Future prospects in antiviral therapy E.H.H. Wiltink
Introduction At present, there is very little t h a t can be done to avoid viral infections, especially when host defences are diminished. Prophylactic measures include prevention of the spread of the virus and the e n h a n c e m e n t of the efficacy of n a t u r a l defence mechanisms. Vaccination and passive immunization with immunoglobulins (antibodies) have been used with success. Interferons have i m m u n o m o d u l a t i n g properties, are part of the natural host defence and have shown to be useful in some antiviral drug regimens. Nevertheless, when a virus has reached its target, symptomatic t r e a t m e n t is the only recourse. However, the use of antiviral drugs creates other problems, the first of which is selectivity. Because a virus uses the biochemical mechanisms of the host cell to produce new viral proteins and genetic material, we must have exact knowledge of the small differences between the target (the virus) and the host (the h u m a n cell). Thus, virus and host cell are intimately connected. Secondly, drug resistance is an increas~ ing problem which requires different approaches in antiviral therapy. Thirdly, the problem of eradicating latent virus must be resolved, since antiviral drugs are not effective against the latent virus. As a result, only a limited n u m b e r of antiviral drugs are available. New analytical techniques have elucidated in more detail the process of a viral invasion and multiplication. This provides an opportunity to use new targets and new approaches in antiviral therapy. Different types of active agents are under investigation in the laboratory. Much work has been done in the field of i m m u n o m o d u l a t i n g drugs, especially in the fight against the h u m a n immunodeficiency virus (HIV). Indeed, most current efforts to develop antiviral drugs are directed against HIV. However, effective and safe new agents are reaching the clinic only very slowly. Thus, the search continues for alternative ways to treat viral infections, for example, combination therapy with different types of antiviral
drugs or with other agents. Sometimes drugs licensed for the t r e a t m e n t of one virus appear to be useful for another. It may also be possible to develop oral formulations of parenteral antiviral drugs, which would greatly facilitate prolonged treatment. For a review of the antiviral drugs currently being used in medical practice, the reader is referred to reference 1. In this article the most promising new agents and approaches against antiviral infections will be discussed, and the replicative cycle of a virus will be described in more detail.
Virus replicative cycle The following steps in viral infection and multiplication can be distinguished and could be regarded as targets for chemotherapeutic attack (Figure 1). - A t t a c h m e n t : the virus binds to the surface membrane of the cell. Specific proteins and cellular receptors are required to make the binding possible. - Penetration: the viral envelope is fused with the cellular membrane and the capsid directly penetrates into the cell. - Uncoating: the protein coat is shed and the genetic material is released into the cell. - Multiplication: RNA and DNA viruses differ and some have multiplication cycles t h a t are quite complicated. In simple terms the following sequence of events takes place, although not for every virus. In RNA viruses, viral RNA is transcribed by virus-coded RNA-polymerase to messenger RNA (mRNA). This m R N A produces new viral genetic material and proteins by translation. Alternatively, RNA is modulated by viral reverse transcriptase to viral DNA, which is integrated by the enzyme integrase in the host cell DNA (the so-called provirus). When the cell is activated m R N A is produced, which codes for viral proteins and viral genetic material. In DNA viruses, in the early transcription period, m R N A is formed
Wiltink EHH. Future prospects in antiviral therapy. Pharm Weekbl [Sci] 1992;14(4A):268-74.
Abstract
Keywords Antiviral drugs Future prospects E.H.H. Wiltink: Academic
Medical Centre, Department of Pharmacy, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands.
268
Two important stumbling blocks to the development of effective and nontoxic antiviral drugs are the intracellular localization of the virus and the fact that a virus uses host cell functions to multiply. Therefore, new antiviral drugs must act on a virus-specific function. Most currently available useful antiviral drugs are the result of compound screening of large numbers of possible agents. Advances in our understanding of the molecular biology and biochemistry of the viral multiplication cycle and new laboratory techniques for determining the molecular sites of action have now made it possible to develop and screen new antiviral drugs in a more purposeful manner. Another possible option in antiviral therapy is combination therapy using drugs that enhance the therapeutic effect or diminish side-effects. The most promising new antiviral drugs are discussed according to the different steps they affect in the viral multiplication process. Combination therapy is also reviewed. Accepted June 1992.
Pharmaceutisch Weekblad Scientific edition
14(4A) 1992
protein gp120, which is responsible for the binding of HIV to the CD4 molecule on the target cells (helper T lymphocytes, macrophages and monocytes). This finding has led to the production of soluble forms of CD4 (recombinant soluble 1. attachment CD4, or rsCD4) by genetic engineering. The idea 2, penetration is very simple. Administration of an excess of soluble CD4 to a patient infected with HIV will bind 3. uncoating the virus extracellularly before it binds to the i RNA 4. multiplicationcell surface. Thus, the virus will be eliminated rVenrsSceriptase before it can infect another cell. DNA Although no toxicity has been seen with solI PmRNA uble CD4 in phase I studies, the clinical results have not been conclusive and more data about enzymes long-term toxicity and efficacy are needed [4]. rsCD4 has a short serum half-life of about 45 min, which is considered to be a disadvantage. ~ provirus RNA potymerase Attachment of the rsCD4 molecule to immunoglobulin (rsCD4-Ig) may provide a variant of i i I I mRNA rsCD4 with a longer serum half-life. In vitro rsCD4-Ig has anti-HIV activity against T cells and macrophages [5]. 1 IRNA prot~ I prole~ It has recently become clear from in vitro exI~I I mRNA periments with rsCD4 and rsCD4-Ig in combi5.viral protein synthesis nation with HIV-1 (strain III B) t h a t CD4 acts in eins two ways. It binds the virus, but it can also irreversibly strip the envelope glycoprotein gp120 from the virus. However, primary isolates of in6. assembly tact sCD4-resistant virions bind sCD4-Ig with greatly reduced affinity, but retain their gp120 relatively well. In addition, very high concentrations of this kind of receptor blocker are required in vivo. These properties may limit the ef7. release ficacy of sCD4 and its derivatives [6 7]. As of this writing, clinical trials with sCD4-IgG have been DNA- VIRUS RNA - VIRUS stopped because of lack of efficacy. Figure 1 from the viral DNA, which will be translated Viral infection to nonstructural proteins such as enzymes. at host-cell level After viral DNA synthesis in the nucleus of the host cell a second transcription (late transcription) and translation (late translation) period takes place in which structural proteins are formed. - Protein synthesis: the proteins of the virus are O_ coded by mRNA. - Virus particle assembly: viral proteins and the viral genome are joined together to form a new virion. - R e l e a s e : newly formed virus leaves the host cell. In all these steps of the multiplication cycle enzymes play an important role. Interference with virus multiplication can be achieved by inhibition 0S03~ of these enzymes [2 3]. New antiviral drugs will be described below according to the multipli- F i g u r e 2 Part of dextran sulfate cation cycle of the virus.
I
',/
\'
\
~
~Z///////////~A
9
/\
~////////f//////~
-CH2
|
New
antiviral
drugs
Attachment
The first step of viral invasion of the cell is att a c h m e n t to the target binding site of the cell membrane. Binding sites have been discovered for the Epstein-Barr virus (EBV), HIV, and the influenza virus. The identification of such binding sites makes it possible to develop drugs to block the interaction and prevent viral infection. Soluble CD4. HIV contains an envelope glyco14(4A) 1992
Pharmaceutisch Weekblad Scientific edition
/
Sulfated polysaccharides. Dextran sulfate (molecular weight 7,000 to 8,000) is another example of a compound that blocks HIV replication in vitro (Figure 2). It is assumed t h a t sulfated oligosaccharides prevent virus adsorption to the host cell. However, in a study with orally administered dextran sulfate the clinical efficacy was disappointing [8]. It is likely t h a t dextran sulfate is not absorbed by this route of administration. Dextran sulfate, like heparin, has anticoagulant properties and may cause coagulation problems
269
in high dosages. The results of another study have suggested t h a t dextran sulfate inhibits the in vivo replication of influenza virus by inhibiting the virus-induced fusion process [9]. Penetration and fusion There are different mechanisms by which a virus may penetrate the host cell. Nonenveloped viruses may enter the cell by direct penetration, whereas enveloped viruses seem to depend on a fusion process between the viral envelope and the cell membrane. Although we are slowly coming to understand the mechanism of virus uptake in the cell, drugs t h a t interfere with this step of viral multiplication are not expected in the very near future. Uncoating When a virus enters the cell the genome must first be disassembled from its protein capsid. This process is called uncoating. The naked viral genome is t h e n available for further multiplication. A n u m b e r of compounds known for m a n y years to have antiviral properties are now understood to interfere with uncoating. For example, arildone, which has been recognized for over a decade as a potential broad-spectrum antiviral agent, acts at the level of uncoating [10]. The uncoating of picorna viruses could be prevented by a compound like WIN 51711. This may lead to a series of agents with antiviral properties against h u m a n rhinovirus 14 [11] and possibly against HIV [12]. Such currently available antiviral drugs as amantadine and related compounds are also likely to interact with the uncoating of the virus.
NH
HOCH2
H
H
ddC Figure 3 Structure of dideoxycytidine 270
Multiplication As noted above, viral multiplication involves m a n y complex processes t h a t are not universal to all viruses. This discussion will be limited to those multiplication processes for which antiviral agents are already being studied in (pre-) clinical investigations. Drugs that affect reverse transcriptase. After uncoating, the actual multiplication takes place. In the case of HIV, an RNA virus, the next step is transcription of viral RNA to DNA, catalyzed by reverse transcriptase. This enzyme offers an important target for influencing the multiplication cycle, because reverse transcriptase is essential and unique for this family of viruses. Zidovudine (AZT) is the first licenced example of a drug t h a t acts at this level. Other dideoxynucleosides, such as 2',3'-dideoxycytidine (ddC) (Figure 3) and 2',3'-dideoxyinosine (ddI) are currently under clinical investigation. These agents all act in the same way. After entering the cell they are phosphorytated to their respective 5'triphosphates by cellular enzymes (thymidine kinase and thymidylate kinases). These triphosphates inhibit viral multiplication by two mechanisms, chain termination and competitive inhibition of cellular nucleoside-5'-triphosphates. When a dideoxynucleoside-5'-triphosphate is added to the end of a growing chain of viral DNA, further elongation is blocked because of the 3'-modification. Since viral reverse trans-
criptase is m u c h m o r e sensitive to these triphosphates t h a n DNA polymerase of the host cell, the rate of viral DNA synthesis is inhibited. This greater sensitivity explains the selective antiviral activity of these drugs [13]. Phase I studies are now being performed with ddC in different dose regimens and in combination therapy with zidovudine, ddC is 70% to 80% absorbed from the gut. It crosses the b l o o d - b r a i n barrier to a much lesser extent t h a n does zidovudine. The toxic effects of ddC are doserelated and include cutaneous eruptions, fever, mouth sores, thrombocytopenia and neutropenia [13 14]. Peripheral neuropathy, observed at all doses, is the major dose-limiting toxicity. The incidence of this effect has been reported to range from 2% to 50% at dosages between 0.005 and 0.09 mg. kg- 1. d-l) [13 14]. ddC given every 8 h in a dose of 0.01 mg/kg seems to be a superior regimen. Since the toxicity profile of ddC differs from t h a t of zidovudine, it is suitable for use in combination therapy with zidovudine at lower dosage (see below).
NH2
O
HN,,'~N~
HOCH2
H
HOCH 2
H
H
H
Figure 4 Structure of dideoxyadenosine (left) and dideoxyinosin (right) 2',3'-Dideoxyadenosine (ddA) was under clinical investigation, but trials were halted because of the formation of carcinogenic metabolites in the stomach, ddI can be regarded as a prodrug of ddA (Figure 4). ddI is metabolized in h u m a n cells to its pharmacologically active moiety dideoxyadenosine triphosphate. In vitro ddI has significant antiviral activity and is less toxic t h a n zidovudine, ddI is acid labile and must be administered in combination with antacids. The bioavailability of ddI administered together with antacids is about 40% and oral doses of 3.2 mg. kg 1. d 1 have shown favourable effects. The toxicity profile of ddI includes painful peripheral neuropathy and pancreatitis as the doselimiting adverse effects. Since these effects differ from those of zidovudine, combination therapy may be possible [15]. 2',3'-Didehydro-2'-3'-dideoxythymidine (2',3'dideoxythymididene, D4T) (Figure 5) shows a great structural resemblance to zidovudine. It has potent anti-HIV activity in vitro, similar to t h a t of zidovudine. Its phosphorylation, however,
Pharmaceutisch Weekblad Scientific edition
14(4A) 1992
0
is totally different and in vitro it is less toxic to zidovudine [16]. Only limited clinical data on D4T are currently available. Another example of a drug that interferes with the viral multiplication cycle is (R,S)-9-[4-hydroxy-2-(hydroxymethyl)butyl]guanine[(_+)2HMHBG] (Figure 6). It has a greater affinity for viral thymidine kinase than does aciclovir and shows a more rapid phosphorylation. The drug is more active against varicella zoster virus than aciclovir. Unfortunately, the bioavailability of orally administered HM-HBG is only about 10%. Attempts have been made to improve the oral absorption by esterification. The esters of (• HBG behave as prodrugs [2 17].
HN~, cH3 bone marrow than
HOCHH~H HH D4T Figure 5 Structure of dideoxythyrnididene
l trigger region
OH
CH2CHCH2CH2OH I
CH2OH (_+) 2HM Figure 6 Structure of % 2 HM-HBG
- HBG
l
In addition to these nucleoside inhibitors of reverse transcriptase, compounds with a different structural formula are under investigation. The nonnucleoside reverse transcriptase inhibitors are in phase I studies, but no clinical data are available yet. Protein synthesis mRNA is stimulated by some trigger to produce viral proteins. Prevention of mRNA translation may offer an opportunity to prevent protein production. Even at an earlier stage mRNA can be deformed. An example of a drug believed to act at this stage is ampligen, which was found several years ago to inhibit the multiplication of HIV in different samples of lymphoid cells. Ampligen synergistically enhances the activity of zidovudine. Ampligen is a so-called mismatched doublestranded RNA (dsRNA) (Figure 7). dsRNA stimulates the production of various kinds of interferons and activates the intracellular enzymes thought to be responsible for natural antiviral activity. The activation of these enzymes inhibits the multiplication of various human viruses by promoting the cleavage of viral RNA (including mRNA) or the interruption of viral protein synthesis. However, in phase I anticancer studies, dsRNAs have shown severe toxicity. In contrast, dsRNAs with periodically mispaired regions, such as ampligen, retain their lymphokine inductive potential, but have very limited toxicity. Ampligen has been investigated in 10 patients with AIDS or AIDS-related complex. Most of
14(4A) 1992
mismatched region
Pharmaceutisch Weekblad Scientific edition
ds RNA Figure 7 Structure of ampligen
these patients improved over 12 to 18 weeks, but the number of patients was too small to draw definitive conclusions [3 18]. In the laboratory considerable success has been achieved with so-called antisense inhibitors for the control of mRNA expression. Agents of a sequence complementary to viral RNA can be used to inhibit virus multiplication. HIV, herpes zoster virus, influenza virus, and vesicular stomatitis virus are affected in vitro by these agents. These chemical substances include oligodeoxyribonucleotides (ODN). Several modified ODNs have been developed, but the problem of transport of these substances into the virusinfected host cell has not yet been resolved [2]. During the translation process of mRNA large molecular weight precursors of viral proteins are synthesized. These large proteins must be modified to mature functional forms. Modifications may involve enzymatically catalyzed cleavage or other changes. It is recognized that, to reproduce itself, HIV needs a crucial enzyme, the HIV pro271
tease. This enzyme breaks two large viral poly- by an increasing number of opportunistic infecproteins into smaller units. Protease inhibitors tions. An explanation for this phenomenon was sought in the ability of DTC to scavenge oxygen are now under investigation in vitro. Proteins may be acylated, glycosylated, sulf- radicals, which are needed to destroy foreign ated or phosphorylated by enzymes of cellular or microbes. Because of these findings, studies with viral origin. Theoretically, virally encoded en- DTC in patients infected with HIV have stopped zymes are unique targets for selective antiviral [unpublished data]. agents [2]. In building the viral envelope of several RNA and DNA viruses parts of the host cell N e w a p p r o a c h e s w i t h current antiviral membranes and virus-specific g]ycoproteins are drugs Since the development of new antiviral drugs combined. By glycosylation, oligosaccharides and polypeptide chains are coupled to glycopro- is time-consuming, it is useful to consider modifireins. This glycosylation could be inhibited by cations of the dosage regimens of currently availagents such as castanospermine or deoxynojiri- able drugs. A few examples are described here. The recommended dose of zidovudine in mycin (DNJ). As a result of the failure to produce glycoprotein, defective virions are formed which patients with advanced HIV infection is CHSCHSCHSCH3 1,500 rag/day. Although effective, this dose is asare incapable of infecting other cells [2]. HIV envelope glycoproteins are heavily glyco- sociated with severe toxicity. The dose-limiting sylated. As mentioned above, the envelope glyco- toxic reaction is neutropenia. The serum half-life HO'" y "'OH protein gp120 is responsible for binding to the of zidovudine is 1 h, whereas the intracellular OH CD4 molecule. N-Butyl-deoxynojirimycin (N- half-life of the 5'-triphosphate form approaches butyl-DNJ) (Figure 8) is a glycosidase inhibitor 3 h. This means that a lower dose of zidovudine Figure 8 Structure of Nwhich may hinder formation of m a t u r e gp120 could be effective while avoiding serious sidebutyl-deoxyand alter its ability to bind to CD4 receptors. effects. nojirimicin A study of 524 patients was u n d e r t a k e n to deWhen HIV-producing cells are incubated and mixed with the CD4+ cell line in the presence of termine whether a lower dose (200 mg orally a glucosidase inhibitor binding between gp120 every 4 h for 4 weeks followed by 100 mg every 4 h thereafter) was as effective as the standard and its CD4 receptors is blocked. N-butyl-DNJ gradually decreases the propor- regimen (250 mg t a k e n orally every 4 h). This retion of infected cells, eventually leading to elimi- duced dose of zidovudine proved to be at least as effective as the standard dose and was less toxic nation of HIV from culture [19]. [22]. This result has been more or less confirmed in a study with low-dose (300 mg/day), mediumAssembly and release At the end of the cycle, newly created viral ma- dose (600 mg/day), and high-dose (1,500 rag/day) terial is joined together to a new virion and zidovudine, either with or without acyclovir leaves the host cell. Several agents may interfere (4,800 g/day). The 300 mg dose t u r n e d to be at with this process. Interferons (especially inter- least as effective as all the higher doses in the feron-alpha) have been found to inhibit HIV mul- short term. The concomitant use of aciclovir did tiplication, and it is believed t h a t they act, at not enhance the antiretroviral effects of zidovudine. least partially, at this step [4]. Unfortunately, the n u m b e r of patients in this study was too small to be conclusive about the Immunomodulation S H5C2\ If Considerable work has been done to improve use of lower doses [23]. In addition, studies conN--C--S--NA the immunological status of patients with HIV ducted over a longer period of time are necessary. H5C2/ infection. Dithiocarb (diethyldithiocarbamate, Nevertheless, these studies indicate t h a t lower DTC, imuthiol) (Figure 9) may restore immune doses of zidovudine can be used. Figure 9 Structure of diFoscarnet has been licensed for the t r e a t m e n t function by modulating T-cell differentiation thiocarb into m a t u r e and immune competent effector of cytomegalovirus (CMV) infection, but prelimcells. Moreover, DTC acts as a potential antiviral inary data have indicated that, when given intraagent by inhibiting a n u m b e r of copper-contain- venously, the drug has antiretroviral activity as well [3]. Combined use of foscarnet and zidovuing enzymes. DTC has been given orally in a dose of dine appears to yield synergistic effects. Some currently used antiviral drugs, such as 10 mg/kg body-weight and intravenously in a dose of 5 mg/kg body-weight once a week. In a pi- ganciclovir and foscarnet, are available only in lot study 8 of 11 patients receiving dithiocarb injectable formulations and the development of orally experienced a significant decrease in such oral formulations would be highly advantageous constitutional symptoms as persistent diarrhoea, for patient treatment. An analogous situation fever and weight loss. At week 16 the CD4+ cell exists with aciclovir, which is only 15% to 20% counts had increased in these patients [20]. How- bioavailable after oral administration. Efforts ever, because of the small n u m b e r of evaluable have been made to develop an analogue, desciclopatients this study was not conclusive. A second vir, with better absorption properties [1]. study concluded t h a t DTC significantly delayed progression to AIDS. Although there was no evi- Combination therapy The disadvantages associated with current andence of a direct antiretroviral effect, there was also no evidence of an increased virus load after tiviral drugs represent a rationale for the use of treatment. No severe side-effects were observed combination therapy. The existence of zidovudine-resistant strains of HIV has been described [21]. In one study the use of DTC was accompanied by Larder et al. in patients undergoing long-term
~ . CHSOH
272
Pharmaceutisch Weekblad Scientific edition
14(4A) 1992
zidovudine t h e r a p y [24]. Zidovudine also causes significant toxicity, notably bone-marrow depression, which necessitates dosage reduction or discontinuation of the drug. Combination t h e r a p y may be synergistic and thus more powerful in suppressing the virus. The toxicity of individual antiviral drugs can be reduced by the use of lower doses of each. The chance of development of resistance to the drugs will also be decreased. An example of such an approach is the use of combined t h e r a p y with zidovudine and ddC in phase I trials. Both drugs are given in an alternating dosage scheme or concurrently in lower doses [22 23]. The goal of these trials is to find an effective antiviral regimen with tolerable toxicity [11]. Haematopoietic growth factors, also called colony-stimulating factors (CSFs), regulate the proliferation and differentiation of haematopoietic progenitor cells and m a t u r e blood cells. A synergistic activity of granulocyte macrophage colony-stimulating factor (GM-CSF) and zidovudine has been observed in vitro [25]. The rationale for using GM-CSF is twofold. Firstly, GMCSF and granulocyte colony-stimulating factor (G-CSF) seem to exert antiretroviral activity by enhancing neutrophil cytotoxicity towards HIVinfected cells [26]. Secondly, GM-CSF (and also erythropoietin) can counteract AIDS-induced or drug-induced bone-marrow suppresion [4]. Neutropenia is the most common side-effect associated with ganciclovir, which is used in the treatment of CMV infections. The combination of GM-CSF and ganciclovir decreases the occurence of neutropenia [27]. Combination therapy with haematopoietic growth factors is currently being under investigation. Ganciclovir in combination with foscarnet has been reported to show increased efficacy against herpes simplex virus type 2 and cytomegalovirus in mice [28]. Although this synergistic effect must still be studied in humans, it may be a means of providing more effective treatment. These examples demonstrate t h a t combination t h e r a p y is a rational way of investigating the t r e a t m e n t of antiviral infections.
Conclusion At this moment an increasing n u m b e r of potential antiviral agents are under investigation in the laboratory. Since our knowledge of viral infection and multiplication is also increasing, we are now able to define more targets and different approaches in our fight against viral infections. Of course, by far the most work is being done on anti-HIV drugs. The reverse transcriptase enzyme is essential and unique for the multiplication ofretroviruses. Nucleoside and nonnucleoside inhibitors of reverse transcriptase are now in clinical trials. Other targets will also be explored. A t t a c h m e n t of HIV to host cells can be prevented by soluble CD4, while sulfated polysaccharides have similar properties in HIV and influenza virus. New antiviral drugs must be selective and this is an important problem. Many agents t h a t are effective in vitro can not be used in vivo because
14(4A) 1992
Pharmaceutisch Weekblad Scientific edition
of toxicity or lack of selectivity. A drug t h a t is active in vitro cannot always reach the target - the virus inside the host cell. Other approaches include enhancing the patient's immunological status with such agents as ampligen and dithiocarb. Interferons also influence the immune response and m a y act on viral assembly and release as well. The most promising approach to treating antiviral infections, however, appears to be the use of combination therapy, which offers m a n y advantages. Combinations of antiviral drugs may be synergistic and m a y also avoid the development of resistence. The administration of lower doses of each drug could reduce toxicity, thereby making therapy more acceptable to the patient. However, although investigation of effective combinations will lead us to new therapeutic regimens, the continuing search for new antiviral drugs will remain necessary.
Acknowledgement The author t h a n k s Eda M a k a t i t a and Marlies Hutters-Wulp for their fine secretarial assistance and Arno Vyth and Joep Lange for their critical review of the manuscript.
References 1 Wiltink EHH, Janknegt R. Antiviral drugs. Pharm Weekbl [Sci] 1991;13:58-69. 2 Sire IS. Virus replication: target functions and events for virus-specific inhibiters. In: Galasso GJ, Whitley RJ, Merigan TC, eds. Antiviral agents and viral diseases of man. 3rd ed. New York: Raven Press, 1990:147. 3 SandstrSm E. Antiviral therapy in human immunodeficiency virus infection. Drugs 1989;38:417-50. 4 Yarchoan R, Mitsuya H, Broder S. Clinical and basic advances in the anti-retroviral therapy of human immunodeficiency virus infection. Am J Med 1989;87: 191-200. 5 Capon DJ, Chamow SM, Mordenti J, Marsters SA, Gregory T, Mitsuya H, et al. Designing CD4 immunoadhesins for AIDS therapy. Nature 1989;337:525-531. 6 Moore JP, Weiss RA, Passive primate protection. Nature 1991;352:371-7. 7 Daar ES, Li XL, Moudgil T, He DD. High concentrations of recombinant soluble CD4 are required to neutralize primary human immunodeficiency virus type 1 isolates. Proc Natl Acad Sci USA 1990;87:6574-78. 8 Abrams DI, Kuno S, Wong R, Jeffords K, Nash M, Molaghan JB, et al. Oral dextran sulfate (UA 001) in the treatment of the acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. Ann Intern Med 1989;110:183-8. 9 Lfischer-Mattli M, Gliick R. Dextran sulfate inhibits the fusion of influenza virus with model membranes, and suppresses influenza virus replication in vivo. Antivir Res 1990;14:39-50. 10 Kim KS, Sapienza VJ, Carp RI. Antiviral activity of arildone on deoxyribo-nucleicacid and ribonucleic acid viruses. Antimicrob Agents Chemother 1980;18:27680. 11 Badger J, Minor I, Kremer MJ, Oliveira MA, Smith TJ, Griffith JP, et al. Structural analysis of a series of antiviral agents complexed with human rhinovirus 14. Proc Natl Acad Sci USA 1988;85:3304-8. 12 Rossmann MG. Antiviral agents targeted to interact with viral capsid proteins and a possible application to human immunodeficiency virus. Proc Natl Acad Sci USA 1988;85:4625-7. 13 Yarchoan R, Perno CF, Thomas RV, Klecker RW, Allain J-P, Wills RJ, et al. Phase I studies of 2',3'dideoxycytidine in severe human immunodeficiency virus infection as a single agent and alternating with zidovudine (AZT). Lancet 1988;1:76-81. 14 Merigan TC, Skowron G, Bozette SA, Richman D, Uttamchandani R, Fischl M, et al. Circulating p24 an-
273
15
16
17
18
19
20
21 22
274
tigen levels and responses to dideoxycytidine in human immunodeficiency virus (HIV) infections. Ann Intern Med 1989;110:189-94. Yarchoan R, Pluda JM, Thomas RV, Mitsuya H, Brouwers P, Wyvill KM, et al. Long term toxicity/activity profile of 2',3'-dideoxyinosine in AIDS or AIDSrelated complex. Lancet 1990;336:526-9. Ho H-T, Hitchcock MJM. Cellular pharmacology of 2',3'-dideoxy-2',3'-didehydrothymidine, a nucleoside analog active against humanimmunodeficiencyvirus. Antimicrob Agents Chemother 1989;33:844-9. Lake-Bakaar DM, Lindborg B, Datema R. Improvement of the absorption of oral (R,S)-9-[4-hydroxy-2-(hydroxymethyl)butyl]guanine, an anti-varicella-zoster virus drug, in rats and monkeys. Antimicrob Agents Chemother 1989;33:110-2. Carter WA, Strayer DR, Brodsky I, Lewin M, Pellegrino MG, Einck L, et al. Clinical, immunological and virological effects of ampligen, a mismatched double-stranded RNA, in patients with AIDS or AIDSrelated complex. Lancet 1987;1:1286-92. Karpas A, Fleet GWJ, Dwek RA, Petursson S, Namgoong SK, Ramsden NG, et al. Aminosugar derivatives as potential anti-human immunodeficiency virus agent. Proc Natl Acad Sci USA 1988;85:9229-33. Lang J-M, Touraine J-L, Trepo C. Randomised, doublebind, placebo controlled trial of dithiocarb sodium ("Imuthiol") in human immunodeficiency virus infec~ tion. Lancet 1988;2:702-6. Reisinger EC, Kern P, Ernst M, Bock P, Flad HD, Dietrich M, et al. Inhibition of HIV progression by dithiocarb. Lancet 1990;335:679-82. Fischl MA, Parker CB, Pettinelli C, Wulfsohn M, Hirsch MS, Collier AC, et al. A randomized controlled
23
24 25
26
27
28
trial of a reduced daily dose of zidovudine in patients with the acquired immunodeficiency syndrome. N Engl J Med 1990;323:1009-14. Collier AC, Bozzette S, Coombs RW, Causey DM, Schoenfeld DA, Spector SA, et al. A pilot study of lowdose zidovudine in human immunodeficiencyvirus infection. N Engl J Med 1990;323:1015-21. Larder BA, Darby G, Richman DD. HIV with reduced sensitivity to zidovudine (AZT) isolated during prolonged therapy. Science 1989;243:1731-4. Hammer SM, Gillis JM. Synergistic activity of granulocyte-macrophage colony-stimulatingfactor and 3'-azido-3~-deoxythymidine against human immunodeficiencyvirus in vitro. Antimicrob Agents Chemother 1987;31:1046-50. Baldwin GC, Fuller ND, Roberts RL, Ho DD, Golde DW. Granulocyte- and granulocyte-macrophage colony-stimulating factors enhance neutrophil cytotoxicity toward HIV-infected cells. Blood 1989;74:16737. Hardy WD, Spector S, Polsky B, Crumpacker C, Holland G, Freeman W, et al. Combined ganciclovir (GCV) and recombinant humangranuIocyte-macrophage-stimulating factor (GM-CSF) vs ganciclovir alone for cytomegalovirus (CMV) retinitis in AIDS, [Abstract]. Thirtieth Interscience Conference on Antimicrobial Agents and Chemotherapy, 1990; New Orleans, abstract 9. Freitas VR, Frase-Smith EB, Matthews TR. Increased efficacy of ganciclovir in combination with foscarnet against cytomegalovirus and herpes simplex virus type 2 in vitro and in vivo. Antiviral Res 1989;12:20512.
P h a r m a c e u t i s c h W e e k b l a d Scientific e d i t i o n
14(4A) 1992
Introduction At present, there is very little t h a t can be done to avoid viral infections, especially when host defences are diminished. Prophylactic measures include prevention of the spread of the virus and the e n h a n c e m e n t of the efficacy of n a t u r a l defence mechanisms. Vaccination and passive immunization with immunoglobulins (antibodies) have been used with success. Interferons have i m m u n o m o d u l a t i n g properties, are part of the natural host defence and have shown to be useful in some antiviral drug regimens. Nevertheless, when a virus has reached its target, symptomatic t r e a t m e n t is the only recourse. However, the use of antiviral drugs creates other problems, the first of which is selectivity. Because a virus uses the biochemical mechanisms of the host cell to produce new viral proteins and genetic material, we must have exact knowledge of the small differences between the target (the virus) and the host (the h u m a n cell). Thus, virus and host cell are intimately connected. Secondly, drug resistance is an increas~ ing problem which requires different approaches in antiviral therapy. Thirdly, the problem of eradicating latent virus must be resolved, since antiviral drugs are not effective against the latent virus. As a result, only a limited n u m b e r of antiviral drugs are available. New analytical techniques have elucidated in more detail the process of a viral invasion and multiplication. This provides an opportunity to use new targets and new approaches in antiviral therapy. Different types of active agents are under investigation in the laboratory. Much work has been done in the field of i m m u n o m o d u l a t i n g drugs, especially in the fight against the h u m a n immunodeficiency virus (HIV). Indeed, most current efforts to develop antiviral drugs are directed against HIV. However, effective and safe new agents are reaching the clinic only very slowly. Thus, the search continues for alternative ways to treat viral infections, for example, combination therapy with different types of antiviral
drugs or with other agents. Sometimes drugs licensed for the t r e a t m e n t of one virus appear to be useful for another. It may also be possible to develop oral formulations of parenteral antiviral drugs, which would greatly facilitate prolonged treatment. For a review of the antiviral drugs currently being used in medical practice, the reader is referred to reference 1. In this article the most promising new agents and approaches against antiviral infections will be discussed, and the replicative cycle of a virus will be described in more detail.
Virus replicative cycle The following steps in viral infection and multiplication can be distinguished and could be regarded as targets for chemotherapeutic attack (Figure 1). - A t t a c h m e n t : the virus binds to the surface membrane of the cell. Specific proteins and cellular receptors are required to make the binding possible. - Penetration: the viral envelope is fused with the cellular membrane and the capsid directly penetrates into the cell. - Uncoating: the protein coat is shed and the genetic material is released into the cell. - Multiplication: RNA and DNA viruses differ and some have multiplication cycles t h a t are quite complicated. In simple terms the following sequence of events takes place, although not for every virus. In RNA viruses, viral RNA is transcribed by virus-coded RNA-polymerase to messenger RNA (mRNA). This m R N A produces new viral genetic material and proteins by translation. Alternatively, RNA is modulated by viral reverse transcriptase to viral DNA, which is integrated by the enzyme integrase in the host cell DNA (the so-called provirus). When the cell is activated m R N A is produced, which codes for viral proteins and viral genetic material. In DNA viruses, in the early transcription period, m R N A is formed
Wiltink EHH. Future prospects in antiviral therapy. Pharm Weekbl [Sci] 1992;14(4A):268-74.
Abstract
Keywords Antiviral drugs Future prospects E.H.H. Wiltink: Academic
Medical Centre, Department of Pharmacy, Meibergdreef 9, 1105 AZ Amsterdam, the Netherlands.
268
Two important stumbling blocks to the development of effective and nontoxic antiviral drugs are the intracellular localization of the virus and the fact that a virus uses host cell functions to multiply. Therefore, new antiviral drugs must act on a virus-specific function. Most currently available useful antiviral drugs are the result of compound screening of large numbers of possible agents. Advances in our understanding of the molecular biology and biochemistry of the viral multiplication cycle and new laboratory techniques for determining the molecular sites of action have now made it possible to develop and screen new antiviral drugs in a more purposeful manner. Another possible option in antiviral therapy is combination therapy using drugs that enhance the therapeutic effect or diminish side-effects. The most promising new antiviral drugs are discussed according to the different steps they affect in the viral multiplication process. Combination therapy is also reviewed. Accepted June 1992.
Pharmaceutisch Weekblad Scientific edition
14(4A) 1992
protein gp120, which is responsible for the binding of HIV to the CD4 molecule on the target cells (helper T lymphocytes, macrophages and monocytes). This finding has led to the production of soluble forms of CD4 (recombinant soluble 1. attachment CD4, or rsCD4) by genetic engineering. The idea 2, penetration is very simple. Administration of an excess of soluble CD4 to a patient infected with HIV will bind 3. uncoating the virus extracellularly before it binds to the i RNA 4. multiplicationcell surface. Thus, the virus will be eliminated rVenrsSceriptase before it can infect another cell. DNA Although no toxicity has been seen with solI PmRNA uble CD4 in phase I studies, the clinical results have not been conclusive and more data about enzymes long-term toxicity and efficacy are needed [4]. rsCD4 has a short serum half-life of about 45 min, which is considered to be a disadvantage. ~ provirus RNA potymerase Attachment of the rsCD4 molecule to immunoglobulin (rsCD4-Ig) may provide a variant of i i I I mRNA rsCD4 with a longer serum half-life. In vitro rsCD4-Ig has anti-HIV activity against T cells and macrophages [5]. 1 IRNA prot~ I prole~ It has recently become clear from in vitro exI~I I mRNA periments with rsCD4 and rsCD4-Ig in combi5.viral protein synthesis nation with HIV-1 (strain III B) t h a t CD4 acts in eins two ways. It binds the virus, but it can also irreversibly strip the envelope glycoprotein gp120 from the virus. However, primary isolates of in6. assembly tact sCD4-resistant virions bind sCD4-Ig with greatly reduced affinity, but retain their gp120 relatively well. In addition, very high concentrations of this kind of receptor blocker are required in vivo. These properties may limit the ef7. release ficacy of sCD4 and its derivatives [6 7]. As of this writing, clinical trials with sCD4-IgG have been DNA- VIRUS RNA - VIRUS stopped because of lack of efficacy. Figure 1 from the viral DNA, which will be translated Viral infection to nonstructural proteins such as enzymes. at host-cell level After viral DNA synthesis in the nucleus of the host cell a second transcription (late transcription) and translation (late translation) period takes place in which structural proteins are formed. - Protein synthesis: the proteins of the virus are O_ coded by mRNA. - Virus particle assembly: viral proteins and the viral genome are joined together to form a new virion. - R e l e a s e : newly formed virus leaves the host cell. In all these steps of the multiplication cycle enzymes play an important role. Interference with virus multiplication can be achieved by inhibition 0S03~ of these enzymes [2 3]. New antiviral drugs will be described below according to the multipli- F i g u r e 2 Part of dextran sulfate cation cycle of the virus.
I
',/
\'
\
~
~Z///////////~A
9
/\
~////////f//////~
-CH2
|
New
antiviral
drugs
Attachment
The first step of viral invasion of the cell is att a c h m e n t to the target binding site of the cell membrane. Binding sites have been discovered for the Epstein-Barr virus (EBV), HIV, and the influenza virus. The identification of such binding sites makes it possible to develop drugs to block the interaction and prevent viral infection. Soluble CD4. HIV contains an envelope glyco14(4A) 1992
Pharmaceutisch Weekblad Scientific edition
/
Sulfated polysaccharides. Dextran sulfate (molecular weight 7,000 to 8,000) is another example of a compound that blocks HIV replication in vitro (Figure 2). It is assumed t h a t sulfated oligosaccharides prevent virus adsorption to the host cell. However, in a study with orally administered dextran sulfate the clinical efficacy was disappointing [8]. It is likely t h a t dextran sulfate is not absorbed by this route of administration. Dextran sulfate, like heparin, has anticoagulant properties and may cause coagulation problems
269
in high dosages. The results of another study have suggested t h a t dextran sulfate inhibits the in vivo replication of influenza virus by inhibiting the virus-induced fusion process [9]. Penetration and fusion There are different mechanisms by which a virus may penetrate the host cell. Nonenveloped viruses may enter the cell by direct penetration, whereas enveloped viruses seem to depend on a fusion process between the viral envelope and the cell membrane. Although we are slowly coming to understand the mechanism of virus uptake in the cell, drugs t h a t interfere with this step of viral multiplication are not expected in the very near future. Uncoating When a virus enters the cell the genome must first be disassembled from its protein capsid. This process is called uncoating. The naked viral genome is t h e n available for further multiplication. A n u m b e r of compounds known for m a n y years to have antiviral properties are now understood to interfere with uncoating. For example, arildone, which has been recognized for over a decade as a potential broad-spectrum antiviral agent, acts at the level of uncoating [10]. The uncoating of picorna viruses could be prevented by a compound like WIN 51711. This may lead to a series of agents with antiviral properties against h u m a n rhinovirus 14 [11] and possibly against HIV [12]. Such currently available antiviral drugs as amantadine and related compounds are also likely to interact with the uncoating of the virus.
NH
HOCH2
H
H
ddC Figure 3 Structure of dideoxycytidine 270
Multiplication As noted above, viral multiplication involves m a n y complex processes t h a t are not universal to all viruses. This discussion will be limited to those multiplication processes for which antiviral agents are already being studied in (pre-) clinical investigations. Drugs that affect reverse transcriptase. After uncoating, the actual multiplication takes place. In the case of HIV, an RNA virus, the next step is transcription of viral RNA to DNA, catalyzed by reverse transcriptase. This enzyme offers an important target for influencing the multiplication cycle, because reverse transcriptase is essential and unique for this family of viruses. Zidovudine (AZT) is the first licenced example of a drug t h a t acts at this level. Other dideoxynucleosides, such as 2',3'-dideoxycytidine (ddC) (Figure 3) and 2',3'-dideoxyinosine (ddI) are currently under clinical investigation. These agents all act in the same way. After entering the cell they are phosphorytated to their respective 5'triphosphates by cellular enzymes (thymidine kinase and thymidylate kinases). These triphosphates inhibit viral multiplication by two mechanisms, chain termination and competitive inhibition of cellular nucleoside-5'-triphosphates. When a dideoxynucleoside-5'-triphosphate is added to the end of a growing chain of viral DNA, further elongation is blocked because of the 3'-modification. Since viral reverse trans-
criptase is m u c h m o r e sensitive to these triphosphates t h a n DNA polymerase of the host cell, the rate of viral DNA synthesis is inhibited. This greater sensitivity explains the selective antiviral activity of these drugs [13]. Phase I studies are now being performed with ddC in different dose regimens and in combination therapy with zidovudine, ddC is 70% to 80% absorbed from the gut. It crosses the b l o o d - b r a i n barrier to a much lesser extent t h a n does zidovudine. The toxic effects of ddC are doserelated and include cutaneous eruptions, fever, mouth sores, thrombocytopenia and neutropenia [13 14]. Peripheral neuropathy, observed at all doses, is the major dose-limiting toxicity. The incidence of this effect has been reported to range from 2% to 50% at dosages between 0.005 and 0.09 mg. kg- 1. d-l) [13 14]. ddC given every 8 h in a dose of 0.01 mg/kg seems to be a superior regimen. Since the toxicity profile of ddC differs from t h a t of zidovudine, it is suitable for use in combination therapy with zidovudine at lower dosage (see below).
NH2
O
HN,,'~N~
HOCH2
H
HOCH 2
H
H
H
Figure 4 Structure of dideoxyadenosine (left) and dideoxyinosin (right) 2',3'-Dideoxyadenosine (ddA) was under clinical investigation, but trials were halted because of the formation of carcinogenic metabolites in the stomach, ddI can be regarded as a prodrug of ddA (Figure 4). ddI is metabolized in h u m a n cells to its pharmacologically active moiety dideoxyadenosine triphosphate. In vitro ddI has significant antiviral activity and is less toxic t h a n zidovudine, ddI is acid labile and must be administered in combination with antacids. The bioavailability of ddI administered together with antacids is about 40% and oral doses of 3.2 mg. kg 1. d 1 have shown favourable effects. The toxicity profile of ddI includes painful peripheral neuropathy and pancreatitis as the doselimiting adverse effects. Since these effects differ from those of zidovudine, combination therapy may be possible [15]. 2',3'-Didehydro-2'-3'-dideoxythymidine (2',3'dideoxythymididene, D4T) (Figure 5) shows a great structural resemblance to zidovudine. It has potent anti-HIV activity in vitro, similar to t h a t of zidovudine. Its phosphorylation, however,
Pharmaceutisch Weekblad Scientific edition
14(4A) 1992
0
is totally different and in vitro it is less toxic to zidovudine [16]. Only limited clinical data on D4T are currently available. Another example of a drug that interferes with the viral multiplication cycle is (R,S)-9-[4-hydroxy-2-(hydroxymethyl)butyl]guanine[(_+)2HMHBG] (Figure 6). It has a greater affinity for viral thymidine kinase than does aciclovir and shows a more rapid phosphorylation. The drug is more active against varicella zoster virus than aciclovir. Unfortunately, the bioavailability of orally administered HM-HBG is only about 10%. Attempts have been made to improve the oral absorption by esterification. The esters of (• HBG behave as prodrugs [2 17].
HN~, cH3 bone marrow than
HOCHH~H HH D4T Figure 5 Structure of dideoxythyrnididene
l trigger region
OH
CH2CHCH2CH2OH I
CH2OH (_+) 2HM Figure 6 Structure of % 2 HM-HBG
- HBG
l
In addition to these nucleoside inhibitors of reverse transcriptase, compounds with a different structural formula are under investigation. The nonnucleoside reverse transcriptase inhibitors are in phase I studies, but no clinical data are available yet. Protein synthesis mRNA is stimulated by some trigger to produce viral proteins. Prevention of mRNA translation may offer an opportunity to prevent protein production. Even at an earlier stage mRNA can be deformed. An example of a drug believed to act at this stage is ampligen, which was found several years ago to inhibit the multiplication of HIV in different samples of lymphoid cells. Ampligen synergistically enhances the activity of zidovudine. Ampligen is a so-called mismatched doublestranded RNA (dsRNA) (Figure 7). dsRNA stimulates the production of various kinds of interferons and activates the intracellular enzymes thought to be responsible for natural antiviral activity. The activation of these enzymes inhibits the multiplication of various human viruses by promoting the cleavage of viral RNA (including mRNA) or the interruption of viral protein synthesis. However, in phase I anticancer studies, dsRNAs have shown severe toxicity. In contrast, dsRNAs with periodically mispaired regions, such as ampligen, retain their lymphokine inductive potential, but have very limited toxicity. Ampligen has been investigated in 10 patients with AIDS or AIDS-related complex. Most of
14(4A) 1992
mismatched region
Pharmaceutisch Weekblad Scientific edition
ds RNA Figure 7 Structure of ampligen
these patients improved over 12 to 18 weeks, but the number of patients was too small to draw definitive conclusions [3 18]. In the laboratory considerable success has been achieved with so-called antisense inhibitors for the control of mRNA expression. Agents of a sequence complementary to viral RNA can be used to inhibit virus multiplication. HIV, herpes zoster virus, influenza virus, and vesicular stomatitis virus are affected in vitro by these agents. These chemical substances include oligodeoxyribonucleotides (ODN). Several modified ODNs have been developed, but the problem of transport of these substances into the virusinfected host cell has not yet been resolved [2]. During the translation process of mRNA large molecular weight precursors of viral proteins are synthesized. These large proteins must be modified to mature functional forms. Modifications may involve enzymatically catalyzed cleavage or other changes. It is recognized that, to reproduce itself, HIV needs a crucial enzyme, the HIV pro271
tease. This enzyme breaks two large viral poly- by an increasing number of opportunistic infecproteins into smaller units. Protease inhibitors tions. An explanation for this phenomenon was sought in the ability of DTC to scavenge oxygen are now under investigation in vitro. Proteins may be acylated, glycosylated, sulf- radicals, which are needed to destroy foreign ated or phosphorylated by enzymes of cellular or microbes. Because of these findings, studies with viral origin. Theoretically, virally encoded en- DTC in patients infected with HIV have stopped zymes are unique targets for selective antiviral [unpublished data]. agents [2]. In building the viral envelope of several RNA and DNA viruses parts of the host cell N e w a p p r o a c h e s w i t h current antiviral membranes and virus-specific g]ycoproteins are drugs Since the development of new antiviral drugs combined. By glycosylation, oligosaccharides and polypeptide chains are coupled to glycopro- is time-consuming, it is useful to consider modifireins. This glycosylation could be inhibited by cations of the dosage regimens of currently availagents such as castanospermine or deoxynojiri- able drugs. A few examples are described here. The recommended dose of zidovudine in mycin (DNJ). As a result of the failure to produce glycoprotein, defective virions are formed which patients with advanced HIV infection is CHSCHSCHSCH3 1,500 rag/day. Although effective, this dose is asare incapable of infecting other cells [2]. HIV envelope glycoproteins are heavily glyco- sociated with severe toxicity. The dose-limiting sylated. As mentioned above, the envelope glyco- toxic reaction is neutropenia. The serum half-life HO'" y "'OH protein gp120 is responsible for binding to the of zidovudine is 1 h, whereas the intracellular OH CD4 molecule. N-Butyl-deoxynojirimycin (N- half-life of the 5'-triphosphate form approaches butyl-DNJ) (Figure 8) is a glycosidase inhibitor 3 h. This means that a lower dose of zidovudine Figure 8 Structure of Nwhich may hinder formation of m a t u r e gp120 could be effective while avoiding serious sidebutyl-deoxyand alter its ability to bind to CD4 receptors. effects. nojirimicin A study of 524 patients was u n d e r t a k e n to deWhen HIV-producing cells are incubated and mixed with the CD4+ cell line in the presence of termine whether a lower dose (200 mg orally a glucosidase inhibitor binding between gp120 every 4 h for 4 weeks followed by 100 mg every 4 h thereafter) was as effective as the standard and its CD4 receptors is blocked. N-butyl-DNJ gradually decreases the propor- regimen (250 mg t a k e n orally every 4 h). This retion of infected cells, eventually leading to elimi- duced dose of zidovudine proved to be at least as effective as the standard dose and was less toxic nation of HIV from culture [19]. [22]. This result has been more or less confirmed in a study with low-dose (300 mg/day), mediumAssembly and release At the end of the cycle, newly created viral ma- dose (600 mg/day), and high-dose (1,500 rag/day) terial is joined together to a new virion and zidovudine, either with or without acyclovir leaves the host cell. Several agents may interfere (4,800 g/day). The 300 mg dose t u r n e d to be at with this process. Interferons (especially inter- least as effective as all the higher doses in the feron-alpha) have been found to inhibit HIV mul- short term. The concomitant use of aciclovir did tiplication, and it is believed t h a t they act, at not enhance the antiretroviral effects of zidovudine. least partially, at this step [4]. Unfortunately, the n u m b e r of patients in this study was too small to be conclusive about the Immunomodulation S H5C2\ If Considerable work has been done to improve use of lower doses [23]. In addition, studies conN--C--S--NA the immunological status of patients with HIV ducted over a longer period of time are necessary. H5C2/ infection. Dithiocarb (diethyldithiocarbamate, Nevertheless, these studies indicate t h a t lower DTC, imuthiol) (Figure 9) may restore immune doses of zidovudine can be used. Figure 9 Structure of diFoscarnet has been licensed for the t r e a t m e n t function by modulating T-cell differentiation thiocarb into m a t u r e and immune competent effector of cytomegalovirus (CMV) infection, but prelimcells. Moreover, DTC acts as a potential antiviral inary data have indicated that, when given intraagent by inhibiting a n u m b e r of copper-contain- venously, the drug has antiretroviral activity as well [3]. Combined use of foscarnet and zidovuing enzymes. DTC has been given orally in a dose of dine appears to yield synergistic effects. Some currently used antiviral drugs, such as 10 mg/kg body-weight and intravenously in a dose of 5 mg/kg body-weight once a week. In a pi- ganciclovir and foscarnet, are available only in lot study 8 of 11 patients receiving dithiocarb injectable formulations and the development of orally experienced a significant decrease in such oral formulations would be highly advantageous constitutional symptoms as persistent diarrhoea, for patient treatment. An analogous situation fever and weight loss. At week 16 the CD4+ cell exists with aciclovir, which is only 15% to 20% counts had increased in these patients [20]. How- bioavailable after oral administration. Efforts ever, because of the small n u m b e r of evaluable have been made to develop an analogue, desciclopatients this study was not conclusive. A second vir, with better absorption properties [1]. study concluded t h a t DTC significantly delayed progression to AIDS. Although there was no evi- Combination therapy The disadvantages associated with current andence of a direct antiretroviral effect, there was also no evidence of an increased virus load after tiviral drugs represent a rationale for the use of treatment. No severe side-effects were observed combination therapy. The existence of zidovudine-resistant strains of HIV has been described [21]. In one study the use of DTC was accompanied by Larder et al. in patients undergoing long-term
~ . CHSOH
272
Pharmaceutisch Weekblad Scientific edition
14(4A) 1992
zidovudine t h e r a p y [24]. Zidovudine also causes significant toxicity, notably bone-marrow depression, which necessitates dosage reduction or discontinuation of the drug. Combination t h e r a p y may be synergistic and thus more powerful in suppressing the virus. The toxicity of individual antiviral drugs can be reduced by the use of lower doses of each. The chance of development of resistance to the drugs will also be decreased. An example of such an approach is the use of combined t h e r a p y with zidovudine and ddC in phase I trials. Both drugs are given in an alternating dosage scheme or concurrently in lower doses [22 23]. The goal of these trials is to find an effective antiviral regimen with tolerable toxicity [11]. Haematopoietic growth factors, also called colony-stimulating factors (CSFs), regulate the proliferation and differentiation of haematopoietic progenitor cells and m a t u r e blood cells. A synergistic activity of granulocyte macrophage colony-stimulating factor (GM-CSF) and zidovudine has been observed in vitro [25]. The rationale for using GM-CSF is twofold. Firstly, GMCSF and granulocyte colony-stimulating factor (G-CSF) seem to exert antiretroviral activity by enhancing neutrophil cytotoxicity towards HIVinfected cells [26]. Secondly, GM-CSF (and also erythropoietin) can counteract AIDS-induced or drug-induced bone-marrow suppresion [4]. Neutropenia is the most common side-effect associated with ganciclovir, which is used in the treatment of CMV infections. The combination of GM-CSF and ganciclovir decreases the occurence of neutropenia [27]. Combination therapy with haematopoietic growth factors is currently being under investigation. Ganciclovir in combination with foscarnet has been reported to show increased efficacy against herpes simplex virus type 2 and cytomegalovirus in mice [28]. Although this synergistic effect must still be studied in humans, it may be a means of providing more effective treatment. These examples demonstrate t h a t combination t h e r a p y is a rational way of investigating the t r e a t m e n t of antiviral infections.
Conclusion At this moment an increasing n u m b e r of potential antiviral agents are under investigation in the laboratory. Since our knowledge of viral infection and multiplication is also increasing, we are now able to define more targets and different approaches in our fight against viral infections. Of course, by far the most work is being done on anti-HIV drugs. The reverse transcriptase enzyme is essential and unique for the multiplication ofretroviruses. Nucleoside and nonnucleoside inhibitors of reverse transcriptase are now in clinical trials. Other targets will also be explored. A t t a c h m e n t of HIV to host cells can be prevented by soluble CD4, while sulfated polysaccharides have similar properties in HIV and influenza virus. New antiviral drugs must be selective and this is an important problem. Many agents t h a t are effective in vitro can not be used in vivo because
14(4A) 1992
Pharmaceutisch Weekblad Scientific edition
of toxicity or lack of selectivity. A drug t h a t is active in vitro cannot always reach the target - the virus inside the host cell. Other approaches include enhancing the patient's immunological status with such agents as ampligen and dithiocarb. Interferons also influence the immune response and m a y act on viral assembly and release as well. The most promising approach to treating antiviral infections, however, appears to be the use of combination therapy, which offers m a n y advantages. Combinations of antiviral drugs may be synergistic and m a y also avoid the development of resistence. The administration of lower doses of each drug could reduce toxicity, thereby making therapy more acceptable to the patient. However, although investigation of effective combinations will lead us to new therapeutic regimens, the continuing search for new antiviral drugs will remain necessary.
Acknowledgement The author t h a n k s Eda M a k a t i t a and Marlies Hutters-Wulp for their fine secretarial assistance and Arno Vyth and Joep Lange for their critical review of the manuscript.
References 1 Wiltink EHH, Janknegt R. Antiviral drugs. Pharm Weekbl [Sci] 1991;13:58-69. 2 Sire IS. Virus replication: target functions and events for virus-specific inhibiters. In: Galasso GJ, Whitley RJ, Merigan TC, eds. Antiviral agents and viral diseases of man. 3rd ed. New York: Raven Press, 1990:147. 3 SandstrSm E. Antiviral therapy in human immunodeficiency virus infection. Drugs 1989;38:417-50. 4 Yarchoan R, Mitsuya H, Broder S. Clinical and basic advances in the anti-retroviral therapy of human immunodeficiency virus infection. Am J Med 1989;87: 191-200. 5 Capon DJ, Chamow SM, Mordenti J, Marsters SA, Gregory T, Mitsuya H, et al. Designing CD4 immunoadhesins for AIDS therapy. Nature 1989;337:525-531. 6 Moore JP, Weiss RA, Passive primate protection. Nature 1991;352:371-7. 7 Daar ES, Li XL, Moudgil T, He DD. High concentrations of recombinant soluble CD4 are required to neutralize primary human immunodeficiency virus type 1 isolates. Proc Natl Acad Sci USA 1990;87:6574-78. 8 Abrams DI, Kuno S, Wong R, Jeffords K, Nash M, Molaghan JB, et al. Oral dextran sulfate (UA 001) in the treatment of the acquired immunodeficiency syndrome (AIDS) and AIDS-related complex. Ann Intern Med 1989;110:183-8. 9 Lfischer-Mattli M, Gliick R. Dextran sulfate inhibits the fusion of influenza virus with model membranes, and suppresses influenza virus replication in vivo. Antivir Res 1990;14:39-50. 10 Kim KS, Sapienza VJ, Carp RI. Antiviral activity of arildone on deoxyribo-nucleicacid and ribonucleic acid viruses. Antimicrob Agents Chemother 1980;18:27680. 11 Badger J, Minor I, Kremer MJ, Oliveira MA, Smith TJ, Griffith JP, et al. Structural analysis of a series of antiviral agents complexed with human rhinovirus 14. Proc Natl Acad Sci USA 1988;85:3304-8. 12 Rossmann MG. Antiviral agents targeted to interact with viral capsid proteins and a possible application to human immunodeficiency virus. Proc Natl Acad Sci USA 1988;85:4625-7. 13 Yarchoan R, Perno CF, Thomas RV, Klecker RW, Allain J-P, Wills RJ, et al. Phase I studies of 2',3'dideoxycytidine in severe human immunodeficiency virus infection as a single agent and alternating with zidovudine (AZT). Lancet 1988;1:76-81. 14 Merigan TC, Skowron G, Bozette SA, Richman D, Uttamchandani R, Fischl M, et al. Circulating p24 an-
273
15
16
17
18
19
20
21 22
274
tigen levels and responses to dideoxycytidine in human immunodeficiency virus (HIV) infections. Ann Intern Med 1989;110:189-94. Yarchoan R, Pluda JM, Thomas RV, Mitsuya H, Brouwers P, Wyvill KM, et al. Long term toxicity/activity profile of 2',3'-dideoxyinosine in AIDS or AIDSrelated complex. Lancet 1990;336:526-9. Ho H-T, Hitchcock MJM. Cellular pharmacology of 2',3'-dideoxy-2',3'-didehydrothymidine, a nucleoside analog active against humanimmunodeficiencyvirus. Antimicrob Agents Chemother 1989;33:844-9. Lake-Bakaar DM, Lindborg B, Datema R. Improvement of the absorption of oral (R,S)-9-[4-hydroxy-2-(hydroxymethyl)butyl]guanine, an anti-varicella-zoster virus drug, in rats and monkeys. Antimicrob Agents Chemother 1989;33:110-2. Carter WA, Strayer DR, Brodsky I, Lewin M, Pellegrino MG, Einck L, et al. Clinical, immunological and virological effects of ampligen, a mismatched double-stranded RNA, in patients with AIDS or AIDSrelated complex. Lancet 1987;1:1286-92. Karpas A, Fleet GWJ, Dwek RA, Petursson S, Namgoong SK, Ramsden NG, et al. Aminosugar derivatives as potential anti-human immunodeficiency virus agent. Proc Natl Acad Sci USA 1988;85:9229-33. Lang J-M, Touraine J-L, Trepo C. Randomised, doublebind, placebo controlled trial of dithiocarb sodium ("Imuthiol") in human immunodeficiency virus infec~ tion. Lancet 1988;2:702-6. Reisinger EC, Kern P, Ernst M, Bock P, Flad HD, Dietrich M, et al. Inhibition of HIV progression by dithiocarb. Lancet 1990;335:679-82. Fischl MA, Parker CB, Pettinelli C, Wulfsohn M, Hirsch MS, Collier AC, et al. A randomized controlled
23
24 25
26
27
28
trial of a reduced daily dose of zidovudine in patients with the acquired immunodeficiency syndrome. N Engl J Med 1990;323:1009-14. Collier AC, Bozzette S, Coombs RW, Causey DM, Schoenfeld DA, Spector SA, et al. A pilot study of lowdose zidovudine in human immunodeficiencyvirus infection. N Engl J Med 1990;323:1015-21. Larder BA, Darby G, Richman DD. HIV with reduced sensitivity to zidovudine (AZT) isolated during prolonged therapy. Science 1989;243:1731-4. Hammer SM, Gillis JM. Synergistic activity of granulocyte-macrophage colony-stimulatingfactor and 3'-azido-3~-deoxythymidine against human immunodeficiencyvirus in vitro. Antimicrob Agents Chemother 1987;31:1046-50. Baldwin GC, Fuller ND, Roberts RL, Ho DD, Golde DW. Granulocyte- and granulocyte-macrophage colony-stimulating factors enhance neutrophil cytotoxicity toward HIV-infected cells. Blood 1989;74:16737. Hardy WD, Spector S, Polsky B, Crumpacker C, Holland G, Freeman W, et al. Combined ganciclovir (GCV) and recombinant humangranuIocyte-macrophage-stimulating factor (GM-CSF) vs ganciclovir alone for cytomegalovirus (CMV) retinitis in AIDS, [Abstract]. Thirtieth Interscience Conference on Antimicrobial Agents and Chemotherapy, 1990; New Orleans, abstract 9. Freitas VR, Frase-Smith EB, Matthews TR. Increased efficacy of ganciclovir in combination with foscarnet against cytomegalovirus and herpes simplex virus type 2 in vitro and in vivo. Antiviral Res 1989;12:20512.
P h a r m a c e u t i s c h W e e k b l a d Scientific e d i t i o n
14(4A) 1992
