Monitoring of Antibodies against Human Immunodeficiency Virus Type 1 p2,5 Core Protein as Prognostic Marker ' Emmanuel Fenouillet, Natalie Blanes, Anne Coutellier, Jacques Demarquest, Willy Rozenbaum, and Jean Claude Gluckman

Laboratoire de Biologie et Genhique des Pathologies Immunitaires, Centre National de la Recherche Scientifique Unite de Recherche Associee 1463; Clinique Medicale I. Groupe Hospitalier Pitie-Salphriere; and Consultation de Maladies Infectieuses, Hiipital Rothschild. Paris. France

Antibodies against human immunodeficiency virus type 1 (HIV-1) structural proteins encoded by env. pol, and gag genes develop in the first weeks after viral infection [1, 2]. The diagnosis of HIV-1 infection currently rests on such antibody reactivity [1], while biologic follow-up of HIV-l-sinfected people is based mainly on the surveillance of blood CD4+ lymphocyte counts, p25 antigenemia, and biochemical parameters such as serum ,82-microglobulin or neopterin [3, 4]. However, it has long been observed that in the course of progression to disease, antibody reactivity against the gag major proteins p25 and p 18 decreases, as usually determined by Western blot, while only limited change in anti-env antibody levels occurs [5]. To ascertain the possible interest of monitoring anti-gag antibody levels as a prognostic marker and clinical correlate, anti-p25 antibody responses of HIV-I-infected patients were evaluated in both cross-sectional and longitudinal studies with an RIA using recombinant p25 and three different ELISAs. Anti-p25 antibody titers were compared with other parameters routinely used to follow HIV-l infection: CD4+ cell counts and p25 antigenemia [4].

Subjects and Methods Patients. A cross-sectional study was done on 130 HIV -L-seropositive patients attending a specialized outpatient clinic: 114

Received 18 December 1991; revised 13 April 1992. Grant support: French Agence Nationale de Recherche sur Ie SIDA and Caisse Primaire d' Assurance Maladie Paris-lle-de-France. Reprints or correspondence: Dr. Emmanuel Fenouillet, Laboratoire de Biologie et Genetique des Pathologies Immunitaires, CERVI, Hopital de la Pitie-Salpetriere, 47 BId. de l'Hopital, 75651 Paris Cedex 13, France. The Journal of Infectious Diseases


© 1992 by The University of Chicago. All rights reserved. 0022': 1899/92/6603-0024$0 1.00

men and 16 women aged 22-58 years (median = 32). Of these, 83 were homosexual men, 22 were intravenous drug users, 23 were infected heterosexually, and 2 had been transfused with contaminated blood. Thirty-three patients were classified as asymptomatic (Centers for Disease Control [CDC] stage II), 35 as isolated persistent generalized lymphadenopathy (stage III), and 62 as stage IV [6]; 36 of these were classified as IV-A, -C-2, or -E (AIDS-related complex [ARC)) and 26 as IV-B, -C-l, or -D (AIDS). Of those with AIDS, 1 had neuropathy, 15 had opportunistic infection(s) (01), and 10 had Kaposi's sarcoma (KS) as the predominant pathology. Thirty-seven had been treated with zidovudine or dideoxyinosine for ;;::.6 months at the time of serum sampling. These serum samples were called panel A. A retrospective longitudinal study was conducted on 56 patients: 49 men and 7 women aged 27-57 (median = 39). Of these, 46 were homosexual men, 2 were intravenous drug users, and 8 were infected by other means. At the end of follow-up, 15 patients were stage II or III (follow-up: 4-9 years, median = 6; last CD4+ cell count: >200/mm 3 in 10, 200/mm 3 in 6, 10 arbitrary units/ml., Patients with AIDS-related complex displayed heterogeneous levels. Patients with AIDS had the lowest values: ::s:;10 units/ml. in 96% of cases. In patients whose CD4+ cell counts eventually fell below 200/mm3 or who developed AIDS (or both), antibodies were initially 1 log unit/5 years, beginning at least 4 years before the index symptom. Because the only point at which CD4+ cell counts significantly differed between progressors and nonprogressors was 1 year before the disease, both initial anti-p25 values and antibody decline seemed to be better long-term prognostic markers than CD4 + cell counts.


Concise Communications

Results Comparison of anti-p25 antibody measurement by RIA or ELISA and anti-gag determination in Western blot. First we systematically compared results obtained by RIA and by ELISA kit I for panel A sera. Next, 10 randomly selected sera from each CDC stage group were used to compare the results obtained with each of the three commercial ELISA kits. This showed that the RIA and the ELISA kits used were all reliable (signal-to-background ratio: 30-50; background ~300 cpm for the RIA and ~0.050 absorbance units for sandwich ELISAs or ~ 1.10 for the competition ELISA). The assays were highly reproducible: The slopes of reference curves correlated from one day to another (P < .001 for the RIA and for the ELISAs) as did anti-p25 antibody levels detected by RIA or by ELISA kit 1 on repeat experiments (P < .001 for both). Anti-p25 antibody level in' a given serum sample was comparable regardless of the technique used (P < .001 for all comparisons). Finally, in each serum sample, anti-p25 antibody levels by ELISA were compared with Western blot scoring against core proteins p55, p40, p25, and p 18. There was a strong correlation between these anti-p25 levels and Western blot scores, especially with respect to p40 (r = .873 for p40, .540 for p18, .742 for p55, and .787 for p25). Anti-p25 antibody levels and clinical staging of HIV-I infection. Antibody levels in panel A ranged from ~ 10- 2 to > 104 units/nil. (median = 40). There was no significant difference in levels between stage II and III (median = 120 and

214 units/ml., respectively; P> . 10) (figure 1A). Levels significantly differed between stage II/III and stage IV (median = 1 unit/ml.) as a whole (P < .001) and, among the latter, between ARC and AIDS (median = 8 and 0.4 units/ml., respectively; P < .001) (figures lA, B). No difference was noted when comparing stage IV patients according to antiretroviral therapy (P > .670) (figure lC). Distribution of ARC patients' anti-p25 antibodies did not differ whether CD4+ cell counts were 200/mm 3 and, in AIDS patients, whether predominant pathology was 01 or KS (data not shown). When these sera were divided at the median for the combined population (40 units/ml.) to segregate individuals with high or low antibody levels, using the median test, levels above cutoff were noted in 67% of stage II/III, 43% of ARC, and only 4% (1 patient) of AIDS patients' sera (P < .001). Other cut points did not add discriminant power. For example, 96% of sera from AIDS patients had antibodies ~10 units and none >500 units while 78% of sera from stage II/III patients had antibodies> 10 units and 12% ~ I unit/ml., This clearly indicated a relationship between anti-p25 antibody levels and the stage of disease progression reached. Predictive value ofanti-p25 antibody levels. Fifteen stage II/III, 11 ARC, and 30 AIDS patients were selected to study evolution of anti-p25 antibodies during the course of HI V-I infection. When measured at the time of diagnosis of the clinical symptom(s), or at the last time point for stage II/III patients, anti-p25 antibodies relative to clinical staging were distributed in a manner similar to that for panel A (data not shown). Because isolated non-Hodgkin's lymphoma occurred at higher antibody levels (J 00-1 000 units/ml.) and CD4+ cell counts [9] than other AIDS conditions, these patients were excluded from the following analyses. Sequential measures of anti-p25 antibodies showed stable levels (slope of decline, 200/mm 3 (group 1; figure 2). In contrast, anti-p25 antibodies progressively declined (> 1 log unit/5 years) for both stage II/III patients (n = 5) and those who progressed to ARC (n = 5), the CD4+ cell counts of whom eventually fell to

Monitoring of antibodies against human immunodeficiency virus type 1 p25 core protein as prognostic marker.

Anti-p25 antibodies were evaluated by cross-sectional analysis of sera from 130 human immunodeficiency virus type 1-infected patients and in a longitu...
602KB Sizes 0 Downloads 0 Views