Re: "Automated" flow cytometry instruments

From: Howard Shapiro (hms@shapirolab.com)
Date: Fri Jan 16 1998 - 11:12:30 EST


Scott Burger inquires about "automated or semi-automated" flow cytometry
instruments for phenotyping (CD3 and/or CD34) on a stat basis to support
the needs of a transplantation/gene therapy clinical service.

What has to be automated?

An automated hematology instrument can essentially suck up a rack of capped
test tubes, do all the necessary sample preparation, run the samples through
a flow cytometer (generally not a fluorescence measurement system, though),
and print out results without operator intervention.

The closest thing to this in the flow area may be Abbott's new system, which
(as I have been told; I haven't seen it or the literature on it) adds
fluorescence measurement capabilities to their Cell-Dyn hematology analyzer.
That represents a fairly expensive solution to the problem.

That's probably not the "small, automated" instrument about which the
transplant docs are said to be enthusiastic; I'd guess they were thinking
more along the lines of the B-D FACScan (now FACSCalibur) and the Coulter
EPICS XL.  Either of these, with appropriate software, could deal with
many samples prepared using another automated or semi-automated sample
prep system from the same manufacturer.  Effective, but not inexpensive.

The less expensive, simplified clinical cytometry systems which come to mind
are the B-D FACSCount flow cytometer and the Biometric Imaging Imagn2000
volumetric capillary (scanning) cytometer  Both of these were originally
designed to do CD4/CD8 counts.  The Biometric Imaging system accepts whole
blood samples, and does automated sample preparation and analysis; in order
to use this for CD3 and or CD34, you'd need to convince Biometric Imaging
to develop the appropriate reagents, sample cassette, and software.  At one
time, they were interested in using the instrument for stem cell enumeration;
if they've given up, it's probably because they can't justify continuing on
economic grounds. To use the FACSCount, you'd similarly have to convince B-D
to develop the reagents and tubes (possibly less difficult than for Biometric
Imaging, because they make the antibodies and labels and wouldn't have to
redesign a sample cassette), and to change the hardware (probably slightly)
and software (possibly substantially).  B-D, like Biometric Imaging, didn't
realize anticipated sales for their instrument for CD4/CD8 counting, and have
considered alternative applications, but they, too, have to envision a sub-
stantial market to justify a redesign effort.

The Microcyte flow cytometer, in the guise in which it was shown at the Rimini
ISAC meeting, is a one-angle scatter, one-color fluorescence measurement system.
Because its optics are highly integrated, it would have to be substantially
reengineered in order to incorporate an additional scatter and/or fluorescence
measurement channel.  The apparatus has limited data analysis capabilities, so
new software would have to be written.  The instrument was designed by a single
individual; adapting it for stem cell/CD3 work would probably require several
person-years of work by several well-paid persons.

Over the past twenty years, I have collaborated with several groups in placing
small "Cytomutt" flow cytometers in areas in which researchers or clinical
staff needed to do well-defined flow cytometric analyses at odd hours.  The
instruments could be built at a relatively low cost, and it only took a few
days to get people up to speed on operating them.  This is still going on in
situations where what is required in the cytometer is unavailable in current
commercial systems (e.g. dual-beam green YAG and red laser excitation, digital
log amps).  However, the CD3 and CD34 assays are now relatively easily done
using the reagents now available for 488 nm argon laser excitation.

Assuming that you have an easy-to-use benchtop cytometer (FACScan, XL, etc.)
available, the best short-term solution to your problem may be to train some
people who are around the transplant service most of the time to prepare
(using current semi-automated apparatus if necessary), and run (using the
benchtop FCM) samples, and to analyze data (using manufacturers' current semi-
automated software packages).  Of course, if you'd like a Cytomutt...

-Howard



This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:51:41 EST