Postgraduate Medical Journal (1988) 64: 552-558

The changing face of chemotherapy

W. Brumfitt and J.M.T. Hamilton-Miller

Department of Medical Microbiology, The Royal Free Hospital School of Medicine, Pond Street, Hampstead,London NW3 2QG, UK.

Summary: The historical development of antibiotics has been summarized. Three distinct phasesare discernible. The first (from historical times to about 1900) involved mostly folk remedies. Thesecond (1900-c. 1940) was ushered in by Paul Ehrlich's development of the concept of 'selectivetoxicity' and saw the establishment of arsenicals and sulphonamides. The third, lasting to the presentday, started with the exploitation of the pioneering studies of Fleming, Dubos and Waksman onantibiotic production by soil fungi. This latest phase has continued with the improvement of naturalproducts by the skills of the medicinal chemist.The properties and evolution of three major groups of antibiotics, penicillins, cephalosporins and

aminoglycosides are fully described.Finally, pathways of possible future evolution of antibiotics are outlined.

Introduction: from pre-history to the 20th century

The idea of using natural or artificial substances toprevent or reverse 'putrefaction of wounds' is avery old one, and has been known since the start ofhistory. Thus, the ancient Egyptians used myrrh(from Commiphora spp.) and frankincense (fromBoswellia spp.), the Assyrians anticipated PaulEhrlich by several centuries by introducing arsenic.The use of mercury, pioneered by the Arabs forsyphilis and championed by Paracelsus, lasted fromthe 16th to the early 20th century. Moses wasadvised to treat leprosy with hyssop (Leviticus,Ch. 14) and the Good Samaritan disinfected thewounds of his assaulted fellow traveller by pouringin oil and wine (Luke, Ch. 11). The Romansdressed wounds with spider's webs dipped in honeyand also observed that keeping drinking water insilver vessels prevented the spread of dysenterythrough their troops. Finally, mouldy bread oftenyields a growth of Penicillium spp. Thus it could bespeculated that the practice of applying mouldybread to wounds was effective by virtue ofantibiotic production.

Plants were a rich source of antimicrobialsubstances. Beside the examples already mentioned,cinchona bark was first used in Peru to combatmalaria, and ipecacuanha was found to be effectivefor amoebic dysentery. The famous British herbalistCulpepper (1826) lists many plants which are activein infections - to mention just two: Carduus

benedictur (the holy thistle) cures 'quartan agues',and lovage 'resisteth poison and infection'. Amodern herbal also gives many plant remedies -for example an extract of the coneflower (Echinaceasp.) is a specific for abscesses. It is interesting tonote that such extracts have been shown toantagonize hyaluronidase. Also, at least one majorpharmaceutical company is exploring the florafound in Madagascar for pharmacological activity(including antimicrobial properties).

Before the germ theory had been properlyformulated it had been realised intuitively forcenturies that many diseases were passed fromperson to person and that this chain could bebroken by the use of appropriate chemicals. Thus,simple antiseptics such as 'hydrated lime' (hypo-chlorite), tincture of iodine and phenol were in wideuse by the mid-19th century.

Ehrlich can justly be called 'The Father ofChemotherapy', and his researches on vital dyessuch as methylene blue and trypan red and thenwith arsenicals prepared the way for the antibioticera which was formally ushered in by Domagk andhis colleagues with the discovery that Prontosilrubrum had antibacterial properties in vivo.

Synthetic compounds: the 1930s, the dawn ofrational chemotherapy

Antifolate agents

The chemotherapeutic era, as generally recognized,

©) The Fellowship of Postgraduate Medicine, 1988

Correspondence: Professor W. Brumfitt, M.D., Ph.D.,F.R.C.P., F.R.C.Path.

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from

THE CHANGING FACE OF CHEMOTHERAPY 553

can be said to have started with the introduction ofthe sulphonamides into clinical use in 1935. Thesewere outstandingly successful at first in the treat-ment of meningitis, streptococcal infections andinfections of the urinary tract. Staphylococcalinfections, however, were found to respond ratherpoorly, due to the neutralization of the antibacterialactivity of the sulphonamides by pus (see below).The active principle of Prontosil rubrum was

found to be sulphanilamide. Due to its relativelylow intrinsic antimicrobial activity (some 40 timesless than moder sulphonamides), large doses ofthis compound had to be used, with undesirableside effects. This provided a stimulus for chemiststo try to improve upon the existing molecule. Theseendeavours were successful in a remarkably shorttime, maximally active sulphonamides such assulphadiazine and sulphathiazole having beenpatented by 1942. This speed was due to the factthat it had been shown that the antimicrobialactivity of a sulphonamide depended upon itsionization characteristics. This allowed rational syn-thetic programmes to be undertaken. Although theintrinsic activity of sulphadiazine could not beimproved upon, further modifications in thesulphonamide molecule have resulted in pharmaco-kinetically superior compounds such as sulfameto-pyrazine. The latter has a half-life of about 100hours and can, therefore, be given once a week.

In this way were born the discipline of medicinalchemistry and the concept of quantitative structure-activity relationships (QSAR), both of which haveplayed and will continue to play an enormouslyimportant role in the development of improvedantibiotics.

Sulphonamides were shown to exert their anti-bacterial activity by inhibiting the biosynthesis ofbacterial folate. Acquired resistance to these com-pounds, which started to emerge soon after theirintroduction, was due to alteration in the targetenzyme. The bacteriostatic action of sulphonamidesis reversed in the presence of substances such as p-aminobenzoate and thymidine, which may be foundin pus.

Research into sulphonamides also produced theanti-leprosy drug diaminodiphenyl sulphone(Dapsone).Sulphonamide fell increasingly out of favour

from the-late 1940s, due to increasing resistanceproblems and the availability of broader spectrumantibiotics.

Other synthetic anti-folate antimicrobial compoundsThe anti-malarial pyrimethamine and the anti-bacterial trimethoprim, both inhibitors of microbial

folate biosynthesis, arose from an area of researchentirely different from that which involved thesulphonamides. Both were found incidentallyduring the research programme by BurroughsWellcome into anti-cancer agents which would actby inhibiting human folate metabolism. Tri-methoprim is a remarkable compound in tworespects. First, it has a unique mode of actionamongst the antimicrobial agents by inhibiting anenzyme (dihydrofolate reductase) present in bothbacterial and human cells. Second, the differentialtoxicity between human and bacterial cells isexplained by small chemical differences in the struc-ture of human and bacterial dihydrofolate reductaseso that the affinity of trimethoprim is about 60,000times greater for the bacterial enzyme. Thus, tri-methoprim in therapeutic doses has negligibletoxicity for host cells. Trimethoprim has enjoyedvery wide usage worldwide in combination withsulphamethoxazole as co-trimoxazole (Septrin andBactrim). However, there is now no doubt that thegreat antibacterial activity and wide therapeuticsuccess of co-trimoxazole is due almost entirely tothe trimethoprim component. Further, many of thearguments in favour of the fixed combination,although probably put forward in good faith, arespecious. In particular the claims for synergy bycombining sulphonamide and trimethoprim andalso prevention of resistance to trimethoprim by thecombination have not withstood scientific investi-gation. Unhappily, trimethoprim is an example ofan excellent drug the value of which has nowbecome eroded by increasing resistance. This is notonly widespread in hospitals where resistance wasdue to an R factor but has subsequently spreadinto the community. To make matters worsebecause of integration of the R factor into thebacterial chromosome a stable resistance results. Inthe United Kingdom resistance includes strains ofStaphylococcus aureus especially those resistant tomethicillin (MRSA). The latter organisms areproving to be a major therapeutic problem since sofew antibiotics are active against them.

Other synthetic compoundsIt is clear from the above that antimicrobial chemo-therapy started by the exploitation of syntheticmolecules. It is notable, however, that following thearsenicals and anti-folate compounds, futureadvances in general mainstream chemotherapy havebeen made almost exclusively through natural pro-ducts and their chemical modification. On the otherhand, some notable contributions have been madein specific areas of chemotherapy by totally syn-thetic compounds. These include the treatment of

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from

554 W. BRUMFITT & J.M.T. HAMILTON-MILLER

tuberculosis where p-aminosalicylate, isoniazid,ethambutol, pyrazinamide and ethionamide are stillused. Other examples are the versatile nitro com-pounds, of which nitrofurantoin, after 30 years,remains invaluable in the treatment of urinaryinfections, the nitro-imidazoles (metronidazole, tini-dazole) widely used both therapeutically andprophylactically when anaerobes are pathogens, andnitrothiazoles in schistosomiasis and other flukeinfestations. It is interesting to note that nitro-furantoin has a safety record which is better docu-mented than any other product used for the treat-ment of urinary infections.

Nalidixic acid and the related 4-quinolones willbe discussed in a later section.

The 1940s: antibiotics from the soil

Fleming's discovery of penicillin, reported in 1928,did not turn into a finding of clinical relevanceuntil the mid-1940s, when the problems involved inproducing and purifying large amounts of the anti-biotic had been solved due to the efforts of Floreyand his colleagues, in collaboration with severalpharmaceutical companies in the USA. However,Fleming's observation stimulated others to searchfor antibiotics in soil organisms. Dubos discoveredtyrothricin - a mixture of the peptide antibioticsgramicidin and tyrocidin (produced by Bacillusbrevis) in 1939, and Waksman reported theisolation of streptomycin (from Streptomycesgriseus) in 1944. While tyrothricin was too toxic forparenteral use, streptomycin revolutionized thetreatment of tuberculosis, and helped Waksman winthe Nobel Prize. In the period between 1947 and1957 chloramphenicol (1947), polymyxins (1947),chlortetracycline (1948), neomycin (1949), oxytetra-cycline (1950), erythromycin (1952), cephalosporinC (1955), vancomycin (1956), novobiocin (1956),rifamycin B (1957) and kanamycin (1957) - toname but a few - were reported. Some of thesesubstances are still in use and will be referred tolater.

The 1960s - the decade of the penicillinsThe isolation of the penicillin 'nucleus' (6APA) in1959 at Beecham Research Laboratories paved theway for the semi-synthesis and marketing of a verylarge number of penicillins. Some of the moresuccessful ones marketed in the 1960s have been:ampicillin, amoxycillin, methicillin, flucloxacillin,carbenicillin and ticarcillin, followed later by mezlo-cillin, azlocillin, piperacillin and mecillinam. Thesecompounds made a great impact initially on both

Gram-positive and Gram-negative pathogens, butresistance (generally caused by the spread ofR-factor mediated fB-lactamases) emerged in Gram-negative strains, particularly in the hospitalpopulation, severely curtailing their usefulness.Compounds such as ampicillin and flucloxacillinremain very useful today for most purposes in thedomiciliary situation, and in hospitals flucloxacillinremains effective for the majority of Gram-positiveinfections. However, the selection of bacterial resis-tance continues relentlessly, and it is predictablethat strains such as MRSA, JK diphtheroids andenterococci will become more widespread (especiallyin compromised patients), further reducing theusefulness of penicillins. Even the recent develop-ment of inhibitors of fl-lactamases (such asclavulanic acid and sulbactam) will probably onlypartially alleviate this decline.Aztreonam is an example of a new class of

fl-lactam, namely the monobactams. This is anexample of a new antibiotic which is remarkable inhaving a range of activity restricted to the Gram-negative organisms. Apart from its activity againsta number of important Gram-negative organismsthe decreased likelihood of disturbing othercommensal flora may prove to be an advantage.

The 1970s - the decade of the cephalosporinsThe first commercially successful semisyntheticcephalosporin (cephalothin) was made in 1962,following the isolation of the 'nucleus' 7ACA fromcephalosporin C (this proved to be much moredifficult that the analogous process in the penicillinseries). However, for the first few years cephalo-sporins were perceived only as more expensivesubstitutes for penicillins. It was not until penicillin-resistant bacterial strains became increasinglycommon that the cephalosporins' intrinsic stabilityto P-lactamases and the fact that this stability couldbe increased greatly in a rational and predictablemanner caused an increase in their use, especially inthe treatment of nosocomial infections caused byGram-negative bacteria. The advent of the so-called'third generation' cephalosporins such as ceftazi-dime which combine very high intrinsic activity,great fl-lactamase stability and activity againstPseudomonas aeruginosa must represent the peak ofachievement in this class of compound. These latestcephalosporins seem set to replace aminoglycosidesas drugs of choice in the treatment of acute sepsisof unknown aetiology, and also seem likely to beeffective alone in the primary treatment of infec-tion, manifested by fever and leukopenia, inimmunocompromised patients. It is interesting tonote that the early cephalosporins (cephaloridine

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from

THE CHANGING FACE OF CHEMOTHERAPY 555

and cephalothin) had a powerful action againstGram-positive bacteria but much less againstanaerobes, certain Gram-negative bacteria andnone against some important nosocomial strains,such as pseudomonads. By contrast, the latest com-pounds (so-called third generation cephalosporins)are highly active against Gram-negative bacteria -including the pseudomonads (especially Ps. aeru-ginosa) and are also more active against anaerobes.However, this greater activity is gained at theexpense of remarkably decreased activity againstthe Gram-positive bacteria. It is not widelyappreciated that no cephalosporin currentlyavailable is active against Streptococcus faecalis.Thus, superinfection with this organism is a wellknown complication of cephalosporin therapy.

The aminoglycosidesAs mentioned above, streptomycin revolutionizedthe treatment of tuberculosis in the 1950s, but theease with which bacteria acquired resistance and ahigh propensity for ototoxicity and nephrotoxicitysoon severely limited the use of this antibiotic formore general use. Neomycin proved too toxic forparenteral use, and kanamycin, although enjoying avery wide market, never really became establishedas an aminoglycoside of first choice due to itsdeficiencies against resistant hospital pathogens.The gentamicin complex which was introduced in1964, on the other hand, rapidly found favour forthe treatment of acute sepsis, due to its broadantibacterial spectrum. This included the muchfeared pathogen Ps. aeruginosa which until thencould only be treated by the polymyxins, the valueof which was limited by their toxicity - especiallynephro- and neurotoxicity. Early clinical use at toolarge a dose gave gentamicin the reputation ofbeing highly ototoxic and nephrotoxic, whereaswhen properly used - i.e. dosage controlled bydetermining peak and trough levels in individualpatients - the incidence of toxicity is acceptablylow. One problem about such monitoring is theincreased time and therefore cost which resultsfrom the need to obtain blood samples in order tocarry out estimations at frequent intervals.

Increasing resistance among Gram-negativebacilli during the 1970s led to intensive investi-gation of the cause of such resistance. The result ofthese studies was the discovery of the large familyof aminoglycoside-inactivating enzymes. Once thebiochemical mechanisms of action of these enzymeshad been made clear it was possible to modify theantibiotic molecules to render them insusceptible toenzymic attack. This process, involving close co-operation between microbiologist, biochemist and

medicinal chemist, was somewhat analogous to thatwhich occurred in respect of the P-lactam anti-biotics and the enzymes which inactivate them, thef-lactamases (see above). In the case of the amino-glycosides the result of these collaborative effortsincluded amikacin (a derivative of kanamycin A inwhich the inactivating enzymes' target sites wereprotected). By contrast dibekacin (a kanamycin Bproduct not available in the United Kingdom) is aproduct where the enzymes' targets have beenremoved completely, thus solving the problem in adifferent way.Amikacin has been in clinical use for some 12

years, and resistance emergence so far has beensmall. This may be because, due to its high cost,usage of amikacin has been relatively small. On theother hand, there is increasing published evidencethat for this antibiotic resistance emergence maynot be connected with total use. For whateverreason, amikacin can still be regarded as an anti-biotic with a future, whereas the other amino-glycosides seem to be drugs the use of which is indecline.

Other groups of antibiotics which have benefitedfrom chemical modification

The rifamycins were isolated from Streptomycesmediterranei but owe their current clinic therapeuticimportance entirely to the skill of the medicinalchemist. Of the original naturally occurring com-pounds rifamycin B had some antimicrobial activitybut this was insufficient for clinical use. The earlymodifications, rifamycin SV and rifamide, createdlittle interest outside Italy, as they were of quite lowantibacterial activity, were somewhat erraticallyabsorbed and had unfamiliar pharmacokinetic pro-files (due to excretion being almost entirely in thebile). However, a later product, rifampicin, hasbecome extremely important in the treatment oftuberculosis, and is increasingly being realised to bean antibiotic of great potential for the managementof infections caused by MRSA and other recalci-trant Gram-positive species.

Lincomycin is another natural product whoseactivity has been successfully increased by chemicalmeans, in the form of clindamycin. Lincomycin hashad a somewhat chequered history, having beenintroduced first as an antistaphylococcal agent.Lincomycin and clindamycin enjoyed a resurgenceof popularity in the early 1970s due to their broadanti-anaerobic spectrum as well as good activityagainst Gram-positive cocci, but since that timetheir association with antibiotic-induced colitis hasreduced their usage substantially.

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from

556 W. BRUMFITT & J.M.T. HAMILTON-MILLER

The macrolides are a large group of naturalproducts of which only erythromycin and, to amuch lesser extent, spiramycin and oleandomycinhave made any significant clinical impact in theUK. Bioavailability has always been a problemwith erythromycin, and chemical modifications -

albeit of a relatively minor nature - have enabledthis to be improved. Thus, the stearate and parti-cularly a recent development roxithromycin arebetter absorbed by mouth than the naturallyoccurring free base. Intramuscular (ethylsuccinate)and intravenous (lactobionate) preparations arealso available. Another ester, the estolate, is now

used with caution because of reports of hepato-toxicity. It has apparently not been possible to alterthe antimicrobial spectrum of the macrolides bychemical modification. These compounds are stillimportant, particularly in certain respiratory infec-tions such as Legionnaires disease and mycoplasmapneumonia. In campylobacter enteritis, erythro-mycin is the drug of choice. For other indicationserythromycins are usually 'second line' antibioticsbut are often used as a substitute for benzylpenicillinwhen the patient is 'penicillin sensitive'.

In the tetracycline series radical changes havebeen made in both the antibacterial and pharmaco-kinetic properties. Beneficial characteristics havebeen introduced into this series by increasing thelipophilicity. Minocycline and doxycycline, whichare semi-synthetic compounds, are the two bestexamples: the former is active against manytetracycline-resistant staphylococci while the latterallows a once-a-day regimen due to its prolongedhalf-life.

In the past 5 years there has been a remarkablechange in the properties and chemotherapeuticpotential of the nalidixic acid group. The firstclinically useful compound in this totally syntheticseries, nalidixic acid, has been available since theearly 1960s. This was used purely as a urinary anti-infective. It has a spectrum limited to the commonGram-negative pathogens, and has to be given in a

6-hourly schedule in order to prevent resistanceemerging. Early attempts to improve on theseproperties met with only modest success -

cinoxacin, rosoxacin, piromidic acid and pipemidicacid have been marketed but have found only asmall therapeutic niche. However, a later phase ofchemical modification in this series has had a

spectacular result, as the fluorinated 4-quinolonefamily has emerged.The synthesis of these compounds probably

represents the most exciting development in anti-microbial chemotherapy since the isolation of6APA. Already several have been marketed such as

norfloxacin, ciprofloxacin, enoxacin, ofloxacin andpefloxacin and many more are under investigation.

This group of compounds is very intrinsically activeagainst a wide range of Gram-negative rods(including Ps. aeruginosa), and have substantialactivity against Gram-positive species as well. Longhalf-lives mean that a twice daily schedule ispracticable. Their unique mode of action - onDNA gyrase - suggests that resistance emergencemay not be common. The quinolones are concen-trated in the tissues due to their lipophilicity, andso the only moderate blood levels that are attainedmust not be taken to imply (as would be the casefor some other antibiotics) poor clinicalperformance. Indeed, clinical trials in many indi-cations show excellent results. It is too soon to bedogmatic as to the place of the quinolones in thetreatment of infectious diseases in general, but thefirst indications are certainly favourable, and havedone nothing to reduce the initial promise of thesedrugs. A fuller assessment can be made in a fewyears time. Meanwhile, we consider it importantthat the quinolones be reserved for serious infec-tions or the treatment of multi-resistant strains. It isto be hoped that use in general practice would beseverely curtailed - otherwise the problem of resis-tance may emerge.

Drugs which have not been modified

Included here are substances such as vancomycinand fusidic acid. These are interesting in that theappearance of new problems, especially the MRSA,have resulted in these drugs being used much morecommonly. In the glycopeptide field other drugssimilar to vancomycin but apparently with betterpharmacokinetic properties and less toxicity haveappeared. Of these the one where investigation ismost advanced is teicoplanin. This should be avail-able for general use by the summer of 1988.

Examples of changes in the morbidity and mortalityof infectious diseases unrelated to antimicrobialchemotherapyWhile not underestimating the impact of anti-microbial agents on the control of infectiousdiseases remarkable changes which are unrelated tothese substances have been observed. The increasingavailability of international statistics makesaccurate analysis of these changes possible.

Early in the 20th century tuberculosis was auniversal scourge well deserving the epithetbestowed upon it by Bunyan of the 'Captain of theMen of Death'. However, a remarkable decline inthe incidence of tuberculosis had already beennoted before the discovery of anti-tuberculosis

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from

THE CHANGING FACE OF CHEMOTHERAPY 557

drugs and, for that matter, before the discovery ofthe causal organism which was identified by Kochin 1882.

Subsequently important factors in the control ofthe disease were improvement in housing, nutritionand sanitation together with the knowledge thattuberculosis was a transmissible disease.However, antibiotics played an important part in

the treatment of tuberculosis. The advent of anti-tuberculosis drugs, notably streptomycin (1944),para-amino salicyclic acid (1946) and isoniazid(1952) revolutionized the treatment of all forms oftuberculosis. From the patient's point of view theoutlook of uncertainty, or of despair, was replacedby one of confident optimism. Long periods of bedrest are now unnecessary and although appropriatesurgical treatment is of great importance in certaincases, it is true to say that efficient chemotherapyleaves only the occasional case where it is required.The discovery of new anti-tuberculosis agents

especially those which are bactericidal and pene-trate cells have resulted in advances in treatment.These include the reduction of duration of treat-ment (which initially was two years and is nowthree months), and oral treatment allowing therapyat home with obvious cost benefits.Problems still remain, for example non-

compliance by patients may lead to the emergenceof resistant organisms. Also lack of awareness ofthe danger of tuberculosis led to a reduction inBCG vaccination of 'at risk' subjects andconsequently of herd immunity.

Tuberculosis is an example of a serious infectionwhere antimicrobial drugs only play a part in itscontrol. However, a more striking example is theglobal eradication of smallpox which has beenachieved without the help of specific antiviralagents. In contrast, an example of persistence of atransmissible disease, in spite of the availability ofetfective chemotherapeutic agents, is malaria.A change in virulence of an organism (as

opposed to host resistance) may also cause adecrease in morbidity and mortality in infectiousdisease. A remarkable example of this is rheumaticfever: this is well known to be associated withstreptococcal sore throat which is therefore anessential precursor of the disease. However,evidence from mortality rates and from necroscopystudies on children leaves little doubt that in theeconomically developed countries of the temperatezone, rheumatic fever in its more serious and easilyrecognized forms has decreased progressively inincidence since at least the beginning of thetwentieth century. This process of decline under-went acceleration after 1930 and has since con-tinued, but foci of higher incidence persist incertain urban areas of the United States and

Britain. Thus, as recently as 1942 recruiting to theAmerican Air Force was almost halted because ofepidemic streptococcal sore throat followed by anincidence of rheumatic fever which was as high as 4per cent. Only a decade later a number of reportsconcerning patients with streptococcal sore throatin the USA and in this country have shown anincidence of rheumatic fever of 0.1 per cent or less.The reason for a change to the benign nature ofsuch infections remains unclear. However, circum-stantial evidence strongly indicates that a decreasein the virulence of Group A f-haemolytic strepto-cocci for humans is an important factor.The fall in incidence of rheumatic fever following

infection by fl-haemolytic streptococcus Group Ahas not been brought about by the availability ofantibiotics. Although it was shown that by treatinga streptococcal sore throat with a 10-day course ofbenzylpenicillin, which eliminates the streptococci,rheumatic fever was prevented, it is now knownthat even in the absence of antibiotic treatmentrheumatic fever rarely complicates streptococcalpharyngitis. This is an example of change in thevirulence of an organism which is unrelated to theuse of antibiotics.

Future prospects

The last 50 years have seen a rapidly evolvingpicture in terms of available antibiotics, and manygroups have experienced rises and falls. Examplesof 'antibiotics of today' are fl-lactams (cephalo-sporins, cephamycins, monobactams and ,f-lactamase inhibitors), aminoglycosides, quinolonesand rifampicin. 'Antibiotics of yesterday' mustinclude streptomycin and sulphonamides. Of greatinterest is the question: which will be the 'anti-biotics of tomorrow'? While one can only speculateon this topic, there are several directions which canbe seen as distinct possibilities for the future evo-lutionary pattern of antibiotic development. Theseinclude:

(a) The revival of 'old' antibiotics in response tonew needs. We see definite prospects for therenaissance of a compound related to novobiocin inresponse to the appearance of MRSA. The nitro-furans also have the advantage of very rare resis-tance acquisition. Nitrofurantoin is still valuable inthe treatment of urinary infections.

(b) Renewed search for new natural products. Ithad been thought that resources capable of pro-ducing new natural antibiotics were exhausted,following the long phase of no further discoveriesafter gentamicin. However, the finding of suchcompounds as the monobactams and teicoplaninshowed that this view was erroneous. Large poten-

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from

558 W. BRUMFITT & J.M.T. HAMILTON-MILLER

tial sources of new natural products remain rela-tively untapped - e.g., the Plant Kingdom.

(c) Production of new chemical entities. To avoidcross-resistance, it would seem logical to concen-trate on molecules whose chemical structure differscompletely from those of existing antibiotics ratherthan to continue to modify existing molecules.Examples of this approach already exist, forinstance in paldimycin and the Du Pont series ofoxazolidinones.

(d) The production of antibiotics with highly

specific ultra-narrow antibacterial spectra should bedeveloped to complement the existing very broadspectrum compounds. The new class of antibioticswould be used once the invading pathogen hadbeen accurately identified. There are alreadyexamples of narrow spectrum compounds in use,notably cefsulodin against Ps. aeruginosa.

It will be an instructive and perhaps humblingexercise to re-read this article in the year 2000, andto see how many predictions have proved true.

Bibliography

Brumfitt, W. & Hamilton-Miller, J.M.T. Principles andpractice of antimicrobial chemotherapy. In: Speight,T.M. (ed) Avery's Drug Treatment, 3rd Edition. AdisPress, Auckland, 1987, pp 1207-1235.

Brumfitt, W., O'Grady, F. & Slater, J.D.H. Benignstreptococcal sore throat. Lancet 1959, ii: 419-423.

Chopra, I., How, T.G.B., Linton, A.H., Linton, K.B.,Richmond, M.H. & Speller, D.C.E. The tetracyclines:prospects at the beginning of the 1980s. J AntimicrobChemother 1981, 8: 5-21.

Culpepper, N. Complete Herbal and English Physician.Gleave & Son, Manchester, 1826.

D'Arcy, P.F. The comparative safety of therapies forurinary tract infection, with special reference to nitro-furantoin. In: Schroder, F.H. (ed) Recent Advances inthe Treatment of Urinary Tract Infections. Royal Societyof Medicine, London 1985, ICSS 97, pp 45-59.

Finland, M., Kass, E.H. & Platt, R. (eds) Trimethoprimsulfamethoxazole revisited. Rev Inf Dis 1982, 4: 196-618.

Good Antimicrobial Prescribing. The Lancet, London,1982.

Kucers, A. & Bennett, N.McK. The Use of Antibiotics,4th Edition. Heinemann Medical, London, 1987.

MacFarlane, G. Alexander Fleming - the Man and theMyth. Chatto & Windus, London, 1984.

Phillips, I. & Williams, J.D. (eds) Macrolides-lincosamides-streptogramins. J Antimicrob Chemother1985, 16 (suppl A): 1-226.

Rolinson, G.N. 6-APA and the development of the f-lactam antibiotics. J Antimicrob Chemother 1979, 5: 7-14.

Sande, M.A. (ed) The use of rifampin in the treatment ofnon-tuberculous infections. Rev Inf Dis 1983, 5 (suppl3): S399-S632.

Sheehan, J.C. The Enchanted Ring. MIT Press,Cambridge, Mass, 1982.

Symposium on Amikacin. Am J Med 1977, 62: 863-966.Waksman, S.A. My Life with the Microbes. ScientificBook Club, London, 1958.

Williams, J.D. (ed) The cephalosporin antibiotics. Drugs1987, 34 (suppl 2): 1-258.

Wise, R.I. & Kory M. (eds) Reassessments of vancomycin- a potentially useful antibiotic. Rev Inf Dis 1981, 3:S199-S300.

copyright. on A

pril 22, 2021 by guest. Protected by

http://pmj.bm

j.com/

Postgrad M

ed J: first published as 10.1136/pgmj.64.753.552 on 1 July 1988. D

ownloaded from