Progressive HIV infection in the presence of a raised CD4+ count: HIV/HTLV-1 co-infection
A F Haeri Mazanderani,1 MB ChB; O
Ebrahim,2 M BChB, MD, DTM&H, FCCP
(SA)
1
Department of Medical Virology, University of
Pretoria & National Health Laboratory Service, Tshwane
Academic Division, Pretoria, South Africa
2
Department of Medical
Microbiology,
University of
Pretoria, Pretoria, South Africa
Corresponding author:
A F
Haeri Mazanderani
([email protected])
There are a number of pathophysiological causes for a normal or raised CD4 count in the context of progressive HIV infection. These include various co-infections, previous splenectomy, and lymphoproliferative disorders. Such circumstances can both confound HIV diagnosis and delay initiation of chemoprophylaxis and highly active antiretroviral therapy (HAART). We describe the case of a patient co-infected with HIV and human T-cell lymphotropic virus type 1 (HTLV-1) who, prior to HAART initiation, was found to have progressive immune deficiency associated with a raised CD4 count.
S Afr J HIV Med 2013;14(2):92-94. DOI:10.7196/SAJHIVMED.904
A 51-year-old married man from Gauteng province, South Africa
(SA), originally from the Northern Cape, tested HIV-positive on
an enzyme-linked immunosorbent assay (ELISA) in March 2002 as
part of a routine medical examination for insurance purposes.
His CD4+
count was 794 cells/μl and his HIV viral load was 19 365
copies/ml. He gave no history of any infectious diseases, but
was receiving treatment for hypertension. On examination he was
generally healthy, with normal vital signs and all systems
proved unremarkable. On account of being asymptomatic with a CD4+ count
within normal range, he was neither initiated on highly active
antiretroviral therapy (HAART) nor chemoprophylaxis, and was
told to follow-up with his general practitioner for regular
immune monitoring (Table 1).
An abnormally high CD4+ count was detected on follow-up in September 2002 prompting T-cell receptor polymerase chain reaction (PCR) studies, which revealed no evidence of a clonal T-cell lymphoproliferative disorder. A bone-marrow biopsy was also performed and showed non-malignant T-cell hyperplasia. No further studies were conducted and expert opinion from HIV clinicians recommended that no antiretroviral therapy (ART) be given at that stage.
In January 2009 the patient
was referred to us with a history of weight loss, fatigue and
night sweats. On examination, he had increased reflexes
affecting the right leg, with weakness in both arms;
otherwise, examination was essentially normal. Magnetic
resonance imaging of the spine revealed a collapsed vertebra at T9; a biopsy showed a
chronic inflammatory process but no granulomata. On account of
the history and clinical presentation, spinal tuberculosis
(TB) was considered and TB treatment was commenced.
Furthermore, despite the high CD4+ count, it was felt that the
patient would benefit
from ART on account of being clinically immune-compromised and
having a high HIV viral load. He was initiated on a regimen of
Truvada (tenofovir/emtricitabine) and efavirenz (EFV) to which he responded well
with a drop in RNA copies/ ml of >2 log10 after three months of treatment and an undetectable
HIV viral load six months thereafter. As the CD4+ count remained above normal
limits, repeat bone marrow and flow cytometry studies were
carried out, identifying a population of T-lymphocytes with abnormal flow characteristics.
T-cell receptor PCR showed the presence of a clonal cell
population and bone marrow histology revealed infiltration by
tumour cells with scattered atypical uninucleated cells and
binucleated Reed-Sternberg cells. Immunophenotypic analysis
showed no overt evidence of a B-cell lymphoproliferative
disorder. Antibodies to human T-cell lymphotropic virus type
1/2 (HTLV-1/2) were
detected by ELISA and the patient was diagnosed with a
smouldering type of adult T-cell leukaemia/lymphoma (ATLL)
secondary to HTLV-1
infection (HTLV-2 not being associated with this condition).
He was treated with four cycles of infusional chemotherapy
consisting of etoposide, vincristine, doxorubicin,
cyclophosphamide and prednisone (EPOCH), which he tolerated
well. Interferon-alpha
therapy was subsequently commenced and mantained three times
per week. At the time of writing, the patient is clinically
well with no neurological deficits, an undetectable HIV viral load and
a CD4+
count of 4 430 cells/μl.
Discussion
HTLV-1 was the first retrovirus to be identified in humans and is structurally related to other viruses within the retroviridae family, such as HIV-1 and HIV-2, sharing similar routes of transmission. Since its discovery in 1979 three additional human deltaretroviruses (HTLV-2, HTLV-3 and HTLV-4) have been found, but only HTLV-1 and HTLV-2 have so far been associated with human disease. Antibodies to HTLV-1 were first identified in SA in 1984 and the first report of isolation of the virus was published in 1988. 1 , 2 Subsequently, a number of seroprevalence studies have been conducted in SA, where HTLV-1 has been found to be endemic in areas of Mpumalanga, the Eastern Cape, Free State and KwaZulu-Natal (KZN). 3 , 4 However, there are no recent representative data regarding prevalence in the general SA population or specific patient subgroups.5
Like other human retroviruses, HTLV-1 causes a lifelong infection of T-lymphocytes, in particular CD4+ cells. However, unlike HIV, the immunological hallmark of HTLV-1-infected individuals is a sustained proliferation of T-cells driven by the HTLV-1-encoded Tax protein. 6 The subsequent transactivation of cellular genes by the Tax-encoded region can result in malignant transformation, although this is rare.7 In the majority of cases, cytotoxic T-cells effectively control the virus by lysis of infected lymphocytes, which in turn results in the release of inflammatory cytokines that can be pathogenic.6 On account of these various pathophysiological mechanisms, HTLV-1 is associated with a diverse range of pathology, including malignant disease, inflammatory syndromes and infective complications.6 A number of these conditions have been described in SA, including ATLL, HTLV-1-associated myelopathy/tropical spastic paraparesis (HAM/TSP) and infectious dermatitis.8-10 Although the life-time risk for HTLV-1-associated diseases in general is considered close to 10%, an indication of a long history of viral-human co-evolution,6 this may be an under-representation when the interaction between HTLV-1 and other infective agents is considered. TB has been found to occur more frequently in patients infected with HTLV-1 and is also thought to be associated with a worse prognosis. 6 HTLV-1 has been shown to up-regulate hepatitis C viral replication and is implicated as a co-factor in the development of hepatocellular carcinoma. Furthermore, two studies have demonstrated an increased rate of cervical carcinoma in HTLV-1-infected patients.7 Whether HIV-1 co-infection with HTLV-1 is associated with a faster progression to AIDS remains a contentious issue, although a number of studies have suggested as much. 11 What is, however, less controversial and perhaps of greater relevance is the effect of HTLV-1 on T-lymphocytes, and in particular, its association with CD4+ lymphocytosis in HIV-1 co-infected patients.12 , 13
In general, lymphocytosis can be classified as belonging to one of two groups: either a reactive polyclonal proliferation, which can be caused by a variety of infective agents, hypersensitivity reactions, autoimmune conditions and splenectomy, or a clonal expansion as a result of a lymphoproliferative disorder. In the context of HIV co-infection, lymphocytosis has been described during early seroconversion associated with CMV, as well as in HIV/HTLV-1 co-infection where CD4+ lymphocytosis can be caused by both a reactive or clonal expansion. Consequently, patients with untreated HIV-1 who are co-infected with HTLV-1 show a dissociation between immunological and virological markers. That is to say, HIV-1/HTLV-1 co-infected patients have been found to progress to AIDS with a high HIV viral load, but in the presence of a normal or higher than normal CD4+ count (both absolute and percentage).12 A recent study in Mozambique demonstrated that co-infected pre-HAART adult patients were seven times more likely to have CD4+ counts >500 cells/μl (median 525 cells/μl) than HIV mono-infected patients.13 However, as these CD4+ cells are likely to be functionally altered, associated with a loss of naive cells and a higher activation pattern, CD4+ lymphocyte counts in HIV-1/ HTLV-1 co-infected patients cannot be considered to be a reliable marker of immunological competence.12 Furthermore, CD4+ counts can be dramatically raised on account of ATLL (i.e. clonal expansion), which occurs in ≤5% of HTLV-1 infections.6 As most cases of ATLL develop in individuals infected early in life through breastfeeding,6 it is probable that our patient was already infected with HTLV-1 when he first presented in 2002 with a CD4+ count of 794 cells/μl. Whether initiation of HAART at this juncture would have prevented the development of ATLL cannot be determined. However, it is thought that zidovudine (AZT) may protect HTLV-1-infected peripheral blood mononuclear cells from immortalisation on account of its genotoxic/mutagenic properties. 14
The last sizeable HTLV-1
seroprevalence study in SA was conducted in northern
KZN in 1993; a prevalence of 2.6% was found among the general
population.4 In the same study an HIV-1
prevalence of 3.5% was noted. As the risk factors for HTLV-1
and HIV are shared, an epidemiological association between
these two retroviruses is to be expected. In 1996, HTLV-1 was
found in 2% of asymptomatic urban black people in the Free
State, but in 6% of HIV-seropositive patients from the same
region.3 More recently, and alarmingly, in a small
retrospective study of 170 HIV-positive plasma specimens
collected between 2007 and 2008 from Limpopo, 24% of specimens
tested positive for HTLV-1/2 antibodies by ELISA. 15 Unfortunately, further testing
to confirm the diagnosis or differentiate between HTLV-1 and
HTLV-2 infection
was not performed. Nevertheless, these findings highlight the
evident gap in current knowledge and the need for clinicians
to be aware of retroviruses other than just HIV.
Conclusion
A CD4+ lymphocyte count cannot always be considered to be a reliable marker of immunological competence in HIV-infected people, especially in patients co-infected with HTLV-1. Normal or raised CD4+ counts in such persons can be on account of reactive or clonal expansion of T-lymphocytes and can confound HIV diagnosis and delay initiation of chemoprophylaxis and HAART. As we lack up-to-date epidemiological data but know that certain areas in SA are endemic for HTLV-1, we suggest maintaining a high index of suspicion of HTLV-1 infection in all HIV-positive adult patients in Southern Africa. In particular, HIV-positive persons who are clinically immune-compromised and have a raised CD4+ count should be tested for HTLV-1, as well as patients who present with clinical features in keeping with ATLL, HAM/TSP or infective dermatitis. As locally available serological tests are unable to differentiate HTLV-1 and -2, a PCR or western blot analysis may be required subsequent to a positive HTLV-1/2 ELISA test to confirm the diagnosis and distinguish between HTLV-1 and -2. Furthermore, the decision to initiate HAART in co-infected patients is better determined by clinical stage and HIV viral load than CD4+ count.
More research is needed to understand the epidemiology of HTLV-1 infection in Southern Africa; not only with regard to co-infections such as HIV-1/ HTLV-1 and TB/ HTLV-1, but also in terms of the wider public health impact, including implications for PMTCT practices and safety of the blood supply.
References
1. Saxinger W, Blattner WA, Levine PH, et al. Human T-cell leukemia virus (HTLV-1) antibodies in Africa. Science 1984;225:1473-1476.
2. Becker WB, Botha MC, Engelbrecht S, Becker MLB. Isolation of human T-lymphotropic virus type I (HTLV-I) from a black South African with Kaposi’s sarcoma. S Afr Med J 1988;73:481-483.
3. Van der Ryst E, Joubert G, Smith MS, et al. HTLV-I infection in the Free State region of South Africa: A sero-epidemiologic study. Cent Afr J Med 1996;42(3):65-68.
4. Bhigjee AI, Vinsen C, Windsor IM, et al. Prevalence and transmission of HTLV-I infection in Natal/ KwaZulu. S Afr Med J 1993;83:665-667.
5. Hlela C, Shepperd S, Khumalo NP, Taylor GP. The Prevalence of Human T-Cell Lymphotropic Virus Type 1 in the General Population is Unknown. AIDS Rev 2009;11:205-214.
6. Verdonck K, Gonzalez E, Van Dooren S, et al. Human T-lymphotropic virus 1: recent knowledge about an ancient infection. Lancet Infect Dis 2007;7:266-281. [http://dx.doi.org/10.1016/S1473-3099(07)70081-6]
7. Taylor GP. The Human T-Lymphotropic Viruses. In : Zuckerman AJ, Banatvala JE, Schoub BD, et al, editors. Principles & Practice of Clinical Virology. 6th ed. West-Sussex: Wiley-Blackwell, 2009:875-896.
8. Jogessar VB, De Bruyn CC, Bhigjee AI, Naicker VL, Bill PLA, Tait D. Adult T-cell leukaemia/lymphoma associated with HTLV-I in Natal. S Afr Med J 1992;81(16):528-529.
9. Schutte C-M, Townsend T, Van Coller R, Olorunju S. Comparison of HTLV-associated myelopathy (HAM) in HIV-positive and HIV-negative patients at a tertiary South African hospital. S Afr Med J 2013;103(1):43-46. [http://dx.doi.org/10.7196/samj.5298]
10. Hlela C. Human T cell lymphotrophic virus 1 associated infective dermatitis in KwaZulu-Natal South Africa [dissertation]. Durban: University of KwaZulu-Natal, 2008.
11. Beilke MA. Retroviral Coinfections: HIV and HTLV: Taking stock of more than a quarter century of research. AIDS Res Hum Retroviruses 2012;28(2):139-147. [http://dx.doi.org/10.1089/aid.2011.0342]
12. Gudo SE, Bhatt NB, Ramalho Bila DR, et al. Co-infection by human immunodeficiency virus type 1 (HIV-1) and human T cell leukemia virus type 1 (HTLV-1): does immune activation lead to a faster progression to AIDS? BMC Infect Dis 2009;9:211. [http://dx.doi.org/10.1186/1471-2334-9-211]
13. Bhatt NB, Gudo SE, Semá C, et al. Loss of correlation between HIV viral load and CD4+ T-cell counts in HIV/ HTLV-1 co-infection in treatment naïve Mozambican patients. Int J STD AIDS 2009;20:863-868. [http://dx.doi.org/10.1258/ijsa.2008.008401]
14. Macchi B, Balestrieri E, Mastino A. Effects of nucleoside-based antiretroviral chemotherapy on human T cell leukaemia/lymphotropic virus type 1 (HTLV-1) infection in vitro. J Antimicrob Chemother 2003;51:1327-1330. [http://dx.doi.org/10.1093/jac/dkg240]
15. Bessong PO, Mathomu LM. Seroprevalence of HTLV1/2, HSV1/2 and Toxoplasma gondii among chronic HIV-1 infected individuals in rural northeastern South Africa. Afr J Microbiol Res 2010;4(23):2587-2591.