HIV Therapy

Written by Dr. Furrat Amen

Introduction

Existing successful combinations of drugs include those used against TB, cancer, nucleoside analogues for HIV treatment, and Hepatitis B amongst many others. This paper deals mainly with the fight against HIV as this is a very interesting and a substanial subject even when taken in isolation. The steps taken in special situations such as pregnancy are detailed as well as the mode of action of the combinations.

The move to combination therapies has been prompted by the emergence of resistance of HIV to zidovudine monotherapy and there are many different ones under investigation - these will be discussed in some detail. Another aim in treating HIV-infected patients is to minimise side-effects from the drugs and methods to do this are mentioned.

Sites of action

The replication of HIV may be interrupted at several stages in its cycle of replication:

1) Virus-cell binding by polyanionic substances
2) Virus-cell fusion by plant lectins
3) Virus fusion/uncoating by bicyclams
4) Reverse transcription by both subtrate and non-subtrate analogues
5) Integration
6) DNA replication
7) DNA to mRNA transcription
8) mRNA translation to by antisense oligonucleotides and ribozymes
9) maturation by proteolysis inhibitors
10) assembly and release of virions

1) Virus cell binding may be inhibited by a variety of methods which include the use of polysulphates, polysulphonates and polycarboxylates. These methods protect CD4+ cells against destruction when fused with gp120 cells. A useful way of administrating such substances would be to use them in vaginal microbicides.

2) Virus fusion and uncoating may be prevented by the adminstration of bicyclams. These are potent inhibitors of HIV replication. Another advantage of these compounds is that they are equally active against HIV1 and 2.

3) Reverse transcriptase inhibitors form the mainstay of current treatment regimens and will be discussed in greater detail in the subsequent text. An example of this group are the acyclic nucleoside phosphonate analogues. They are taken up by cells and are phosphorylated. The so formed compound will then inhibit the HIV reverse transcriptase reactions by becoming incorporated into the chain, being formed, and terminating it. A further advantage of the acyclic nucleoside phosphonate analogues is that their intracellular half-life is very long and this obviously has advantages with respect to infrequent dose requirements and helps with compliance.

TIBO, HEPT and TSAO are non-substrate analogues which interfere with reverse transcriptase. The problem with this group of compounds is that resistance occurs readily and so it is necessary implement strategies to ameliorate such problems. Switching drugs, combining drugs with mutually antagonistic resistance patterns, and using high killing concentrations of the drug to prevent and virus remaining viable are all possible methods of reducing problems with resistance (12).

Combination therapy

Combination therapy for HIV infection has become progressively more important in the treatment of HIV infection, and there has been a move towards early treatment for asymptomatic patients. The primary infection in the majority of HIV patients will consist of an acute illness and this is a marker for a worse prognosis in the long term. If a primary infection also presents with very high levels of viraemia at the time of seroconversion this allows for wide dissemination of the disease eg. into the central nervous system. Another point to note at the time of the primary infection, is that the viral population is homogenous.

The goal has become to treat the patients who are infected so as to make them long term non-progressers. This is achieved by aiming for three areas of control: keeping viral replication down, keeping the CD4+ count within the normal range, and maintaining normal lymph nodes. Zidovudine is limited in its ability to limit the viraemia associated with HIV infection, so it may be necessary to add lamivudine with or without the addition of a non-nucleoside inhibitor of reverse transcriptase and a protease inhibitor.

When considering the use of such 'combination therapy' it is possible to follow two paths. 'Convergent therapy' uses drugs, with actions, which have the same viral target. It there is no steep drop in the viraemia the selection pressures will result in simultaneous mutations in enzymes which may decrease enzyme activity.

'Divergent therapy' targets different sites of viral replication. Only protease inhibitors can reduce the production of infectious virons by latently infected cells when they start replicating. When a virus escapes the effects of protease inhibitors they can be targeted by reverse transcriptase inhibitors. Alternatively, when proviral DNA is synthesised after failure of reverse transcriptase inhibitor activity, the protease inhibitors act later in the cycle.

Using combinations of four drugs may create problems related to treatment. There may be increased toxicity and decreased compliance as the regimen becomes more and more difficult and confusing to stick to (10). Drug treatment also selects for resistant mutants and a chronic HIV infection has high replication rates which are likely to produce such mutants in a short space of time. Single stranded RNA viruses eg. HIV also lack proof reading mechanisms which also contributes to the formation of mutants.

When combining antiretroviral therapies it is optimal treatment to achieve synergy and to have drugs with differing toxic effects. An example of this can be seen with the drugs zalcitabine and didanosine which are seldom used in combination. This is because they are both thought to cause neuropathy. This was also thought to be the case with didanosine with stavudine, but the combination proved to be well tolerated and to have a low risk of inducing neuropathy in patients.

Toxicity of all these treatments is minimised via a number of methods. Zidovudine may be given in alternating or intermittent treatments. The problem is that alternating therapy is no better than monotherapy in HIV suppression. When intolerance develops to zidovudine, it is better to replace it with zalcitabine or didanosine.

They are many toxic effects and each drug has its own particular problems, but when combinations of drugs are considered there are some drugs which together cause marked toxic side-effects. Zidovudine when combined with interferon-alpha may result in a neutropenia - this merits a reduction in dose. Malaise occurs in 40% of patients on this combination, and anorexia occurs in 44%. These figures are in excess of those for zidovudine therapy used on its own. A flu-like illness also may manifest. Other features, of using this combination, include increased incidences of asthenia, fever, and myalgia.

Zidovudine and aciclovir are synergistic in their actions, but unfortunately have increased side-effects. Haematological toxicity occurs in 52% of those on the combination, as opposed to 46% of patients on zidovudine monotherapy. It is necessary to modify the dose of zidovudine in 69% due to toxic effects, when used in combination. This may be unfavourable compared with 46% requiring dose modification because of toxic side-effects when on zidovudine alone (14).

Saquinavir is a hydroxyethylamin transition-state analogue of the HIV-protease cleavage site. Saquinavir is one of the new protease inhibitor drugs and this too may be used in combinations eg. saquinavir, zidovudine, and zalcitabine. This particular combination has proved to be synergistic when compared to monotherapy. Any two of these drugs performed similarly. It was found, however, that this triple therapy was more effective than double therapy when used in patients who had previous long term treatment using the drug zidovudine, when compared with zalcitabine monotherapy.

The combination decreases cell associated infectious titres of HIV, plasma titres of viral RNA, levels of neopterin and beta-2 microglobulin, and increase CD4+ cell counts more than during dual therapy. Further studies are required to investigate the effects on long term progression and survival (28).

The combination of saquinavir and zalcitabine significantly increased the time to the first AIDS defining event and death (15).

Another method of bypassing rapidly forming HIV resistance is by the used of inhibitors of cellular deoxynucleotide (dNTP) metabolism eg. hydroxyurea may be used to decrease HIV replication. Hydroxyurea is an inhibitor of ribonucleotide reductase and therefore dNTP synthesis - this in turn diminishes HIV replication.

HIV replication is decreased by treatment of cells with an excess of thymidine. Drugs which target HIV enzymes generate resistance, but this is not a problem encountered with targeting host cell synthesis of dNTP which is required by HIV for reverse transcription. There may be, however, a problem of side-effects to such treatment.

DUTP is a toxic metabolite because it is used by DNA polymerases. This causes cell death by means of the action of uracil glycosylase-mediated excision repair. DUTP pyrophosphatase (dUTPase) catalyses the hydrolysis of dUTP. Human endogenous retrovirus K (HERV-K) is a defective retrovirus whose dUTPase may be needed for efficient replication of HIV. The advantages of targeting HERV-K are that it may not be important in cellular metabolism, and its resistance is independent of HIV. The targeting may be achieved in two ways: by the use of antisense oligonucleotide to block the HERV-K genes' expression, or to design a deoxyuridine which blocks HERV-K dUTPase without affected the host cell's cellular enzymes. HIV lacks a dUTPase gene itself an so the replication would by reduced (1).

Primary monocyte dervived macrophages act as a reservoir for virus in the body. Glutathione, in vitro, suppresses HIV replication in a dose dependent fashion. Glutathione also decreases expression of gp120 in chronically infected macrophages and therefore may affect release of virus from infected cell. Glutathione may have its effects in vivo if adequate concentrations are reached and so may be a useful part of future combination therapies.

Reverse transcriptase inhibitors and protease inhibitors may be used in two drug combinations. The reverse transcriptase inhibitors comprise zidovudine, didanosine (ddI), stavudine (d4T), zalcitabine (ddC) and lamivudine (3TC). Extensive use of nucleoside analogues may result in side-effects. HIV protease inhibitors are SC-52151, saquinavir, indinavir, BMS-186,318 (BMS-PI). Protease resistant mutants have been isolated in patients over the time they have taken these drugs.

Synergy is when a combination of drugs:

Affects more than one cell type,
Affects cells in different stages of activation,
Stops virus replication at different stages,
Is used with compounds improving drug uptake, and
Affects activation levels or increases levels of an active metabolite.

In vitro experiments

Other synergistic combinations are: ddI and AZT; and ddC and AZT. AZT and d4T may be used in experiments in T-cell line cells. In a 20:1 combination moderate synergy is displayed in the course of experimentation. In peripheral blood mononuclear cells (PBMC) a 5:1 ratio of the drugs is required to achieve synergy.

D4T and ddC display high synergy, and no toxicity was displayed at the highest concentrations used for experimentation. The combination of BMS-PI and d4T was found to have additive effects when used in T-cell lines and in PBMCs synergy became apparent at ratios of 1.3:1. D4T and saquinavir display synergy at 10:1 and 20:1 ratios in PBMCs, which is better than the former combination. BMS-PI and ddI or AZT are also synergistic. Indinavir and d4T are additive in their effects when combined 40:1. Indinavir and ddI are synergistic when 120:1 and 240:1 combinations are utilised.

BMS-PI and saquinavir are synergistic in T-cell lines in combinations of 75:1 and 25:1, while their effects are only additive in PBMCs. BMS-PI and SC52151 also display synergy when used together. When all these experiments are taken together is may be seen that BMS-PI is very effective in T-cell lines when combined with protease inhibitors saquinavir or SC-52151. None of the combinations tested showed cytotoxicity in vitro.

The problem with short term assays, in vitro, of these types are that they do not address the problems of resistance and the risks of toxicity in patients. It is possible on the basis of the results to suggest that d4T and saquinavir, d4T and AZT, and d4T and ddI displayed significant combination synergy and so deserve further clinical evaluation (7).

SC-52151

The late stage of HIV replication requires virus-encoded aspartyl protease for processing of structural protein and replicative enzyme precursors. Inhibition would result in the production of immature non-infectious virus. SC-52151 is a non-nucleoside, urea-based peptidomimetic antiretroviral compound which may be used alone or in combination with reverse transcriptase inhibitors to produce synergy.

Synergy is also displayed with ddI, ddC, or d4T against AZT-resistant HIV. Additional synergy is achieved when SC-52151 is used in triple therapy combination with AZT and alpha-glucosidase inhibitor SC-48334 as multiple points in the life cycle may be affected. In chronic or latent HIV infection, in which AZT or reverse transcriptase inhibitors had no effect, this compound produces a dose dependent decrease in virus production (8).

AZT and 3TC have synergistic effects forming the restoration of AZT sensitivity with the occurrence of a 3TC resistance mutation.

The Tat protein is encoded for by HIV and is needed for replication of the virus. Therefore, it would be a good target for therapeutic intervention. It activates viral gene expression at the level of transcription. It has been found in vitro that a variety of kinase inhibitors inhibit Tat-activated transcription, and so such compounds may become useful additions to combination therapy (16).

Ritonavir and saquinavir are both protease inhibitors which are synergistic because when used together the bioavailability of saquinavir is increased. It is also beneficial that their resistance patterns do not overlap. There is a resultant increase in T-cell function and antigen presenting cell function. After 24 weeks of use, the percentage of CD8+ lymphocytes expressing CD28 was increased from 32% to 48%, in line with the figures for seronegative individuals. These figures are indicative of increases in T-cell function . PBMCs in HIV positive patients produce less IL12 and these figures were increased after 24 weeks.

Some patients did not improve their cell mediated immunity on treatment, and so may need a longer period of therapy exceeding 24 weeks. It does appear that the combination would improve clinical outcome but further clinical studies are required to confirm this (20).

Immunotoxin kills HIV infected activated cells and cyclosporine inhibits activation of latently infected cells - therefore taken together they provide for therapy that completely suppresses HIV production. Cyclosporine acts by inhibiting T cell activation and blocks the interaction between viral Gag and cellular cyclophilin A. FK506, an unrelated compound, acts by inhibiting T cell activation by binding to calcineurin.

SDZ NIM811 is a non-immune suppressant analogue of cyclosporine which also has anti-HIV activity which when added to the combination reduces HIV production to undetectable levels, as measured by viral p24 levels. Cyclosporine inhibits cellular proliferation and Gag-cyclophilin interactions in CD25- cells thereby achieving the optimal anti-HIV activity of these compounds (21).

Reducing Side-Effects

The use of zidovudine is limited by its toxic effects on bone marrow which in turn cause anaemia and neutrophilia. This means that patients may require transfusions, dose reductions, or interruption of treatment. AZT has a direct inhibitory effect on growth of human granulocyte-macrophage colony forming units (CFU-GM) and BFU-E.

The use of beta-D-uridine reverses the toxic effects and protects from AZT toxicity in human CFU-GM cells without affected anti-HIV activity. The action of beta-D-uridine is dependent on its activation to its nucleotides. Once the step is completed then the pharmacological effect becomes apparent and may be protective against the toxic effects of the combination of AZT and FLT which is synergistic but would otherwise cause an unacceptable degree of bone marrow toxicity. IL1, IL3 and GM-CSF overcome toxicity of AZT by stimulating growth of bone marrow cells.

HIV increases demand on haematopoietic cells and advanced HIV infection is known to cause anaemia, neutropenia and thrombocytopenia. AZT therapy, itself, is myelosuppressive of three cell lines and many other drugs used in the treatment of HIV patients exhibit similar effects: ganciclovir, foscarnet, antifungal agents, interferon alpha, sulphonamides, dihydrofolate reductase inhibitors, pentamidine, and antineoplastic therapies.

Recombinant erythropoietin (EPO) raises haematocrits and decreases the number of transfusions required in HIV patients with anaemia. Blood transfusion is associated with accelerated mortality in AIDS patients, and although confounding factors do exist for this finding, it may still be the case that HIV progression may be accelerated by activating HIV replication, and transfusion related immunosuppression.

A study of 279 AIDS patients, conducted double blind, showed that EPO significantly increases mean haematocrit levels and decreases the mean number of units of blood transfusions compared with placebo of normal saline. Similar effects are found in zidovudine treated patients and those not on zidovudine.

GM-CSF use in neutropenic patients can result in an increase in levels of neutrophils, monocytes, eosinophils, and, to a lesser extent, lymphocytes. The effect is dose dependent and also allows for continued zidovudine therapy when it would otherwise have been halted as a result of drug's associated neutropenia. GM-CSF may also be used in conjunction with ganciclovir in CMV retinitis and in this combination it reduces the proportion of patients who are neutropenic. It also helps patients with Non-Hodgkin's Lymphoma (NHL) in AIDS who are suffering from the side-effects of cyclophosphamide, doxorubicin, vincristine and prednisolone - it decreases neutropenia, the number of febrile episodes, and the duration of admission to hospital.

G-CSF can increase survival in neutropenic AIDS patients. IL2 increases T and B lymphocyte numbers and may initiate their differentiation in vitro, although clinical trial have not proved any increase in quality of life scores even after fourteen months of treatment.

Intermediate doses of IL2 given subcutaneously may be combined with zidovudine and didanosine in patients who have an asymptomatic HIV infection. This maintains an elevated CD4 cell count and a low HIV viraemia according to studies conducted six months following initiation of treatment. This combination does also cause the mild, IL2-associated, side-effects of capillary leak and flu-like symptoms (3).

Contradictory evidence exists in another study which found - IL2 activates T-cells and could increase the activity of HIV. Viral loads of HIV increase, with IL2, transiently with patients that have high CD4 counts, but more worryingly this effect is sustained in those who have a low CD4 count.

IL3 can increase neutrophil counts, but the effects are minor when compared with those of G-CSF and GM-CSF - therefore its clinical usefulness is under scrutiny. IL3, GM-CSF and M-CSF (macrophage colony stimulating factor) have the unwelcome side-effect, in vitro, of increasing viral replication.

GM-CSF increases HIV replication, in vitro, in monocyte and macrophages - not in T cell lines. There is conflicting evidence for this effect in vivo studies. The studies may be conducted by measuring p24 levels in patients. GM-CSF has the more favourable action of enhancing the in vitro zidovudine anti-retroviral effect by increasing intracellular concentrations of zidovudine. It does this by increasing cell uptake, and increases the activity of thymidine kinase which phosphorylates zidovudine to its active moiety. These effects have yet to be proved in vivo.

Combinations of drugs used in AIDS related illnesses.

NHL affects 2% of HIV patients and standard chemotherapy results in poor prognosis for these patients. Complete remission is achieved in 50% of patients and median survival is nine months - most have a progressive lymphoma or succumb to opportunistic infection. Poor prognostic indicators include CD4 counts <100/microL, lymphomatous bone marrow invasion and a history of AIDS opportunistic infections.

A differing combination of chemotherapy of patients is the use of cyclophosphamide, doxorubicin, and etoposide (CDE). Toxic effects are myelosuppression and the potentiatation of opportunistic disease. Unfortunately, the use of G-CSF does not appear to improve the lymphopenia or thrombocytopenia in patients being treated with this combination.

DdI use significantly increases leukocyte, neutrophil, red blood cell, and platelet counts. It is fortunate that the toxic effects, peripheral neuropathy and pancreatitis, do not coincide with the toxicity of CDE therapy. In clinical studies the effect found, in vitro, in improving neutrophil counts has yet to be substaniated. CDE also decrease CD4 and CD8 counts and these counts are not improved by adding ddI to the combination.

CDE did result in nausea and vomiting in 72% of patients and stomatitis in 56%, 40% due to the chemotherapy and 16% due to herpes simplex infection (18).

In vitro studies conducted on patients with Kaposi's sarcoma (KS) found that IFN alpha and retinoids have a synergistic antiangiogenesis effect. To this end, all-trans-retinoic-acid (ATRA) was used in clinical trials. Previously this compound was found to inhibit KS related spindle cell lines. It down regulates IL6 receptors which are responsible for the autocrine stimulation of KS. Plasma drug exposure of this drug decrease with chronic daily use because of up regulation of oxidative catabolism and an increase in cellular ATRA binding protein. Unfortunately no equivalent response was found in vivo (19).

Resistance and vertical transmission

Mutation of the viral genome leads to resistance. In the case of HIV, the resistance to reverse transcriptase inhibitors arises as a result of mutations in the reverse transcriptase genes. In the mother, there are three possibilities to account for resistance to therapy: the mother may have acquired a resistant form of HIV, resistance to HIV may have arisen as a result of treatment during pregnancy, or resistance may have developed spontaneously as a result of long term infection. A mutated virus is generally less efficient at replicating. Antiretroviral drugs increase the resistance of the virus and so vertical transmission increases and by the same virtue a pregnant women not taking antiretroviral therapy will decrease vertical transmission.

Resistance to zidovudine is stepwise even when taking short courses. The administration of zidovudine also reduces its efficacy in the future. When the mother is taking the antiretroviral therapy the baby has an increase possibility of receiving a resistant virus. Such a mother will also have increased transmission rates over a lady which presents with a sensitive virus.

If many antiretrovirals are directed at the same molecular target the virus may not mutate sufficient whilst retaining the function of the target molecule - this is the principle upon which convergent therapy works. It is also the case that viruses with multiple resistance grow less well that the wild type strain. Virus that undergoes the commonest resistance mutation for lamivudine (3TC), namely M184V, continue in their sensitivity to zidovudine even if mutated at the site which normally confers resistance to zidovudine.

A combination of zidovudine and nevirapine decrease resistance by mutation at tyrosine position 181. Such combination therapy is important in the newborn and in pregnancy especially in the context of prolonged treatment during one pregnancy or multiple pregnancies. It is also important not to use nevirapine at the time of delivery for more than a few weeks, as a highly resistant virus is produced.

The dilemma occurs when it is necessary to tailor treatment to a particular pregnant women. It is necessary to bet on the time of delivery. If the delivery occurs too soon then there has been insufficient time for a decrease in maternal plasma RNA levels, and for the drug to cross the placenta. Alternatively, if the delivery occurs too late then the plasma viraemia will have risen again, a mutation will have occurred at position 181Y and the virus will be fully resistant. It is also necessary to use a high dose of nevirapine in sufficient killing concentrations. This may cause a severe rash. The same principles of judging the length of treatment also apply to the administration of zidovudine (2).

Conclusion

It is important to achieve a balance between the number of drugs to suppress HIV production, in a combination, and the questions of compliance and side-effects for the patient. Combinations are available for asymptomatic patients who have HIV infection as well as later on in the illness when complications arise such as NHL in AIDS.

Combination can also be structured to include agents which limit side-effects and toxicity of the commonly used drugs or improve their uptake and action. Special regimens may be implemented to deal with pregnant women who have HIV infection so as to reduce their risk of vertical transmission and such plans include Caesarean section and no breast feeding in addition to pharmacological therapy.

The main indication for combination therapy remains, however, the prevention of resistance of HIV to therapy and in this vein such drugs may achieve the target of making patients, infected with HIV, non-progressors that do not die of AIDS despite remaining infected. In this way HIV infection will become an irrelevancy and patients will live a normal life.

Furrat Amen