 |
 |
 |
 |
We are using FITC-CD34, PE-CD38 and PER-CP-HLA-DR on the Coulter Elite. PE
spilling into PER-CP has been the biggest aggrevation. The compensation
problems are as follows--
Using PE-CD38 alone and examining the PE vs PER-CP histogram for a
"rectangular shape" does not give sufficient control. In other words, we
over- or undercompensate according to cells stained with [PE and PER-CP]
versus PER-CP alone. For that reason, we are considering the following
approach -- Run cells stained with a PER-CP-HLA-DR to find the "true value".
Then, adjust the compensation values for PE into PER-CP using cells stained
with [PER-CP + PE] by seeking a compensation value for PE that gives the same
percentage positivity for PER-CP when run alone. Playing with this approach
has shown that minute changes in compensation for PE have large effects on
PER-CP positivity. That is, with a double stained population, changing the
compensation for PE --> PER-CP from 29.0% to 28.5% increased the PER-CP
positive population from 21% to 27%! OUCH!! In both cases, the PER-CP vs PE
histograms for cells stained with PE alone looked fine and were almost
indistinguishable.
By the same token, compensation is exquisitely sensitive to antibody
concentration: Doubling the PE-antibody concentration led to a 2.5-fold
decrease in PER-CP positive cells.
Does anybody have an opinion on this unorthodox approach to compensation? Is
anyone else having similar problems?
??
Thank You in Advance
Doug Dooley, Barb Openlander and Mike Plunkett
American Red Cross
Bone Marrow Research
Portland Oregon
|
|
|
|