successful, or that the transduction was 
unsuccessful, or that clonal evolution occurred which 
favored the untransduced cells, thus making the 
marking uninterpretable. 
4. The CLL cell is marked. There are three 
possibilities: both markers are positive, suggesting 
that the infused purged cells contributed to the 
origin of relapse, and that neither of the 
fractionation steps were effective. Another 
possibility is that the LNL6 is positive in the 
normal and leukemic population, but that the GINa is 
positive in the normal but not the leukemic 
population. This result suggests that the CD19 purge 
was not effective but that the CD34 purge is 
effective. The other possibility, that the LNL6 is 
negative and the GINa is positive in the leukemic 
population and that the LNL6 is positive in the 
normal population, suggests that the CD19 purge is 
ineffective, and that for some reason, the clonal 
evolution of the leukemic population suppressed the 
cells which were double virus positive. 
Therefore, the analysis identifies eight possible 
types of cells, allows unambiguous interpretations as 
to the efficacy of the two purging steps with respect 
to the removal of the leukemic population, and 
permits assessment of the effect of the two purging 
steps on the positive selection of the early normal 
progenitor cells. 
In addition to the PCR analysis above, the use of 
interphase centromeric FISH for the trisomy 12 
technique can be carried out by Dr. Michael Andreeff. 
In this context of long term remissions after intensive 
therapy, it becomes important to determine whether the 
relapse emerges from cells remaining in the infused 
marrow or results from residual disease which remains 
after systemic therapy used to prepare the patient for 
transplantation. This question becomes even more 
important to resolve and to measure now that 
immunological methods for removing or separating the 
neoplastic B cells from the autologous marrow and 
peripheral blood of CLL patients are available (9) . 
No good animal model or in vitro model exists which would 
permit us to identify the origin of relapse in CLL 
patients treated with intensive therapy and autologous 
stem cell transplants (residual systemic disease or 
leukemia cells contaminating infused autologous stem 
cells marrow) . Recently, Anderson and Rosenberg, at the 
NIH (11) have used replication incompetent retroviruses 
of the N2 series, first developed by A. Dusty Miller of 
[392] 
Recombinant DNA Research, Volume 16 
