| |
Encephalitis Developed After HAART Interruption
|
| |
| |
"Persistence of replication-competent HIV in the central nervous system despite long-term effective highly active antiretroviral therapy'
CORRESPONENCE
AIDS: Volume 19(2) 28 January 2005
Lambotte, Oliviera; Chaix, Marie-Laurec; Gasnault, Jacquesa; Goujard, Cécilea;Lebras, Philippea; Delfraissy, Jean Françoisa; Taoufik, Yassineb
aDepartment of Internal Medicine
bLaboratory of Immunology, Bicêtre Hospital, AP-HP, Bicêtre, France
cLaboratory of Virology, EA MRT Necker Hospital, AP-HP, Paris, France.
"...This is the first report of acute HIV meningoencephalitis after a programmed HAART interruption after long-term effective HAART. The diagnosis was supported by the high CSF-to-plasma HIV viral load ratio, the exclusion of other frequent causes of acute meningoencephalitis, and the rapid response to HAART. Replication-competent HIV appeared to have persisted in the CNS for more than 5 years of HAART, despite being undetectable in the patient's plasma during this period..."
The HIV load in cerebrospinal fluid (CSF) usually falls to undetectable levels during highly active antiretroviral therapy (HAART) [1]. The ability of replication-competent virus to persist in the central nervous system (CNS) during long-term effective HAART has not been demonstrated [2].
A 43-year-old woman was tested HIV-1 seropositive in 1990. In June 1993, treatment with zidovudine was started. Didanosine was added in February 1996. She was started on HAART (zidovudine, lamivudine, and indinavir) in September 1996. The treatment was highly effective on the viral load, which became undetectable, with an increase in the CD4 T-cell count above 500 × 106 cells/l. Treatment was changed (renal lithiasis) in December 1998, with stavudine, lamivudine, and nelfinavir. In May 2002, after 5.5 years of effective HAART with undetectable plasma virus, she was enrolled in a structured treatment interruption protocol. The CD4 cell count was 820 × 106 cells/l. The patient had always been asymptomatic before structured treatment interruption.
Eight days after HAART withdrawal (day 8), she had fever, myalgia and arthralgia. On day 21, she was admitted to an intensive care unit for fever and altered consciousness. Brain magnetic resonance imaging supported the diagnosis of encephalitis. CSF analysis showed 12 cells/mm3 (100% lymphocytes), 1.59 g/l protein and a normal glucose concentration. Aciclovir and amoxicillin therapy was started. Bacteriological studies of blood and CSF were negative, as was polymerase chain reaction analysis of CSF for herpesvirus groups 1 to 6 and enterovirus. The HIV-RNA load was 113,000 copies/ml in the CSF and 17 100 copies/ml in plasma. Antiretroviral therapy was resumed on day 28, with stavudine, lamivudine, nevirapine and nelfinavir. Her clinical status improved within 48 h, with defervescence and the disappearance of confusion, although memory disorders persisted. CSF sampled on day 32 was normal, with the exception of mild proteinorachia and 13,300 HIV-RNA copies/ml (180 copies/ml in plasma). The long-term outcome was favourable. The latest CSF sample, taken on day 60, contained 280 HIV-RNA copies/ml. The plasma viral load was below the detection limit, and has remained so for the past 2 years.
CSF viral sequences in June 2002 (during the meningoencephalitis episode) were compared with contemporary plasma viral sequences, and with sequences in two frozen plasma samples taken in 1996 and 1992. The comparison was based on the C2V4 region of the Env gene, after amplification, cloning, and phylogenetic analysis, as previously described [3]. Antiretroviral genotypic resistance was analysed using the TruGene HIV-1 assay (Visible Genetics Inc., Suwanee, GA, USA).
Significant viral diversity was observed in the clones isolated from the pre-HAART plasma samples from 1992 and 1996 [intravariability of 2.9 ± 1.1% and 2.16 ± 1.6% (±standard deviation), respectively]. After more than 5 years of effective HAART, and after HAART withdrawal, the plasma clones were more homogenous (intravariability 0.24 ± 0.35%). The intravariability of CSF clones was also low (0.4 ± 0.3%). Phylogenetic analysis based on the neighbour-joining method was performed in the C2V4 region (Fig. 1). High bootstrap values supported the clustering of the pre-HAART plasma clones with the clones isolated from plasma in 2002. The 14 CSF clones clustered together, but were distinct from the plasma clones, suggesting that the CSF virus evolved independently. No strains corresponding to pre-HAART plasma strains were found in the CSF. The maximum likelihood and parsimony methods supported these findings (not shown). Independent evolution of the CNS virus was also supported by genotypic resistance analysis. The main viral strain in the 1992 plasma sample was 'wild-type', with no resistance-associated mutations in the Pol gene. The main viral strain in the 1996 plasma sample (just before HAART initiation) and in the 2002 plasma sample (during the viral rebound of the structured treatment interruption) shared the following mutations associated with resistance to nucleoside reverse transcriptase inhibitors: 41L, 67N, 210W, and 215Y. No non-nucleoside reverse transcriptase inhibitor or protease inhibitor resistance mutations were found. The main viral clone in the 2002 CSF sample had the same pattern of mutations, but also the 184I mutation. This latter mutation might have arisen because of the poor CNS bioavailability of lamivudine [4].
This is the first report of acute HIV meningoencephalitis after a programmed HAART interruption after long-term effective HAART. The diagnosis was supported by the high CSF-to-plasma HIV viral load ratio, the exclusion of other frequent causes of acute meningoencephalitis, and the rapid response to HAART. Replication-competent HIV appeared to have persisted in the CNS for more than 5 years of HAART, despite being undetectable in the patient's plasma during this period. This suggests that the CNS can behave as an independent reservoir for HIV, and that caution is required during programmed HAART interruption.
References
1 Gunthard HF, Havlir DV, Fiscus S, Zhang ZQ, Eron J, Mellors J, et al. Residual human immunodeficiency virus (HIV) type 1 RNA and DNA in lymph nodes and HIV RNA in genital secretions and in cerebrospinal fluid after suppression of viremia for 2 years. J Infect Dis 2001; 183:1318-1327.
2 Blankson JN, Persaud D, Siliciano RF. The challenge of viral reservoirs in HIV-1 infection. Annu Rev Med 2002; 53:557-593.
3 Lambotte O, Gubler B, Nasreddine N, Wallon C, Goujard C, Rouzioux C, et al. The lymphocyte HIV reservoir in patients on long-term HAART is a memory of virus evolution. AIDS 2004; 18:1147-1158.
4 Schrager LK, D'Souza MP. Cellular and anatomical reservoirs of HIV-1 in patients receiving potent antiretroviral combination therapy. JAMA 1998; 280:67-71.
|
|
| |
| |
|
|
|
