[Minutes, Advisory Committee to the Director GP-160 Meeting (1992: Bethesda, Maryland)]

132

1 to show that it's strongly prognostic for clinical
2 outcome in these patients. And in the treatments that
3 we have used so far, the nucleoside analogs that have
4 been shown to have at least some clinical effect, the
5 CD4 has been responsive. So it's not a ridiculous
6 idea at all to consider CD4 as a surrogate.
7 Next slide. But there are certainly
8 problems with looking at CD4 as an endpoint in these
9 trials. First, two general problems with surrogate
10 endpoints, not just for CD4, but for any type of
11 intermediate endpoint.
12 First of all, the clinical effect may or
13 may not be mediated through the surrogate or may be
14 mediated only partially. This is especially true when
15 you have different regimens with different mechanisms
16 of action. They may work differently.
17 But a second point that's a little
18 different, but equally important, is that if you look
19 at the intermediate endpoint, you have no handle on
20 adverse effects that may come to bear against
21 ultimately what you're looking at. There are lots of
22 examples of this in other areas of medicine that I
NEAL R. GROSS
COURT REPORTERS AND TRANSCRIBERS
1323 RHOOE ISLAND AVENUE, N.W.
[phone] WASHINGTON, D.C. 20005 [phone]

I