Re: ploidy analysis
T. Vincent Shankey (tshanke@bsd.medctr.luc.edu)
Fri, 30 Aug 1996 11:59:09 -0500 (CDT)
Arnold,
Your e-mail message raises the issue of the correct technique to
use to identify DNA aneuploid populations by flow cytometry. While you
comment on a specific situation with a specific software program, the
fundamental issue is quite independent of software.
The authors of the DNA cytometry consensus (Cytometry 14:472, 1993)
discussed this issue in some depth. Our first consideration was that all
tumors consist of a variable mixture of tumor plus normal (vessels,
muscle, connective tissue, macrophages, the normal cellular compartment
that the tumor arose from) or reactive cells (neutrophils, lymphs).
Providing that the cell or nuclear isolation technique used does not
somehow isolate only normal or only tumor (which is highly unlikely), the
sample used for flow cytometric analysis includes both tumor and normal
or reactive elements. These provide an internal DNA diploid marker
to use to determine the DNA Index of the tumor. Use of markers which
identify the tumor plus its normal precursors (ie cytokeratins) likely
provide the best DNA content measurement (chanel number) for the tumor
and its DNA diploid precursor, all in the same sample.
You have also raised the issue of mixing a portion of a "normal" sample
plus a portion of "tumor" or of overlaying histograms from these two
types of samples. The authors of the consensus felt that either approach
was likely to introduce artifacts. DNA binding dyes such as PI are
sensitive to chromatin configuration, the ratio of dye to DNA, and
techniques used to treat cells or nuclei (acid, proteolytic enzymes).
These variables can cause small changes in the peak channel measured for
two identical samples. For paraffin-embedded specimens, these factors are
further confounded by the degree of fixation of the tissue. In this
context, comparison of peak PI fluorescence of samples from two different
normal paraffin-embedded tissues will usually give different peak
fluorescence chanels.
Finally, there is the issue of "so what". Provided that careful
cytometric measurements are made on a sample, revealing a near diploid,
DNA aneuploid tumor, what does this do in the grand scheme of things. In
many types of tumors, a tumor with a DI up to 1.2
has no significantly different prognosis than a DNA diploid tumor (ie.
breast tumors). In many types of tumors, the finding of a near-diploid
DNA aneuploid population has not been demonstrated to carry a separate
prognostic significance.
Vince Shankey
Loyola Univ Med Cntr
CD ROM Vol 2 was produced by staff at the Purdue University Cytometry Laboratories
and distributed free of charge as an educational service to the cytometry community.
If you have any comments please direct them to Dr. J. Paul Robinson, Professor & Director, PUCL,
Purdue University, West Lafayette, IN 47907. Phone:(317) 494-0757; FAX (317) 494-0517; Web http://www.cyto.purdue.edu
EMAIL robinson@flowcyt.cyto.purdue.edu
