See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/8480095

Highly Active Antiretroviral Therapy and Viral Response in HIV Type 2 Infection

Article  in  Clinical Infectious Diseases · July 2004

DOI: 10.1086/421390 · Source: PubMed

CITATIONS READS
42 20

14 authors, including:

Christopher Mullins Geoffrey Eisen

Harvard University Northwestern University
22 PUBLICATIONS   495 CITATIONS    27 PUBLICATIONS   1,855 CITATIONS   

SEE PROFILE SEE PROFILE

Stephen Popper Abdoulaye Dieng Sarr

University of California, Berkeley U.S. Department of Health and Human Services
73 PUBLICATIONS   2,795 CITATIONS    116 PUBLICATIONS   1,771 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Serum Resistance in Moraxella catarhalis View project

Vertebrate lectins View project

All content following this page was uploaded by Christopher Mullins on 06 January 2016.

The user has requested enhancement of the downloaded file.

 MAJOR ARTICLE HIV/AIDS

Highly Active Antiretroviral Therapy and Viral

Response in HIV Type 2 Infection
Christopher Mullins,1 Geoffrey Eisen,1 Stephen Popper,6 Abdoulaye Dieng Sarr,1 Jean-Louis Sankalé,1
Judith J. Berger,7 Sharon B. Wright,2 Hernan R. Chang,3 Gérard Coste,5 Timothy P. Cooley,3,4 Peter Rice,3
Paul R. Skolnik,3 Margaret Sullivan,3 and Phyllis J. Kanki1
1
Department of Immunology and Infectious Diseases, Harvard University School of Public Health, 2Division of Infectious Diseases, Beth Israel
Deaconess Medical Center, 3Center for HIV/AIDS Care and Research, Department of Medicine, Boston University Medical Center, 4Sections of
Hematology/Oncology and Infectious Diseases, Department of Medicine, and Boston University School of Medicine and Boston Medical Center,
Boston, 5Department of Infectious Diseases, Cambridge Hospital, Cambridge, Massachusetts; 6Department of Microbiology and Immunology,

Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015

Stanford University School of Medicine, California; and 7Department of Infectious Diseases, St. Barnabas Hospital, Bronx, New York

Human immunodeficiency virus type 2 (HIV-2), the second human retrovirus known to cause AIDS, is endemic
to West Africa but is infrequently found outside this region. We present a case series of 10 HIV-2—infected
individuals treated in the United States. Physicians applied the principles of highly active antiretroviral therapy
(HAART), normally used in treating HIV type 1, with modifications considered appropriate for treating HIV-
2. CD4+ cell count, HIV-2 virus load, and clinical status were found to correlate well, providing evidence that
HIV-2 virus load is useful in managing treatment of patients with HIV-2 who are receiving therapy. However,
HAART regimens with predicted efficacy for treatment of HIV type 1 infection are not as efficacious for
treatment of HIV-2. Controlled clinical trials of HIV-2–infected patients receiving various HAART regimens
are needed to provide therapeutic guidance to the medical community.

HIV type 2 (HIV-2) was identified in 1985 in Senegal,
West Africa, as the second immunodeficiency virus to
infect humans [1–3]. Like HIV type 1 (HIV-1), HIV-
2 is known to cause AIDS; however, HIV-2 is com-
paratively less pathogenic than HIV-1. Since its iden-
tification, the epidemic of HIV-2 infection has
remained largely confined to West Africa [4]. However,
evidence of an HIV-2 epidemic has been reported in
India [4], along with modest prevalence in Europe and
the United States [5]. HIV-2 is less transmissible than
HIV-1, with a 5- to 10-fold lower rate of heterosexual
transmission [6, 7] and a 20- to 30-fold lower rate of
mother-to-child transmission [8–10]. The natural his-
tory of HIV-2 infection is distinguished by a longer
asymptomatic phase than that of HIV-1 infection [11],
Received 5 November 2003; accepted 10 February 2004; electronically published
25 May 2004.
Reprints or correspondence: Dr. Phyllis Kanki, Harvard School of Public Health,
Dept. of Immunology and Infectious Diseases, 651 Huntington Ave., Boston, MA
02115 (pkanki@hsph.harvard.edu).
Clinical Infectious Diseases 2004; 38:1771–9
 2004 by the Infectious Diseases Society of America. All rights reserved.
1058-4838/2004/3812-0017$15.00
with 85% of HIV-2–infected individuals remaining free
of disease after 18 years of infection [12].
Even though HIV-2 demonstrates a more attenuated
phenotype compared with that of HIV-1, it is still ca-
pable of causing AIDS. Clinical experience with HIV-
2 infection and its treatment is limited because of re-
stricted geographic distribution, lower incidence of
infection as compared with HIV-1, and slower pro-
gression to disease. Given the similarities between HIV-
2 and HIV-1 and the attenuated natural history of HIV-
2 infection, efficacy of the response to HAART might
be predicted.
PATIENTS AND METHODS
Study population. Since 1994, the Department of
Immunology and Infectious Diseases at the Harvard
University School of Public Health (Boston, MA) has
collaborated with several United States–based physi-
cians to assist with diagnosis, virus characterization,
and virus load monitoring of HIV-2–infected patients.
Histories of serological testing for HIV, virus loads,
CD4+ cell counts, and antiretroviral therapy (ART) were
obtained. Clinical illness and treatment were docu-

HIV/AIDS • CID 2004:38 (15 June) • 1771

mented. This research was approved as exempt under 45 CFR
46.101(b)(4) by the Institutional Review Board of the Harvard
University School of Public Health (EX-02-232).
HIV-2 diagnosis. All patients’ serum samples were ana-
lyzed for antibody reactivity to whole-virus lysate on HIV-1
and HIV-2 immunoblots to confirm infection. An in-house,
whole-virus HIV-1 and HIV-2 immunoblot was performed with
a combined-antigen format. Results were interpreted on the
basis of World Health Organization recommendations [13].
Recombinant immunoblots that used the highly sensitive re-
combinant envelope peptides 566 and 996 for HIV-1 and HIV-
2, respectively [14], were run when differentiation between
HIV-2 and HIV-1 was required.
In cases in which serological testing could not positively
distinguish the infecting virus, confirmation was performed by
diagnostic PCR for HIV-1 and HIV-2, as previously described
[15, 16]. Diagnostic PCR was performed on PBMC-derived
DNA. Primers for HIV-1 env and gag, and HIV-2 env and gag,
were used in nested PCR reactions. The specificity of the PCR
product was confirmed by Southern blot test.
HIV-2 virus load. Plasma RNA virus load was measured
for sequential samples from each subject by means of an in-
house method that has a lower detection limit of 100 copies/
mL and is linear over 4 logs [17]. Briefly, virus was pelleted
from 0.5 mL of cryopreserved plasma, and viral genomic RNA
was isolated via guanidinium isothiocyanate precipitation fol-
lowed by alcohol precipitation. Reverse transcriptase-PCR was
performed with a GeneAmp rTth RNA PCR kit (Applied Bio-
systems); primers were used that were specific to gag in the
presence of a coamplifying internal control template. Products
were quantified by fluorescence-labeled primer read on an ABI
373XL automated sequencer (Applied Biosystems) and were
analyzed by Genescan software, version 2.1 (Applied Bio-
systems).
Data analysis. Clinical, therapeutic, and laboratory results
were plotted on individual patient timelines. Disease status was
categorized with use of the Centers for Disease Control and
Prevention (CDC) revised classification system for HIV infec-
tion [18]. CD4+ cell counts and virus loads were analyzed by
Stata software, version 6.0 (Stata Institute). These data and their
transformations were not normally distributed; therefore, the
Wilcoxon rank-sum (Mann-Whitney) test was used for data
analyses.
CASE STUDIES
Ten patients are presented in this case series. Demographic and
clinical data for each of these patients are given in table 1.
Patient 1. Patient 1 was a 51-year-old man from Guinea
who received a diagnosis of infection with HIV-2 in 1997 (figure
1A). Baseline CD4+ cell count was 560 cells/mL. In June 2000,
1772 • CID 2004:38 (15 June) • HIV/AIDS
the patient’s CD4+ cell count had dropped to 193 cells/mL, and
ART was initiated, consisting of a regimen of a combination
of lamivudine and zidovudine, ritonavir, and indinavir. The
patient was unable to tolerate ritonavir, and in September 2000,
ritonavir was discontinued, and the indinavir dose was in-
creased. In April 2001, neutropenia prompted discontinuation
of ART for 1 week; it was resumed with a combination of
lamivudine and zidovudine and indinavir. A recurrence of neu-
tropenia 2 months later prompted initiation of a new regimen,
with didanosine, stavudine, and nelfinavir. Throughout this
period, the patient’s CD4+ cell counts remained !200 cells/mL.
In April 2002, peripheral neuropathy required discontinuation
of ART. Two weeks later, ART was resumed with abacavir, ten-
ofovir, and a combination of lopinavir and ritonavir. During
the next 12 months, the patient’s CD4+ cell count increased to
540 cells/mL, and the patient’s virus load decreased to 3.17 log10
Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015
copies/mL.
Patient 2. Patient 2 was a woman from Cape Verde who
received a diagnosis of infection with HIV-2 in January 2001
(figure 1B). At that time, she presented with thrush, diarrhea,
and oral pain. Immunoblot revealed antibody reactivity to pol
and env antigens with no reactivity to gag antigens. Lack of
reactivity to the p26 core protein has previously been associated
with more-rapid disease progression in cases of HIV-2 infection
[19]. In February 2001, the patient’s CD4+ cell count was !200
cells/mL, and ART was initiated with zidovudine, lamivudine,
and nelfinavir. Clinical improvement was noted during the
course of the year, with resolution of the presenting symptoms.
In February 2002, the patient self-reported nonadherence dur-
ing the previous 2 months. The patient’s CD4+ cell count re-
mained low at 204 cells/mL, with her virus load remaining near
4.5 log10 copies/mL. The patient has since reported good
adherence.
Patient 3. Patient 3 was a 49-year-old man from Cape
Verde who received a diagnosis before August 1999 of HIV-1/
HIV-2 dual infection (figure 1C). His baseline CD4+ cell count
was 150 cells/mL, and ART was initiated with a combination
of lamivudine and zidovudine and the nonnucleoside reverse-
transcriptase inhibitor (NNRTI) nevirapine. Our laboratory
(Harvard School of Public Health; Boston, MA) confirmed that
the patient had serological test results positive for HIV-2, neg-
ative for p26, with reactivity to env antigens only, and negative
for HIV-1. Nevirapine therapy was stopped, and abacavir was
added to the patient’s regimen; the patient’s baseline HIV-2
virus load at this time was 4.7 log10 copies/mL. Adherence
problems were noted, including the continued taking of nev-
irapine despite instructions to the contrary from the physician
and nonadherence to the prescribed therapy for a 2-week period
in April 2001 and again for a month in June 2002. HIV lip-
odystrophy developed in May 2001. The patient’s CD4+ cell
Table 1. Epidemiologic, laboratory, and clinical summary of patients with HIV-2.

Demographic characteristics Clinical course

Duration of
Country Sex, age Year of CDC therapy, CD4+ count, Virus load, Treatment Treatment
a b
Patient of origin in years diagnosis class months Drug regimen cells/mL copies/mL problems success
1 Guinea M, 51 1997 B3 … … … … … …
… … … … 3 AZT/3TC-indinavir-ritonavir 193 ND + ⫺
… … … … 6 AZT/3TC-indinavir ND ND + ⫺
… … … … 2 ddl-d4T-nelfinavir 141 4.88 + ⫺
… … … … 10 ABC-tenofovir-lopinavir/ 191 4.96 ⫺ ⫺
ritonavir
… … … … … Final values 540 3.17 … …
2 Cape Verde F, 34 2001 B3 … … … … … …
… … … … 27 AZT-3TC-nelfinavir 150 ND + ⫺
… … … … … Final values 218 4.49 … …
3 Cape Verde M, 49 1999 B3 … … … … … …
… … … … 1 AZT/3TC-nevirapine 117 ND + ⫺

Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015

… … … … 45 AZT/3TC-ABC 137 4.74 + ⫺
… … … … … Final values 155 4.48 … …
4a Cape Verde M, 44 1999 A3 … … … … … …
… … … … 1 AZT/3TC 76 ND ⫺ +
… … … … 6 AZT/3TC-nelfinavir 54 5.44 ⫺ ⫺
… … … … 28 AZT/3TC-saquinavir 205 4.66 ⫺ ⫺
… … … … … Final values 308 4.62 … …
4b Cape Verde F, NA 1999 A1 … … … … … …
… … … … … … 746 2.19 … …
… … … … … Final values 1012 2.65 … …
5 Cape Verde M, 43 NA C3 … … … … … …
… … … … 15 ddI-d4T-nelfinavir- 98 3.24 ⫺ ⫺
saquinavir
… … … … … Final values 45 3.49 … …
6 Ivory Coast M, 35 1999 C3 … … … … … …
… … … … 8 AZT/3TC-ABC 20 ND ⫺ ⫺
… … … … … ddI-d4T-ABC-ritonavir- 26 ND ⫺ +
indinavir
… … … … … Final values 364 3.51 … …
7 Cape Verde M, 44 1997 C3 … … … … … …
… … … … 33 AZT/3TC-indinavir 250 ND ⫺ ⫺
… … … … 6 Tenofovir-indinavir ND ND ⫺ ⫺
… … … … 12 ABC-indinavir ND ND ⫺ ⫺
… … … … 2 ddI-d4T-lopinavir/ritonavir 38 5.02 ⫺ ⫺
… … … … 7 ddI-d4T-tenofovir- 60 5.04 ⫺ ⫺
efavirenz
… … … … 9 ABC-saquinavir-ritonavir ND 5.27 ⫺ ⫺
… … … … … Final values ND 4.74 … …
8 Sierra Leone F, 35 1999 A1 …
… … … … 6 AZT/3TC ND ND ⫺ +
… … … … … Final values 639 3.06 … …
9 Cape Verde M, 29 2002 A2 … … … … … …
… … … … 3 AZT/3TC/ABC 225 4.64 ⫺ +
… … … … … Final values 345 3.27 … …

NOTE. 3TC, lamivudine; ABC, abacavir; AZT, zidovudine; AZT/3TC, Combivir; AZT/3TC/ABC, Trizivir; CDC, Centers for Disease Control and Prevention; d4T,
stavudine; ddI, didanosine; lopinavir/ritonavir, Kaletra. +, problems reported; ⫺, no problems reported.
a
Self-reported compliance and/or intolerance from side effects.
b
Antiretroviral therapy success, as defined by viral suppression below the level of detection (!100 copies/mL).
 Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015

Figure 1. Time lines depicting CD4+ cell count (squares), plasma HIV-2 load (triangles), treatment, and illness over time. Breaks in treatment are
indicated by a dashed line. AZT, zidovudine; AZT/3TC/ABC, Trizivir; 3TC, lamivudine; AZT/3TC, Combivir; ddI, didanosine; d4T, stavudine; ABC, abacavir;
lopinavir/ritonavir, Kaletra; TB, tuberculosis. (Continued on the next 2 pages.)

1774
Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015

(Continued.)

1775
Figure 1.

Figure 1.
counts have remained !250 cells/mL, and his virus loads have
remained 14.5 log10 copies/mL.
Patients 4a and 4b. Patient 4a (figure 1D) was a man
from Cape Verde, and patient 4b was his female partner. The
man received his diagnosis in October 1999 when he was eval-
uated for neurological symptoms (including aphasia, imbal-
ance, and bowel incontinence). Serological testing results for
patient 4a showed HIV-2 env only and were negative for p26.
ART was initiated with a combination of lamivudine and zi-
dovudine, and nelfinavir was added in December 1999. The
patient’s neurological symptoms improved during the course
of the year. Treatment with the protease inhibitor (PI) was
suspended shortly thereafter for 2 weeks during treatment for
an obstruction of the small bowel. In June 2001, nelfinavir was
replaced with saquinavir. The patient remained clinically stable,
although the patient’s HIV-2 virus load did not decrease during
therapy.
The patient’s partner (patient 4b) was also clinically evalu-
ated and received a diagnosis of HIV-2 infection. The results
of an immunoblot demonstrated typical HIV-2 reactivity to all
major viral antigens. The patient remained healthy without
receiving ART. The patient’s virus loads have consistently re-
mained at 2 log10 copies/mL, and the patient’s CD4+ cell counts
have consistently been 1800 cells/mL.
Patient 5. Patient 5 was a man from Cape Verde. His
history included Pneumocystis carinii pneumonia, cytomega-
lovirus-associated retinitis, and pulmonary tuberculosis. The
patient was free of illness at the time of initial evaluation in
July 2000. Immunoblot analysis revealed reactivity to HIV-2
env only. HAART, consisting of 2 nucleoside reverse-transcrip-
tase inhibitors (NRTIs) (didanosine and stavudine) and 2 PIs
(saquinavir and nelfinavir) was initiated. No reduction in virus
1776 • CID 2004:38 (15 June) • HIV/AIDS
Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015
(Continued.)
load was noted after 5 months of observation, and the patient’s
CD4+ cell counts remained !100 cells/mL after 1 year of ob-
servation. The patient relocated outside of the United States.
Patient 6. Patient 6 was a man from the Ivory Coast who
received a diagnosis of infection with HIV-2 in 1999 after pre-
senting with CNS toxoplasmosis, uveitis, and thrush (figure
1E). Therapy was initiated with a combination of lamivudine
and zidovudine and abacavir; it was changed the next year to
abacavir, stavudine and didanosine, and ritonavir and indinavir.
The patient’s virus load increased from 2.5 to 3.5 log10 copies/
mL during the next year, and the patient’s CD4+ cell count
increased to 1300 cells/mL. The patient recovered from his op-
portunistic infections and remained stable through February
2003.
Patient 7. Patient 7 was a man who originally received a
diagnosis of infection with HIV-1 in January 1997 (figure 1F).
ART, consisting of a combination of lamivudine and zidovudine
and indinavir, was initiated during the same month. The pa-
tient’s HIV-1 virus load had remained undetectable through
1999, with CD4+ cell counts of 1200 cells/mL. In November
1999, with the patient’s CD4+ cell count decreasing, the NRTIs
were replaced with tenofovir. Oral thrush developed in May
2001, and abacavir replaced the tenofovir. HIV-1 virus load
remained undetectable throughout, and our laboratory con-
firmed HIV-2 infection. HIV-2 virus load was measured at 4.88
log10 copies/mL. In April 2002, ART was changed to stavudine
and didanosine and a combination of lopinavir and ritonavir.
In June 2002, with continued depressed CD4+ cell counts, ther-
apy was changed, with the PIs being dropped and the addition
of tenofovir and efavirenz. Therapy was changed again, in Feb-
ruary 2003, to abacavir, saquinavir, and ritonavir. Although the
patient is currently asymptomatic, treatment has failed to com-
pletely control the patient’s virus load or boost his CD4+ cell
counts.
Patient 8. Patient 8 was a woman from Sierra Leone who
received a diagnosis of HIV-2 infection in March 1999, when
she was 14 weeks pregnant. Results of serological testing showed
HIV-2 env only. Therapy with a combination of lamivudine
and zidovudine was initiated the next month. The patient re-
ceived an intravenous infusion of zidovudine (1350 mg) during
labor in October 1999. The patient was enrolled in ACTG
protocol 316 [20] and received either 200 mg of nevirapine or
placebo during labor. ART was discontinued postpartum be-
cause of her high CD4+ cell counts and undetectable HIV-2
virus load. In June 2003, the patient’s virus load was 3.06 log10
copies/mL.
Patient 9. Patient 9 was a man from Cape Verde who
received a diagnosis of infection with HIV-1 in January 2002
(figure 1G). The patient’s history included gonorrhea (in 1997);
positive test results for toxoplasmosis IgG, anti-hepatitis A IgG,
and hepatitis B core IgG antibody; and negative test results for
anti–hepatitis C antibody (in 2002). After repeated undetectable
HIV-1 virus load measurements, a plasma sample was evaluated
by our laboratory and found to be HIV-2 positive (p26 antibody
negative). The patient’s HIV-2 virus load in February 2002 was
4.64 log10 copies/mL with a CD4+ cell count of 225 cells/mL.
ART was initiated in March 2002 with zidovudine/lamivudine/
ABC (Trizivir). During the next 3 months, we observed a de-
cline in the virus load to a low of 3.27 log10 copies/mL, with
an increase of CD4+ cell count to 345 cells/mL.
RESULTS
Our laboratory has aided in the diagnosis and monitoring of
120 cases of HIV-2 infection in people living in the United
States. We have selected 10 patients to describe on the basis of
the availability of follow-up data, clinical histories, and labo-
ratory correlates. We compared median CD4+ cell count and
median virus load values between sick and healthy individuals.
We stratified our study population by clinical status: patients
who were identified as CDC class A1 and A2 were considered
asymptomatic, and patients identified as belonging to the re-
maining CDC classes were considered symptomatic.
Median CD4+ cell counts were lower among symptomatic
individuals (median CD4+ cell count, 175 cells/mL; range, 53–
286 cells/mL) compared with the asymptomatic group (median
CD4+ cell count, 639 cells/mL; range, 302–1012 cells/mL). The
difference between the medians was found to be statistically
significant (P ! .0167, by Wilcoxon rank-sum test).
Median HIV-2 virus loads of symptomatic and asymptomatic
individuals were compared. Asymptomatic individuals dem-
onstrated a lower median virus load (median virus load, 2.498
log10 copies/mL; range, 1.699–3.846 log10 copies/mL), compared
with symptomatic individuals (median virus load, 4.679 log10
copies/mL; range, 3.018–4.917 log10 copies/mL). The difference
between the medians fell slightly short of statistical significance
(P p .0527, by Wilcoxon rank-sum test).
Three individuals were asymptomatic (patients 4b, 8, and 9)
and had normal CD4+ cell counts and undetectable or low virus
loads. Patients 8 and 9 were provided with ART. Patient 8
received ∼5 months of treatment with a combination of la-
mivudine and zidovudine during her pregnancy. Seven of the
individuals whom we studied were symptomatic with HIV dis-
ease (patients 1, 2, 3, 4a, 5, 6, and 7). All of these individuals
exhibited depressed CD4+ cell counts and high virus loads. The
symptomatic individuals were evaluated for increases in CD4+
cell count and reduction of virus load. Three of 7 symptomatic
patients (patients 1, 4, and 6) demonstrated improvement as
measured by CD4+ cell counts. None of the symptomatic pa-
tients demonstrated a reduction of 12 log10 copies/mL in virus
Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015
load during the observation period. In 4 of 7 patients with
symptomatic cases, the patients’ conditions failed to improve.
In 2 of these patients (patients 2 and 3), self-reported non-
adherence to therapy for periods varying from 2 weeks to 2
months was noted.
DISCUSSION
HIV-2 is endemic to Africa, where the cost of antiretroviral
medications has been prohibitive for most people; thus, there
is little experience with the use of antiretroviral regimens for
treating HIV-2. The lack of a commercially available HIV-2
virus load assay has also been an obstacle for HIV-2 treatment
in all settings. In case studies, zidovudine and other NRTIs have
been shown to be effective in treating HIV-2 infection [21, 22].
As with HIV-1, the earliest methods for monitoring the ef-
fectiveness of therapy vis-à-vis HIV-2 were based on clinical,
immunologic [23], and proviral quantitation [22]. Additional
methods now include viral phenotypic and genotypic resistance
screening and monitoring of virus load. As is the case in patients
infected with HIV-1, the development of mutations conferring
resistance occurs in patients infected with HIV-2 [24]. The lack
of studies undertaken to reveal associations between clinical
and virologic parameters and disease or therapeutic efficacy has
resulted in uncertainty in the treatment and management of
HIV-2–infected patients.
HIV-2, like HIV-1, is a causative agent of AIDS. Although
the HIV-2 epidemic is generally limited to West Africa, cases
of HIV-2 infection are likely to be seen by physicians outside
of this region because of emigration and contact with individ-
uals from regions of endemicity. Individuals infected with HIV-
2 in the United States may not immediately receive a correct
diagnosis because of clinicians’ lack of experience with HIV-2
diagnosis in the United States. In many cases, HIV-2 infection
HIV/AIDS • CID 2004:38 (15 June) • 1777
is only discovered after an initial diagnosis of HIV-1, with more
discriminating tests being ordered only after other clinical cor-
relates fall outside of expected values (i.e., low CD4+ cell count
with undetectable HIV-1 virus load or average CD4+ cell count
with low CD4+ cell percentage). This was the case in 3 of 10
patients in our series (patients 3, 7, and 9), for whom a mis-
diagnosis of HIV-1 infection was discovered as a result of the
increasing severity of symptoms and the lack of detection of
HIV-1 viremia.
Determining appropriate treatments and laboratory corre-
lates has become more important with the potential for im-
plementing HAART regimens in regions of the world where
HIV-2 is endemic. However, treatment of HIV-2 infection re-
mains an uncertain endeavor. Most physicians opt to treat pa-
tients with modified HIV-1 HAART regimens, with modifi-
cations based on predictions from in vitro studies. NNRTIs
have been shown to be largely ineffective against HIV-2 in vitro
[25, 26], whereas NRTIs have been found to be as effective
against HIV-2 as they are against HIV-1 in vitro [27–31]. The
development of NRTI resistance in HIV-2 has been reported
at positions in the HIV-2 reverse-transcriptase homologous to
positions in HIV-1 [32, 33]. In vitro, PIs have mixed effec-
tiveness against HIV-2 [34]. Equal potency against the protease
of both viruses has been demonstrated for saquinavir, ritonavir,
and nelfinavir, and reduced potency against HIV-2 protease
(compared with potency against HIV-1 protease) has been ob-
served with indinavir. Investigation of the variations between
the HIV-2 and HIV-1 protease sequences has in part revealed
the reasons for differential PI resistance in both virus infections
[35]. In at least one study, it has been demonstrated that the
resistance mutations in HIV-1 result in sequences more similar
to those found in HIV-2 in these homologous regions [36].
This suggests that HIV-2 may possess constitutive PI resistance,
suggesting clinical responses to PI-containing regimens may be
less effective, compared with the efficacy of these regimens in
treating HIV-1 infection.
The majority of patients in our study received several dif-
ferent HAART regimens, most of which included ⭓1 PI. Prob-
lems with tolerability and toxicity to the prescribed drugs and
poor adherence to the regimens were noted for 4 of 10 patients.
In such instances, it is likely that this accounts for the lack of
viral suppression as defined by the absence of detectable virus
load. In 5 other patients (none of whom reported adherence
problems), viral suppression was not observed, despite minor
clinical or immunological improvement. Similar results were
noted in a recent case series of 18 patients from Ivory Coast
[37]. The authors of that study present limited evidence that
suggests that indinavir is superior to nelfinavir in controlling
virus load.
Although our study is limited by small numbers, the results
are surprising and important with respect to HIV-2 clinical
1778 • CID 2004:38 (15 June) • HIV/AIDS
management. Prospective natural history studies from West
Africa have documented the slower course of disease in indi-
viduals infected with HIV-2. In comparison to HIV-1–infected
individuals, individuals infected with HIV-2 demonstrate sta-
tistically significant lower virus loads during the first 8 years
after infection, with ∼30-fold lower virus loads during this time
[17]. These data suggest that viral suppression of HIV-2 with
antiretroviral drug regimens should be readily achieved. In all
8 patients who demonstrated detectable HIV-2 virus loads and
clinical symptomatology, various HAART regimens failed to
suppress HIV-2 virus loads below detection. Although HIV-2
virus loads in many of our patients were considered high for
HIV-2 infection, such levels would be considered average or
low in patients with HIV-1 infection.
Further clinical studies of HIV-2 HAART are clearly war-
ranted. Our preliminary observations suggest that drug regi-
mens with documented efficacy in treating HIV-1 infection may
Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015
not demonstrate similar efficacy for treatment of HIV-2 infec-
tion. Although in vitro studies have demonstrated similar in-
hibition of virus for most NRTI and PI classes of drugs, the
clinical experience noted with these drugs in our described
series suggests that further evaluation is required. Controlled
clinical trials of HIV-2—infected patients who are receiving
various HAART regimens are clearly needed to provide ther-
apeutic guidance to the medical community.
Acknowledgments
Financial support. US Army Medical Research and Ma-
terial Command 17-95-C-5005; Roche Pharmaceuticals, Gilead
(to P.R.S.); Abbott Laboratories, Allergan, Boehringer Ingel-
heim Pharmaceuticals, Bristol-Myers Squibb Company, Gilead
Sciences, Glaxo Smith Kline, Johnson and Johnson, Ligand
Pharmaceuticals, Pharmacia, Sarawak MediChem Pharmaceu-
ticals, Serono Laboratories (to P.R.S.).
Conflict of interest. P.J.K. consulted for Acambis. J.J.B.
received a speaker fee from Roche Pharmaceuticals. T.P.C. was
a consultant for Abbott Laboratories, Bristol-Myers Squibb,
Gilead Sciences, Glaxo Smith Kline, Merck, and Vertex Phar-
maceuticals, and he was on the Speakers’ Bureau for Bristol-
Myers Squibb and Glaxo Smith Kline. All other authors: No
conflict.
References
1. Barin F, M’Boup S, Denis F, et al. Serological evidence for virus related
to simian T-lymphotropic retrovirus III in residents of West Africa.
Lancet 1985; 2:1387–9.
2. Clavel F. Isolation of a new human retrovirus from west African patients
with AIDS. Science 1986; 233:343–6.
3. Kanki PJ. New human T-lymphotropic retrovirus related to simian T-
lymphotropic virus type III (STLV-IIIAGM). Science 1986; 232:238–43.
 4. Kanki PJ, DeCock K. Epidemiology and natural history of HIV-2. AIDS
1994; 8:S85–93.
5. Smallman-Raynor M, Cliff A. The spread of human immunodeficiency
virus type 2 into Europe: a geographical analysis. Int J Epidemiol
1991; 20:480–9.
6. Horsburgh CR Jr, Holmberg SD. The global distribution of human
immunodeficiency virus type 2 (HIV-2) infection. Transfusion 1988;
28:192–5.
7. Kanki PJ, Travers KU, MBoupS, et al. Slower heterosexual spread of
HIV-2 than HIV-1. Lancet 1994; 343:943–6.
8. Adjorlolo-Johnson G, De Cock KM, Ekpini E, et al. Prospective com-
parison of mother-to-child transmission of HIV-1 and HIV-2 in Ab-
idjan, Ivory Coast. JAMA 1994; 272:462–6.
9. Morgan G, Wilkins HA, Pepin J, Jobe O, Brewster D, Whittle H. AIDS
following mother-to-child transmission of HIV-2. AIDS 1990; 4:
879–82.
10. Matheron S, Courpotin C, Simon F, et al. Vertical transmission of HIV-
2. Lancet 1990; 335:1103–4.
11. Marlink R, Kanki P, Thior I, et al. Reduced rate of disease development
after HIV-2 infection as compared to HIV-1. Science 1994; 265:
1587–90.
12. Kanki PJ. Human immunodeficiency virus type 2 (HIV-2). AIDS Rev
1999; 1:101–8.
13. Acquired immunodeficiency syndrome (AIDS). Proposed WHO cri-
teria for interpreting results from Western blot assays for HIV-1, HIV-
2, and HTLV-I/HTLV-II. Wkly Epidemiol Rec 1990; 65:281–3.
14. Gueye-Ndiaye A. Cost-effective diagnosis of HIV-1 and HIV-2 by re-
combinant-expressed envpeptide (566/996) dot-blot analysis. AIDS
1993; 7:475–81.
15. Grankvist O, Bredberg-Raden U, Gustafsson A, et al. Improved de-
tection of HIV-2 DNA in clinical samples using a nested primer-based
polymerase chain reaction. J Acquir Immune Defic Syndr 1992; 5:
286–93.
16. Sarr AD, Hamel DJ, Thior I, et al. HIV-1 and HIV-2 dual infection:
lack of HIV-2 provirus correlates with low CD4+ lymphocyte counts.
AIDS 1998; 12:131–7.
17. Popper SJ, Sarr AD, Travers KU, et al. Lower human immunodeficiency
virus (HIV) type 2 viral load reflects the difference in pathogenicity
of HIV-1 and HIV-2. J Infect Dis 1999; 180:1116–21.
18. 1993 revised classification system for HIV infection and expanded sur-
veillance case definition for AIDS among adolescents and adults.
MMWR Morb Mortal Wkly Rep 1992; 41(RR-17):1–19.
19. Popper SJ, Sankale JL, Thior I, et al. Antibodies to the HIV type 2
core protein p26 and Vpx: association with disease progression. AIDS
Res Hum Retroviruses 1998; 14:1157–62.
20. Mirochnick M, Dorenbaum A, Blanchard S, et al. Predose infant nev-
irapine concentration with the two-dose intrapartum neonatal nevi-
rapine regimen: association with timing of maternal intrapartum nev-
irapine dose. J Acquir Immune Defic Syndr 2003; 33:153–6.
View publication stats
21. Ayanian JZ. HIV-2 infection in the United States. N Engl J Med
1989; 320:1422–3.
22. Clark NM. Immunologic and virologic response of HIV-2 infection to
antiretroviral therapy. AIDS 1998; 12:2506–7.
23. van der Ende ME. HIV-2 infection in 12 European residents: virus
characteristics and disease progression. AIDS 1996; 10:1649–55.
24. van der Ende ME. Antiviral resistance of biologic HIV-2 clones ob-
tained from individuals on nucleoside reverse transcriptase inhibitor
therapy. J Acquir Immune Defic Syndr 2000; 25:11–8.
25. Debyser Z. An antiviral target on reverse transcriptase of human im-
munodeficiency virus type 1 revealed by tetrahydroimidazo-[4,5,1-jk]
[1,4]benzodiazepin-2 (1H)-one and -thione derivatives. Proc Natl Acad
Sci USA 1991; 88:1451–5.
26. Witvrouw M. Activity of non-nucleoside reverse transcriptase inhibi-
tors against HIV-2 and SIV. Antivir Chem Chemother 1999; 10:211–7.
27. Cox SW. Comparison of the sensitivities of primary isolates of HIV
type 2 and HIV type 1 to antiviral drugs and drug combinations. AIDS
Res Hum Retroviruses 1994; 10:1725–9.
28. Mitsuya H. 3-Azido-3-deoxythymidine (BW A509U): an antiviral
agent that inhibits the infectivity and cytopathic effect of human T-
lymphotropic virus type III/lymphadenopathy-associated virus in vitro.
Proc Natl Acad Sci USA 1985; 82:7096–100.
Downloaded from http://cid.oxfordjournals.org/ by guest on November 21, 2015
29. Mitsuya H. Inhibition of the in vitro infectivity and cytopathic effect
of human T-lymphotrophic virus type III/lymphadenopathy-associated
virus (HTLV-III/LAV) by 2,3-dideoxynucleosides. Proc Natl Acad Sci
USA 1986; 83:1911–5.
30. Mitsuya H. Strategies for antiviral therapy in AIDS. Nature 1987; 325:
773–8.
31. Mitsuya H. Inhibition of infectivity and replication of HIV-2 and SIV
in helper T-cells by 2,3-dideoxynucleosides in vitro. AIDS Res Hum
Retroviruses 1988; 4:107–13.
32. Perach M. Resistance to nucleoside analogs of selective mutants of
human immunodeficiency virus type 2 reverse transcriptase. J Virol
1995; 69:509–12.
33. Perach M. Analysis of HIV-2 RT mutants provides evidence that re-
sistance of HIV-1 RT and HIV-2 RT to nucleoside analogs involves a
repositioning of the template-primer. J Mol Biol 1997; 268:648–54.
34. Pichova I. Peptide inhibitors of HIV-1 and HIV-2 proteases: a com-
parative study. Arch Biochem Biophys 1997; 341:62–9.
35. Kiso Y. Design and synthesis of substrate-based peptidomimetic human
immunodeficiency virus protease inhibitors containing the hydroxy-
methylcarbonyl isostere. Biopolymers 1996; 40:235–44.
36. Weber J, Majer P, Litera J, et al. Potency comparison of peptidomimetic
inhibitors against HIV-1 and HIV-2 proteinases: design of equipotent
lead compounds. Arch Biochem Biophys 1997; 341:62–9.
37. Adje-Toure CA, Cheingsong R, Garcia-Lerma JG, et al. Antiretroviral
therapy in HIV-2-infected patients: changes in plasma viral load, CD4+
cell counts, and drug resistance profiles of patients treated in Abidjan,
Cote d’Ivoire. AIDS 2003; 17:S49–54.
HIV/AIDS • CID 2004:38 (15 June) • 1779