Appeal No. 2002-0091 Page 7
Application No. 08/479,884
Thus, we would agree with the examiner that the claims could not be
practiced without undue experimentation, if the claims were directed to hGH
variants having a binding affinity the same as that of wild-type hGH. But that is
precisely what the claims are not directed to: claim 88 requires that the claimed
hGH variants have a binding affinity different from the affinity of wild-type human
growth hormone. The claimed variants can thus bind a target better or worse than
wild-type hGH, but any variants that bind with the same affinity are outside the
scope of the claims.
The examiner has conceded that the specification is enabling with respect
to the exemplified variants. The examiner has apparently concluded, therefore,
that the specification enables those skilled in the art to use hGH variants that bind
their target worse than, as well as better than, wild-type hGH. The exemplified
variants have Kd(variant)/Kd(wt) ratios varying from 0.6 to over 100. Since the
examiner has concluded that those skilled in the art could make and use any of
these variants, it follows that they could use other variants having similar binding
affinities. The evidence of record suggests that most hGH variants will have a
Kd(variant)/Kd(wt) ratio within this range, i.e., from somewhat better binding than
wild-type to much worse binding.
The examiner’s analysis also seems to assume that an enabling disclosure
must allow those skilled in the art to predict what binding affinity will result from a
given amino acid substitution. We disagree; enablement does not necessarily
require predictability. In the instant case, for example, the specification shows that
96% of the exemplified variants have binding affinities that put them within the
Page: Previous 1 2 3 4 5 6 7 8 9 10 Next
Last modified: November 3, 2007