Rd. Blumenthal et al., COMPARISON OF EQUITOXIC RADIOIMMUNOTHERAPY AND CHEMOTHERAPY IN THE TREATMENT OF HUMAN COLONIC-CANCER XENOGRAFTS, Cancer research, 54(1), 1994, pp. 142-151
The therapeutic efficacy of 5-fluorouracil (5-FUra; 0.6 mg/day x 5 day
s) + leucovorin (LV; 1.8 mg/day x 5 days) and of I-131-labeled MN-14 a
nticarcinoembryonic antigen IgG (275 muCi single dose) was evaluated i
n size-matched (0.3-0.7 cm3) s.c. LoVo, HT-29, DLD-1, HCT-15, LS174T,
and MOSER, GW-39, and WidR human colonic tumors. These lines express v
arying amounts of carcinoembryonic antigen and exhibit varying degrees
of in vitro responsiveness to 5-FUra. Unlike radioimmunotherapy (RAIT
), multiple cycles of chemotherapy were feasible over a 3-week period.
However, no therapeutic advantage to a second cycle of 5-FUra/LV admi
nistration was found. Therefore, it is reasonable to compare single cy
cles of both treatment modalities. RAIT was statistically more effecti
ve in 5 of 8 tumor lines (LoVo, LS174T, MOSER, WidR, and GW-39). In 1
other line (DLD-1), RAIT was marginally more efficacious, but tumors r
esponded well to both therapies. The lack of a statistical difference
between the 2 modalities of treatment may indicate that the efficacy o
f the 2 treatments is equivalent, or the relatively large variability
within the treatment groups may have prevented significance given the
number of animals evaluated. RAIT and 5-FUra/LV were equally efficacio
us in the HT-29 and the HCT-15 tumor lines. Of the 5 xenografts that r
esponded better to RAIT, 3 lines (LS174T, GW-39, and WidR) demonstrate
d a greater percentage of tumors responding over a 5- to 6-week period
. The other 3 lines (LoVo, MOSER, and DLD-1) exhibited a similar perce
nt of tumors responding to both therapies, but a greater growth inhibi
tion in those RAIT-treated tumors that responded. In vitro responsiven
ess to 5-FUra/LV did not directly correlate with in vivo responsivenes
s (r2 = -0.664), since LS174T and LoVo tumors, with rapid growth rates
(0.05-0.36 cm3/day), were not highly responsive to therapy. Growth in
hibition from RAIT also did not correlate with total tumor carcinoembr
yonic antigen content (r2 = 0.003), an observation that may be due to
additional variables, such as accessibility of antigen and innate radi
osensitivity of the tumor. RAIT was most effective in the fastest grow
ing tumor lines (LS174T, GW-39, MOSER, WidR, and LoVo). These preclini
cal results suggest an advantage to radioantibody therapy over one of
the most commonly used forms of chemotherapy to treat colorectal cance
r. These studies also highlight the need to establish criteria that wi
ll enable the selection of therapeutic modalities in patients.