AIDS RESEARCH AND HUMAN RETROVIRUSES Volume 8, Number 7, 1992 Mary Ann Liebert, Inc., Publishers

Quantitation of Human Immunodeficiency Virus (HIV) with Respect to Disease Stage and Zidovudine (AZT) Therapy SHARANJIT S. VED

BRAT,1

LESLYN P. SHEPHERD,1 ABDELHAMID PHILLIP F. PIERCE,2 and KIKI B. HELLMAN3

BOULBERHANE,'

ABSTRACT in 115 seropositive individuals was undertaken to evaluate the potential for HIV transmission as a nosocomial infection through the use of medical devices that may come in contact with the peripheral blood of HIV-infected individuals. The virus burden in the peripheral blood was estimated from the level of: plasma HIV p24 antigenemia; plasma viremia; p24 antigen in peripheral blood mononuclear cell (PBMC) lysates as indicators of productive infection; and frequency of latently infected cells. Negligible HIV levels were observed in the plasma and PBMC lysates of the majority of samples except for late-stage patients with certain opportunistic infections and/or lack of zidovudine (AZT) therapy. Some individuals on AZT therapy and at late-stage of disease may show antigenemia without plasma viremia or alternatively, plasma viremia may be observed without plasma antigenemia. PBMC lysate data indicated that the frequency of productively infected cells was less than one in 20,000 PBMCs for the majority of samples irrespective of status on AZT therapy or disease stage. HIV was detected in >95% of the cocultures and within 14 days for most of the samples, again regardless of the stage of disease or status on AZT therapy. The frequency of latently infected cells in this cohort ranged from 125 to 3125 per million PBMCs and was calculated to be as high as 2.5% of the helpter T-cell (CD4+ cell) population in the peripheral blood. The average latently infected cell frequency was 2-3-fold higher in early stage patients not on AZT than in late-stage patients on AZT therapy.

Quantitation of HIV

INTRODUCTION

A

infections is the transmis¬ sion of infectious agents by multiple use of inappropri¬ ately sterilized medical devices.' As the number of HIVinfected individuals began to rise significantly, concern over the risk of patient to patient transmission of HIV via medical devices during regular patient care was raised by public health authori¬ ties.2'3 The modes of HIV transmission that have been docu¬ mented to date require direct contact with HIV-infected body fluids. The level of risk for HIV transmission by medical devices is expected to be dependent upon the virus burden of the body fluid with which a device comes in contact. The purpose of this study, therefore, was to quantify HIV in a representative body fluid, such as peripheral blood, and to major cause of nosocomial

evaluate the potential for HIV transmission via medical devices coming in contact with the blood of infected individuals. The reusable medical devices which come in contact with blood directly or through body tissues, and which therefore may be potential vectors for HIV transmission, include: surgical and dental instruments, catheters, implants, hemodialyzers, fiberop¬ tic instruments used for exploratory purposes, especially those which pass through the veins, endoscopes used for the purposes of tissue biopsy, or flexible fiberoptic endoscopes with accesso¬ ries used in diagnosis and therapy. '~4 For any retroviral infection, possible contributors to the infectious virus load in a fluid are expected to be free virus particles, virus particles in host cells, or viral genome incorpo¬ rated in host cell DNA. Studies on the natural history of HIV infection without intervention of any therapy have demonstrated

'Braton Biotech, Inc.. Rockville, MD 20850.

2Georgetown University School of Medicine, Washington, DC 20007.

'Center for Devices and Radiological Health. Food and Drug Administration, Rockville. 1271

MD 20850.

VED BRAT ET AL.

1272

free HIV antigens in the plasma of only 20-30% of the infected individuals and for a very limited time period (i.e., either before seroconversion or at a very late stage of HIV disease). 5~y Recent studies have also shown the intracellular replication of HIV in monocytic cells without virus shedding into the intercellular environment.10" Moreover, in most cases, the virus is found primarily in a latent state in the peripheral blood mononuclear cells (PBMCs) either as integrated proviral DNA in the host cell genome or as a silent infection with unintegrated HIV DNA that I2~14 replicates only when the host cell is activated to proliferate. Consequently, to determine the total virus load in the periph¬ eral blood of HIV seropositive individuals at different stages of the HIV disease, this study was designed to include a combina¬ tion of parameters. These parameters are the: (a) level of plasma HIV p24 antigenemia which could result from shedding of intact infectious or noninfectious virus particles or virus proteins from acutely or persistent chronically infected cells circulating in peripheral blood or residing at some other site; (b) infectivity of plasma for normal stimulated PBMC cultures as an indication of plasma viremia resulting from only the free infectious virus particles; (c) level of p24 antigen in the lysates of PBMCs as a measure of intracellular productive infection in blood cells; (d) number of days required for HIV detection in cocultures as an indirect estimate of the frequency of PBMCs latently and productively infected with HIV; and (e) the direct determination of frequency of such latently infected PBMCs. Moreover, the need for the development of methodology to distinguish be¬ tween these categories for identifying and measuring the HIV load in the peripheral blood, and especially, the need for quantifying the latently infected cells has been well-recognized by other investigators1516 and is addressed in this study. The goal of the study was to determine the levels of these components of the virus load in a characteristic population of HIV patients visiting a hospital or medical center and to compare these levels for individuals at different stages of the HIV disease. The study of blood samples from individuals on zidovudine (AZT) treatment was included particularly, since a large portion of the HIV seropositive patient population is expected to be on AZT or other antiviral therapy and there has been a growing concern over the infectiousness of individuals from this cate¬

gory.17

Immuno-Assay (EIA, Organon Teknika)

and Western Blot

(WB, Dupont Co., Wilmington, DE) assay; and (c) availability of data on helper T-cell (CD4) counts by flow cytometric

analysis of the blood samples. These tests were performed at Georgetown University Medical Center as a part of the general

clinical evaluation of these individuals. Case histories of these individuals were recorded during regular clinic visits and in¬ cluded information on: (a) opportunistic infections (OIs) such as Pneumocystis carimi pneumonia (PCP), cytomegalovirus (CMV), varicella zoster (VZ), candida esophagitis (CE), herpes simplex virus (HSV), and toxoplasmosis (TP); (b) other mani¬ festations such as Kaposi sarcoma (KS), wasting disease, immune thrombocytopenia purpura (ITP), chronic renal failure (CRF), and dementia; and (c) information on therapy such as AZT, pentamidine, or other therapy. Clinical history and CD4 count were used as the most reliable predictors of Walter Reed (WR) stage of HIV disease as follows: WR1 individuals with CD4 count above 650/µ1 of blood; WR2 with CD4 count of 401-650; WR3 with CD4 count of 301-400; WR4 and CD4 count of 201-300; WR5 with CD4 below 200 and no OIs; and WR6 with CD4 count below 200 and/or with OIs (Table 1 ). In order to determine the effect of AZT therapy on total virus load, samples were included from three categories of individu¬ als: (a) on AZT therapy (68 samples) for at least 2 months, (b) not receiving AZT therapy either because of their higher CD4 count or because they had donated the blood sample immedi¬ ately prior to enrollment in the hospital for AZT treatment (46 samples), and (c) who had discontinued the AZT therapy due to certain clinical complications (7 samples) for at least 1 month. One of the individuals in category 'c' was able to resume AZT therapy during the course of this study and the subsequent sample was included in category (a). Blood samples from 20 low-risk HIV seronegative individu¬ als were used as controls and as a source of P-phytohemagglutinin (P-PHA) stimulated cells for the cocultures. All clinical samples were given a number code by George¬ town University Medical Center and were assayed in a blind fashion.

Sample processing Blood samples were processed to separate PBMCs using Ficoll-Hypaque (Bionetics, Kensington, MD) or leukoprep

MATERIALS AND METHODS

Subjects seropositive participated in

A cohort of 115 HIV

volunteers at different this study. The cohort comprised 50% black and 50% white individuals with a gender distribution of 85% males and 15% females. These individuals were patients at Georgetown University Medical Center and were in different high-risk categories. The male donors included homosexuals (90%), intravenous (i.v.) drug users (5%), and hemophiliacs (5%). Approximately, 50% of the females were i.v. drug users. Single samples were obtained from most of the individuals and multiple samples were obtained from only a few individuals with special clinical conditions. Eligibility for participation in the study was based on: (a) informed consent; (b) seropositivity for HIV by both Enzyme

stages of HIV disease

tubes (Becton Dickinson, Lincoln Park, NJ) as described else¬ where. lx Both fresh and frozen ( —70°C) PBMC samples were used for cocultures with P-PHA stimulated fresh normal PBMCs. Plasma samples were aliquoted and stored at —70°C for later testing of HIV p24 antigen and HIV antibody activity.

COCULTURE STUDIES Virus detection and isolation Normal fresh PBMCs isolated from leukophoresed buffy coat cells by Ficoll-Hypaque were stimulated with P-PHA (Bur¬ roughs ISWellcome) and used for cocultures as described else¬ where. Cocultures of 5 X 105-5 x IO6 PBMCs from seropositve individuals and 5 x 106 freshly P-PHA stimulated normal PBMCs were established and maintained in complete

HIV

QUANTITATION,

DISEASE STAGES AND AZT THERAPY

Table I. Different Plasma HIV p24 Levels

as a

Function

1273 of

Disease Stage

Plasma HIV P24 level

WR1

AZT Therapy Status

(pg/ml) Total

HIV disease

stage

and

AZT

CD4

count

>650

WR2

401-650

WR3

301-400

WR4

201-300

WR5

200 pg/ml) plasma antigenemia without plasma viremia, one case

with CRF and the other with acute VZ infection (Table 2 and cases of low-level antigenemia (21-60 pg/ml) without plasma viremia (categories 9, 10, and 11 in Table 4). One half of the 50 individuals negative for plasma viremia were also on AZT therapy (Table 2). Five of the plasma viremia positive samples were from individuals not on AZT therapy (categories 8, 13, 14, and 15 in Table 4). The other four plasma viremia positive samples were from donors who had CD4 counts < 100 and had concurrent acute VZ or HSV infection. They were also on AZT therapy (Category 7 in Table 4) suggesting that such therapy was not effective for these individuals at that stage. However, one of the plasma viremia-positive individuals with acute VZ infection and with AZT therapy temporarily suspended during the period of sample collection, became non viremic on resumption of therapy (Table 2 and category 11 in Table 4) showing the positive effect of therapy despite the presence of an opportunistic infection such as VZ.

category 12 in Table 4) and 7

antigen expression in sample PBMC lysates as a measure of cell-associated chronic productive infection in the peripheral blood

HIV p24

Before initiating this portion of the study the sensitivity of the cell and assay was determined on cultured cell lines of monocytic origin. It was found that HIV p24 antigen could be detected easily in a lysate of mixtures containing 2 x 106known uninfected cells and 100 productively infected cells by this procedure with O.D. readings equivalent to 125 pg/ml. Data on the 57 PBMC samples tested showed that 55 had negligible levels (0-20 pg/ml of lysate) of HIV p24 antigen expression, indistinguishable from the HIV negative controls

one

sample from a patient

(Table 3). These 55 samples with undetectable intracellular HIV p24 antigen expression are distributed randomly with respect to disease stage. There were three samples from WR1 stage patients, 9 from WR2 stage, 2 from WR3, 4 from WR4, 10 from WR5, and as many as 27 samples from 25 WR6 stage patients (Table 3). Of the 55 samples, 23 were from individuals who were not AZT therapy at the time of sample collection (Table 3). It is clear from the data that, irrespective of the stage of the

even on

Table 3. HIV p24 Antigen Expression in PBMC Lysates with Respect to HIV Disease Stage and AZT Therapy Cellular HIV p24

HIV disease

Positive

Negative

( >20 pg/ml)

(0-20 pg/ml of lystate)

AZT+

stage

AZT-

WR1 WR2 WR3 WR4 WR5 WR6 SUBTOTAL TOTAL

antigen

1

AZT+

AZT-

1 0 1 3 6

2 9 1 1

2^b

_6

4

Total

3 9 2 4 10 29

23

32

55

57

"Samples from patient with VZ infection, but with AZT therapy on hold for the first sample and reinstituted for the second sample. bIncludes data on PBMCs of 4 viremic patients with concur¬ rent VZ or

HSV infection.

VED BRAT ET AL.

1276 or the status of AZT therapy, the majority of HIV seropositive individuals including most of the last stage patients do not have productively infected PBMCs at a frequency greater

disease

than one in 20,000 cells. Intracellular HIV expression above the detection limit was observed only in special clinical situtions as represented by two such HIV p24 antigen-positive samples obtained from the individual with concurrent VZ infection (Table 3). One sample, obtained when the AZT therapy was on hold and the plasma HIV level was high, also showed exceptionally high HIV p24 antigen expression in the PBMC lysate (> 1000 pg/ml of lysate) (Table 4 category 13). On resumption of therapy for at least two months, decrease in plasma p24 level (from 650 to 60 pg/ml) and loss of viremia were accompanied by a decrease in the PBMC lysate p24 level to 240 pg/ml (Table 4 category 11). However, this HIV p24 level in PBMCs was still higher than that in the remainder of the population tested. On comparing of the data for plasma HIV p24 antigen level, plasma viremia and PBMC lysate HIV p24 antigen level from the same sample, it is clear that plasma antigenemia can be found with or without plasma viremia (categories 13 to 15 and 9 to 12, respectively, in Table 4). Similarly, plasma p24 antigen can be detectable with or without detectable PBMC p24 antigen expres¬ sion (categories 11 and 13 categories 9, 10, and 12, respectively, in Table 4). On the other hand, it is possible that a plasma antigenemia negative sample can exhibit plasma viremia with undetectable p24 antigen level in PBMCs (Category 7 in Table 4).

Coculture studies

as a measure

of latent HIV infection

Coculture studies were performed in order to determine: ( 1 ) the frequency of detectable HIV-positive cases based on activa¬

tion of latent virus in cocultures; (2) the time required for HIV antigen detection in a particular coculture as a reflection of the frequency of latently infected cells in that sample; and (3) the actual frequency of infected cells in a given sample. The data were correlated with the stage of disease and status of AZT

therapy. Frequency of detectable HIV positive cases. Induction of latent HIV was detected easily in cocultures of samples from 95 of 98 individuals as determined by HIV p24 antigen expression in culture supernatant fluids and, in some cases, lysates of cultured cells. HIV was not detected in samples from one individual whose CD4 counts fluctuated erratically between 900 and 300 over a period of 2 to 3 months. Failure to detect HIV in two individuals was attributed to the use

of frozen normal

PBMCs18 additional samples from those individuals for repeat

cocultures with fresh PBMCs were not available. Taking these factors into consideration, it is apparent from this study that HIV can be isolated from >95% of the HIV-infected individuals irrespective of the stage of the disease or status of AZT therapy. Coculture time required for HIV antigen detection as a reflection of the frequency of latently infected cells in a sample. Cumulatively, HIV was detected in cocultures of 75.8% of the samples by day 7 and in most of the samples by day 14 (Table 5). The average number of days of coculture required to detect HIV initially was 10.55 for samples at early clinical stages (WR1WR3) and 7 days for those samples at late stages of the disease (WR4-WR6) (Table 5). Analysis of these data indicates that coculture time for initial HIV positivity is significantly shorter for samples from late stage patients as opposed to early stage

patients (p

Quantitation of human immunodeficiency virus (HIV) with respect to disease stage and zidovudine (AZT) therapy.

Quantitation of HIV in 115 seropositive individuals was undertaken to evaluate the potential for HIV transmission as a nosocomial infection through th...
2MB Sizes 0 Downloads 0 Views