HIV-1 seroreversion in an HIV-1-seropositive patient treated during acute infection with highly active antiretroviral therapy and mycophenolate mofetil

Seroreversion for anti-HIV-1 antibodies has been documented in babies from infected mothers as a result of the loss of maternal antibodies [1–3], and in late-stage AIDS patients [4], probably because of the loss of immune cells. Other supposed cases of seroreversion could be attributed to sample mix-ups and other errors in testing [5]. Temporary seronegativity was reported in a Dutch patient [6], but it was unclear whether the patient presented with an early HIV-1 infection or with end-stage AIDS. Consistent seronegativity has been described in a few HIV-1-infected patients [7–10]. Here we describe a case of seroreversion after the start of highly active antiretroviral therapy (HAART) during acute HIV infection, with seronegativity being sustained for almost one year, after which the patient discontinued drug therapy leading to an immediate increase in both the viral load and antibody titre. A Dutch homosexual man developed fever, a sore throat and a skin rash in October 2000. HIV-1 RNA tested positive on 7 November, anti-HIV-1 antibodies were undetectable. Three weeks later, a Western-blot showed reactivity with all categories of HIV-1 antigens, and serology was positive for HIV-1 antibodies, hepatitis A virus IgM and total antibodies, hepatitis B surface and hepatitis B core antibodies, cytomegalovirus IgG, Epstein–Barr virus viral capsid antigen-IgG and anti-Epstein–Barr nuclear antigen. In December 2000, HAART with five drugs (didanosine/lamivudine/abacavir/nevirapine/indinavir and ritonavir) and 2 g/day mycophenolate mofetil (MMF) was started as part of a trial studying the efficacy of the addition of MMF to HAART in treatment-naive patients [11]. The plasma HIV-1-RNA load had then already declined to 680 copies/ml, with a CD4 cell count of 1.63 × 109 cells/l. Sequencing of the pro/pol region revealed a normal subtype B virus. Deletions in the nef gene or in the long-term repeat, which could have contributed to a low-replicating viral phenotype [12], were not detected. The patient also had no 32 basepair deletion in the CCR5 gene (data not shown) [13], suggesting that his low viral load could be attributed to strong anti-HIV-1 T-cell responses [14]. To monitor the efficacy of therapy, plasma HIV-1 RNA was measured using the NucliSens assay (bioMérieux, Boxtel, the Netherlands), until the detection level of the assay, 400 copies/ml, was reached, after which quantification was performed using an ultrasensitive adaptation of the NucliSens assay, with a detection limit of 5 copies/ml [15]. The anti-HIV-1 antibody titre in serum was measured simultaneously using the IMx System (IMx System HIV-1/HIV-2 III Plus; Abbott Laboratories, Abbott Park, IL, USA). Two weeks after the start of HAART, the plasma HIV-RNA load was found to be below the detection limit of 5 copies/ml (Fig. 1). Surprisingly, serum anti-HIV-1 antibody titres were decreasing in parallel. Within 2 months, the patient became seronegative (Fig. 1). Follow-up showed that seronegativity persisted for almost one year, with a concomitant undetectable plasma HIV-RNA load. At that time the patient decided to stop therapy, which resulted in an immediate increase in both plasma HIV-RNA and serum anti-HIV-1 antibodies. Viral RNA could be detected as early as 6 days after ending HAART, whereas HIV-specific antibodies became detectable again within one month (Fig. 1). To test whether this regained seropositivity was caused by superinfection with a second virus, we sequenced a pol gene fragment one year after infection. The fragment clustered strongly with the sequence obtained at presentation (bootstrap value 100; data not shown), indicating viral reactivation.Fig. 1. Changes in plasma HIV-1 RNA and serum antibody levels over time in an HIV-1 seroreverter (subject MAN004).: Open symbols are below the detection limit of the assay (5 copies/ml for HIV-1 RNA), or negative in the IMx System antibody assay (ratio ≤ 1). Arrows indicate the start and end of therapy with highly active antiretroviral therapy (HAART) and mycophenolate mofetil (MMF; 2 g/day, after 12 weeks 1 g/day). MMF and HAART were stopped after 48 and 56 weeks of therapy, respectively. CD4 cell counts remained high: 1.63 × 109/l at therapy start, and 1.50 × 109/l before ending HAART. During HAART, CD4 cell counts ranged between 1.34 and 2.36 × 109/l. —♦— HIV-1 RNA; —♦— HIV-1 antibodies.To our knowledge, this is the first report of a true and stable seroreversion in an HIV-1-infected patient without AIDS. Eight other patients with acute HIV-1 infection in the same trial, of whom three also received MMF, did not serorevert. Although their plasma HIV-RNA reduction was comparable to that of the patient described here, their anti-HIV-1 antibody titres were either stable or slowly increasing. MMF has been shown to decrease B-cell counts in transplant patients [16]. It does not affect the anti-cytomegalovirus response [17], but can reduce antibody formation when antigen and MMF are given simultaneously [18,19], which is not the case in our patient. However, MMF treatment could have contributed to the seroreversion observed here, although this is probably not a general mechanism, as the other HIV-1-infected patients receiving MMF did not show a decrease in HIV-1 antibody titres. The patient was fully able to mount an antibody response to HIV-1, given the positive serology results before and after HAART and the completely reactive Western blot. Furthermore, he showed antibody responses to a variety of viruses, indicating that he has no B-cell defect, unstable antibodies or fast antibody clearance. Probably, strong HIV-specific T-cell responses, the early potent antiviral therapy, and the MMF treatment together resulted in decreased viral antigen exposure and the abrogation of antibody formation. There were no competing interests. The study in which this patient participated was financially supported by a non-commercial, private foundation that does not wish to be named.