Signals

From Signals, Fall 1995
A Clinician's View-Using Viral Load to Treat HIV Dise

ise

ignials interviewed Dr. Marcus
Conant to get his perspective
on viral load and to learn
how he is using bDNA results in
decisions to start and adjust therapy
for his AIDS patients. (The views
expressed are Dr. Conant's and do
not reflect the views of the Chiron
Reference Testing Laboratory)
In the last 6 months, the number of
physicians who are measuring their
patients' viral loads has increased dramatically. What do you think has
caused this increase? I think the initial questions we had-how useful
is bDNA, how reproducible is it,
should it be used-are all being answered. The Mellors paper showed
that patients with bDNA levels of
>100 KEq/mL* progressed to AIDS
much more quickly than patients
with levels of <10 KEq/mL.
And retrospective results from
ACTG 116B and 117 have given us
the first evidence that therapyinduced reductions in viral load
were associated with improved
clinical outcomes. In our experience, patients do better if you
reduce their viral burdens.
But you were measuring viral load
even before some of this evidence was
available. I'm curious why? I was
trained in a mindset not to use a

guide treatment decisions. So right
now, I'm "practicing medicine" based
on what I do know I'm developing
and revising my own guidelines as I
go along. (See box.)
So what would you say to those physicians who are still waiting for more
evidence before they start measuring
their patients' viral loads? I think the
"practicing" physician needs to ask
himself: Do I really believe that
AIDS is a chronic viral disease, and
that the progression of the disease
is related to the amount of virus
present in my patient? If so, if I
measure the amount of virus, and
reduce it with therapy, is the patient likely to do better than the
patient in whom I am not making
the attempt to reduce the viral
burden as much as possible?
The dilemma that the physician faces then is this: Am I going

to act now with the data I have? Or
am I going to wait for doubleblind, placebo-controlled trials that
may take 3-5 years?
For some physicians, I know cost is an
issue. How would you address this concern? The whole issue of cost needs
to be put into perspective. Viral load
can help physicians more cost-effectively use expensive antiretroviral
therapies. For example, with bDNA,
the physician can now know in less
than a month whether the therapy
he has prescribed is effective in
reducing viral load. If the viral load
does not drop, the physician can
stop an ineffective therapy that
would have cost over $2,500/year.
That's a significant savings for the
patient or insurance company
Sometimes the bDNA results
are a major factor in decisions not
to prescribe antiretrovirals, so the

cost of the test is offset by the savings from not prescribing drugs.
Ultimately though, the cost
decision should be made by the
patient, not the doctor.
Do you have any closing thoughts? It's
time to look at the situation from a
different perspective and ask, "Is
there any evidence that not using
viral load to manage patient therapy
is more effective than using it?" We
still have some unanswered questions
that are being studied, and someday,
we'll have those answers. In the
meantime, I think it's appropriate
that we "practice medicine" based on
the information we do have.
Suggested reading:
Ho. Time to hit HIV, early and hard N Engl J Med
1995;333:450-451
Volberding. HIV quantification: clinical applications.
Lancet 1996; 347:71-73

Dr. Marcus Conant is Medical Director of the
Conant Medical Group, a clinical practice in
San Francisco with over 3,000 HIV+ patients.
He has been using viral load to individualize
patient therapy for the last 12 months.
So how do you relate the idea of
"practicing medicine" to using viral
load to individualize therapy for
AIDS patients? Let me answer by
telling you what we know and
don't know about viral load.
We know that bDNA measures
viral load, and I think the evidence is
pretty clear that the test is reproducible. It may be even more reproducible than CD4 counts. Viral load
doesn't appear to vary by the time of
the day, or whether the patient went
to the gym in the morning, or what
they had for breakfast. First off,
bDNA is a reliable and reproducible
method for measuring viral load.
Secondly, we have shown that
you can reduce viral load with

1 THINGS PATHOLOGISTS SHOULD KNOW, FROM PAGE 1

USING bONA TO INDIVIDUALIZE THERAPY
In an interview with Dr. Marcus Conant, Signals asked, "How are you using bDNA to
manage patient therapy?" He responded, "The greatest value of the test is indicating to
me and to the patient whether to initiate or adjust therapy." Dr. Conant then outlined the
following decision tree that approximates how he uses bDNA in his practice.
Get baseline on all patients
bDNA < 10 KEq/mL bDNA > 10 KEq/mL (regardless of CD4)
CD > 50 or CD4 <500I

recently begun to offer viral load
quantification.
3. bDNA is highly reproducible because it
uses signal amplification.
PCR targets a specific region of
a nucleic acid sequence and then
makes a large number of copies of
that sequence before measuring the
quantity. It amplifies the target.
This method requires highly
trained technicians and excellent
laboratory procedures to avoid
contamination of new reactions
from previously amplified targets.
In contrast, bDNA does not
change the number of molecules in
the sample. Instead, it attaches
probes and branched DNA molecules to each HIV RNA molecule to
increase the molecule's size and
adds a chemiluminescent substrate
to generate light. Chiron's assay
then measures the light signal
which is directly proportional to
the HIV RNA level that is present
in the patient sample (Figure 1).
Consequently, bDNA is the
most reproducible method for
measuring viral load.2 In fact, the
Community Programs for Clinical
Research on AIDS (CPCRA) recently selected bDNA for their multicenter viral load study sponsored
by the NIAID. In selecting bDNA,
the CPCRA cited several advantages over PCR:
Direct detection of nucleic
acids at their physiologic concentration
Elimination of false positives
secondary to contamination
Reproducibility of test results
4I. New HIll subtypes are appearing in
the U.S., and bDNA measures all known
subtypes accurately.
There is increasing evidence
that new subtypes of HIV- 1 are
appearing in the U.S.'4' As these
new variants emerge, it is important that the method used to measure viral load detects and
quantifies these subtypes. In a

study comparing bDNA to PCR,
samples of known subtypes with
varying levels of virus were measured. The study concluded that
bDNA accurately measured each of
the subtypes tested, while the
accuracy and sensitivity of PCR
was impaired by genetic variation.2

5. In controlled settings, PCR results
correlated closely with bDNA results.
Using technicians who were
highly skilled in PCR and using
standardized PCR methods, the
virology laboratories of the ACTG
found that their PCR method was
comparable to bDNA. Branched
DNA was more reproducible than
PCR, and PCR had a somewhat
lower cut-off level of sensitivity
than the first generation bDNA
assay. Chiron's second generation
bDNA assay, currently being developed, has essentially the same sensitivity as PCR methods.
6. Outside of controlled settings, PCR
results show significant variability from
lab to lab.
Not all PCR methods used to
quantify HIV-1 RNA are created
equal. Each laboratory develops
and optimizes its own method,
standards, controls, and reagents,
and there is generally no standardization between laboratories. Therefore, consistency of performance
from day-to-day and between
batches of reagents is difficult for
the end-user or physician to determine. And, without standardized
reagents and methods, PCR results
from different labs are not interchangeable.
At the Chiron Reference Testing
Lab (CRTL) all reagents, standards,
and controls are manufactured by
Chiron and have been carefully
tested and standardized to ensure

consistent performance on a lot-tolot and day-to-day basis.
7. bDNA viral load quantitfication is
available now through CRTL.
If your future plans call for inhouse viral load testing, it makes
sense to start using bDNA today
when ordering viral load tests.
bDNA is
Quantitative
Accurate for all HIV-1
subtypes
Sensitive
Reproducible
Standardized
Available now
Ordering information is provided
on page 6.
The bDNA method is similar
to an ELISA in its basic approach
and uses a 96-well microplate. To
further ensure accuracy and reproducibility, the assay is designed for
testing in duplicate, and the RNA
level is determined using a 4-point
standard curve.,
The bDNA assay can be performed by laboratory technicians
in a routine clinical lab after minimal training. Requiring less handson time per result and better
through-put than PCR, bDNA may
be a more cost-effective way for
laboratories to bring viral load
testing in-house in the future.

treatment or a laboratory test until
someone designs a trial and does
the studies to show it is efficacious.
But AIDS is an epidemic that requires immediate solutions, and it
sometimes takes too long to validate
every new approach before we start
using it. Most of my patients don't
have the time to wait.
I believe we should do the
studies to further validate the clinical utility of viral load. But in the
meantime, I think it's appropriate to
act on the evidence that is available
and "practice medicine" the way we
did before the age of double-blind,
placebo-controlled clinical trials.
What do you mean by that? What our
forefathers meant 200 years ago
when they talked about the "practice of medicine" was that a physician would take everything he's
learned from his education, his experience managing other patients,
what he's learned from his colleagues, and what he knows about
the individual patient. He would
take all this knowledge and experience and apply it as best as he
could in managing his patients.

many of the antiretrovirals. You can
get a half log to one log reduction
with drugs like AZT, ddC, 3TC,
and ddl. And a 2 to 3 log decrease
in viral load with the new protease
inhibitors. Perhaps with combinations of those drugs, we might get
even greater reductions.
We also know that patients with
high virus levels progress to AIDS
more rapidly than patients with low
levels. And, we now have some preliminary evidence that reducing
viral load with existing antiretrovirals is associated with improved
short-term clinical outcomes.
Those are the things we know.
What don't we know? We haven't
done the prospective clinical trials
to prove that using viral load to
manage patient therapy will result
in improved clinical outcomes over
the long-term. I do have some
anecdotal evidence from my own
practice that when I'm successful at
reducing my patients' bDNA levels,
I see clinical improvement, and my
patients feel better.
The other thing we don't know
is exactly how to use these assays to

References:
1 Third Conference on Retroviruses & Opportunistic Infections, January, 1996
2 Todd et al. Performance characteristics for the quantitation of plasma HIV-1 RNA using branched DNA signal
amplification technology J Acquir Immune Dc/Ic Syndr Hum Retroviral 1995;19:s35-s44
3 Artenstein et al. Multiple introductions of HIV-1 subtype E into the western hemisphere. Lancet 1995;346:1197
4 Brodine et al. Detection of diverse HIV-1 genetic subtypes in the USA. Lancet 1995;346:1198-1199
5 Dale et al. The emerging genetic diversity of HIV.JAMA 1996;275:210-216
6 Pachl et al. Rapid and precise quantification of HIV-1 RNA in plamsa using a branched DNA (bDNA) signal
amplification assay. J Acquit Immune Defic Syndr Hum Retroviral 1995;8:446-454
combinations of new and standard antivirals, have been effective in significantly reducing viral load. Dr. Johnathan Kaplan of the CDC described
the findings as being the "hottest news" of the conference. Reports also
show a substantial increase in CD4+ cell levels.
The new drugs, protease inhibitors, operate by attacking the virus at
the late stage of replication and blocking the production of protease-the
enzyme that allows HIV to replicate. Standard drugs, known as nucleoside analogs, attack the virus earlier in its reproductive cycle.

Some of the new drugs and their current status
" Saquinavir (Invirase) Hoffman-LaRoche
" Ritonavir (Norvir) Abbott
" Indinavir (Crixivan) Merck
" Nelfinavir (Viracept) Agouron
" VX478 Glaxo-Wellcome & Vertex

Approved/Available
Pending FDA review: 2/29
Pending FDA review: 3/1
Still testing
Still testing
Signals page 3

*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.