C. Leport et al., Immunoblot profile as predictor of toxoplasmic encephalitis in patients infected with human immunodeficiency virus, CL DIAG LAB, 8(3), 2001, pp. 579-584
In order to define more accurately human immunodeficiency virus-infected pa
tients at risk of developing toxoplasmic encephalitis (TE), we assessed the
prognostic significance of the anti-Toxoplasma gondii immunoglobulin G (Ig
G) immunoblot profile, in addition to AIDS stage, a CD4(+) cell count < 50/
mm(3), and an antibody titer greater than or equal to 150 IU/ml, in patient
s with CD4 cell counts < 200/mm(3) and seropositive for T. gondii. Baseline
serum samples from 152 patients included in the placebo arm of the ANRS 00
5-ACTG 154 trial (pyrimethamine versus placebo) were used. The IgG immunobl
ot profile was determined using a Toxoplasma lysate and read using the Koda
k Digital Science 1D image analysis software. Mean follow-up was 15.1 month
s, and the 1-year incidence of TE was 15.9%. The cumulative probability of
TE varied according to the type and number of anti-T. gondii IgG bands and
reached 65% at 12 months for patients with IgG bands of 25 and 22 kDa, In a
Cox model adjusted for age, gender, Centers for Disease Control and Preven
tion (CDC) clinical stage, and CD4 and CD8 cell counts, the incidence of TE
was higher when the IgG 22-kDa band (hazard ratio [HR] = 5.4; P < 0.001),
the IgG 25-kDa band (HR = 4.7; P < 0.001), or the IgG 69-kDa band (HR = 3.4
; P < 0.001) was present and was higher for patients at CDC stage C (HR = 4
.9; P < 0.001), T. gondii antibody titer and CD4 cell count were not predic
tive of TE. Thus, detection of IgG bands of 25, 22, and/or 69 kDa may be he
lpful for deciding when primary prophylaxis for TE should be started or dis
continued, especially in the era of highly active antiretroviral therapy.