Statement of William F. McCarthy, Ph.D.

decrease in viral PCR copy when in reality there was no such
reduction or if there was a reduction, not one as dramatic.]; 7.)
There was no mention that some of the gpl60 Phase I patients were
on AZT [In any circumstance this should be mentioned because it is
an indication of the health status for the gpl60 Phase I patients.
In addition, if any of the 15 patients presented in Amsterdam were
taking AZT and any of their PCR copy data were generated during AZT
use, this would create problems with determining whether the effect
was caused by gpl60 or by AZT: I have been told by infectious
disease physicians that AZT does reduce viral load.]; and 8.) In
his presentations subsequent to Amsterdam, LTC Redfield states that
viral load increases in the natural history group but not in his
gpl60 Phase I patients. My 20 August 92 memo shows that this is not
correct - there was no significant difference between the two
groups (gp160 patients versus the natural history historic control
patients) in terms of change in PCR copy from baseline (using both
LTC Redfield's baseline and the baseline Dr. Vahey preferred ( Dr.
Vahey's baseline contained only pre-vaccination PCR copy data -
thus two separate analyzes were done, one for each type of
baseline). As mentioned early, it was determined that the patients
in the natural history historic control group were not comparable.
Infectious disease physicians told me that this lack of
comparability gave a worst case scenario for the natural history
historic control group because they were infected longer and/or had
lower CD4 counts and therefore would in all likelihood have higher
viral load than the gpl60 Phase I patients. Even with this worst
case scenario my 20 August 92 memo results indicated no significant
difference between the two groups. In addition, there was no
significant difference between baseline and endpoint for either
group in terms of PCR copies. (Again, using both definitions for
baseline - two separate analyses were done). When LTC Redfield
stated that it is known that viral burden increases in a natural
history cohort but does not in the gpl60 Phase I patients, this is
also a misleading statement. Comparing a natural history group of
patients who have been infected longer to the gpl60 Phase I
patients who have been infected for a shorter amount of time is not
appropriate, according to the various infectious disease physicians
that I have talked with. From a statistical point of view, in order
to have a meaningful comparison, one would need to compare two
groups that had been infected for approximately the same amount of
time (i.e., compare two groups with early HIV infection).

6