Antivirals — Questions

First-line treatment

The first-line agents are intravenous ganciclovir or its oral prodrug valganciclovir at induction dosing. Valganciclovir is suitable for non-severe disease where gut absorption is reliable; intravenous ganciclovir is preferred for severe, sight-threatening or life-threatening disease, or where absorption is uncertain.

Deciding the duration

Treatment is given for a minimum of two weeks and continued until the viral load has cleared below the local threshold (commonly to undetectable on consecutive samples) and the patient has responded clinically. An important rule is not to switch agents within the first two weeks if the patient is improving, since the viral load may take that long to fall. Where reasonable, reducing immunosuppression is a useful adjunct, but it does not replace the antiviral (so option E is wrong). Letermovir (B) is prophylaxis only, foscarnet (C) is not first-line and needs monitoring, and aciclovir-class drugs (D) are not effective treatment for CMV disease.

Salvage for resistant CMV

With a confirmed UL97 mutation (ganciclovir resistance), the active options are:

• Maribavir, an oral inhibitor of the UL97 kinase itself, licensed specifically for refractory or resistant post-transplant CMV and not myelosuppressive (though resistance to it, via UL97 and UL27, can emerge);
• Foscarnet, which acts on the polymerase and does not need UL97 activation, so it retains activity (limited by nephrotoxicity).

Reducing immunosuppression is a valuable adjunct to let immunity help, but not a substitute for an active drug.

Why the distractors are wrong

Letermovir (B) has a low barrier to resistance and is a prophylactic agent, not salvage. Simply increasing ganciclovir (C) does not overcome a true UL97 mutation. Stopping all antivirals (D) abandons effective salvage options. Aciclovir-class drugs (E) are not effective treatment for CMV disease at all.

Daily suppressive valaciclovir lowers recurrences and asymptomatic shedding and cuts transmission to a susceptible partner by about half.

Because it reduces but does not abolish shedding, the protection is partial, so the higher figures overstate it and it is certainly better than no effect.

Why A is correct

Letermovir inhibits the CMV terminase (UL56) complex, the machinery that cleaves and packages the viral genome, a target unique to CMV. Two properties make it valuable for prophylaxis in seropositive allogeneic stem-cell recipients: it is not myelosuppressive (unlike ganciclovir), so it does not impair engraftment, and it significantly reduces clinically significant CMV when given from transplant through about week 14.

The practice-critical caveats

• It is specific to CMV and has no activity against herpes simplex or varicella-zoster virus, so a patient on letermovir still needs an aciclovir-class drug to cover those herpesviruses (option B is wrong).
• It has a low barrier to resistance (selecting UL56 mutations), so it is a prophylactic agent, not a treatment for established disease or a salvage drug (options C and D are wrong).
• It is metabolised through and inhibits CYP3A, raising calcineurin inhibitor (tacrolimus, cyclosporine) levels, so doses are adjusted (option E is wrong).

Haemagglutinin tethers budding virions to surface sialic acid; neuraminidase cleaves it to free them. Blocking neuraminidase traps progeny at the surface and halts spread.

These are protease inhibitors. A virus that translates its genome as one polyprotein must cut it into functional units with a viral protease; blocking that protease leaves the polyprotein uncut and the virions immature.

Because no host enzyme copies RNA from RNA, an RdRp inhibitor can be highly selective for the virus. Sofosbuvir (hepatitis C) and remdesivir (SARS-CoV-2) act here.

Why C

Ganciclovir and maribavir must not be combined. Maribavir inhibits the UL97 kinase, the very enzyme that phosphorylates (activates) ganciclovir; blocking it makes ganciclovir ineffective, so the two are antagonistic and are never given together.

The other pairs

The remaining combinations are not contraindicated on this basis: ganciclovir with foscarnet (different targets, sometimes combined in resistant disease), ganciclovir with letermovir, and letermovir or maribavir with valaciclovir (which covers herpes simplex and varicella-zoster, a gap left by both letermovir and maribavir).

Antiviral treatment is indicated for complicated varicella and for any immunocompromised patient, in whom the disease can progress and disseminate. Aciclovir should be started as early as possible, given intravenously when the disease is severe.

Uncomplicated chickenpox in a healthy child is self-limiting and does not require antivirals.

Antivirals shorten the acute illness and reduce acute pain when started early, within the first two to three days of the rash, but they do not prevent postherpetic neuralgia.

That established neuropathic pain is managed separately, with agents such as gabapentin or pregabalin, tricyclic antidepressants, and topical lidocaine.

Aciclovir must first be activated by the viral thymidine kinase, and resistance, which arises mainly during prolonged exposure in advanced HIV infection, is usually due to mutation of that kinase.

Foscarnet and cidofovir inhibit the viral DNA polymerase directly, without needing viral activation, so they retain activity against thymidine-kinase-deficient resistant strains.

a. Cytomegalovirus (CMV) disease, here CMV hepatitis; the owl’s-eye inclusions are its cytopathic hallmark.

b. First-line is ganciclovir (intravenous) or oral valganciclovir, with the full blood count monitored for neutropenia. When disease is refractory or resistant (the viral load fails to fall, with a confirmed UL97 mutation), switch to foscarnet or cidofovir, or to maribavir for resistant disease; reducing immunosuppression where possible aids control.

a. Oral valganciclovir for 6 months (intravenous ganciclovir if oral therapy is not tolerated). Treating central-nervous-system-involved symptomatic disease improves hearing and neurodevelopmental outcomes.

b. Neutropenia; the full blood count is monitored through the course.

c. Asymptomatic infants, and those with isolated hearing loss without other disease, are generally not treated, because the benefit does not outweigh the toxicity; the decision is specialist-led.

a. Influenza (an orthomyxovirus).

b. Start a neuraminidase inhibitor (oseltamivir), continued even beyond 48 hours in severe or hospitalised disease, while recognising that the evidence in critical illness is limited; baloxavir is an alternative. The mainstay is supportive intensive care (oxygenation, lung-protective ventilation), treatment of bacterial co-infection, and appropriate infection-prevention precautions.

• Mechanism. Cidofovir is an acyclic nucleotide (phosphonate) analogue that already carries the equivalent of a first phosphate, so cellular kinases alone activate it to the diphosphate, which inhibits the viral DNA polymerase as a delayed chain terminator. Its long intracellular half-life allows dosing only once every week or two.
• Activity against resistant CMV. Because it needs no viral kinase, it stays active against UL97-mutant ganciclovir-resistant CMV, although UL54 polymerase mutations can confer cross-resistance.
• Limiting toxicity. Its dose-limiting nephrotoxicity is reduced by giving probenecid with intravenous saline hydration to cut proximal-tubular uptake of the drug.

• Mechanism. Letermovir inhibits the CMV terminase complex (UL56, UL89 and UL51), which cleaves and packages replicated viral DNA into capsids; the target is unique to cytomegalovirus.
• Place in therapy. Oral and well tolerated, used for CMV prophylaxis (notably in seropositive allogeneic stem-cell transplant recipients), with no cross-resistance to the polymerase-targeting agents.
• Caveat. A low genetic barrier (UL56 resistance) makes it a prophylactic rather than a treatment agent, and it inhibits CYP3A (interaction with cyclosporine).

• a. Neuraminidase inhibitors. Active-site mutations in the neuraminidase reduce drug binding; the H275Y substitution lowers oseltamivir and peramivir susceptibility but spares zanamivir. Haemagglutinin mutations that reduce dependence on neuraminidase are a second route.
• b. Cap-dependent endonuclease inhibitors. The polymerase acidic (PA) I38T substitution reduces baloxavir susceptibility and emerged in nearly 10% of treated patients in trials, sometimes with viral rebound.

• Activation. Aciclovir is an inert prodrug until the viral thymidine kinase of HSV (herpes simplex virus) or VZV (varicella-zoster virus) monophosphorylates it; cellular kinases then add the second and third phosphates to make aciclovir triphosphate. Host thymidine kinase barely activates it, so active drug concentrates only where the virus is replicating.
• Inhibition. Aciclovir triphosphate inhibits the viral DNA polymerase far more strongly than host polymerases and, lacking a 3’-hydroxyl, acts as an obligate chain terminator that freezes the enzyme in a dead-end complex.
• Selectivity. The two layers, selective activation by a viral kinase and selective inhibition of the viral polymerase, together make aciclovir toxic to the virus but not to the host.
• Side effects. Aciclovir is very well tolerated. The main concern is crystalluria with renal injury at high intravenous doses (prevented by good hydration and slow infusion), with rare neurotoxicity in renal impairment; it is not marrow-toxic at ordinary doses.

• Zanamivir. A neuraminidase inhibitor given by inhalation; useful where oseltamivir resistance is a concern, since it retains activity against H275Y.
• Peramivir. A neuraminidase inhibitor given intravenously, for patients unable to take the oral or inhaled routes.
• Baloxavir marboxil. Inhibits the polymerase acidic (PA) cap-dependent endonuclease, blocking the polymerase’s cap-snatching; a single oral dose, active against neuraminidase-inhibitor-resistant strains.
• Favipiravir. A broad-spectrum inhibitor of the viral RNA-dependent RNA polymerase, used for influenza in some settings.

• Mechanism. Foscarnet is a pyrophosphate analogue, not a nucleoside analogue. It inhibits the viral DNA polymerase directly, binding the pyrophosphate-binding site and blocking release of pyrophosphate during chain elongation, and it is not incorporated into the DNA.
• No activation step. It requires no intracellular phosphorylation by a viral kinase, so it stays active against thymidine-kinase-deficient (aciclovir-resistant) HSV (herpes simplex virus) and VZV (varicella-zoster virus) and UL97-mutant (ganciclovir-resistant) CMV. This is why it is the principal second-line agent for resistant herpesvirus disease.
• Toxicities. Nephrotoxicity is the most frequent. It also causes electrolyte disturbance, notably hypocalcaemia from chelation of ionised calcium (paraesthesia, tetany, seizures), with hypomagnesaemia and hypokalaemia, plus painful genital ulceration. It is given intravenously only.

• Thymidine kinase mutations (commonest). Loss of the enzyme (TK-negative), reduced production (TK-partial), or altered substrate specificity (TK-altered) prevent the first phosphorylation; TK-negative is the usual mechanism.
• DNA polymerase mutations (less common). Changes in the UL30 polymerase reduce binding of the drug.
• Context and consequence. Resistance is mainly a problem in the immunocompromised (advanced HIV, allogeneic transplant). TK-deficient isolates remain susceptible to foscarnet and cidofovir, which need no viral kinase.

• Mechanism. Oseltamivir is an oral prodrug whose active form inhibits the influenza neuraminidase, the enzyme that cleaves sialic acid to release progeny virions, leaving new virus tethered to the cell. It needs no intracellular activation and is active against influenza A and B.
• Clinical use. Treatment and post-exposure prophylaxis of influenza, of most value in severe or complicated disease and in those at high risk of complications; benefit is greatest when started within 48 hours.
• Limitations. The benefit is modest (it shortens symptoms by about a day, with little effect on hospitalisation or mortality), the 48-hour window limits real-world use, gastrointestinal upset is common, and the H275Y neuraminidase mutation can confer resistance.

A complete answer covers how resistance arises, how common it is and in whom, and what to use instead.

Mechanism

Aciclovir is a guanosine analogue that must first be activated by the viral thymidine kinase (TK) and then by cellular kinases; aciclovir-triphosphate inhibits the viral DNA polymerase and terminates the growing DNA chain. Resistance therefore arises in two ways:

• Thymidine-kinase mutations (much the commonest): TK-deficient, TK-low-producing, or TK with altered substrate specificity, all of which prevent activation of the drug.
• DNA polymerase mutations (less common), which can also confer cross-resistance to other polymerase inhibitors.

Prevalence

Resistance is rare in immunocompetent people (around 1% or less) but much more common in the profoundly immunocompromised, such as those with advanced HIV or haematopoietic stem-cell transplants, where prolonged high-level replication under drug pressure selects resistant virus.

Alternative treatment

• Foscarnet (a pyrophosphate analogue) and cidofovir (a nucleotide analogue) inhibit the viral DNA polymerase directly and do not need thymidine kinase, so they remain active against TK-mutant virus; both are limited by nephrotoxicity.
• Reducing immunosuppression where possible, topical options such as cidofovir gel, and newer helicase-primase inhibitors are additional considerations.

Both scenarios combine active and passive measures, but the neonatal aim is to interrupt vertical transmission before chronic infection is established, and the transplant aim is to prevent reactivation under profound immunosuppression.

(a) Neonate born to an HBsAg-positive mother

Without prophylaxis, transmission from HBeAg-positive mothers approaches 90% and over 90% of infected infants become chronic; combined active and passive prophylaxis prevents ~95% of cases.

• Antenatal (NDoH 2019): universal first-trimester HBsAg screening; if positive, quantitative HBV DNA, HBeAg and ALT. Maternal tenofovir from 28 to 32 weeks if HBV DNA exceeds 200,000 IU/mL, continued to 12 weeks postpartum if the only indication is transmission prevention.
• At birth: monovalent HBV birth-dose vaccine within 24 hours (10 micrograms intramuscular), co-administered with BCG and OPV, plus HBIG within 12 to 24 hours only if the mother is HBeAg-positive (the cost-rationed South African policy). EPI doses follow at 6, 10 and 14 weeks.
• Delivery and feeding: caesarean section is not indicated and breastfeeding is not contraindicated in a suppressed mother.
• Follow-up: post-vaccination serology (HBsAg and anti-HBs) at 9 to 18 months; anti-HBs of 10 mIU/mL or more confirms protection.

(b) HBsAg-positive patient before haematopoietic stem cell transplant (HSCT)

HSCT combines conditioning, prolonged neutropenia and graft-versus-host immunosuppression, and essentially all HBsAg-positive patients reactivate without prophylaxis, with high mortality.

• Pre-transplant: HBsAg and anti-HBc total (the latter to detect occult HBV, which still needs prophylaxis); if HBsAg-positive add HBV DNA, HBeAg/anti-HBe, fibrosis and HCC screening, with parallel HCV and HIV testing.
• Prophylaxis: entecavir or tenofovir (high genetic barrier), started before conditioning with HBV DNA ideally undetectable, and continued for at least 12 months after immunosuppression ends, often indefinitely given late-reactivation risk. Lamivudine is acceptable only for short courses at low baseline HBV DNA.
• Monitoring: HBV DNA and ALT every 1 to 3 months on treatment, and monthly for 3 months after stopping, watching for a severe late flare.

A complete answer covers the first-line nucleoside analogues, the agents for resistant disease, the newer class, and the resistance mechanism.

First-line. Aciclovir, its better-absorbed prodrug valaciclovir, and famciclovir (the penciclovir prodrug) are activated by the viral thymidine kinase and terminate viral DNA synthesis.

Resistant disease. Thymidine-kinase-deficient, aciclovir-resistant HSV, seen mainly in the immunocompromised, is treated with foscarnet or cidofovir, which need no viral kinase.

Newer option. The helicase-primase inhibitors (pritelivir, amenamevir) act on a different target and retain activity against aciclovir- and foscarnet-resistant HSV.

Aciclovir resistance. Most resistance is loss or alteration of the viral thymidine kinase, removing the activation step shared by aciclovir and famciclovir; DNA polymerase mutations are a less common route.

A complete answer defines both prevention strategies with their advantages and disadvantages, then links late-onset disease to the prophylactic approach.

Universal (prophylactic) therapy

Prophylaxis gives an antiviral (typically valganciclovir, or letermovir in seropositive allogeneic stem-cell recipients) to every at-risk patient for a fixed period after transplant, regardless of the viral load.

• Advantages: simple to deliver, needs no surveillance monitoring, reliably prevents early disease, and may additionally blunt the indirect immunomodulatory effects of CMV (rejection, other infections).
• Disadvantages: drug cost and toxicity (ganciclovir myelosuppression), selection of resistance, and, above all, late-onset disease once the drug stops.

Pre-emptive therapy

Pre-emptive therapy withholds the drug and instead monitors the viral load at regular (usually weekly) intervals, starting treatment only when replication crosses a threshold.

• Advantages: spares drug exposure, toxicity, cost and resistance, and treats only those who actually need it.
• Disadvantages: depends on reliable, frequent, standardised viral-load testing, can be outrun by rapidly rising viraemia between tests, and does not suppress the indirect effects in patients who never cross the threshold. It is often preferred in stem-cell transplant to avoid marrow toxicity.

Late-onset CMV disease

Late-onset disease is CMV disease appearing after a fixed prophylactic course ends, characteristically beyond three to six months. It occurs because prophylaxis suppresses replication throughout, so the patient is never exposed to enough virus to build CMV-specific T-cell immunity; when the drug stops, an unprimed and still-immunosuppressed patient becomes vulnerable. The highest-risk group is the seronegative recipient of a seropositive organ (D positive, R negative), in whom up to a third may be affected.

It is managed by surveillance after prophylaxis stops (switching to a pre-emptive approach for a period), risk-adapted prophylaxis duration, immune monitoring to judge when protective immunity has recovered, and secondary prophylaxis after a first episode. This trade-off, prophylaxis preventing early disease but causing late disease, is a central argument for individualising the strategy to the patient and the transplant centre.

Hepatitis B virus (HBV) is the prototype of the Hepadnaviridae (Baltimore class VII), unusual in using reverse transcription despite being a DNA virus and in maintaining a persistent nuclear mini-chromosome that current treatment cannot eliminate.

Replication cycle and drug targets

Attachment and entry. Low-affinity binding to heparan sulfate proteoglycans is followed by high-affinity binding of the myristoylated preS1 region to NTCP (sodium taurocholate co-transporting polypeptide), encoded by SLC10A1, delivering the nucleocapsid by endocytosis. Target: bulevirtide, a preS1-derived peptide that blocks NTCP (licensed for hepatitis delta; limited in HBV monoinfection).

cccDNA formation. The relaxed-circular DNA (rcDNA) is repaired into covalently closed circular DNA (cccDNA), which assembles with histones into a mini-chromosome that is the persistence reservoir. Target: cccDNA itself, by CRISPR/Cas9, base editors or epigenetic silencing, all investigational.

Transcription. Host RNA polymerase II transcribes cccDNA into the 3.5 kb pregenomic RNA (pgRNA) and the surface, core, polymerase and HBx messages. Targets: small interfering RNA (siRNA) and antisense oligonucleotides (JNJ-3989, VIR-2218, bepirovirsen) degrade HBV transcripts, lowering HBsAg.

Encapsidation and reverse transcription. pgRNA and polymerase are packaged into nucleocapsids nucleated at the epsilon signal, and polymerase reverse-transcribes pgRNA into rcDNA. Targets: nucleos(t)ide analogues (NUCs) (tenofovir, entecavir) inhibit the polymerase reverse-transcriptase; capsid assembly modulators block encapsidation.

Assembly, egress and integration. Mature nucleocapsids either recycle to the nucleus to amplify cccDNA or are enveloped and secreted as Dane particles alongside vast excess subviral HBsAg, using the ESCRT pathway. HBV DNA integration (X and S sequences) is a replication side-product that continues HBsAg production and contributes to HCC.

Effect on viral markers

Novel agents: capsid assembly modulators lower HBV DNA but not HBsAg; siRNA and ASO reduce HBsAg by 1 to 2 log10, breaking the decoy; immune modulators (TLR7/8 agonists, PD-1 inhibitors, therapeutic vaccines) restore exhausted T cells; cccDNA-targeting tools alone could achieve complete cure.

Endpoints

Sustained virological response (HBV DNA undetectable), HBeAg seroconversion (allowing finite therapy in HBeAg-positive disease), functional cure (sustained HBsAg loss, the realistic ceiling) and complete cure (cccDNA eradication, not achievable). The self-replenishing cccDNA mini-chromosome is the principal barrier to cure, compounded by integrated HBsAg output and T cell exhaustion, which is why the pipeline combines direct-acting and immune-restorative strategies.

A complete answer covers what genotyping and RAS testing once decided, how the pan-genotypic era narrowed both, and how South Africa organises them.

Genotyping

Eight major genotypes (1 to 8) differ by ~30 to 35% at the nucleotide level. Historically genotype drove regimen choice and duration in the interferon and early direct-acting antiviral (DAA) era, and genotype 3 carries prognostic weight through steatosis and faster fibrosis. In the pan-genotypic era (sofosbuvir/velpatasvir, glecaprevir/pibrentasvir) routine pre-treatment genotyping is often unnecessary, retained mainly for treatment-failure work-up, surveillance, and distinguishing reinfection from relapse.

South African genotype distribution

South Africa is pan-genotypic with a distinctive signature: genotype 5a was first identified here and stays largely confined to southern Africa. In the general population genotypes 1 and 5 co-dominate, with genotype 4 rising; in people who inject drugs, genotype 1a (~73%) and 3a (~15%) dominate and genotype 5 has not been detected.

Resistance-associated substitutions

RAS are amino acid changes that reduce DAA binding or efficacy, present at baseline or selected during therapy.

NS5A RAS matter most because they persist for years off therapy, whereas sofosbuvir’s high genetic barrier makes NS5B resistance rare. RAS testing is mainly indicated after DAA failure, to guide retreatment; it is not needed before first-line pan-genotypic therapy.

South African operational context

Genotyping runs on the NHLS network and RAS testing is centralised at a single national reference centre using polymerase-gene sequencing, reserved for treatment failure and selected complex cases. Pre-treatment RAS testing is not routine, since pan-genotypic regimens hold high cure rates across baseline RAS profiles.

A complete answer explains why current treatment suppresses but does not cure hepatitis B virus (HBV), then maps the cure pipeline onto the specific viral and immune barriers.

Three levels of cure are recognised: sustained virological response (HBV DNA undetectable on therapy, achievable now but not curative), functional cure (sustained HBsAg loss off therapy, the realistic target, reached in 3 to 7% on pegylated interferon-alpha and ~1% per year on tenofovir), and complete cure (cccDNA eradication, not achievable with any current agent).

Viral factors

The covalently closed circular DNA (cccDNA) mini-chromosome is the central barrier: a stable nuclear form that is the template for all viral transcripts, self-replenishing through nucleocapsid recycling, and untargeted by any licensed drug. Integrated HBV DNA cannot replicate but continues producing HBsAg and HBx indefinitely, so even a perfect cccDNA-clearing agent may leave residual HBsAg. Massive HBsAg over-production as subviral particles floods the circulation at 1,000 to 10,000 times the concentration of complete virions, acting as a decoy that exhausts HBV-specific B and T cells. Genotype heterogeneity (ten genotypes, A to J) and precore/basal core promoter mutants further modify antigen expression and immune visibility.

Immunological factors

Chronic antigen exposure drives HBV-specific T cell exhaustion, with upregulation of the inhibitory receptors PD-1, TIM-3, LAG-3 and CTLA-4 and loss of effector function, so the response cannot clear infection even under pharmacological suppression. Soluble HBeAg acts as a tolerogen, crossing the placenta to induce neonatal tolerance and dampening T cell costimulation. HBV also achieves innate immune evasion, with HBx and polymerase inhibiting the MAVS, IRF3/7 and STAT1 signalling that would drive type I interferon, earning it a “stealth virus” reputation. Regulatory T cell expansion, myeloid-derived suppressor cells and NK cell dysfunction reinforce the immunosuppressive intrahepatic environment.

The cure pipeline

The likely route to functional cure is combination therapy: an HBsAg-lowering agent (small interfering RNA or antisense oligonucleotide, such as bepirovirsen, VIR-2218 or JNJ-3989) to break the decoy effect, paired with an immune restorer (PD-1 checkpoint inhibitor, TLR7/8 agonist or therapeutic vaccine) to let the host clear residual cccDNA-expressing hepatocytes. Capsid assembly modulators and entry inhibitors (bulevirtide) reduce cccDNA replenishment, while direct cccDNA-targeting tools (CRISPR/Cas9, base editors) remain investigational and are the only strategies that could achieve complete cure.

A complete answer covers the activation pathway of each drug, their shared target, and how the difference explains when each is used.

Aciclovir and its selectivity

Aciclovir is a guanosine analogue. Its first phosphorylation is performed by the viral thymidine kinase, which is present only in infected cells, and cellular kinases then complete the conversion to aciclovir-triphosphate. This triphosphate inhibits the viral DNA polymerase and terminates the DNA chain. Selectivity arises because the activating first step happens essentially only in infected cells and because the drug acts on the viral, not the cellular, polymerase, so uninfected tissue is spared.

Cidofovir

Cidofovir is a nucleotide (monophosphate) analogue of cytosine that already carries a phosphonate group, so it does not need the viral thymidine kinase: cellular kinases alone activate it. It too inhibits the viral DNA polymerase. Being thymidine-kinase-independent, it stays active against TK-mutant virus.

Clinical translation

• Aciclovir (and its prodrug valaciclovir) is first-line for HSV because it is safe and selective.
• Cidofovir is reserved as salvage for TK-deficient aciclovir-resistant HSV in the immunocompromised (and is also used against cytomegalovirus); it is nephrotoxic and needs hydration and probenecid, which keeps it second-line.
• So the activation difference explains the roles: TK-dependence gives aciclovir its safety and front-line place, while TK-independence lets cidofovir work where resistance has emerged, at the cost of greater toxicity.

A complete answer gives the clinical trigger, the genes tested and what each implies, and the resulting choice of agent.

When to suspect resistance

Resistance is suspected when the viral load fails to fall, plateaus, or rises despite adequately dosed therapy, in a patient who is adherent and has had cumulative antiviral exposure (generally beyond two weeks of correct treatment and around four to six weeks of total exposure). Important caveats: a failure of the load to fall in the first two weeks alone is not reliable evidence, because the response can be slow, and resistance is more likely with prolonged or repeated drug exposure and in the most immunosuppressed patients. Genotyping is most informative when the viral load is reasonably high (above about 1000 copies/mL).

Genotypic testing and what it implies

Resistance is confirmed by sequencing the viral genes that the drugs target:

• UL97 (the viral kinase that activates ganciclovir): mutations here are the commonest and emerge first, conferring ganciclovir and valganciclovir resistance.
• UL54 (the DNA polymerase): mutations confer broader resistance, including cross-resistance to foscarnet and cidofovir and additional ganciclovir resistance.
• UL56 is sequenced if letermovir is failing, and UL27 in selected maribavir-refractory cases.

How it guides management

The mutation directs the switch. Isolated UL97 ganciclovir resistance is treated by changing to foscarnet or maribavir; broader UL54 resistance narrows the options and may need combinations or investigational agents, alongside reducing immunosuppression where possible to let immunity help clear the virus.

RNA genomes sit just below an error threshold that caps them near 30 kilobases; coronaviruses are the exception, with an nsp14 proofreading exonuclease. Pushing the rate past the threshold drives the population to extinction.

Treatment is decided by phenotype. Any central nervous system involvement, defined broadly to include abnormal neuroimaging, microcephaly, chorioretinitis or hearing loss, is treated for six months; symptomatic disease without central nervous system involvement is treated for six weeks. The drug of choice is oral valganciclovir at 16 mg/kg per dose twice daily, started before four weeks of age. A randomised trial established that six months preserves hearing and neurodevelopment better than six weeks. Asymptomatic infants with normal hearing are not treated.

Herpes simplex encephalitis is a medical emergency, so intravenous aciclovir is started empirically the moment it is suspected, because delay worsens outcome and the drug is safe; the standard course is 10 mg/kg every 8 hours for 14 to 21 days.

Waiting for PCR loses critical time, oral therapy gives inadequate central nervous system levels, and ganciclovir and foscarnet are neither first-line for HSV nor free of greater toxicity.

Active HBV co-infection (HBsAg positive) demands a nucleos(t)ide analogue started before or with the DAA and continued for at least 12 months after DAA completion. Hepatitis C normally suppresses HBV replication; rapid DAA-driven HCV clearance lifts that suppression, and HBsAg-positive patients reactivate almost universally without cover, sometimes with severe flares. Tenofovir disoproxil fumarate, tenofovir alafenamide, or entecavir are the high-genetic- barrier agents of choice.

A baseline HBV DNA below 2,000 IU/mL (A) does not remove reactivation risk in an HBsAg-positive patient. Deferral (B) is wrong: the infections are safely co-treated. Lamivudine monotherapy (C) has an unacceptable HBV resistance rate. Vaccination (D) is futile in someone already HBsAg positive.

Ganciclovir and its alternatives

• Ganciclovir and valganciclovir: the dose-limiting toxicity is myelosuppression, especially neutropenia (in roughly a third of patients), which is monitored by the full blood count and managed by dose adjustment or a granulocyte colony-stimulating factor. This matters particularly after stem-cell transplant, where it can impair engraftment.
• Foscarnet: limited by nephrotoxicity and electrolyte wasting (calcium, magnesium, potassium, phosphate), needing renal and electrolyte monitoring and hydration.
• Cidofovir: limited by nephrotoxicity (proximal tubular injury), given with probenecid and hydration, which restricts its use.

This complementary toxicity profile is useful: a patient who is neutropenic may be switched from ganciclovir to foscarnet, and one who is developing renal impairment away from foscarnet. The distractors misassign the toxicities (B), invent ones that do not apply (C, D), or wrongly claim these agents are free of toxicity (E).

Cytomegalovirus lies beyond aciclovir’s reach because it makes no thymidine kinase to activate the drug, relying instead on the UL97 kinase that ganciclovir exploits.

Aciclovir is active, to a decreasing degree, against HSV-1, then HSV-2, then VZV.

M2 is a proton channel that acidifies the virion interior to allow uncoating; amantadine and rimantadine block it. Near-universal resistance has retired them clinically, but they remain the classic uncoating inhibitor.

Thymidine-kinase-deficient HSV is cross-resistant to the kinase-dependent drugs, but foscarnet needs no viral kinase and stays active (as do cidofovir and pritelivir).

Valaciclovir, famciclovir, penciclovir and ganciclovir all depend on a viral kinase and are escaped.

The H275Y neuraminidase mutation reduces oseltamivir and peramivir binding but spares zanamivir, which remains active.

The adamantanes (amantadine, rimantadine) act on the M2 channel and are not used because of separate near-universal resistance.

Baloxavir inhibits the polymerase acidic (PA) subunit’s cap-dependent endonuclease, blocking the cap-snatching that primes viral messenger RNA synthesis.

Neuraminidase cleavage is the target of oseltamivir and zanamivir, and the M2 channel of the adamantanes; influenza has no reverse transcriptase, and fusion is not the target.

A brivudine metabolite irreversibly inhibits dihydropyrimidine dehydrogenase, the enzyme that clears 5-fluorouracil, so co-administration causes fatal fluoropyrimidine toxicity, and the two must be separated by several weeks.

The other classes have no comparable interaction with brivudine.

Why A

Maribavir penetrates the central nervous system and the eye poorly, so it should not be used for CMV encephalitis (or retinitis). It is an oral UL97-kinase inhibitor reserved for refractory or resistant CMV, and is effective for the other manifestations listed, but central-nervous-system or ocular disease calls for an agent that reaches those compartments, such as ganciclovir or foscarnet.

Foscarnet chelates ionised calcium, and co-administered pentamidine potentiates the resulting hypocalcaemia, which can cause paraesthesia, tetany or seizures.

The other agents listed do not share this interaction.

Brincidofovir is a lipid-ester prodrug of cidofovir, well absorbed orally and, by not concentrating in the renal tubule, largely free of the nephrotoxicity that limits cidofovir (its dose-limiting toxicity is diarrhoea instead).

It shares cidofovir’s mechanism and broad DNA-virus activity, and is now licensed for smallpox rather than a herpesvirus.

Molnupiravir’s active form is incorporated into viral RNA and base-pairs ambiguously, so mutations accumulate until the genome is no longer viable, a strategy called lethal mutagenesis or error catastrophe.

It is teratogenic and reserved for high-risk patients when alternatives are unsuitable; the other mechanisms belong to other drug classes.

SVR12 is undetectable HCV RNA 12 weeks after the last dose of direct-acting antiviral therapy, and is equivalent to virological cure with a lifetime late-relapse rate well below 1%. The defining feature is persistence of suppression off therapy, not on it. SVR4 predicts SVR12 with about 99% positive predictive value. Cure does not confer sterilising immunity, so reinfection remains possible in ongoing-risk groups, and HCC surveillance continues in cirrhotic patients.

A single on-treatment undetectable result (A) is not SVR12. Anti-HCV antibody persists for life and does not serorevert (B). A ten-fold fall at week 12 (C) was the interferon-era early virological response, now irrelevant. Core antigen (E) is an RNA surrogate, not the defining cure endpoint.

Ritonavir has no useful antiviral activity here; it is a CYP3A inhibitor that blocks the metabolism of nirmatrelvir and props up its levels (pharmacokinetic boosting), which is also the source of Paxlovid’s many drug interactions.

Nirmatrelvir is the active inhibitor of the SARS-CoV-2 main protease (Mpro).

Letermovir is a moderate CYP3A inhibitor, so it raises levels of the calcineurin inhibitors, particularly cyclosporine (and tacrolimus and several statins), requiring monitoring and dose adjustment.

The other interactions listed are not features of letermovir.

Letermovir blocks the terminase complex that cuts concatemeric DNA and packages it into preformed capsids, an assembly/packaging step distinct from the DNA polymerase that ganciclovir inhibits.

The same enzyme both matures the virus and disarms innate immunity: NS3/4A cleaves the four downstream non-structural polyprotein junctions, and it also cleaves the adaptors MAVS and TRIF, abolishing RIG-I and TLR3-driven type I interferon induction in the infected hepatocyte. The -previr protease inhibitors (glecaprevir, voxilaprevir, grazoprevir, and the obsolete telaprevir and boceprevir) block the active site and, as a bonus, spare MAVS and TRIF.

The structural junctions are cut by host signal peptidase, not NS3/4A, so D is wrong, and lamivudine has no HCV activity. NS3 does carry a separate helicase domain (B), but that is not the immune-evasion function, and sofosbuvir targets NS5B. NS3/4A disables rather than activates innate immunity, so C is wrong, and the -asvirs target NS5A. miR-122 is bound by the HCV RNA itself (E), the target of miravirsen, not by NS3/4A.

Trifluridine is phosphorylated by cellular rather than viral kinases, so it does not depend on the viral thymidine kinase and stays active against thymidine-kinase-deficient, aciclovir-resistant HSV.

It does not act on the terminase or on viral entry, and it is not selectively activated by the virus.

The only meaningful sofosbuvir escape mutation, S282T in the NS5B polymerase, cripples viral replication, so resistant virus is unfit and rarely survives, giving sofosbuvir a high barrier to resistance.

The other explanations are inaccurate.

Intravenous ganciclovir is preferred when gut absorption cannot be relied on (vomiting, gut graft-versus-host disease, critical illness) or when disease is severe or sight-threatening; oral valganciclovir suits stable patients and outpatient prophylaxis.

Ganciclovir is not obsolete; only the old oral ganciclovir capsule was superseded. A UL97 mutation calls for a different drug rather than a different route, and prophylaxis is usually oral.

The targets

• Letermovir inhibits the terminase (UL56) complex, which cleaves and packages the viral genome, a target unique to CMV. It is used for prophylaxis only and has no activity against other herpesviruses.
• Ganciclovir, foscarnet and cidofovir all act on the DNA polymerase (UL54).
• Maribavir inhibits the UL97 kinase.

Aciclovir lacks the 3’-hydroxyl needed to add the next nucleotide, so once it is incorporated DNA synthesis halts absolutely and the polymerase is frozen in a dead-end complex.

Ganciclovir, penciclovir and cidofovir retain a hydroxyl equivalent and are delayed (non-obligate) terminators that allow a few further bases before stalling; foscarnet is a pyrophosphate analogue that is not incorporated into the DNA at all.

Ganciclovir and zidovudine are both myelosuppressive, so together they cause severe additive neutropenia, often precluding combined use without growth-factor support.

Foscarnet adds nephrotoxicity rather than marrow toxicity, and the other pairings do not share this effect.

The covalently closed circular DNA (cccDNA) mini-chromosome is the cure barrier. Formed when host enzymes repair the incoming relaxed-circular DNA, it is the template for all viral transcripts, self-replenishing through nucleocapsid recycling, untargeted by any licensed agent, and persists for the lifespan of the long-lived hepatocyte. This is why the realistic ceiling is functional cure (sustained HBsAg loss) rather than complete cure (cccDNA eradication), and why investigational tools (CRISPR/Cas9, base editors, capsid assembly modulators, siRNA plus immune modulators) are built around depleting or silencing it.

The polymerase does mutate (A), but that drives resistance, not the cure barrier. The envelope (C) is an antibody target, not a cure obstacle. Tenofovir (D) has essentially no established resistance after a decade of use. Integration (E) contributes HBsAg and HCC risk but is a replication side-product; cccDNA persistence is the principal barrier.

Sofosbuvir/velpatasvir for 12 weeks is a pan-genotypic first-line regimen, with sustained virological response (SVR12) above 95% across all eight genotypes, and is the standard in the South African NDoH 2019 guideline. Glecaprevir/pibrentasvir (8 weeks in treatment-naive non-cirrhotic patients) is the other first-line option, and sofosbuvir plus daclatasvir is the low-cost generic alternative.

Pegylated interferon plus ribavirin (A) is the obsolete pre-DAA standard. Telaprevir and boceprevir (C) are withdrawn first-generation protease inhibitors. The drugs in D are antiretrovirals with no HCV activity. Ribavirin monotherapy (E) is not curative and is only ever an adjunct to DAAs.

Lamivudine resistance is mediated by mutations in the YMDD motif: M204V or M204I, often stabilised by L180M. This is the prototype nucleoside-analogue resistance pattern, reaching ~70 to 80% by five years and making lamivudine unsuitable as first-line monotherapy.

Adefovir resistance (B) is N236T (D domain) and A181T/V, not M204V. Entecavir (C) needs the lamivudine backbone (M204V/I) plus an entecavir-specific mutation, so N236T alone is wrong and naive resistance is very rare. Tenofovir (D) has no clinically established resistance, and M250V is an entecavir mutation. Telbivudine (E) selects M204I (cross-resistant with lamivudine), not T184G. First-line for chronic HBV is tenofovir or entecavir, and tenofovir remains fully active against lamivudine-resistant mutants.

The 45-year-old man (E) has HBeAg-negative chronic hepatitis with HBV DNA above the 2,000 IU/mL threshold, sustained ALT elevation, and APRI 1.4 indicating significant fibrosis, so treatment is indicated. He meets two independent NDoH 2019 criteria (the HBeAg-negative pattern and fibrosis evidence); start tenofovir or entecavir.

Patient A is phase 1 (immune tolerant): high DNA but normal ALT, no fibrosis, under 30, so the immune-tolerant threshold is not met, monitor. Patient B has coexisting HBsAg and anti-HBs with undetectable DNA and no active disease, investigate rather than treat. Patients C and D have no active HBV infection (isolated anti-HBc, and vaccine-induced immunity respectively). Treatment is also indicated regardless of these thresholds for cirrhosis or APRI above 2, HIV co-infection, pregnancy with HBV DNA above 200,000 IU/mL, and planned immunosuppression.

Sofosbuvir is a uridine nucleotide analogue prodrug that the HCV NS5B RNA-dependent RNA polymerase incorporates into the growing RNA chain, where its modified 2’-fluoro-2’-methyl sugar terminates chain extension. Its high genetic barrier makes clinical resistance rare, and it is the backbone of most pan-genotypic regimens.

The other options misassign the target. Ganciclovir inhibits the cytomegalovirus DNA polymerase (after UL97 phosphorylation), not HIV integrase. Simeprevir is an HCV NS3/4A protease inhibitor, not a neuraminidase inhibitor. Cabotegravir is an HIV-1 integrase strand-transfer inhibitor, not an HBV reverse transcriptase inhibitor. Lenacapavir is an HIV-1 capsid inhibitor, not a SARS-CoV-2 protease inhibitor.

Entecavir and tenofovir need several mutations to fail, so resistance is slow, whereas lamivudine’s single mutation emerges quickly; this high genetic barrier is why they are first-line.

None of these agents clears covalently closed circular DNA (cccDNA), so therapy stays long-term, and the other statements are incorrect.

A single M2 mutation confers adamantane resistance at no cost to viral fitness, so resistance spread to near-universal levels and the class was abandoned.

The adamantanes acted on the M2 proton channel of influenza A only (not B), and toxicity was not the reason for their fall.

Hepatitis C is an RNA virus with no latent DNA form, so sufficient suppression eradicates it, whereas hepatitis B persists as stable nuclear covalently closed circular DNA (cccDNA) that current drugs cannot clear.

The contrast is the biology of the reservoir, not drug potency, and the other statements are wrong.

Aciclovir and its relatives act only on the virus’s lytic replication. By the time mononucleosis presents the viral load has already peaked, and the symptoms are driven by the host T-cell response, not by ongoing viral replication. So antiviral treatment reduces oropharyngeal shedding but shortens neither the illness nor its course.

Epstein-Barr virus is a herpesvirus and is susceptible to aciclovir-class drugs in its lytic phase; the limitation is the immune-mediated nature of the disease, not drug resistance or absorption.

• UL97 (commoner). Mutations impair the phosphotransferase that activates ganciclovir. UL97-mutant CMV stays susceptible to foscarnet and cidofovir (which need no kinase) and to maribavir.
• UL54 (DNA polymerase). Mutations can confer cross-resistance across ganciclovir, cidofovir and sometimes foscarnet; combined with a UL97 mutation they give high-level resistance.
• Switching. Confirm by genotyping the UL97 and UL54 genes. Letermovir (terminase) and maribavir (UL97 kinase) act at targets the polymerase mutations do not affect.

• Mechanism. Maribavir is a direct competitive inhibitor of the CMV (cytomegalovirus) UL97 protein kinase; it is not phosphorylated and does not act on the DNA polymerase. Blocking UL97 impairs viral DNA synthesis and the nuclear egress of new capsids.
• Indication. Oral treatment of post-transplant CMV refractory to or resistant against ganciclovir, valganciclovir, foscarnet and cidofovir.
• Interaction with ganciclovir. Ganciclovir needs UL97 to be activated, and maribavir inhibits UL97, so maribavir antagonises ganciclovir and the two should not be combined.

• Indication. Treatment and prevention of cytomegalovirus (CMV) disease in the immunocompromised (retinitis, gastrointestinal disease, transplant CMV), and human herpesvirus 6 disease.
• Mechanism. Monophosphorylated in CMV-infected cells by the viral UL97 protein kinase (not a thymidine kinase); cellular kinases complete the triphosphate, which inhibits the viral DNA polymerase as a delayed chain terminator.
• Principal toxicity. Bone-marrow suppression, above all neutropenia (and thrombocytopenia), requiring full-blood-count monitoring; additive with zidovudine.
• Dosing. Intravenous 5 mg/kg twice daily for induction; oral valganciclovir gives equivalent exposure for maintenance and less severe disease.

• Indication. Intravenous treatment of COVID-19. Its clinical benefit has proved marginal, so use is limited and several authorities, South Africa included, advise against routine use.
• Mechanism. A prodrug of an adenosine analogue whose triphosphate the viral RNA-dependent RNA polymerase (RdRp) incorporates as a delayed chain terminator, stalling RNA synthesis.
• Note. It is incorporated efficiently enough to largely escape the coronavirus proofreading exonuclease, which is why it has any activity against a proofreading virus.

• Mechanism. Tenofovir is an acyclic nucleotide (adenosine monophosphate) analogue activated by cellular kinases to the diphosphate, which the HBV polymerase incorporates as a chain terminator. Its high genetic barrier to resistance makes it first-line; it suppresses replication but does not clear nuclear covalently closed circular DNA (cccDNA), so therapy is long-term.
• Side effects. The disoproxil prodrug (TDF) causes renal tubular toxicity and reduced bone mineral density; the alafenamide prodrug (TAF) has much less renal and bone effect. Tenofovir is also active against HIV, which matters in co-infection.

• Indication. HSV (herpes simplex virus) disease (genital herpes, herpes labialis, suppression) and VZV (varicella-zoster virus) disease (herpes zoster), and a step-down option in some severe presentations.
• Mechanism. The L-valyl ester prodrug of aciclovir, hydrolysed after absorption to aciclovir; the mechanism is then identical, with thymidine-kinase activation and obligate chain termination.
• Reason for use. Much better oral bioavailability, three to five times that of aciclovir, allowing less frequent dosing.

• Mechanism. A guanosine analogue with a debated, probably several-fold mechanism: depletion of the guanosine pool through inhibition of inosine monophosphate dehydrogenase, interference with messenger RNA capping, direct polymerase inhibition, and lethal mutagenesis (error catastrophe).
• Uses (for context). Now narrow: some viral haemorrhagic fevers (notably Lassa fever) and chronic hepatitis E in the immunocompromised; the historic roles in respiratory syncytial virus and hepatitis C are largely abandoned.
• Side effects. A dose-dependent haemolytic anaemia, and teratogenicity severe enough to contraindicate it in pregnancy and require contraception in both partners.

• Mechanism. Famciclovir is the oral prodrug of penciclovir, which the viral thymidine kinase activates; penciclovir triphosphate inhibits the viral DNA polymerase as a non-obligate (delayed) chain terminator and has a long intracellular half-life.
• Spectrum. HSV (herpes simplex virus) and VZV (varicella-zoster virus); it shares aciclovir’s thymidine-kinase dependence and therefore its resistance.
• Side effects. Very well tolerated, with headache and nausea the main complaints.

A complete answer covers the polymerase-targeting agents, the two newer agents with novel targets, and the resistance logic that selects between them.

First-line. Ganciclovir and its oral prodrug valganciclovir are first-line. They are activated by the viral UL97 kinase, and their dose-limiting toxicity is bone-marrow suppression.

Agents for resistant or intolerant disease. Foscarnet (a pyrophosphate analogue) and cidofovir (a nucleotide analogue) need no viral kinase, so both remain active against UL97-mutant ganciclovir-resistant CMV; both are nephrotoxic.

Newer agents. Letermovir inhibits the viral terminase and is used for prophylaxis, with no cross-resistance to the polymerase drugs but a low resistance barrier. Maribavir inhibits the UL97 kinase and is used for refractory or resistant disease, and must not be combined with ganciclovir.

Resistance-directed choice. UL97 mutations (commonest) retain foscarnet, cidofovir and maribavir; UL54 polymerase mutations can cross-resist the polymerase agents but not letermovir or maribavir. Genotyping the UL97 and UL54 genes guides the switch.