Treatment of HIV-2 infection

INTRODUCTION — Although HIV-1 infection is associated with most of the global AIDS pandemic, HIV-2 is an important cause of disease in West Africa, where it is endemic [1]. HIV-2 infection is also found in areas with cultural and socio-economic ties to West Africa, including the United States. Although multiple studies suggest that HIV-2 is generally less pathogenic than HIV-1, the majority of people living with HIV-2 will eventually have disease progression, and mortality rates among persons with HIV-1 and HIV-2 who have advanced disease appear similar when adjusting for CD4 cell count, age, and sex [1-16]. Thus, antiretroviral therapy plays an important role in patient management.

This topic will address the approach to treatment in persons with HIV-2 infection. The epidemiology, natural history, clinical manifestations, and diagnosis of HIV-2 infection are discussed in detail elsewhere. (See "Epidemiology, transmission, natural history, and pathogenesis of HIV-2 infection" and "Clinical manifestations and diagnosis of HIV-2 infection".)

GOALS OF THERAPY FOR HIV-2 INFECTION — As with HIV-1 infection, the following benchmarks are critical goals of antiretroviral therapy (ART) [17,18]:

●Reduced morbidity and mortality

●Improvement in quality of life

●Durable suppression of HIV-2 viral load

●Restoration of immune function (as indicated by CD4 cell count)

●Prevention of HIV-2 transmission to uninfected sex or needle sharing partners

●Prevention of mother to child transmission

●Prevention of drug resistance

However, there are scant data on HIV-2 infection related to these basic principles. For example, the selection of ART for patients with HIV-2 is complicated by the absence of published randomized clinical trial data. Routine HIV-2 viral load testing is available in the United States at the University of Washington [19] and the New York State Department of Health [20]. However, drug resistance assays for HIV-2, which are considered key to the management of HIV-1 infection, are mainly relegated to research laboratories [18]. It is also unknown as to whether treatment-induced viral suppression of HIV-2 is associated with decreased transmission. (See "Epidemiology, transmission, natural history, and pathogenesis of HIV-2 infection", section on 'Risk of transmission'.)

INDICATORS OF IMMUNE FUNCTION — The CD4 cell count is the main indicator of immune function in patients who are infected with HIV-1 or HIV-2, and is a strong predictor of disease progression and survival [18,21,22]. Although patients with HIV-2 infection tend to have a slower immunologic decline than those with HIV-1, once the CD4 cell count is <200 cells/microL, patients with HIV-2 are also at risk for morbidity and mortality related to opportunistic infections [1-16]. Interestingly, levels of viremia in patients with HIV-2 tend to be much lower than those seen with HIV-1 infection, even at stages of advanced immunosuppression [23]. (See "Epidemiology, transmission, natural history, and pathogenesis of HIV-2 infection", section on 'Lower levels of viremia'.)

In patients with HIV-1 infection, it has been well-established that untreated patients with a CD4 count <200 cells/microL are at high risk for opportunistic infections and mortality. Guidelines by the United States Department of Health and Human Services [18], and the World Health Organization [24], recommend starting antiretroviral therapy (ART) irrespective of CD4 cell count to reduce the risk of serious AIDS- and non-AIDS-related complications. The CD4 cell count is used to guide whether to initiate chemoprophylaxis to prevent opportunistic infections. Discussions of when to initiate ART in patients with HIV-2 and the use of chemoprophylaxis to prevent opportunistic infections are found below. (See 'When to start antiretroviral therapy' below and 'Prevention and treatment of opportunistic infections' below.)

CLASSES OF ANTIRETROVIRAL MEDICATIONS — There are multiple classes of antiretroviral medications for the treatment of HIV-1 including (table 1). (See "Overview of antiretroviral agents used to treat HIV".)

However, there are no antiretroviral medications approved by the US Food and Drug Administrations for the treatment of HIV-2. In addition, HIV-2 has intrinsic resistance to certain classes of antiretroviral medications (ie, NNRTIs, fusion inhibitors), while within other classes (eg, PIs), not all agents are considered equally potent. (See 'Nonnucleoside reverse transcriptase inhibitors' below and 'Fusion inhibitors' below and 'Protease inhibitors' below.)

Our approach to antiretroviral therapy for patients with HIV-2 infection is described below. (See 'Selection of an initial antiretroviral regimen' below.)

HIV-2 DRUG RESISTANCE — Drug resistance data are important in guiding the selection of antiretroviral therapy regimens. Generally speaking, HIV drug resistance occurs when viral replication continues in the face of ongoing drug pressure. However, there are some important differences between HIV-1 and HIV-2 in terms of drug resistance patterns [25,26]. Much of the data below has emerged from specialized research laboratories; commercial HIV-2 drug resistance testing is not clinically available. Basic concepts and principles regarding HIV drug resistance are found elsewhere. (See "Interpretation of HIV drug resistance testing".)

Nucleoside reverse transcriptase inhibitors — Most data indicate that HIV-1 and HIV-2 are equally susceptible to all nucleoside reverse transcriptase inhibitors (NRTIs) currently in clinical use [27-33]. However, resistance data suggest that tenofovir (or abacavir) may be favored as a first-line agent over zidovudine, when available [33]. (See 'Assays for drug resistance testing' below.)

Drug resistance is prevalent in treatment-experienced patients with virologic failure. Commonly seen mutations include those that confer high-level resistance to lamivudine/emtricitabine (eg, M184V) and most NRTIs (eg, Q151M pathway) [34,35]. In vitro data suggest that as few as two NRTI mutations in HIV-2 (eg, Q151M and M184V) can confer full resistance to zidovudine and lamivudine [33,36].

In contrast, tenofovir and abacavir may maintain partial activity in the presence of mutations known to confer decreased susceptibility to other NRTIs (eg, K65R or Q151M alone or together); however, class-wide nucleoside resistance is demonstrated when all three major NRTI mutations are present (ie, K65R, Q151M and M184V) [33,37,38].

Drug resistance data also suggest some important differences between HIV-1 and HIV-2:

●Patients failing thymidine analogues (eg, zidovudine or stavudine) often do not have the expected thymidine analog mutations (TAMs) associated with the use of these drugs; this suggests that alternative resistance pathways may exist for HIV-2 compared with HIV-1 [34,39-42].

●Some NRTI mutations are selected more frequently in patients with HIV-2 compared with those with HIV-1 (eg, K65R or Q151M) [41,42].

●Two key NRTI thymidine analog mutations (TAMS) (M41L and T215Y), which have a negative impact on HIV-1 drug susceptibility to NRTIs, appear to have no effect on HIV-2 susceptibility to zidovudine [33].

Detailed information on thymidine analog mutations is found elsewhere. (See "Interpretation of HIV drug resistance testing", section on 'Nucleoside reverse transcriptase inhibitors'.)

Protease inhibitors — The protease inhibitor (PI) class has variable activity against HIV-2. PIs with the greatest in vitro activity include: lopinavir, darunavir, and saquinavir [18,43-45]. Drug susceptibility data suggest that atazanavir, amprenavir, indinavir, nelfinavir, and tipranavir have lower potency against HIV-2 isolates compared with HIV-1, and should be avoided [27,43-48].

The HIV-2 protease contains natural polymorphisms that are commonly associated with HIV-1 drug resistance, including major mutations (eg, V32I/L, M46I/V, I47V) and minor mutations (eg, L10VI, E35G/R, Q58E, A71V/I, G73A/T) [26,34,37,46,47,49-53]. These mutations are the likely reason for variable PI potency against HIV-2 [54]; in addition, these mutations may increase the risk of virologic failure in patients with HIV-2 during PI-based ART [39,53,55]. Detailed information on PI drug resistance mutation is found elsewhere. (See "Interpretation of HIV drug resistance testing", section on 'Protease inhibitors'.)

Integrase inhibitors — In vitro data suggest that integrase inhibitors (raltegravir, elvitegravir, dolutegravir, bictegravir, and cabotegravir [56]) have potent anti-HIV-2 activity [57-64]. Studies that have evaluated genotypic resistance among integrase inhibitor-naive patients with HIV-2 infection found no key mutations associated with resistance to integrase inhibitors at the pre-treatment baseline [58,65].

However, similar to HIV-1, the development of integrase inhibitor resistance during ART is a concern for HIV-2 infection [48,66]. In one small study of patients with HIV-2, virologic failure during raltegravir-containing combination ART was associated with the emergence of key signature mutations that confer raltegravir resistance [67]. Additionally, the key HIV-1 integrase inhibitor resistance mutations, Q148R and N155H, confer resistance to HIV-2 [59,67]. Although another key integrase inhibitor mutation, Y143C, does not confer significant resistance to HIV-2 by itself, it can cause resistance in conjunction with ancillary mutations [62,64]. Emerging data also demonstrate that the integrase inhibitor resistance mutations G118R and 231ins (5 amino acid insert at IN codon 231) confer resistance to first and second generation INSTIs [68,69]. (See "Interpretation of HIV drug resistance testing", section on 'Factors contributing to resistance' and "Interpretation of HIV drug resistance testing".)

CCR5 coreceptor antagonists — CCR5 coreceptor antagonists are effective against HIV-1 strains, which use the CCR5 coreceptor for cell entry. In vitro data and anecdotal reports suggest that CCR5 antagonists, such as maraviroc, may also be effective against HIV-2 as well [70-73]. However, the clinical effectiveness of this drug class remains uncertain since HIV-2 may use other co-receptors for cellular entry (eg, CXCR4, CCR1, CCR2b, CCR3, CXCR6 and GPR15) with greater affinity than HIV-1 [74-76]. Furthermore, the susceptibility of CCR5-tropic isolates from patients with HIV-2 and AIDS was lower than isolates from asymptomatic patients [77]. (See "Interpretation of HIV drug resistance testing", section on 'CCR5 antagonists'.)

Viral tropism assays to determine treatment candidacy for CCR5 inhibitors are not routinely available for clinical care, but assays for HIV-2 infection are being evaluated [78].

Nonnucleoside reverse transcriptase inhibitors — HIV-2 is intrinsically resistant to nonnucleoside reverse transcriptase inhibitors (NNRTIs), including second generation drugs, such as etravirine, rilpivirine, and doravirine [79,80]. Thus, this drug class should not be used in HIV-2 therapy and should be avoided in dual HIV-1/HIV-2 infection as well [81,82].

Fusion inhibitors — HIV-2 is intrinsically resistant to enfuvirtide and should not be used for treatment of patients with HIV-2 or dual infection [27,83,84].

Attachment inhibitors — Limited data suggest HIV-2 is resistant to the attachment inhibitor fostemsavir [85].

Post-attachment inhibitors — Limited in-vitro data suggests the post-attachment inhibitor, Ibalizumab, inhibits HIV-2 replication [86].

Capsid inhibitors — Limited in vitro data suggest the capsid inhibitor, lenacapavir, has activity against HIV-2 but is reduced compared to HIV-1 [87].

Investigational drugs — Limited in vitro data suggest the investigational nucleoside reverse transcriptase translocation inhibitor, islatravir (MK-8591, EFdA), has more potent activity against HIV-2 than HIV-1 [88]. By contrast, in vitro data suggest that HIV-2 is intrinsically resistant to maturation inhibitors, which are in development [89].

WHEN TO START ANTIRETROVIRAL THERAPY — There are no data regarding the optimal time to initiate antiretroviral therapy (ART) in HIV-2 infection, and guideline panels vary in their recommendations [18,82,90-92]. It is helpful to examine benchmarks for HIV-1 infection and how they may apply to HIV-2 infection.

HIV-1 infection — ART should be initiated in patients with HIV-1 infection, regardless of CD4 cell count. A detailed discussion of when to initiate ART is found elsewhere. (See "When to initiate antiretroviral therapy in persons with HIV" and "Use and impact of antiretroviral therapy for HIV infection in resource-limited settings", section on 'Initiation of ART'.)

HIV-2 infection — All patients with HIV-1/HIV-2 dual infection should initiate antiretroviral therapy. (See "When to initiate antiretroviral therapy in persons with HIV".)

In addition, we recommend initiating ART in patients with HIV-2 monoinfection who:

●Have an AIDS defining illness (see "The natural history and clinical features of HIV infection in adults and adolescents", section on 'AIDS-defining conditions')

OR

●Have a comorbid condition that may benefit from ART (eg, HIV-associated nephropathy and neurocognitive disorders, thrombocytopenia, chronic hepatitis B virus infection) (see "HIV-associated nephropathy (HIVAN)" and "HIV-associated cytopenias" and "HIV-associated neurocognitive disorders: Management" and "Treatment of chronic hepatitis B in patients with HIV")

OR

●Are pregnant (see "Antiretroviral selection and management in pregnant individuals with HIV in resource-rich settings" and "Prevention of vertical HIV transmission in resource-limited settings")

OR

●Are in a discordant sexual partnership with an HIV-2-negative individual or share needles with an injection partner who is HIV-2-negative.

We also suggest initiating ART in patients who have:

●A CD4 cell count below normal (ie, less than approximately 750 copies/microL)

OR

●A detectable plasma HIV-2 RNA (see "Clinical manifestations and diagnosis of HIV-2 infection", section on 'Testing for HIV-2 infection')

A shared decision-making approach should be used when deciding to start ART in asymptomatic patients with HIV-2 who have a normal CD4 cell count and an undetectable HIV-2 viral load. Many patients with HIV-2 who have an undetectable HIV-2 viral load and a normal CD4 count may never progress to AIDS or death [16,93-95]. Thus, the risks of taking ART must be weighed against the unknown benefits of improving clinical outcomes and preventing HIV-2 transmission.

CLINICAL DATA ON ANTIRETROVIRAL THERAPY OF HIV-2

Importance of viral suppression — There are conflicting data as to whether immune recovery (as measured by increases in CD4 cell counts), in response to antiretroviral therapy (ART), is as robust in HIV-2 as in HIV-1 infection [35,39,93,94,96,97]. However, among patients who attain viral suppression on ART, significant increases in CD4 counts are usually observed [96,98]. Viral suppression is also associated with a decreased risk of drug resistance [34,35].

HIV-2 monoinfection

Observational data — As in HIV-1 treatment, the majority of studies have utilized three-drug combination regimens. Observational studies for the treatment of HIV-2 have mainly included zidovudine-lamivudine in combination with the ritonavir-boosted protease inhibitor, lopinavir [99]. Subsequent studies from West Africa have included tenofovir disoproxil fumarate-lamivudine plus lopinavir/ritonavir as the primary regimen [100]. Antiretroviral medications were often chosen based on restricted availability in locales within West Africa, where many of these studies were performed [101,102].

The largest study to date included 422 patients with HIV-2 infection from West Africa [103]. Those treated with a boosted protease inhibitor (PI)-based regimen (which was typically zidovudine-lamivudine plus ritonavir-boosted lopinavir) had significantly higher increases in CD4 count compared with those treated with a triple NRTI regimen, or an unboosted PI-based regimen (191, 110, and 133 cells/microL, respectively). In this study, viral load data was generally not available.

Another large study from Senegal followed 291 participants for 926 person-years of follow-up over 13 years; most were treated with tenofovir disoproxil fumarate-lamivudine plus lopinavir/ritonavir starting in 2014 [100]. Over the study period, 76.7 percent of plasma viral loads were suppressed in patients receiving ART, and the median CD4 cell count gain during the first two years on study was 84 cells/microL.

Three smaller case series from Europe (76 patients), France (29 patients), and the Gambia (45 patients) also support the effectiveness of lopinavir/ritonavir as part of combination ART [23,96,99]. Within these cohort studies, ART led to viral suppression in the majority of patients with significant improvements in CD4 cell counts.

There is little clinical data on the use of other HIV-2 "active" PIs, such as darunavir. However, available data suggest that protease inhibitors should be given with low-dose ritonavir (or cobicistat) to boost serum levels of the parent drug (eg "pharmacokinetic boosting"). As an example, use of unboosted indinavir or nelfinavir was associated with high rates of virologic failure [34,35] and multiclass drug resistance [34], while ritonavir-boosted indinavir was associated with high rates of viral suppression in one small cohort [39]. (See "Overview of antiretroviral agents used to treat HIV", section on 'Protease inhibitors (PIs)'.)

Scant data are available on the use of other antiretroviral combinations. Similar to patients with HIV-1, triple NRTI regimens are inferior to PI-based regimens in HIV-2 treated patients [99,103]. By contrast, an increasing number of reports suggest that integrase inhibitors are clinically effective [57,58,104-106]. (See 'Integrase inhibitors' above.)

Only a few studies have looked at the efficacy of second-line or salvage regimens for HIV-2 infection. These regimens mainly used combinations that included boosted darunavir and dolutegravir; however, outcomes were generally suboptimal [107-109]. Thus, consultation with an HIV-2 expert is warranted when determining treatment options in the patient with virologic failure on antiretroviral therapy.

Clinical trials — Published data from 30 patients enrolled in a trial evaluating a single-tablet regimen of elvitegravir-cobicistat-emtricitabine-tenofovir disoproxil fumarate showed 28 of 30 patients had viral suppression/maintenance of control at 48 weeks (modified intention to treat), and the median increase in CD4 count was approximately 161 cells/microL [98]. One individual developed virologic failure and had multidrug-resistant HIV-2 (reverse transcriptase mutation K65R, integrase mutations G140S and Q148R) detected at week 48. Another pilot trial evaluating 30 patients with HIV-2 who received tenofovir disoproxil fumarate-emtricitabine plus raltegravir demonstrated a median CD4 count increase of 87 cells/microL, and 27 of 28 patients had a viral load of <40 copies/mL at 48 weeks [110]. The one individual with virologic failure had integrase inhibitor resistance mutations (E92Q, T97A, and Y143CGHR).

Several other clinical trials in ART-naïve patients have completed enrolment and are awaiting results; these include a trial comparing tenofovir disoproxil fumarate-emtricitabine with either raltegravir or lopinavir/ritonavir, as well as an open-label, single-arm trial of dolutegravir combined with either tenofovir disoproxil fumarate-emtricitabine or abacavir-lamivudine [111,112].

HIV-1 and HIV-2 dual infection — The only data on ART for HIV-1/HIV-2 dual infection is based on a few case reports and case series; outcomes were poor, immunologic and virologic failure rates were high, and ARV resistance (usually related to HIV-2) was common [55,113-116]. Consultation with an HIV-2 expert is warranted for dually-infected patients.

ASSAYS FOR DRUG RESISTANCE TESTING — In patients with HIV-1 infection, drug resistance testing is recommended at initial diagnosis, to assess for transmitted drug resistance (TDR), and at virologic failure, to assess for acquired drug resistance. (See "Overview of HIV drug resistance testing assays".)

A similar approach to testing is desirable for persons with HIV-2. However, drug resistance testing assays are not commercially available for HIV-2 infection, which hampers the ability to select the optimal medication regimen for treatment initiation or modification. Fortunately, transmitted drug resistance has rarely been reported in HIV-2 infection [117-120]. However, virologic failure in an patient with HIV-2 on ART is associated with acquired drug resistance that affects multiple classes of drugs [34,42,121].

Drug resistance testing may be obtained from research facilities such as the University of Washington (contact Dr. Geoffrey Gottlieb, [email protected]).

Interpretation of HIV-2 drug resistance genotyping can be obtained using the Stanford University HIV-2 Drug Resistance Database [26] or the European Union HIV-2 GRADE Drug Resistance Database [122].

SELECTION OF AN INITIAL ANTIRETROVIRAL REGIMEN

General approach — As noted above, there are no published randomized clinical trials evaluating the selection of antiretroviral medications for persons with HIV-2 infection, so the optimal strategy has not yet been defined. Guidance on antiretroviral selection for patients with HIV-2 largely depends upon in vitro drug susceptibility data from research laboratories and observational data from cohort studies and small single-arm trials. (See 'HIV-2 drug resistance' above and 'HIV-2 monoinfection' above.)

Until further data are available, two nucleoside reverse transcriptase inhibitors (NRTIs) should form the backbone of therapy and should be used in combination with a third agent from a different class, similar to the treatment of HIV-1. However, in contrast to HIV-1, the third drug should not include any agent from the nonnucleoside reverse transcriptase (NNRTI) class since HIV-2 is intrinsically resistant to these medications. Instead, an integrase strand transfer inhibitor (eg, raltegravir, elvitegravir, dolutegravir, or bictegravir), or alternatively, an HIV-2 active boosted protease inhibitor (eg, lopinavir, darunavir), should be used.

Simplicity (pill burden and dosing schedule) of the ART regimen is likely to improve adherence and reduce the risk of virologic failure. Thus, a single-tablet regimen is preferred, when possible. However, there are no data on using a two-drug regimen with an integrase inhibitor backbone (eg, dolutegravir plus lamivudine) for HIV-2 infection, and pending additional data these two drug regimens should not be used.

Preferred regimens — Based upon available in vitro and observational data, several antiretroviral regimens can be used for the treatment of HIV-2 infection. (See 'Clinical data on antiretroviral therapy of HIV-2' above.)

We suggest a nucleoside combination (tenofovir disoproxil fumarate-emtricitabine, tenofovir alafenamide-emtricitabine, or abacavir-lamivudine) be used with an integrase strand transfer inhibitor (raltegravir, elvitegravir/cobicistat, dolutegravir, or bictegravir). An alternative option is to use a nucleoside combination with a boosted protease inhibitor (ritonavir-boosted darunavir, ritonavir-boosted lopinavir). (See "Overview of antiretroviral agents used to treat HIV".)

Single-tablet INSTI-based regimens for HIV-2 include abacavir-lamivudine-dolutegravir, elvitegravir-cobicistat-emtricitabine-tenofovir disoproxil fumarate, elvitegravir-cobicistat-emtricitabine-tenofovir alafenamide, or bictegravir-emtricitabine-tenofovir alafenamide. Coformulated darunavir-cobicistat-emtricitabine-tenofovir alafenamide is available as a single-tablet protease inhibitor-based regimen.

If an abacavir regimen is being considered, patients must first be tested for the presence of HLA-B*57:01. Abacavir is contraindicated in patients who are HLA-B*57:01-positive. (See "Abacavir hypersensitivity reaction".)

The choice of regimen also depends upon the presence or absence of comorbid conditions (eg, reduced kidney function, hepatitis B virus infection), desire to become pregnant, and potential drug-drug interactions. More detailed discussions of how to select a regimen in women and patients with comorbid conditions are found elsewhere. (See "HIV and women" and "Selecting antiretroviral regimens for treatment-naïve persons with HIV-1: General approach", section on 'Additional considerations' and "Selecting antiretroviral regimens for treatment-naive persons with HIV-1: Patients with comorbid conditions".)

PATIENT MONITORING

HIV RNA assays — Viral load testing should be used to monitor response to therapy in patients with HIV-1, HIV-2, or with dual infection. A discussion of viral load monitoring during antiretroviral therapy (ART) is found elsewhere. (See "Patient monitoring during HIV antiretroviral therapy".)

However, unlike HIV-1, the availability of HIV-2 RNA testing is quite limited. It can be obtained from the following laboratories [19,20,123,124]:

●In the United States contact:

University of Washington Lab Medicine Community Services: (phone 1-800-713-5198 or [email protected]) https://dlmp.uw.edu/site/contact

●New York State Dept. of Health Lab: (phone 518-473-6007)

http://www.wadsworth.org/programs/id/bloodborne-viruses/clinical-testing/hiv-2-nucleic-acid

CD4 cell counts and clinical monitoring — In the absence of these basic tools for monitoring treatment responses in patients with HIV-2, clinical assessment and CD4 count monitoring take on additional importance. Declining CD4 cell counts on ART would be consistent with virologic failure and risk of disease progression. Any evidence of thrush would also be an early clinical indicator of a failing regimen. New opportunistic infections or clinical AIDS-associated events may also occur in the first months after initiation of ART related to advanced AIDS. Clinical and laboratory monitoring every three to six months is advised based on stage of infection. (See "Primary care of adults with HIV".)

Adverse events — Adverse events related to antiretroviral medications, and the type of monitoring needed for patients receiving ART, are discussed in detail elsewhere. (See "Overview of antiretroviral agents used to treat HIV" and "Patient monitoring during HIV antiretroviral therapy".)

PREVENTION AND TREATMENT OF OPPORTUNISTIC INFECTIONS — Until further data are available specifically for HIV-2, clinicians should follow guidelines for the prevention and treatment of opportunistic infections for HIV-1 [125]. (See "Overview of prevention of opportunistic infections in patients with HIV".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: HIV treatment in nonpregnant adults".)

SUMMARY AND RECOMMENDATIONS

●Goals of therapy – Although HIV-1 infection is associated with most of the global AIDS pandemic, HIV-2 is an important cause of disease in certain regions of the world where it is endemic, such as West Africa. (See 'Introduction' above.)

As in HIV-1 infection, the main goals of therapy are to attain viral suppression and improvement in CD4 cell counts, reduce mortality and morbidity, and reduce transmission to others. (See 'Goals of therapy for HIV-2 infection' above.)

●Indicators of immune function – The CD4 cell count is the main indicator of immune function in patients with HIV-2. Although patients with HIV-2 infection tend to have a slower immunologic decline than those with HIV-1, once the CD4 cell count is <200 cells/microL, patients with HIV-2 are at high risk for morbidity and mortality from opportunistic infections, as seen in patients with HIV-1 infection. (See 'Indicators of immune function' above.)

●HIV-2 drug resistance – HIV-2 has intrinsic resistance to some entire classes of antiretroviral medications, such as the nonnucleoside reverse transcriptase inhibitors (NNRTIs) and the fusion inhibitors. Within the nucleoside reverse transcriptase inhibitor (NRTI) class, tenofovir appears to have the best drug resistance profile, while within the protease inhibitor (PI) class, lopinavir, darunavir, and saquinavir have superior in vitro potency against HIV-2 compared with other PIs. Integrase strand transfer inhibitors have broad potent activity against HIV-2. (See 'Classes of antiretroviral medications' above.)

Drug resistance testing assays are not commercially available, which hampers the ability to check for resistance mutations prior to treatment initiation or during virologic failure on treatment. However, clinicians may obtain drug resistance testing from certain research facilities in the United States and Europe. (See 'Assays for drug resistance testing' above.)

●When to initiate antiretroviral therapy – Our approach to initiating antiretroviral therapy (ART) in patients with HIV- 2 is based in large part on data in patients with HIV-1 and vary to some extent based upon a patient's symptoms, risk for transmission, viral load, and CD4 count:

•For patients with HIV-2 infection who are symptomatic or have comorbidities that may improve with HIV therapy, we recommend initiation of ART (Grade 1B). For pregnant women and those who are in a discordant sexual partnership with an HIV-2-negative individual, we recommend initiation of ART to prevent transmission (Grade 1C). Therapy should be initiated regardless of the CD4 count or viral load. (See 'When to start antiretroviral therapy' above.)

•For asymptomatic patients who have a detectable viral load and/or a CD4 count below the lower limit of normal (eg, <750 cells/microL) we also suggest initiating ART (Grade 2C). However, there are no data regarding the optimal time to initiate therapy in such patients. (See 'When to start antiretroviral therapy' above.)

•For asymptomatic patients who have an undetectable viral load and a normal CD4 count, a shared decision-making approach should be used to determine when antiretroviral therapy should be initiated. Such patients may never progress to AIDS or death, so the risks of taking ART must be weighed against the unknown benefits of improving clinical outcomes and preventing HIV-2 transmission. (See 'When to start antiretroviral therapy' above.)

●Selecting an ART regimen – For treatment-naïve patients with HIV-2 infection, we suggest a three-drug combination therapy regimen with two nucleoside reverse transcriptase inhibitors (NRTIs) plus an integrase strand transfer inhibitor (Grade 2C). An alternative regimen should include two NRTIs plus a ritonavir-or cobicistat-boosted protease inhibitor active against HIV-2 (eg, lopinavir/ritonavir, darunavir/ritonavir, or darunavir/cobicistat). The use of these regimens is based upon in vitro and observational data from small nonrandomized trials. (See 'Selection of an initial antiretroviral regimen' above.)

●Patient monitoring – Because HIV-2 viral load and resistance testing are often not available, patient monitoring is usually dependent on clinical symptomatology and CD4 count monitoring. Clinical and laboratory monitoring every three to six months is desirable depending on the stage of disease. (See 'Patient monitoring' above.)