Antiviral drug resistance is not a random pharmacological misfortune but an inevitable evolutionary consequence of applying selection pressure to a replicating organism that generates genetic diversity faster than any other biological system. Understanding the principles governing resistance emergence — mutation rate, quasispecies structure, and fitness cost — provides the conceptual framework for interpreting resistance test results, designing combination regimens, and anticipating which pathogens pose the greatest long-term resistance threat.
Viral Mutation Rates and Error-Prone Replication. Ribonucleic acid (RNA) viruses replicate using RNA-dependent RNA polymerases (RdRp) that lack the 3'-to-5' exonuclease proofreading activity present in cellular deoxyribonucleic acid (DNA) polymerases. This absence of proofreading produces mutation rates of approximately 10-4 to 10-6 substitutions per nucleotide per replication cycle — six to ten orders of magnitude higher than cellular DNA replication error rates. For a virus with a genome of 10,000 nucleotides replicating at 1010 copies per day (as occurs in untreated human immunodeficiency virus (HIV) infection), this means every possible single-nucleotide substitution is generated multiple times daily within a single infected host, and many two-nucleotide combination mutants exist at low frequency before any drug is administered.1 DNA viruses such as herpesviruses replicate with higher fidelity, generating mutation rates of approximately 10-7 to 10-8 per nucleotide per cycle, producing a smaller preformed mutant spectrum but still generating drug-resistant variants under sustained antiviral selection pressure.
Quasispecies: The Mutant Swarm Model. Because RNA virus populations replicate at high mutation rates, the viral population within a single host is not a homogeneous collection of identical genomes but rather a dynamic swarm of closely related but genetically distinct variants — a quasispecies — distributed around a dominant consensus sequence. The quasispecies concept, developed by Eigen and Biebricher in the 1970s and applied to RNA virus biology by Holland and colleagues, profoundly changes the interpretation of viral drug resistance: resistance does not arise de novo by mutation after drug exposure but rather is selected from preexisting variants already present at low frequency in the quasispecies before therapy begins. The clinical implication is that drug-resistant minority variants present at frequencies as low as 1% to 20% of the viral population can rapidly become the dominant population under drug selection pressure, particularly when a resistance mutation imposes minimal fitness cost on the virus.3
Fitness Cost: The Determinant of Resistance Spread. The fitness cost of a resistance mutation — defined as the reduction in viral replicative capacity conferred by the mutation in the absence of drug — is the primary determinant of whether resistance spreads within and between patients. A resistance mutation that impairs the function of its target protein (for example, by distorting the active site geometry required for efficient enzyme catalysis) will reduce viral replication efficiency; this fitness cost means resistant variants replicate more slowly than wild-type virus and are outcompeted when drug pressure is removed. Conversely, resistance mutations that alter drug-binding geometry without disrupting catalytic function impose negligible fitness cost and may spread efficiently. The adamantane resistance mutation serine-to-asparagine at position 31 (S31N) in the influenza matrix protein 2 (M2) ion channel and the lamivudine resistance mutation methionine-to-valine/isoleucine at position 184 (M184V/I) in HIV reverse transcriptase (RT) illustrate these contrasting fitness profiles: S31N spread globally without drug pressure because it imposes near-zero fitness cost, while M184V/I is highly fit-costly for HIV and resuppressed rapidly upon lamivudine cessation in most patients.6
Combination Therapy: The Pharmacological Response to Quasispecies Diversity. The quasispecies model explains why combination antiviral therapy — the simultaneous use of multiple drugs targeting different viral proteins or mechanisms — is the pharmacological strategy required to durably suppress RNA viruses with high mutation rates. For a virus to replicate in the presence of two drugs targeting different proteins, it must simultaneously carry resistance mutations at both targets. The probability of a viral variant carrying two independent resistance mutations is approximately the product of the individual mutation frequencies: if each occurs at 10-5, the probability of simultaneous double mutation is approximately 10-10 — below the daily viral production rate for most pathogens, but not zero. Adding a third drug target reduces this probability to approximately 10-15, below the threshold for preformed triple-resistant variants even in heavily infected patients. This mathematical rationale underlies triple-drug regimens for HIV, the two-drug backbone requirements for hepatitis C virus (HCV) direct-acting antiviral (DAA) therapy, and the move toward combination strategies for emerging viral pathogens with pandemic potential.1
Single-drug antiviral therapy against high-mutation-rate RNA viruses selects for resistance predictably and rapidly. In untreated HIV infection, resistant variants to any single antiretroviral agent pre-exist in the quasispecies before treatment begins; the drug eliminates wild-type virus and the resistant variant expands to dominance within weeks to months. This principle applies equally to influenza (oseltamivir monotherapy in immunocompromised patients), HCV (NS3 inhibitor monotherapy), and any RNA virus with error-prone replication. Two or more mechanistically distinct drugs simultaneously are required to prevent resistance emergence.
Resistance testing has transformed antiviral management by enabling targeted selection of active drugs, minimizing exposure to ineffective agents with their associated toxicities, and providing surveillance data that guides empiric treatment recommendations at the population level. Two categorically different testing approaches exist — genotypic and phenotypic — each with distinct turnaround times, interpretive challenges, and clinical utilities.
Genotypic Resistance Testing. Genotypic resistance testing sequences the viral genome at drug target regions and compares the detected mutations against a curated database of mutations known or predicted to confer reduced drug susceptibility. For human immunodeficiency virus (HIV), genotypic testing covers the reverse transcriptase (RT), protease, and integrase genes, with additional sequencing of the gp41 envelope gene when entry inhibitor therapy is planned. The turnaround time for genotypic testing is typically 1 to 3 weeks for clinical laboratory sequencing. Standard genotypic testing reliably detects mutations present in more than 15% to 20% of the viral population; minority variants below this threshold are missed by conventional Sanger sequencing, which was the historical gold standard. Next-generation sequencing (NGS) — also termed deep sequencing or ultra-deep sequencing — can detect minority variants present at frequencies as low as 1% of the viral population, substantially increasing sensitivity for pre-existing resistance mutations that may not be apparent on standard genotype but could become dominant under selective drug pressure.4 The clinical significance of minority variants detected only by NGS at frequencies below 5% remains an area of active investigation, but variants above 5% to 10% are increasingly recognized as clinically meaningful for certain drug classes, particularly non-nucleoside reverse transcriptase inhibitors (NNRTIs).
Phenotypic Resistance Testing. Phenotypic resistance testing directly measures the ability of the patient's virus to replicate in the presence of increasing drug concentrations in cell culture, expressing results as the fold-change in the inhibitory concentration required to reduce viral replication by 50% (IC50) relative to a susceptible reference strain. Clinical cutoffs — defined as fold-change values below which treatment response is expected — have been established for most approved antiretroviral agents through correlation of baseline phenotype with virologic outcomes in clinical trials. Phenotypic testing is particularly valuable for highly mutated viruses in treatment-experienced patients, where multiple interacting mutations may produce complex resistance patterns that are difficult to predict from genotypic interpretation rules alone. The major limitations of phenotypic testing are longer turnaround time (3 to 4 weeks), greater cost, and technical complexity requiring specialized laboratory infrastructure. Virtual phenotype, which uses a database of paired genotype-phenotype results to predict phenotypic susceptibility from a genotypic sequence, provides phenotype-equivalent information at genotyping turnaround times and cost.4
Resistance Testing in Clinical Practice: HIV and Beyond. For HIV, resistance testing is recommended at the time of HIV diagnosis (transmitted resistance prevalence is approximately 10% to 15% in resource-rich settings), before initiating or changing antiretroviral therapy (ART), and at the time of virologic failure — defined as a confirmed plasma HIV ribonucleic acid (RNA) above 200 copies per milliliter while on therapy. Testing should ideally be performed while the patient is receiving the failing regimen, because wild-type virus re-emerges as the dominant population rapidly after selective drug pressure is removed, and the resistance mutations that drove failure may fall below the detection threshold of standard genotyping within weeks of stopping the selecting drug. For hepatitis C, resistance-associated substitution (RAS) testing guides treatment selection in patients with prior treatment failure, particularly for NS5A (nonstructural protein 5A) inhibitor-based regimens, where RAS testing may determine regimen choice and treatment duration. For cytomegalovirus (CMV), UL97 (CMV phosphotransferase gene) and UL54 (CMV DNA polymerase gene) genotypic testing guides management of suspected ganciclovir-resistant CMV, as discussed in the opportunistic infections module.5
HIV: order at diagnosis, before therapy initiation, and at virologic failure (on the failing regimen). Standard genotype detects mutations >15-20% frequency; NGS detects minorities >1%. Phenotype reserved for complex treatment-experienced patients. HCV: RAS testing before NS5A inhibitor re-treatment after failure. CMV: UL97 + UL54 genotype when viral load fails to decline ≥1 log10 after 2 weeks adequate ganciclovir.
Human immunodeficiency virus (HIV) drug resistance is the most extensively characterized antiviral resistance system in clinical medicine, with decades of surveillance data, validated resistance interpretation algorithms, and a pharmacological toolkit broad enough to construct fully active regimens for most treatment-experienced patients regardless of resistance history. Understanding the mechanistic basis of resistance to each antiretroviral drug class enables clinicians to anticipate cross-resistance patterns, interpret resistance reports, and design rational salvage regimens.
Nucleoside Reverse Transcriptase Inhibitor Resistance. Nucleoside reverse transcriptase inhibitors (NRTIs) are prodrugs activated by cellular kinases to triphosphate forms that compete with natural deoxynucleoside triphosphates (dNTPs) for incorporation into the growing deoxyribonucleic acid (DNA) chain, causing chain termination. Nucleoside reverse transcriptase inhibitor (NRTI) resistance arises through two distinct mechanisms. Discrimination mutations reduce the ability of HIV reverse transcriptase (RT) to incorporate the NRTI triphosphate relative to the natural dNTP substrate; M184V/I (methionine-to-valine/isoleucine at position 184) is the archetypal discrimination mutation, conferring high-level lamivudine (3TC) and emtricitabine (FTC) resistance with only modest reduction in replicative capacity. Excision mutations — principally the thymidine analog mutations (TAMs) — enhance the RT's ability to remove the incorporated chain-terminating NRTI through a pyrophosphorolysis mechanism, restoring chain elongation; Thymidine analog mutation (TAM) accumulation produces broad NRTI cross-resistance. The K65R (lysine-to-arginine at position 65) mutation, selected by tenofovir, abacavir, and didanosine, reduces the incorporation of multiple NRTIs but is not selected by zidovudine (ZDV) — indeed ZDV suppresses K65R emergence, explaining part of the rationale for ZDV-containing salvage regimens in specific situations.2
Non-Nucleoside Reverse Transcriptase Inhibitor (NNRTI) and Integrase Inhibitor Resistance. Non-nucleoside reverse transcriptase inhibitors (NNRTIs) bind a hydrophobic pocket adjacent to the RT active site, inducing a conformational change that restricts the flexibility required for polymerization. Single mutations in the NNRTI binding pocket — including K103N (lysine-to-asparagine at position 103, selected by efavirenz and nevirapine), Y181C (tyrosine-to-cysteine at 181), and G190A (glycine-to-alanine at 190) — confer high-level resistance across first-generation NNRTIs because the binding pocket geometry is altered by even a single amino acid substitution. Second-generation NNRTIs (etravirine, rilpivirine) have greater binding flexibility that allows them to maintain activity against many single-mutation NNRTI-resistant variants, though their activity is eroded by accumulation of multiple NNRTI resistance-associated mutations. Integrase strand transfer inhibitors (INSTIs) target the HIV integrase enzyme by chelating the magnesium ions required for strand transfer. First-generation INSTIs (raltegravir, elvitegravir) are vulnerable to single-signature resistance mutations (N155H, Q148H/R/K, Y143C/R) that directly alter the active site. Second-generation INSTIs (dolutegravir, bictegravir, cabotegravir) have higher genetic barriers to resistance, requiring the accumulation of two or more resistance mutations to meaningfully reduce susceptibility — dolutegravir failures with emerging resistance are rare in integrase strand transfer inhibitor (INSTI)-naive patients and have primarily been described in patients with highly complex treatment histories or subtherapeutic drug levels.7
Protease Inhibitor and Entry Inhibitor Resistance. HIV protease inhibitor (PI) resistance is characterized by the accumulation of multiple primary mutations in the active site or substrate binding cleft of the viral protease enzyme, supplemented by secondary or accessory mutations that restore replication capacity lost by primary mutations. Because clinically significant PI resistance typically requires 3 to 5 or more accumulated mutations, boosted PIs (ritonavir-boosted or cobicistat-boosted darunavir, lopinavir) have high genetic barriers to resistance. The entry inhibitor maraviroc inhibits the C-C chemokine receptor type 5 (CCR5) co-receptor and is active only against CCR5-tropic HIV; tropism testing (determination of whether the patient's virus uses CCR5, CXCR4 (C-X-C chemokine receptor type 4), or both co-receptors) is required before prescribing maraviroc, as treatment failure due to pre-existing CXCR4-tropic or dual/mixed-tropic virus will occur without this assessment. Ibalizumab, a monoclonal antibody targeting the cluster of differentiation 4 (CD4) receptor, and fostemsavir, an attachment inhibitor, provide additional options for multi-drug-resistant HIV that is no longer responsive to conventional antiretroviral classes.7
Aim for 2 to 3 fully active agents; count second-generation INSTIs (dolutegravir, bictegravir) as fully active unless INSTI resistance mutations are documented. Include a boosted PI (darunavir/r) as a backbone even with some PI mutations unless ≥3 primary PI mutations are present. Optimized background regimen activity score ≥2 associated with virologic suppression. Novel agents (ibalizumab, fostemsavir, lenacapavir) provide additional fully active options when conventional class coverage is exhausted. Always check for archived resistance in prior resistance test results — mutations driven below detection threshold by current regimen may re-emerge if that drug class is reintroduced.
Resistance in the hepatitis viruses illustrates contrasting evolutionary trajectories: hepatitis B virus (HBV) resistance emerged as a dominant clinical problem during the era of sequential nucleoside analog monotherapy and has been substantially controlled by the shift to high-barrier agents, while hepatitis C virus (HCV) resistance has become largely clinically irrelevant for most patients through the development of pangenotypic direct-acting antiviral (DAA) regimens with near-universal cure rates, though resistance-associated substitution testing retains specific roles in re-treatment planning.
HBV Drug Resistance: The rtM204 Pathway. Hepatitis B virus replicates through a ribonucleic acid (RNA) intermediate that is reverse-transcribed by the HBV polymerase (which functions as both a reverse transcriptase and a deoxyribonucleic acid (DNA)-dependent DNA polymerase) — a mechanism that, like human immunodeficiency virus (HIV), generates significant genetic diversity through error-prone replication. The critical distinction is that HBV has a much smaller genome and replicates at lower mutation rates than HIV, producing a more constrained mutational landscape. The archetypal HBV resistance pathway involves the rtM204V/I mutation (methionine-to-valine or isoleucine at reverse transcriptase position 204), which confers resistance to lamivudine (3TC) and other L-nucleoside analogs (telbivudine, emtricitabine) and typically co-occurs with rtL180M (a compensatory mutation that restores replication capacity impaired by rtM204V alone). The prevalence of HBV lamivudine resistance reached 70% after five years of monotherapy, representing one of the most dramatic examples of antiviral resistance emergence in clinical virology. Adefovir resistance (rtA181V/T, rtN236T) and the combination rtA181T/rtN236T double mutant emerged during adefovir monotherapy, with approximately 30% resistance prevalence by year five.8
High-Barrier HBV Agents: Entecavir and Tenofovir. Entecavir and tenofovir disoproxil fumarate (TDF), or its prodrug tenofovir alafenamide (TAF), represent the current standard of care for chronic HBV treatment, selected for their high genetic barriers to resistance. Entecavir resistance requires three simultaneous mutations — rtM204V/I plus rtL180M (already present in lamivudine-resistant strains) plus one of several additional mutations at positions rtI169, rtT184, rtS202, or rtM250 — making de novo entecavir resistance extraordinarily rare in treatment-naive patients; however, patients with pre-existing lamivudine resistance are at substantially elevated risk of developing entecavir resistance because the required rtM204V/I and rtL180M mutations are already present. Tenofovir resistance has not been definitively documented in clinical practice after more than 15 years of widespread use in HBV-infected patients, providing the highest resistance barrier of any approved antiviral agent in HBV management.8 Current guidelines recommend tenofovir or entecavir as first-line monotherapy for chronic HBV, with preference for tenofovir in patients with prior lamivudine or telbivudine treatment history due to the elevated entecavir resistance risk in that setting.
HCV Resistance-Associated Substitutions. Hepatitis C virus is an RNA virus with one of the highest mutation rates in virology — approximately 10-3 to 10-4 substitutions per nucleotide per replication cycle — and generates extensive pretreatment diversity at all drug target sites. The three primary HCV drug target classes are NS3/4A (nonstructural protein 3/4A) serine protease inhibitors, NS5A (nonstructural protein 5A) phosphoprotein inhibitors, and NS5B (nonstructural protein 5B) RNA-dependent RNA polymerase inhibitors (either nucleoside inhibitors or non-nucleoside inhibitors). Resistance-associated substitutions (RASs) at NS3 (including specific substitutions in genotype 1a), NS5A (including substitutions across multiple genotypes), and NS5B (rare sofosbuvir-associated substitutions) reduce susceptibility to agents targeting each protein. For modern pangenotypic regimens such as sofosbuvir-velpatasvir (SOF/VEL) and sofosbuvir-velpatasvir-voxilaprevir (SOF/VEL/VOX), baseline resistance-associated substitution (RAS) testing is not routinely required because cure rates exceed 95% even in the presence of known baseline RASs for most patient populations — the activity of sofosbuvir (an NS5B nucleoside inhibitor with exceptionally high resistance barrier) anchors these combinations against NS5A or NS3 RAS-mediated failures.9 RAS testing retains clinical utility for NS5A inhibitor-based re-treatment after prior failure, where NS5A RASs can persist for years and influence regimen selection and duration.
Never use lamivudine, telbivudine, or adefovir monotherapy for HBV — resistance rates exceed 30-70% by 5 years. Tenofovir (TDF or TAF) is the preferred first-line agent: no confirmed resistance in clinical practice after 15+ years. Entecavir is appropriate for treatment-naive patients but avoid in prior lamivudine-treated patients (rtM204V/I pre-existing mutations create elevated entecavir resistance risk). For HBV/HIV co-infected patients: tenofovir-based regimens treat both viruses simultaneously.
Herpesvirus drug resistance is uncommon in immunocompetent hosts but is a clinically significant problem in immunocompromised patients receiving prolonged antiviral therapy, where sustained drug selection pressure combined with deficient host immune surveillance allows resistant variants to emerge and persist. Because clinically significant herpesvirus resistance occurs almost exclusively in immunocompromised patients, resistance should be considered whenever antiviral therapy appears to be failing in this population.
Herpes Simplex Virus (HSV) Resistance: Thymidine Kinase Mutations and Treatment. Acyclovir-resistant herpes simplex virus (HSV) emerges almost exclusively in immunocompromised patients — particularly those with acquired immunodeficiency syndrome (AIDS), hematopoietic stem cell transplant (HSCT) recipients, and patients receiving prolonged acyclovir prophylaxis. The predominant resistance mechanism is loss-of-function mutation in the viral thymidine kinase (TK) gene, producing TK-null or TK-partial phenotypes in which viral TK enzyme is absent or has reduced activity, preventing phosphorylation of acyclovir to its active monophosphate form. Because acyclovir, valacyclovir, penciclovir, and famciclovir all require viral TK for activation, TK-null mutations confer cross-resistance to the entire TK-dependent drug class simultaneously. Less commonly, mutations in the viral deoxyribonucleic acid (DNA) polymerase gene (UL30 in HSV-1) confer acyclovir resistance through an activation-independent mechanism, potentially producing cross-resistance to foscarnet.10 Clinically, acyclovir-resistant HSV presents as progressive, atypical, non-healing mucocutaneous lesions in immunocompromised patients despite adequate acyclovir therapy. Standard management is foscarnet (90 mg per kilogram intravenously every 12 hours), which does not require viral TK for activation and directly inhibits the viral DNA polymerase at the pyrophosphate-binding site. Cidofovir is an alternative for foscarnet-refractory or TK-null/DNA polymerase double-mutant HSV, as cidofovir diphosphate is generated by cellular enzymes independently of viral TK.
Cytomegalovirus (CMV) Resistance: Phosphotransferase and Polymerase Mutation Patterns. Ganciclovir-resistant CMV emerges most commonly in solid organ transplant (SOT) and HSCT recipients who receive prolonged ganciclovir or valganciclovir prophylaxis or treatment, particularly those with high viral loads, inadequate drug exposure due to subtherapeutic dosing or poor absorption, or profound T-cell immune deficiency preventing immune-mediated viral clearance. The primary resistance pathway involves mutations in the UL97 (CMV phosphotransferase) gene — particularly mutations at codons 460, 594, 595, and the region around codon 520 — which impair UL97-mediated phosphorylation of ganciclovir to its monophosphate without affecting UL54 (CMV DNA polymerase) function. UL97-only mutations confer ganciclovir resistance while leaving foscarnet and cidofovir fully active, because neither agent requires UL97 for activation. Secondary resistance involving the UL54 (CMV DNA polymerase) gene may produce cross-resistance to foscarnet, cidofovir, or both, depending on the specific mutation location within the polymerase domain; combined UL97 plus UL54 mutations represent the most clinically challenging resistance scenario.5 Maribavir, a benzimidazole riboside that inhibits UL97 kinase through a binding site distinct from ganciclovir, retains activity against many UL97 resistance mutations (particularly those at codons 460 and 595) and is FDA-approved for treatment of refractory or resistant CMV infection and disease in transplant recipients.
Varicella-Zoster Virus (VZV) Resistance and Clinical Recognition. Acyclovir-resistant varicella-zoster virus (VZV) is rare but well-documented in immunocompromised patients, arising through the same TK gene mutation mechanism as acyclovir-resistant HSV. Clinical presentation is atypical — progressive verrucous or hyperkeratotic skin lesions that fail to evolve through the expected vesicular stages, often without the dermatome distribution of typical zoster, particularly in AIDS patients with very low cluster of differentiation 4 (CD4) counts. The absence of a typical vesicular eruption and failure to heal despite adequate acyclovir therapy in an immunocompromised patient should prompt VZV resistance testing (TK gene sequencing) and empiric switch to foscarnet. Because foscarnet does not require viral TK activation, TK-null VZV strains remain fully foscarnet-susceptible. Cidofovir is an additional option for foscarnet-refractory VZV resistance, though clinical experience is limited. The resistance monitoring principle applicable across all three herpesviruses is that antiviral failure in an immunocompromised patient warrants virologic rather than pharmacokinetic evaluation as the first diagnostic step, since inadequate antiviral levels (due to poor absorption, dosing errors, or drug interactions) and true drug resistance produce clinically identical presentations and require categorically different management responses.10
Suspect resistance when: progressive HSV/VZV lesions despite 10+ days adequate acyclovir in an immunocompromised patient, or CMV viral load fails to decline ≥1 log10 after 2 weeks adequate ganciclovir. Confirm: HSV/VZV — viral TK and DNA polymerase gene sequencing from lesion swab. CMV — UL97 + UL54 genotype from plasma or tissue. Always check drug levels and adherence first to exclude pharmacokinetic failure before attributing to resistance.
The coronavirus disease 2019 (COVID-19) pandemic demonstrated with unprecedented clarity that resistance considerations must be integrated into antiviral development, deployment, and surveillance strategies from the earliest stages of a pandemic response — not as an afterthought after widespread antiviral use has already established selection pressure. The pharmacological lessons from influenza, human immunodeficiency virus (HIV), and hepatitis C virus (HCV) resistance provide a ready blueprint for anticipating and mitigating antiviral resistance in emerging viral threats.
Surveillance Infrastructure for Emerging Resistance. Effective antiviral resistance surveillance requires three integrated components: systematic virologic monitoring of treated patients to detect phenotypic or virologic treatment failure, molecular characterization of viral isolates from failing patients to identify the causative resistance mutations, and epidemiological tracking of resistance mutation prevalence in treatment-naive patients to detect transmitted resistance. For established pathogens, these functions are served by national and international networks — the World Health Organization (WHO) Global Influenza Surveillance and Response System (GISRS) for influenza, the Stanford HIV Drug Resistance Database (HIVdb) for HIV, and the European Association for the Study of the Liver (EASL) resistance panel for HCV — that maintain curated mutation databases, distribute consensus resistance interpretation algorithms, and publish annual surveillance reports.11 For novel emerging pathogens, building this surveillance infrastructure from scratch during an active outbreak is both urgent and challenging; the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic demonstrated that genomic sequencing networks established for other purposes (influenza surveillance, clinical microbiology) could be rapidly repurposed to track emerging antiviral resistance, though the absence of validated resistance interpretation algorithms meant that new mutations had to be characterized functionally before their clinical significance could be assessed.
SARS-CoV-2 Resistance: Early Lessons. Resistance to SARS-CoV-2 antivirals has emerged at a clinically meaningful rate for some agents but not others, reflecting the differing genetic barriers of available drugs. Nirmatrelvir (the active component of nirmatrelvir-ritonavir) targets the viral main protease (Mpro), and resistance mutations in the Mpro gene have been identified in immunocompromised patients receiving prolonged or repeated nirmatrelvir-ritonavir courses — mirroring the pattern of herpesvirus resistance emerging preferentially in immune-deficient hosts on sustained antiviral therapy. Remdesivir resistance (mutations in the ribonucleic acid (RNA)-dependent RNA polymerase nsp12 subunit) has been documented in immunocompromised patients, again paralleling the established pattern of resistance emergence under prolonged drug selection in the absence of immune clearance. The SARS-CoV-2 experience has reinforced a general principle: resistance emergence risk is highest in severely immunocompromised patients who cannot clear virus immunologically and who therefore require prolonged antiviral therapy, creating sustained selection pressure without the immune co-factor that eliminates residual low-level viral replication in immunocompetent hosts.1
Combination Antiviral Strategies for Pandemic Preparedness. The pharmacological response to the threat of antiviral resistance in pandemic contexts follows the same principle established for HIV and HCV: combination therapy targeting multiple viral proteins simultaneously raises the genetic barrier to resistance to a level that exceeds the replicative capacity of the pathogen to generate double or triple resistant mutants under normal selective conditions. For influenza pandemic preparedness, the combination of baloxavir (cap-dependent endonuclease inhibitor targeting the polymerase acidic (PA) protein) with a neuraminidase inhibitor targets two distinct and mechanistically non-overlapping viral proteins, substantially reducing the probability of resistant variant emergence compared with either agent alone. For coronaviruses, combining an Mpro inhibitor (nirmatrelvir) with an RNA-dependent RNA polymerase inhibitor (remdesivir or molnupiravir) provides complementary mechanisms that have been investigated in animal models and clinical trials. Stockpiling strategies for pandemic antivirals must account for resistance probability, including maintaining diversity in the stockpile (drugs with different mechanisms) rather than concentrating stockpiles in a single agent whose resistance would leave the entire strategy compromised.11
1. Establish resistance surveillance infrastructure before widespread antiviral deployment, not after. 2. Stockpile mechanistically diverse agents — single-drug stockpiles create single-point-of-failure resistance risk. 3. Prioritize combination therapy for high-risk or immunocompromised patients with the greatest resistance emergence probability. 4. Monitor immunocompromised patients receiving antiviral therapy most intensively — they are the primary source of resistance emergence for most antivirals. 5. Validate resistance interpretation algorithms for each new pathogen before widespread clinical deployment of resistance testing.
Domingo E, Sheldon J, Perales C. Viral quasispecies evolution. Microbiol Mol Biol Rev. 2012;76(2):159–216.
doi:10.1128/MMBR.05023-11Coutsinos D, Invernizzi CF, Xu H, et al. Template usage is responsible for the preferential acquisition of the K65R reverse transcriptase mutation in subtype C variants of human immunodeficiency virus type 1. J Virol. 2009;83(4):2029–2033.
doi:10.1128/JVI.01349-08Vignuzzi M, Stone JK, Arnold JJ, Cameron CE, Andino R. Quasispecies diversity determines pathogenesis through cooperative interactions in a viral population. Nature. 2006;439(7074):344–348.
doi:10.1038/nature04388Paredes R, Clotet B. Clinical management of HIV-1 resistance. Antiviral Res. 2010;85(1):245–265.
doi:10.1016/j.antiviral.2009.09.015Limaye AP, Babu TM, Boeckh M. Progress and challenges in the prevention, diagnosis, and management of cytomegalovirus infection in transplantation. Clin Microbiol Rev. 2020;34(1):e00043-19.
doi:10.1128/CMR.00043-19Clavel F, Hance AJ. HIV drug resistance. N Engl J Med. 2004;350(10):1023–1035.
doi:10.1056/NEJMra025195Wensing AM, Calvez V, Ceccherini-Silberstein F, et al. 2019 update of the drug resistance mutations in HIV-1. Top Antivir Med. 2019;27(3):111–121.
IAS-USA Drug Resistance Mutations 2019Lok AS, McMahon BJ, Brown RS Jr, et al. Antiviral therapy for chronic hepatitis B viral infection in adults: a systematic review and meta-analysis. Hepatology. 2016;63(1):284–306.
doi:10.1002/hep.28280Pawlotsky JM. Hepatitis C virus resistance to direct-acting antiviral drugs in interferon-free regimens. Gastroenterology. 2016;151(1):70–86.
doi:10.1053/j.gastro.2016.04.003Piret J, Boivin G. Antiviral resistance in herpes simplex virus and varicella-zoster virus infections: diagnosis and management. Curr Opin Infect Dis. 2016;29(6):654–662.
doi:10.1097/QCO.0000000000000288Hurt AC. The epidemiology and spread of drug resistant human influenza viruses. Curr Opin Virol. 2014;8:22–29.
doi:10.1016/j.coviro.2014.04.009