re CD34 enumeration

From: D. Robert Sutherland (rob.sutherland@utoronto.ca)
Date: Fri Apr 20 2001 - 17:47:39 EST


Hi Tracie,

In general, the successful practise of single platform flow methods
requires a high level of precision when it comes to such mundane issues
as pipetting. This requirement takes on huge importance when applied to
CD34+ cell enumeration because we are dealing with 'rare event' analysis
and the consequences of inaccuracy could be disastrous for patients
undergoing 'Armageddon' therapies such as  myeloablation and stem cell
transplantation.

Single platform methods should utilise positive displacement/reverse
pipetting techniques, such that accuracy is maintained when pipetting
relatively viscous samples such as peripheral blood and apheresis
products. Such accuracy requirements apply to both Stem-Kit and ProCOUNT
methodologies since samples usually have to be pipetted for both
dilution and aliquoting purposes.  The Stem-Kit additionally requires
accurate pipetting of the counting beads.  We use an electronic pipettor
(that can be set to perform reverse pipetting) for performing dilutions
(when needed), for subsequent sample pipetting, and for pipetting
counting beads (Stem-CountTM, part of the Stem-Kit).

Regarding the two kits, BDB have developed specialised software for
semi-automated data acquisition and analysis of ProCOUNT stained samples
(on BDFacs instruments) and this product can accurately measure CD34+
cells in most fresh PB and apheresis samples. However, the presence of
platelet aggregates, non-viable cells and other debris that stain
non-specifically with CD34PE conjugates can cause the software to 'flag'
even some fresh samples and BDB have issued a directive to ProCOUNT
users to the effect that manual gating of such flagged samples must be
performed by an experienced flow cytometrist.

Unfortunately, manual analysis of such listmode files (sent to us by
people who have experienced this problem) does not always generate
reliable data. A recent study by Gutensohn et al of 90 fresh apheresis
samples (Transfusion 39: 1220, 1999) found that 23% of them were
'flagged',  and even after manual analysis, significant differences were
noted between ProCOUNT and both the German Reference and ISHAGE
protocols (Stem-Kit is closely modelled on the single platform ISHAGE
protocol, as originally described by Keeney et al, Cytometry (Comm Clin
Cytom) 34: 61, 1998). Other problems have been noted by others when
analysing cord blood samples with this kit (Logdberg et al, Blood 96:
(Suppl 1), 2000 (abstr 1645).

If you are using the range of samples indicated (PB, apheresis, cord
blood and marrow), I would not recommend ProCOUNT since strictly
speaking, it is marketed for use on fresh peripheral blood and apheresis
products only.  In contrast, the ISHAGE protocol and its single platform
derivative, can be used to enumerate CD34+ cells in all of the samples
you listed.

You indicated you were having problems with reproducibiliy using the
Stem-Trol control cells, supplied in the Stem-Kit.  If you are confident
that these problems are not due to pipetting errors, the problems could
lie elsewhere.  Specifically, it is important to resuspend the Stem-Trol
cells adequately prior to pipetting.  This also applies to the
Flow-Count beads.  If methods are used that are too 'violent', you can
end up pipetting air bubbles (particularly wrt  the beads).  Inadequate
resuspension of either product will lead to erroneous results and
warrant the discarding of the remainder of either product since its
assayed value will no longer be correct.  These and other issues
pertaining to enumerating CD34+ cells have been detailed in Current
Protocols in Cytometry Unit 6.4, 1999 (maybe Paul Robinson will give you
a free copy!) and the ISHAGE Graft Evaluation web-site
(www.ishage.org/committees/Committees/Graft_Evaluation/graft.htm).

In some respects, a more valid method involves comparing the white blood
cell count from a hematology analyser, with the absolute CD45+ cell
count from the flow cytometer.  This can be readily obtained since the
Stem-Kit contains antibodies to CD45 and CD34 in addition to the
counting beads). On PB samples, the numbers obtained with the two
methods should be very similar.  If not, you need to either go to
pipetting school, or check that the assayed bead (or Stem-Trol) count
has not been compromised by inadequate storage and improper handling
procedures as described above.   On appropriately diluted apheresis
samples, the results should also be very similar.  Larger differences
between methods might be expected on CB and marrow samples due to the
presence of nucleated red blood cells, that are counted as nucleated
cells by the hematology analyser, but excluded by CD45 staining on the
cytometer.

Thus I would proceed as follows: on a fresh PB sample, compare the
absolute CD45 count from flow, with the absolute leukocyte count from
the hematology analyser.  If these numbers look very similar, then the
beads and pipetting issues discussed above are probably OK. If not, a
fresh batch of beads should be acquired and used appropriately.  If
results are now similar, discard the old beads. Now run Stem-Trol. If
you obtain the 'wrong' answer, a new batch of Stem-Trol should be
assessed.

HTH

cheers,

Rob Sutherland
Casey Nayar
University Health Network,
Princess Maragaret Hospital,
Toronto
Tel:416 946 2218



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:15 EST