Signals

From Signals, Spring 1993
Studies Point to Clinical Utility of Measuring Viral Load

A panel of prominent AIDS
researchers presented evidence of the clinical utility
of measuring viral load in plasma
and offered guidelines for using
viral load in making patient treatment decisions.
The evidence comes from retrospective studies of viral load in
subsets of patients who had participated in a number of AIDS
Clinical Trial Group (ACTG) studies. This new information was presented during a panel discussion
on the "Clinical Correlates of Virologic Measurement" at the National
Conference on Human Retroviruses in Washington D.C. in February
of this year.
Viral load predicted clinical progression
Brooks Jackson, MD from Case
Western University Medical Center
concluded that:
High baseline viral load was
independently predictive of
clinical progression in patients
with advanced disease.
(Asymptomatic patients and

patients with less advanced
stages of disease were not
included in these studies.)
Changes in viral load from
baseline were predictive of
antiviral treatment effect and
clinical progression.
Preliminary data suggested
that the development of drug
resistance and the presence of
mutations that cause resistance
were associated with increases
in viral load.
Dr. Jackson's conclusions were
based on analysis of the viral load
in subsets of patients who had
been enrolled in several ACTG
studies (Figure 2).
In these studies, changes in
viral load predicted changes in
clinical outcome and reflected the
effect of treatment. For example,
in ACTG studies 116B and 117,
patients on AZT who received ddl
experienced a marked drop in
viral load and also showed clinical
benefit.
As Dr. Jackson explained, how
ever, the viral load changes alone
do not explain the efficacy of treatment, and the long-term effects of
reducing viral load are not yet
known.
Using viral load in treatment decisions
Patricia Reichelderfer, PhD of
the National Institutes of Health
discussed what is known about the
interrelationship of viral load and
CD4+ cell counts and how viral
load could be used in managing
patient treatment.
According to Dr. Reichelderfer,
there is an inverse relationship between CD4+ cell counts and viral
load-the greater the increase in
viral load, the greater the decrease
in CD4+ cell counts. For many
patients, she believes a goal of
therapy should be to move them
from a state of high viral load/low
CD4+ cell counts to a state of low
viral load/high CD4+ cell counts
(Figure 3).
She also suggested that changes
in viral load and CD4+ cell counts
could be used to compare the treatment effect of different drugs. For
example, in ACTG study 143, most
of the patients who had been on
AZT only started out with high
viral loads and low CD4+ cell
counts (quadrant D). But after ddl
was added, most patients' viral
loads declined and CD4+ cell
counts increased (a move toward
quadrant A).
However, Dr. Reichelderfer
pointed out that individuals do not
always behave as the group data
would suggest. For example, there
are pathogenic factors, such as viral
phenotype and resistance, that can
modify the inverse relationship
between viral load and CD4+ cell
counts, and these factors may then

explain those individual patients in
quadrants B & C.
Viral threshold at low levels of virus
Dr. Reichelderfer introduced
the concept of a "viral threshold"
below which the viral load may be
too small to impact the CD4+ cell
count. For example, in ACTG
studies 116B and 117, patients
with viral loads of <10 KEq/mL*
had a much smaller antiviral
response than those with viral
loads of >10 KEq/mL. She postulated that response to antiviral
therapy may be less apparent
below this viral threshold of
10 KEq/mL.
Again looking at data from
ACTG studies 116B and 117, Dr.
Reichelderfer examined disease
progression in patients with very
low CD4+ cell counts. From those
studies, it appears that decreases in
viral load led to increases in CD4+
cell counts for those patients with
CD4+ cell counts >50. But the disease continued to progress in those
patients with CD4+ cell counts <50

Viral Load

LOW

HIGH

cC-)
o
+
0
C

A
C

B

D

0
Figure 3. A goal of therapy should be to move
patients from quadrant D to quadrant A.
in spite of decreases in viral load.
Dr. Reichelderfer concluded
her talk with the hope that this
research will stimulate thinking on
new ideas for clinical trial design
and new ways to individualize
patient management using both
viral load measurement and CD4+
cell counts.
Suggested reading:
Ho et al. Rapid turnover of plasma virions and CD4
lymphocytes in HIV-1 infection. Nature
1995;373:123-126
Ho et al. Quantitation of human immunodeficiency
virus type 1 in the blood of infected persons. N Engl J
Med 1989;321:1621-1625
Mellors et al. Quantitation of HIV-1 RNA in plasma
predicts outcome after seroconversion. Ann Intern
Med 1995;122:573-579

116A Comparison of ddl and AZT in therapy of HIV+ patients (AZT naive or <4 mo.
on AZT therapy)
116B Comparison of ddl and AZT in therapy of HIV+ patients (>4 mo. on AZT
therapy)
117 Phase II efficacy study comparing ddl and AZT in patients who had been on
long-term AZT therapy
143 Phase I and II open label study to evaluate the antiviral potential of combination AZT and ddl in patients with asymptomatic HIV disease
229 Phase I and II randomized, double-blind study comparing AZT plus
saquinavir versus AZT plus ddC versus AZT plus saquinavir plus ddC
Figure 2. In ACTG studies, HIV viral load was predictive of clinical progression and associated with
the development of drug resistance.

T1-011 Signals, Wintcr 1-994
Viral Load Versus CD4+ Cell Counts

Fromi Signals, 5pring 1995
HIV Pathogenesis
Impacts Therapy
everal recent studies reveal that the long, clinically latent phase that
characterizes HIV infection is not a period of inactivity, but a dynamic
process in which cells are being infected and dying at a high rate and
in large numbers.
This new breakthrough in the understanding of HIV and the implications for patient therapy were summarized in the January 27 edition of
Science by Dr. John Coffin from the Tufts University School of Medicine.
Using potent inhibitors to block production of HIV and precise assays
to measure viral load, researchers determined that the dynamics of HIV
could be described by a simple steady-state model in which infection, cell
death, and cell replacement are all in balance.
Dr. Coffin reported that in every patient studied (with amazingly concordant results using six different compounds), viral load declined by as
much as 99% from baseline within 1-2 weeks of initiating treatment. The
change in viral load also preceded a rapid increase in CD4+ cells (Figure 6).
This finding highlights the fact that the effect of antivirals on viral load can
be assessed within two weeks of initiating therapy.
In most patients, however, Dr. Coffin explained that viral load began to increase after the initial decline with the growth due entirely to mutants resistant
to the drugs used. And again, the change in viral load was met with a parallel
change in CD4+ cells, which gradually declined to near their starting level.
According to Dr. Coffin, the unique feature of HIV that drives the progression of disease is the rapid replication of the virus, which causes a high
level of mutation. He also argues that the real disease caused by HIV may
occur during the period when almost nothing seems to be happening, and
that the final collapse of the immune system may be caused by an accumulation of damage over the entire multiyear course of the infection, not the
result of some new event.
If this is true, Dr. Coffin suggests that it may be possible to extend the
latent period by permanently reducing viral load. He speculates that even a
two-fold reduction would be highly significant clinically, especially by comparison with current therapy. And it is possible that some compounds and
combination of compounds currently in trials might well be capable of
achieving a clinically useful response.

s more and more studies
cast doubt on the value of
CD4+ cell counts as the
most useful marker to guide
drug therapy, monitoring viral
load, as determined by measuring HIV-1 RNA levels in plasma,
may provide the clinician with
additional and more useful information than has been previously
available.
In general, there is an inverse
relationship between HIV-1 RNA
levels and CD4+ cell counts. But,
as shown in the scatter plot diagram (Figure 4), patients with
identical CD4+ cell counts can
have a wide range of HIV-1 RNA
levels. This variance is explained
by the fact that these parameters
are measuring two different
things-HIV- 1 RNA levels measure the amount of circulating
virus while the CD4+ cell counts
indicate the status of the immune
system.
Changes in viral load may
also signal the need to change a
patient's treatment plan. Figure 5
is representative of how viral
load and CD4+ cell counts might
be used to monitor a patient on
therapy. In this example, viral
load is increasing while the
CD4+ cell count shows little or
no significant change. And if
CD4+ cell counts, instead of viral
load, had been used to adjust
therapy, different treatment deci
1000
E
v 100
ci:
z
10

0
0
0 0 0 0
0
0 o 0

0 200 400 600 800 1000
CD4+ Cells/mL

1200

1400

Figure 4. Viral load may be an earlier indicator of disease state than CD4+ cell counts. Patients with
identical CD4+ cell counts can have a wide range of HIV-1 RNA levels.

1000
E
X~100 -'ci
z
10

-f- HIV-1 RNA KEq/mL --- CD4+/mm 3

In===*

600
500..

-- ------ f U

400
300 E
+
200 o

CD4+ cells

\~
0
Cv
Cll
U)
CD
m
E
O
C
CD
Cm
C=
co
fU

mutant virus

100,I

0 2 4 6 8 10
Months of Treatment

12 14 16

f wild-type virus
7-14
Days After Start of Therapy

Figure 5. Changes in viral load may signal the need for a change in the individual's treatment plan. In
this case the viral load is increasing while the C04+ cell count shows little or no significant change.

sions may have been made.
Using viral load, in addition to
other laboratory markers and the
patient's clinical status, may provide the best possible treatment
information.

Suggested reading:
Cao et al. Clinical evaluation of branched DNA
(bDNA) signal amplification for quantifying HIV-i
in human plasma. AIDS Res Human Retroviruses
1995;11:353-361
Dewar et al. Application of branched DNA signal
amplification to monitor human immunodeficiency
virus type 1 burden in human plamsa. J Infect Dis
1994;170:1172-1179

Figure 6. Viral load in the blood of HIV infected patients fell sharply within 2 weeks after introducing
therapy but increased to near starting levels with the rise due entirely to mutant virus.
Adapted from J. Coffin, SCIENCE, January 27, 1995.
Suggested reading:
Ho et al. Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection. Nature 1995;373:123-126
Wei et al. Viral dynamics in human immunodeficiency virus type 1 infection. Nature 1995;373:117-126
Coffin. HIV population dynamics in vivo: implications for genetic variation, pathogenesis and therapy. Science
1995;267:483-489

Signals page 2

'HIV-1 RNA kilo equivalents/milliliter of plasma as measured by branched DNA (bDNA) direct quantification. A kilo equivalent - 1,000 molecules of HIV-1 RNA.