Appeal No. 2001-2401 Application 08/277,225 1316 (Fed. Cir. 2000). Thus, with respect to claims 1 and 20 which define a competitive assay format, we do not find the combination of Pollema and Friguet to support a prima facie case of obviousness. With respect to claims 11 and 23, our review of the examiner's rejections and analysis in the present case have been made difficult, as the examiner has failed to provide analysis as to why each of the claims argued separately by appellants is rejected. The examiner has failed to separately argue the rejection of each claim in the Answer. As best we can determine Woods is relied on by the examiner solely for the purpose of rejection of the sandwich assay claims 11 and 23. As with the assay of Friguet, the sandwich assay of Woods would reasonably appear to be conducted when equilibrium between the antigen and antibody is reached. For example, the assay time of Example 1 of Woods requires an incubation time of one hour. Woods, column 7, line 15. We do not find that the examiner has provided an indication of an appropriate reason, suggestion or motivation, in either Woods or Pollema, to conduct the sandwich assay of Woods at a time other than after equilibrium has been reached between the antigen and antibody. In our view, the only suggestion to combine the cited references comes from appellants' disclosure. We agree with appellants that the examiner has not provided evidence of sufficient motivation to combine Pollema and Woods. Nor do we find that Freytag, describing an affinity-column-mediated immunoenzyomemetric assay with a dwell time of 75-120 seconds (page 1497, column 2) to overcome the 12Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 NextLast modified: November 3, 2007