re: CD34+ subsets & CFU-GM

From: D. Robert Sutherland (rob.sutherland@utoronto.ca)
Date: Mon Jan 26 1998 - 18:12:43 EST


Re Sharon's questions and the responses of Scott and William, it is our
experience that an increasing number of laboratories are using only
CD34+ cell enumeration to routinely assess auto- and allo-graft
quality.  As Scott indicated at least in part, the reasons for this are
many and varied but in our view, they nevertheless remain valid at this
time and perhaps bear restating. 

Colony-forming assays have been described in the ongoing dialogue as
'functional assays', but in fact only measure relatively 'late' or
single, lineage-committed progenitors of the myeloid and erythroid
lineages.  Typically, the only colonies 'scored' in these assays for
autograft assessment are colony-forming units for granulocytes and
macrophages (CFU-GM).  Unfortunately, due to differences in the way
these assays are performed between different centres, including widely
differing culture conditions (leucocyte-conditioned medium versus
recombinant growth factors etc.), and the subjectivity involved in
identifying various colony types, it is difficult to make meaningful
comparisons on data from different institutions.  Except in the hands of
a few specialist laboratories, CFC assays are not used to measure what
many consider the most relevant progenitor subset, namely megakaryocyte
progenitors (CFU-MK, BFU-MK).  Similarly, most labs do not measure
lymphoid-lineage progenitors and because of 'subjectivity' issues, most
labs still performing graft assessment by CFC assays do not even score
so-called 'mixed' lineage progenitors (CFU-mix, or CFU-GEMM).  

In vitro assays for precursors of the lineage-restriced CFCs such as
CFU-Blast, Long Term Culture-Initiating Cells, Cobblestone Area-Forming
Cells, etc., rely upon technologies that are even more esoteric and
generally beyond the technical capabilites of all but a few dozen
laboratories worldwide.  Finally, there exists in 1998, no in vitro
assay capable of identifying and quantitating the tiny fraction of
marrow leucocytes that are responsible for long term, multilineage
engraftment in vivo.  As Scott already indicated, an additional
limitation of colony-forming cell assays for hematopoietic progenitors
is the 10-14 day interval required for assay read-out.  This facet alone
precludes the assay any role in optimising the timing of apheresis
collections, a potentially useful thing to do for heavily pre-treated
cancer patients in whom obtaining an adequate does of PBSC may be
problematic.  Finally, CFC assays cannot be used to monitor 'on-line',
the yield of PBSC in the apheresis collection.  Thus, while all of the
above assays have played and will continue to play a vital role,
primarily in the research laboratory environment, we feel that their
potential utility in the routine measurement of engraftment potential is
limited.  

Studies performed in Seattle in the late 80s have clearly shown that
reinfusion of purified marrow CD34+ cells results in long-term,
multilineage hematopoietic recovery following myeloablative therapy. 
Thus the impetus to develop flow-based protocols for enumerating CD34+
cells began.  This mode of assessment takes less than 1 hour, is less
susceptible to subjectivity, 'operator', and (hopefully) nowadays,
'reagent' issues.  The CD34+ population encompasses the earliest stem
cells and maturing, lineage-committed progenitors of all lineages, and
can be used to 'time' apheresis collections and monitor the yield of
PBSCs 'on-line'.  It is also suitable for the determination of optimal
timing for apheresis collections.  Finally, the development of both
'positive' and 'negative' selection techniques to enrich for CD34+ cells
has driven the need for rapid, accurate and reliable flow techniques to
assess the purity of such supensions in both research and clinical
laboratory settings.  

Despite the above, and it is easy to tell which side of the fence we sit
on, both CFC assays and CD34+ cells are merely surrogate markers that
either alone or even in combination, are incapable of answering all of
our questions.  Hematopoietic cell transplantation is unlikely to become
a risk-free therapy any time soon, particularly in the allogeneic
setting.  This despite the fact that the best flow techniques for
measuring CD34+ cells, such as the ISHAGE protocol, even when properly
performed, can occasionally fail, or generate flawed data.  Recent
experience has indicated that flow techniques are more likely to
generate flawed or inaccurate data on purged or otherwise manipulated
samples of the type eluded to by William.  One potential reason for this
is that while the gating strategy published in the original version of
the ISHAGE Guidelines (J. Hematother. 5:213,1996) is at least able to
distinguish non-CD34+ dead cells from true CD34+ cells, it is not
capable of distinguishing dead CD34+ cells from viable CD34+ cells.  
Ultimately it is the viable CD34+ cells reinfused which will ensure
engraftment.  By incorporating 7-AAD and fluorescent counting beads into
the basic method, one can delineate dead from viable CD34+ cells and
generate an absolute viable CD34+ cell count using only flow
cytometery.  Experience with this modification has clearly indicated
that purged/manipulated marrow and PB/apheresis samples and post-thawed
cord blood samples have the greatest potential to generate seriously
flawed data.  The simple inclusion of this viability stain renders the
method highly reliable at quantitating viable CD34+ cells (Keeney et al
Comm. Clin. Cytomery, in press 1998).   It should be noted, however,
that CFC assays done pre-cryopreservation/manipulation will also suffer
the same limitation, i.e. may have little bearing on actual number of
functional stem cells infused to the patient.  So whilst those
proclaiming the virtues of functional assays may feel vindicated by
Scott and William's comments if, as we believe, the quality of the
post-cryopreservation stem cells is crucial, then the addition or
replacement of CD34 assays with CFC assays before storage is unlikely to
solve the problem.  

We all love anecdotes and over the years they have played a major role
in advancing the 'state of the art' in many fields, including this one. 
However, we could do with a little more specific information on the
example William cited to support his continued committment to routine
CFC assays.  Allografts purged of T lymphocytes have been shown to fail
on many occasions and the reasons for this have little or nothing to do
with the presence of sufficient viable CD34+ cells in the graft. 
Questions arising from his correspondence include, but are not
necessarily limited to:
1.	What was the clinical diagnosis in this case and how much prior
therapy had the recipient received.
2.	How closely matched were donor and recipient?
3.	Which protocol did the flow lab use and was a viabilty dye included
so that only viable CD34+ cells were enumerated.  
4.	Which method of T cell purging was used and did it involve
cytotoxicity.  When cytotoxic regimens are used it common for 'innocent
bystanders' including CD34+ cells to be killed also. 
5.	Did any other CFC assays set up by the same operator at the same time
using the same assay media generate poor colony growth?

Rob Sutherland, Mike Keeney, Ian Chin-Yee, Rakash Nayar.
Oncology Research, The Toronto Hospital, and London Health Sciences
Centre, London Ontario Canada.



This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:51:42 EST