Phase III randomized clinical trials have greatly contributed to our u
nderstanding of the pathobiology of neoplastic disease and, particular
ly, to therapeutic progress. However, randomized Phase III studies are
no better than or are critically dependent on Phase I and Phase II st
udies for positive therapeutic leads that are compelling enough to tes
t in the Phase III arena. The variables involved in the series of rand
omized trials that led to the curative treatment of acute lymphocytic
leukemia also resulted in an understanding of the principles of cancer
therapy in therapeutic research. These principles, when applied to Ho
dgkin's disease in non-Hodgkin's lymphoma, testis cancer, childhood so
lid tumors, and others, resulted in a substantial cure rate for those
diseases. However, for the adult epithelial common solid tumors, a sec
ond strategy, adjuvant chemotherapy, was required. This has resulted i
n a 20% reduction in mortality in patients with node positive and node
negative breast cancer. Tamoxifen has been similarly effective in pat
ients with postmenopausal breast cancer. In colon cancer, adjuvant che
motherapy with fluorouracil plus levamisole has decreased mortality to
a comparable degree. New agents, modulations, combination chemotherap
y, and biotherapeutics are being addressed to the adjuvant situation w
hich has proven effective in a variety of neoplastic diseases. A third
strategy is neoadjuvant chemotherapy. This involves the use of chemot
herapy first for patients with solid tumors, designed to down-stage th
e primary tumor, thus making it more susceptible to less radical surge
ry and to organ- or limb-sparing procedures in osteogenetic sarcoma an
d in head and neck cancer. For example, neoadjuvant chemotherapy has n
ot resulted in an increased survival as compared with the appropriate
control but has allowed for important quality-of-life contributions, s
uch as limb-sparing and radical surgery-sparing procedures. In additio
n to new agents and combination chemotherapy, dose is a critical varia
ble. This is most evident clinically in the transplantation arena. Com
parative studies recently completed, for example, in patients with adj
uvant breast cancer and with acute leukemia indicate that dose is a si
gnificant factor in tumor control.