Damjanov and Meropol review the status of several oral
fluoropyrimidine drugs that are currently undergoing clinical testing
for the treatment of colorectal cancer. Meta-analysis indicates an 8%
improvement in the overall response rate and a survival advantage of
0.8 months with infusional vs bolus fluorouracil (5-FU)arguably
small differences. The incidence of grade 3-4 diarrhea, mucositis,
and myelosuppression is < 10% with infusional 5-FU, although
hand-foot syndrome is commonly associated with this therapy. The
prevailing strategy for the development of oral 5-FU drugs employs
multiple daily dosing for periods of 2 to 4 weeks followed by a 1- or
2-week break in order to mimic the protracted infusional schedule.
Although preclinical studies of uracil plus tegafur (UFT) documented
higher intratumoral 5-FU levels following administration of UFT vs
ftorafur alone, UFT has not been directly compared to ftorafur in
clinical trials. The benefit of combining leucovorin with UFT vs UFT
alone has also not been investigated. Based on the activity seen in
phase II studies, however, clinical development in Western countries
has involved UFT plus oral leucovorin calcium (Orzel) on a monthly schedule.
Two large randomized trials comparing UFT/leucovorin with a monthly
schedule of bolus 5-FU/leucovorin demonstrated apparent equivalence.
Although patients in the UFT arms had significantly less neutropenia
and infection, quality of life was similar for the two combinations.
Since the Food and Drug Administration requires evidence that each
component of a fixed-combination tablet must contribute to the
therapeutic effect, the new drug application for Orzel was recently withdrawn.
Through inactivation of dihydropryimidine dehydrogenase, eniluracil
permits complete oral bioavailability of 5-FU, markedly prolongs its
half-life, prevents the formation of catabolites that might
contribute to clinical toxicity or interfere with 5-FU cytotoxicity,
and circumvents a potential resistance mechanism.[1,2] Initial
studies indicated that single daily dosing of eniluracil (3.7
mg/m²) rendered oral 5-FU completely bioavailable, and the
effect on 5-FU clearance is similar to that seen with eniluracil in
an oral dosage of 20 mg bid.
Current studies employ a fixed-combination tablet of eniluracil/5-FU
in a 10:1 ratio (with the 5-FU dose of 1.0 to 1.15 mg/m² bid for
28 of 35 days). Ongoing phase III trials are comparing eniluracil/5-FU
with either infusional 5-FU or monthly bolus 5-FU/leucovorin.
Oral administration of doxifluridine (5 ×-deoxy-5-fluorouridine)
was supplanted by the use of capecitabine (Xeloda). The currently
approved schedule for this agent is 2,500 mg/m² po in two
divided doses for 14 of 21 days.
Results from two large, randomized trials of capecitabine vs monthly
bolus 5-FU/leucovorin were presented at the 1999 American Society of
Clinical Oncology meeting. The efficacy data for the American trial
pertained to a subset of 534 of 605 randomized patients. A response
advantage in favor of capecitabine was reported, but no significant
difference in either median time to progression or survival was seen.
In the second trial, the response rates ascertained by an independent
review committee were similar19% and 15%, respectively, for the
capecitabine and 5-FU/leucovorin recipientsand median time to
progression was comparable.
An appealing feature of capecitabine therapy is the potential for
increased activation of 5-FU in tumors that contain high levels of
thymidine phosphorylase (TP). Thymidine phosphorylase activity in
human tumors from 11 different sites of origin showed that, in
general, the tumor tissue had higher activity than the corresponding
In this study, intratumoral 5-FU levels were measured in 19
colorectal cancer patients who received capecitabine, 1,255 mg/m
× po bid for 5 to 7 days prior to elective surgery. In 11
patients, the median 5-FU level in primary tumor tissue was 2.9-fold
higher than in adjacent normal mucosa. In 10 paired metastatic and
normal liver samples, the median 5-FU in tumor was 1.4-fold higher.
These two studies support the potential for selective activation of
5-FU in tumors. However, colorectal cancer patients with high TP
levels were unlikely to respond to 5-FUbased therapy,
suggesting that high TP expression may be a poor prognostic
factor. Whether tumors with high TP expression are sensitive to
capecitabine is therefore an intriguing question that is being
addressed in clinical studies.
As yet, it is unclear whether any one of the oral fluoropyrimidines
is superior to the others, and whether the currently employed
schedules are indeed optimal. There are strong regional preferences
outside the United States for a variety of infusional 5-FU regimens
(eg, high-dose weekly 24- to 48-hour infusions, twice-monthly
combined-bolus/infusional regimens, and chronomodulated regimens).
Few clinical studies of oral fluoropyrimidine agents have tried to
mimic these intermittent high-dose regimens.
The contribution of leucovorin to infusional 5-FU regimens is
unclear. The results of a randomized trial by the Southwest Oncology
Group suggested no advantage to the addition of low-dose leucovorin
to protracted infusional 5-FU. Preliminary data from a European
phase III trial suggest an advantage in median progression-free
survival with the addition of high dose leucovorin to a high dose
weekly 24-hour infusion, at the cost of more diarrhea.
Oral agents representing other classes of drugs that may be useful in
colorectal cancer treatment are also in clinical development. These
compounds include irinotecan (Camptosar), 9-nitrocamptothecin (which
acts by topoisomerase-I targeting), thalidomide (an antiangiogenesis
agent), and marimastat (an inhibitor of matrix metalloproteinases).
Studies seeking to draw correlations between the molecular profile of
a tumor and benefit with a specific therapy are a high priority.
Since resistance to many chemotherapy agents is multifactorial,
prospective trials designed to evaluate the importance of a single
determinant of sensitivity may not tell a complete story.
Clearly, prospective trials that investigate the value of selecting
therapy on the basis of the tumors molecular profile need to be
performed. Emerging technologies hold forth the promise that not just
a few, but dozens, hundreds, and even thousands of molecular targets
will be quantifiable. The implications for how to design clinical
trials to test the utility of such complex molecular profiles is staggering.
1. Zhang R, Soong S-J, Liu T, et al: Pharmacokinetics and tissue
distribution of 2-fluoro-b-alanine in rats: Potential relevance to
toxicity pattern of 5-fluorouracil. Drug Metab Dispos 20:113-119, 1992.
2. Cao S, Baccanari DP, Rustum RM, et al: 2-Fluoro-b-alanine: Effects
on the antitumor activity and toxicity of 5-fluorouracil. Biochem
Pharmacol 59:953-960, 2000.
3. Miwa M, Ura M, Nishida M, et al: Design of a novel oral
fluoropyrimidine carbamate, capecitabine, which generates
5-fluorouracil selectively in tumours by enzymes concentrated in
human liver and cancer tissue. Eur J Cancer 34:1274-1281, 1998.
4. Schüller J, Cassidy J, Dumont E, et al: Preferential
activation of capecitabine in tumor following oral administration to
colorectal cancer patients. Cancer Chemother Pharmacol 45:291-297, 2000.
5. Metzger R, Danenberg K, Leichman CG, et al: High basal level gene
expression of thymidine phosphorylase (platelet-derived endothelial
cell growth factor) in colorectal tumors is associated with
nonresponse to 5-fluorouracil. Clin Cancer Res 4:2371-2376, 1998.
6. Leichman CG, Fleming TR, Muggia FM, et al: Phase II study of
fluorouracil and its modulation in advanced colorectal cancer: A
Southwest Oncology Group study. J Clin Oncol 13:1303-1311, 1995.