(Last updated: April 8, 2015; last reviewed: April 8, 2015)
|Rating of Recommendations: A = Strong; B = Moderate; C = Optional
Rating of Evidence: I = Data from randomized controlled trials; II = Data from well-designed nonrandomized trials or observational cohort studies with long-term clinical outcomes; III = Expert opinion
Antiretroviral (ARV) regimens currently recommended for initial therapy of HIV-infected patients have a high likelihood of achieving and maintaining plasma HIV RNA levels below the lower limits of detection (LLOD) of currently used assays (see What to Start). Patients on antiretroviral therapy (ART) who do not achieve this treatment goal or who experience virologic rebound often develop resistance mutations to one or more components of their regimen. Based on surveillance data for HIV patients in care in selected cities in the United States in 2009, an estimated 89% of the patients were receiving ART, of whom 72% had viral loads <200 copies/mL.1 Many patients with detectable viral loads are non-adherent to treatment. Depending on their treatment histories, some of these patients may have minimal or no drug resistance; others may have extensive resistance. Managing patients with extensive resistance is complex and usually requires consultation with an HIV expert. This section of the guidelines defines virologic failure in patients on ART and discusses strategies to manage these individuals.
Virologic Response Definitions
The following definitions are used in this section to describe the different levels of virologic response to ART.
Virologic suppression: A confirmed HIV RNA level below the LLOD of available assays
Virologic failure: The inability to achieve or maintain suppression of viral replication to an HIV RNA level <200 copies/mL
Incomplete virologic response: Two consecutive plasma HIV RNA levels ≥200 copies/mL after 24 weeks on an ARV regimen in a patient who has not yet had documented virologic suppression on this regimen. A patient’s baseline HIV RNA level may affect the time course of response, and some regimens will take longer than others to suppress HIV RNA levels.
Virologic rebound: Confirmed HIV RNA ≥200 copies/mL after virologic suppression
Virologic blip: After virologic suppression, an isolated detectable HIV RNA level that is followed by a return to virologic suppression
ART Treatment Goals and Virologic Response
The goal of ART is to suppress HIV replication to a level below which drug-resistance mutations do not emerge. Although not conclusive, the evidence suggests that selection of drug-resistance mutations does not occur in patients with HIV RNA levels persistently suppressed to below the LLOD of current assays.2
Viremia “blips”—defined by viral suppression followed by an isolated detectable HIV RNA level and subsequent return to undetectable levels—are not usually associated with subsequent virologic failure.3 In contrast, there is controversy regarding the clinical implications of persistent HIV RNA levels between the LLOD and <200 copies/mL in patients on ART. Furthermore, viremia at this threshold is detected with some frequency by commonly used real-time polymerase chain reaction (PCR) assays, which are more sensitive than PCR-based viral load platforms used in the past.4-6 Findings from a large retrospective analysis showed that, as a threshold for virologic failure, HIV RNA levels of 200 copies/mL and <50 copies/mL had the same predictive value for subsequent rebound to >200 copies/mL.7 Two other retrospective studies also support the supposition that virologic rebound is more likely to occur in patients with viral loads >200 copies/mL than in those with low-level viremia between 50 to 199 copies/mL.8,9 However, other studies have suggested that viremia at this low level (<200 copies/mL) can be predictive of progressive viral rebound10,11 and can be associated with the evolution of drug resistance.12
Persistent HIV RNA levels ≥200 copies/mL are often associated with evidence of viral evolution and accumulation of drug-resistance mutations.13 This association is particularly common when HIV RNA levels are >500 copies/mL.14 Therefore, persistent plasma HIV RNA levels ≥200 copies/mL should be considered virologic failure.
Causes of Virologic Failure
Virologic failure can occur for many reasons. Data from patient cohorts in the earlier era of combination ART suggested that suboptimal adherence and drug intolerance/toxicity accounted for 28% to 40% of virologic failure and regimen discontinuations.15,16 Presence of pre-existing (transmitted) drug resistance may also be the cause of virologic failure.17 Virologic failure may be associated with both patient- and regimen-related factors, as listed below:
ARV Regimen-Related Factors
Management of Patients with Virologic Failure
Assessment of Virologic Failure
If virologic failure is suspected or confirmed, a thorough assessment that includes consideration of the factors listed in the Causes of Virologic Failure section above is indicated. Often the causes of virologic failure can be identified, but in some cases, the causes are not obvious. It is important to distinguish among the causes of virologic failure because the approaches to subsequent therapy differ. The following potential causes of virologic failure should be explored in depth:
Suboptimal Adherence. Assess the patient’s adherence to the regimen. Identify and address the underlying cause(s) for incomplete adherence (e.g., drug intolerance, difficulty accessing medications, depression, active substance abuse) and, if possible, simplify the regimen (e.g., decrease pill count or dosing frequency). (See Adherence.)
Medication Intolerance. Assess the patient’s tolerance of the current regimen and the severity and duration of side effects, keeping in mind that even minor side effects can affect adherence. Management strategies to address intolerance in the absence of drug resistance may include:
Suspected Drug Resistance. Perform resistance testing while the patient is still taking the failing regimen or within 4 weeks of regimen discontinuation if the patient’s plasma HIV RNA level is >1,000 copies/mL (AI), and possibly even if between 500 to 1,000 copies/mL (BII). (See Drug-Resistance Testing.) In some patients, resistance testing should be considered even after treatment interruptions of more than 4 weeks, recognizing that the lack of evidence of resistance in this setting does not exclude the possibility that resistance mutations may be present at low levels (CIII). Evaluate the extent of drug resistance, taking into account the patient’s past treatment history and prior resistance-test results. Drug resistance is cumulative; thus, all prior treatment history and resistance test results should be considered when evaluating resistance. Genotypic or phenotypic testing provides information relevant for selecting nucleoside reverse transcriptase inhibitors (NRTIs), NNRTIs, PIs, and INSTIs. Additional drug-resistance tests for patients experiencing failure on a fusion inhibitor (AII) and viral tropism tests for patients experiencing failure on a CCR5 antagonist (BIII) are also available. (See Drug-Resistance Testing.)
Approach to Patients with Confirmed Virologic Failure.
Once virologic failure is confirmed, every effort should be made to assess whether suboptimal adherence and drug-drug or drug-food interactions may be contributing to the inadequate virologic response to ART. If virologic failure persists after these issues have been adequately addressed, resistance testing should be performed, and the regimen should be changed as soon as possible to avoid progressive accumulation of resistance mutations.18 In addition, several studies have shown that virologic responses to new regimens are greater in individuals with lower HIV RNA levels10,19 and/or higher CD4 cell counts at the time of regimen changes.10,19 Discontinuing or briefly interrupting therapy in a patient with viremia may lead to a rapid increase in HIV RNA and a decrease in CD4 cell count and increases the risk of clinical progression;20,21 therefore, this strategy is not recommended (AI). See Discontinuation or Interruption of Antiretroviral Therapy.
Ideally, a new ARV regimen should contain at least two, and preferably three, fully active drugs whose predicted activity is based on the patient’s drug treatment history, resistance testing, or the mechanistic action of a new drug class (AI).10,22-31 Despite drug resistance, some ARV drugs (e.g., NRTIs) may contribute partial ARV activity to a regimen,21 while other agents (e.g., enfuvirtide [T20], NNRTIs, the INSTI raltegravir [RAL]) likely will not.32-34 Using a “new” drug that a patient has never used previously does not ensure that the drug will be fully active; there is potential for cross-resistance, particularly among drugs from the same class. In addition, archived drug-resistance mutations may not be detected by standard drug-resistance tests, particularly if testing is performed when the patient is not taking the drug in question. Therefore, both treatment history and prior and current drug-resistance test results must be considered when designing a new regimen. When designing a new ART regimen, drug potency and viral susceptibility are more important factors to consider than the number of component drugs.
In general, patients who receive at least three active drugs, selected based on a review of the patient’s treatment history and past and most current drug-resistance test results, experience better and more sustained virologic response than those receiving fewer active drugs in the regimen.23,24,26,27,35,36 However, there are increasing data in treatment-naive and treatment-experienced patients showing that an active pharmacokinetically enhanced PI plus one other active drug or several partially active drugs will effectively reduce viral load in most patients.37-40 Active drugs are ARVs that, based on resistance test results and treatment history, are expected to have antiviral activity equivalent to that seen when there is no resistance to the specific drugs; ARVs with partial activity are those predicted to reduce HIV RNA but to a lesser extent than when there is no underlying drug resistance. The activity of a given drug must be uniquely defined for each patient. Active drugs may be newer members of existing drug classes that are active against HIV isolates that are resistant to older drugs in the same classes (e.g., etravirine [ETR], darunavir [DRV] and tipranavir, and dolutegravir [DTG]) An active drug may also be one with unique mechanisms of action (e.g., the fusion inhibitor T20, the CCR5 antagonist maraviroc in patients with no detectable CXCR4-using virus). In the presence of certain drug resistance mutations, some ARVs such as DTG, ritonavir-boosted DRV, and ritonavir-boosted lopinavir (LPV/r) need to be given twice daily instead of once daily to achieve higher drug concentrations necessary to be active against the less sensitive virus.41,42
Addressing Detectable Viral Load in Different Clinical Situations
HIV RNA above the LLOD and <200 copies/mL. Confirm that levels remain above the LLOD and assess adherence, drug-drug interactions (including those with over-the-counter products and supplements), and drug-food interactions. Patients with HIV RNA typically below the LLOD with transient increases in HIV RNA (i.e., blips) do not require a change in treatment (AII).5 Although there is no consensus on how to manage patients with persistent HIV RNA levels above the LLOD and <200 copies/mL, the risk of emerging resistance is believed to be relatively low. Therefore, these patients should maintain on their current regimens and have HIV RNA levels monitored at least every 3 months to assess the need for changes in ART in the future (AIII).
HIV RNA ≥200 and <1,000 copies/mL. Confirm that HIV RNA levels remain in this range and assess adherence and potential drug-drug interactions (including those with over-the-counter products and supplements) and drug-food interactions. In contrast to patients with HIV RNA levels persistently <200 copies/mL, those with levels persistently ≥200 copies/mL often develop drug resistance, particularly with HIV RNA levels >500 copies/mL.8,9 Persistent plasma HIV RNA levels in the 200 to 1,000 copies/mL range should be considered as virologic failure, and resistance testing should be attempted, particularly if HIV RNA >500 copies/mL. When resistance testing can successfully be performed and no resistance is detected, manage the patient as outlined below in the section on HIV RNA >1,000 copies/mL and no drug resistance identified . If drug resistance is detected, manage the patient as outlined below in the section on HIV RNA >1,000 copies/mL and drug resistance identified. When resistance testing cannot be performed because of low-level viremia, the decision whether to empirically change ARVs should be made on a case-by-case basis.
HIV RNA >1,000 copies/mL and no drug resistance identified. This scenario is almost always associated with suboptimal adherence. Conduct a thorough assessment to determine the level of adherence and identify any drug-drug and drug-food interactions. Consider the timing of the drug-resistance test (e.g., was the patient mostly or completely ART-non-adherent for more than 4 weeks before testing). If the current regimen is well tolerated and there are no significant drug-drug or drug-food interactions, it is reasonable to resume the same regimen. If the agents are poorly tolerated or there are important drug-drug or drug-food interactions, consider changing the regimen. Two to four weeks after treatment is resumed or started, repeat viral load testing; if viral load remains >500 copies/mL, perform genotypic testing to determine whether a resistant viral strain emerges (CIII).
HIV RNA >1,000 copies/mL and drug resistance identified. The availability of newer ARVs, including some with new mechanisms of action, makes it possible to suppress HIV RNA levels to below the LLOD in most of these patients. The options in this setting depend on the extent of drug resistance present and are addressed in the clinical scenarios outlined below.
Management of Virologic Failure in Different Clinical Scenarios
First Regimen Failure
Failing an NNRTI plus NRTI regimen. Patients failing an NNRTI-based regimen often have viral resistance to the NNRTI, with or without lamivudine (3TC) and emtricitabine (FTC) resistance. Although several options are available for these patients, several studies have explored the activity of a pharmacokinetically boosted PI with NRTIs or an INSTI.43-45 Two of the studies found that regimens containing a ritonavir-boosted PI (PI/r) combined with NRTIs were as active as regimens containing the PI/r combined with RAL.43,45 Two studies also demonstrated higher rates of virologic suppression with use of a PI/r plus NRTIs than with a PI/r alone.44,45 On the basis of these studies, even patients with NRTI resistance can often be treated with a pharmacokinetically boosted PI plus NRTIs or RAL (AI). Although LPV/r was used in these studies, it is likely that other pharmacokinetically boosted PIs would behave similarly. Although data are limited, the second-generation NNRTI ETR or the other INSTIs (i.e., elvitegravir [EVG] or DTG) combined with a pharmacokinetically boosted PI may also be options in this setting.
Failing a pharmacokinetically boosted PI plus NRTI regimen. In this scenario, most patients will have either no resistance or resistance limited to 3TC and FTC.46,47 Failure in this setting is often attributed to poor adherence, drug-drug interactions, or drug-food interactions. A systematic review of multiple randomized trials of PI/r first-line failure showed that maintaining the same regimen, presumably with efforts to enhance adherence, is as effective as changing to new regimens with or without drugs from new classes.48 In this setting, resistance testing should be performed along with an assessment of overall adherence and tolerability of the regimen. If the regimen is well tolerated and there are no concerns regarding drug-drug or drug-food interactions, the regimen can be continued with adherence support and viral monitoring. Alternatively, if poor tolerability or interactions may be contributing to virologic failure, the regimen can be modified to include a different pharmacokinetically boosted PI plus NRTIs—even if not all of the NRTIs are fully active—or to a new non-PI-based regimen that includes more than two fully active agents (AII).
Failing an INSTI plus NRTI regimen. Virologic failure with a regimen consisting of RAL plus two NRTIs or with EVG/cobicistat/tenofovir disoproxil fumarate/FTC may be associated with emergent resistance to 3TC and FTC and possibly the INSTI.49 Viruses with INSTI resistance often have virus still susceptible to DTG.19 In contrast, persons failing DTG plus two NRTI first-line therapy in clinical trials have not yet been shown to develop phenotypic resistance to DTG.49 There are no clinical trial data to guide therapy for first-line INSTI failures, although one can likely extrapolate from the data for NNRTI failures. Thus, patients with first-line INSTI failure should respond to a pharmacokinetically boosted PI plus NRTIs (AII). A pharmacokinetically boosted PI plus an INSTI may also be a viable option in patients with no INSTI resistance (BII). In the setting the virus is found to have resistance to RAL and EVG but remains susceptible to DTG, DTG can be used in combination with a pharmacokinetically boosted PI. If no resistance is identified, the patient should be managed as outlined above in the section on virologic failure without resistance.
Second-Line Regimen Failure and Beyond
Drug resistance with treatment options allowing for full virologic suppression. Depending on treatment history and drug-resistance data, one can predict whether or not to have a fully active pharmacokinetically boosted PI to include in future regimens. For example, those who have no documented PI resistance and previously have never been treated with an unboosted PI are likely to harbor virus that is fully susceptible to ARVs in the PI class. In this setting, viral suppression should be achievable using a pharmacokinetically boosted PI combined with either NRTIs or an INSTI—provided the virus is susceptible to the INSTI. If a fully susceptible pharmacokinetically boosted PI is not an option, the new regimen should include at least two, and preferably three, fully active agents, if possible. Drugs to be included in the regimen should be selected based on the likelihood that they will be active as determined by the patient’s treatment history, past and present drug-resistance testing, and tropism testing if a CCR5 antagonist is being considered.
Multidrug resistance without treatment options allowing for full virologic suppression. Use of currently available ARVs has resulted in a dramatic decline in the number of patients who have few treatment options because of multi-class drug resistance.50,51 Despite this progress, there remain patients who have experienced toxicities and/or developed resistance to all or most currently available drugs. If maximal virologic suppression cannot be achieved, the goals of ART will be to preserve immunologic function, prevent clinical progression, and minimize increasing resistance to drug classes that may eventually include new drugs that may be important for future regimens. Consensus on the optimal management of these patients is lacking. If resistance to NNRTIs, T20, EVG or RAL are identified, there is rarely a reason to continue these drugs, as there is little evidence that keeping them in the regimen helps delay disease progression (BII). Moreover, continuing these drugs, in particular INSTIs, may allow for further increasing resistance and within-class cross resistance that may limit future treatment options. It should be noted that even partial virologic suppression of HIV RNA to >0.5 log10 copies/mL from baseline correlates with clinical benefits.50,52 Cohort studies provide evidence that continuing therapy, even in the presence of viremia and the absence of CD4 count increases, reduces the risk of disease progression.53 Other cohort studies suggest continued immunologic and clinical benefits with even modest reductions in HIV RNA levels.54,55 However, all these potential benefits must be balanced with the ongoing risk of accumulating additional resistance mutations. In general, adding a single fully active ARV to the regimen is not recommended because of the risk of rapid development of resistance (BII).
Patients with ongoing viremia who lack sufficient treatment options to construct a fully suppressive regimen may be candidates for research studies or expanded access programs or may qualify for single-patient access of an investigational new drug as specified in Food and Drug Administration regulations: http://www.fda.gov/AboutFDA/CentersOffices/OfficeofMedicalProductsandTobacco/CDER/ucm163982.htm. Information about these programs may also be available from the sponsoring pharmaceutical manufacturer.
Isolated Central Nervous System (CNS) Virologic Failure and New Onset Neurologic Symptoms
Presentation with new-onset CNS signs and symptoms has been reported as a rare form of virologic failure. These patients present with new, usually subacute, neurological symptoms associated with breakthrough of HIV infection within the CNS compartment despite plasma HIV RNA suppression.56,57 Clinical evaluation frequently shows abnormalities on MRI brain imaging and abnormal cerebrospinal fluid (CSF) findings with characteristic lymphocytic pleocytosis. When available, measurement of CSF HIV RNA shows higher concentrations in the CSF than in plasma, and in most patients, evidence of drug-resistant CSF virus. Drug-resistance testing of HIV in CSF, if available, can be used to guide changes in the treatment regimen according to principles outlined above for plasma HIV RNA resistance (CIII). In these patients it may be useful to consider CNS pharmacokinetics in drug selection (CIII). If CSF HIV resistance testing is not available, the regimen may be changed based on the patient’s treatment history or on predicted drug penetration into the CNS58-60 (CIII). This “neurosymptomatic” CNS viral escape should be distinguished from: (1) other CNS infections that can induce a transient increase in CSF HIV RNA (e.g., herpes zoster61), (2) incidental detection of asymptomatic mild CSF HIV RNA elevation likely equivalent to plasma blips,62 or (3) relatively common chronic, usually mild, neurocognitive impairment in HIV-infected patients without evidence of CNS viral breakthrough.63 None of these latter conditions currently warrant a change in ART.64
In summary, the management of treatment-experienced patients with virologic failure often requires expert advice to construct virologically suppressive regimens. Before modifying a regimen, it is critical to carefully evaluate the cause(s) of virologic failure, including incomplete adherence, poor tolerability, and drug and food interactions, as well as review HIV RNA and CD4 cell count changes over time, treatment history, and drug-resistance test results. If HIV RNA suppression is not possible with currently approved agents, consider use of investigational agents through participation in clinical trials or expanded/single-patient access programs. If virologic suppression is still not achievable, the choice of regimens should focus on minimizing toxicity and preserving treatment options while maintaining CD4 cell counts to delay clinical progression.