50
ICCS 2011
P14
Using GemStone
TM
in the Routine Analysis of
Clinical Stem Cell List-mode Data
Bruce W Greig
1
, David Miller
2
, Don Herbert
3
, C. Bruce
Bagwell
3
1
Vanderbilt University Medical Center, Nashville,
TN, USA,
2
U S Labs, Brentwood, TN, USA,
3
Verity
Software House, Topsham, ME, USA
Background
: Given the subjectivity of human-derived
analysis there has long been a desire in clinical
fow cytometry to automate and thus standardize
the analysis of list mode data. This holds true for
stem cell analysis. Since the stem cell count is
predictive for engraftment, accurate results are
imperative. Preliminary fndings show considerable
potential for performing this analysis successfully
using Gemstone software. The primary goal of this
project is to test the feasibility of using GemStone
in automating and standardizing the evaluation of
the level of CD34+ stem cells in clinical samples
from stem cell transplantation programs.
Methods:
This study evaluates results from two institutions,
one using a single-platform method (fow cytometry
results only, n=58), and the other a dual-platform
(fow cytometry plus hematology results, n=50. All
data were compared WinList
TM
(Verity Software
House) or FACSDiva
TM
(Becton Dickinson) manually
vs GemStone
TM
(automated) analysis. Specimens
for single-platform method are cord-blood and
bone marrow. Specimens for dual-platform method
are peripheral blood, apheresis products, and
bone marrow. The scope of the study is limited to
identifying the extent that GemStone analysis can
be as good as or better than current manual gating
approaches for analysis of stem cell specimens.
Results:
Initial results show that automated detection
of viable CD34+ stem cells by GemStone is in
excellent correlation with manual gating methods.
Conclusions:
GemStone has the potential to answer
this need with high correlation to current methods.
Additional model development and studies to confrm
fully automated processing are warranted.
POSTER ABSTRACTS