If the definition is so clear, then why didn't Tefferi just refer to the (alleged) non-CR's he refers to as PR's? If "a CR requires 4 specific criteria to be met," and Imetelstat clearly does not meet the 4 criteria (or even three, as you and jq and Clark and Dew and Feuerstein allege), then why does Tefferi refer to them as CR's? He could just as easily have referred to them as PR's.
Because:
1) If your question is whether an incomplete CR could be counted as a full PR - the answer is that what is missing from the incomplete CR in this abstract is the clinical response. But that is also required to claim a full PR. So he can't more accurately describe the incomplete CRs as full PRs - because they aren't full PRs.
2) If your question is what differentiates incomplete CR as described in the abstract vs incomplete PR as described in the abstract the answer is the PR has easier criteria for BM/Peripheral Blood. So claiming an incomplete CR as an incomplete PR would not be accurately describing the BM and/or Peripheral Blood benefit for those patients.
If you're going to tell me that the PR's did meet the criteria of 3 of four specific criteria, but the CR's only met two of those criteria, then you are, in fact, saying that the PR is of a higher order of significance (because it met more criteria) than a CR, which, we all know, is patently absurd.
No sure what you are getting at - but I would suggest that rigor mortis has set in on this horse. Truly not much point to further beating it. We just need to wait - although there may be room for interpretation even then since it will be updated data and the numbers won't exactly match this prelim abstract.
He doesn't refer to them as CR's - he clearly refers to them as CR's with respect to two of the criteria. That's much more precise (and stronger for the drug) in this case than referring to them as PR's.
The abstract is written in what I would term abstractese rather than in English. In abstractese, it's non-ambiguous and clear. Only if you try to read the language out of context, assuming it is language from a paper rather than an abstract, is there any issue here at all.