med 1979; 300:
369–70.
19. Chen SC, Gilbert GL. Cat scratch disease: past and present.
J Paediat Child Health 1994; 30: 467–9.
20. Marshall BJ. The Campylobacter pylori story. Scand J Gast-
roenterol 1988; 146(Suppl): 58–66.
21. Anderson JL, Muhlestein JB, Carlquist J et al. Randomized
secondary prevention trial of azithromycin in patients
with coronary artery disease and serological evidence for
Chlamydia pneumoniae infection: The Azithromycin in
Coronary Artery Disease: Elimination of Myocardial
Infection with Chlamydia (ACADEMIC) study. [See
comment]. Circulation 1999; 99: 1540–7.
22. Selzman CH, Netea MG, Zimmerman MA, et al. Athero-
genic effects of Chlamydia pneumoniae: refuting the inno-
cent bystander hypothesis. J Thoracic Cardiovas Surg 2003;
126: 688–93.
23. Ronchetti F, Ronchetti R, Guglielmi F, et al. Detection of
Chlamydia pneumoniae in cholesteatoma tissue: any patho-
genetic role? Otol Neurotol 2003; 24: 353–7.
24. Little CS, Hammond CJ, MacIntyre A, Balin BJ, Appelt
DM. Chlamydia pneumoniae induces Alzheimer-like
amyloid plaques in brains of BALB ⁄ c mice. Neurobiol
Aging 2004; 25: 419–29.
25. Dalton R. Genetic sleuths rush to identify anthrax strains
in mail attacks. Nature 2001; 413: 657–8.

2004 Copyright by the European Society of Clinical Microbiology and Infectious Diseases, CMI, 10 (Suppl. 4), 10–17
Antibacterial drug discovery in the 21st century 17
26. Brook I, Elliott TB, Pryor HI, et al. In vitro resistance of
Bacillus anthracis Sterne to doxycycline, macrolides and
quinolones. Int J Antimicrob Agents 2001; 18: 559–62.
27. Shen Y, Zhukovskaya NL, Zimmer MI, et al. Selective
inhibition of anthrax edema factor by adefovir, a drug for
chronic hepatitis B virus infection. Proc Natl Acad Sci USA
2004; 101: 3242–7.
28. Ellis R, Lancaster DJ, Cephawhatchamacallums. N Engl J
Med 1989; 320: 872.
29. Kang J, Wang L, Chen XL, Triggle DJ, Rampe D. Interac-
tions of a series of fluoroquinolone antibacterial drugs
with the human cardiac K+ channel HERG. Molec Phar-
macol 2001; 59: 122–6.
30. French G. Safety and tolerability of linezolid. J Antimicrob
5 Chemother 2003, 51: ii45–53.
31. Zhanel GG, Walters M, Noreddin A, et al. The ketolides: a
critical review. Drugs 2002; 62: 1771–804.
32. Dimasi JA. Risks in new drug development: approval
success rates for investigational drugs. Clini Pharmacol
Therapeutics 2001; 69: 297–307.
33. Keim KL, Calcagno J, Leventer SM, Harris HW. Unlocking
the value of failed CNS trials. Curr Drug Disc 2003; 3: 29–33.
34. DiMasi JA, Hansen RW, Grabowski HG, Lasagna L.
Research and development costs for new drugs by thera-
peutic category. A study of the US pharmaceutical
industry. Pharmacoeconomics 1995; 7: 152–69.
35. Projan SJ. Infectious diseases in the 21st century: increas-
ing threats, fewer new treatments and a premium on
prevention. Curr Opin Pharmacol 2003; 3: 457–8.
6 36. IMS Health World Review 2003.
37. DiMasi JA, Hansen RW, Grabowski HG. The price of
innovation: new estimates of drug development costs.
J Health Econom 2003; 22: 151–85.
38. Bush K. Why it is important to continue antibacterial drug
discovery. ASM News 2004; 70: 282–7.
39. Projan S, Shlaes D, Update: antibacterial drug discovery: is
it all downhill from here? Clin Microbiol Infect 2004;
8 10(Suppl. 4): 18–22.
40. Powers JH. Antimicrobial drug development–the past, the
present, and the future. Clin Microbiol Infect 2004; 10(Suppl.
4): 23–30.