leukemic hematopoieti-c progenitor cells from patients with 
CML have many differences from normal cells, and may behave 
differently during hematologic reconstitution and as targets 
for gene transfer. 5 In the CML protocol peripheral blood 
stem cells are collected while the patient is on G-CSF 
during recovery from high-dose daunorubicin and ara-C 
chemotherapy given to decrease the percentage of bcr-abl+ 
cells, and bone marrow is harvested after 5-f luorouracil 
treatment. 6-8 The conditioning regimen in the CML protocol 
involves very-high dose cyclophosphamide and total body 
irradiation, and would be considered ablative. 
In the myeloma protocol, peripheral blood cells for 
transplantation are collected after treatment with high-dose 
cyclophosphamide and G-CSF, and bone marrow is harvested 
after 5-f luorouracil administration. Conditioning consists 
of very-high dose melphalan and total body irradiation, and 
is ablative. The breast cancer protocol does not include 
peripheral blood stem cell harvest, but the marrow is 
collected soon after completion of standard multi-agent 
chemotherapy. The conditioning regimen (ICE) is not 
completely ablative. If there are major differences in the 
pattern of autologous reconstitution or in the efficiency of 
gene transfer between the three patient groups, then the 
results will guide future gene therapy protocols. 
Disease-directed gene therapy of both congenital deficiency 
states and hematopoietic malignancies, or gene therapy 
strategies designed to protect normal hematopoietic stem 
cells from anticancer therapies will only be possible if 
long-term reconstituting stem cells can be transduced. 9 ' 10 
Marking studies such as this one will lay the groundwork for 
assessing the impact of source of target cells (i.e. 
peripheral blood versus bone marrow) , pretreatment of the 
donor with chemotherapy or growth factors, and type of 
conditioning regimen on gene transfer efficiency. Pilot 
studies using a neutral marker gene are necessary to assess 
the feasibility and safety of most gene transfer procedures 
before using therapeutic vectors that could themselves 
perturb hematopoiesis. Therapeutic vectors for all three 
diseases are in preclinical development in our laboratory. 
An interleukin-6 autocrine-paracrine loop may be central to 
the development and propagation of multiple myeloma. 11-13 It 
is possible that this loop could be interrupted with soluble 
gpl30 (part of the interleukin 6 receptor complex) produced 
locally in the marrow by a retroviral vector. Retroviral 
delivery of leukemia inhibitory factor (LIF) to the marrow 
microenvironment may give normal progenitor and stem cells 
an advantage over leukemic cells. 14 Transduction of normal 
progenitor or stem cells from patients with breast cancer 
with the multidrug resistance gene (MDR1) might permit the 
[ 8 ] 
Recombinant DNA Research, Volume 16 
