Adjuvant therapy
Despite substantial improvements in surgical technique and
postoperative care, colorectal cancer continues to kill 95 000
people in Europe alone each year.
Of the annual 150 000 newly diagnosed cases, about 80%
have no macroscopic evidence of residual tumour after
resection. More than half of patients, however, develop
recurrence and die of their disease. This is a result of occult
viable tumour cells that have metastasised before surgery and
which are undetectable by current radiological techniques (the
limit of detection of standard computed tomography is about
1cm3, equivalent to 109 cells).
Adjuvant treatment (chemotherapy and radiotherapy) has
developed as an auxiliary weapon to surgery and is aimed at
eradicating these micrometastatic cancer cells before they
become established and refractory to intervention. As the
presence of the primary tumour can exert an inhibitory
influence on micrometastases, theoretically the removal of the
tumour might stimulate growth of any residual cells, increasing
the proliferating fraction and rendering them more susceptible
to the cytotoxic effects of the widely used cytotoxic agent,
fluorouracil.
It is reasonable to predict therefore that the earlier
chemotherapy is started after surgery, the greater the potential
benefit, although this has not yet been formally addressed in
adjuvant trials. Implicit in this belief is a necessity for a
multidisciplinary effort between surgeon, oncologist, and the
community care team to provide seamless, streamlined cancer
care for the individual patient.
Pharmacology of fluorouracil
Fluorouracil has remained the cornerstone chemotherapy for
colorectal cancer for over 40 years. It is a prodrug that is
converted intracellularly to various metabolites that bind to the
enzyme thymidylate synthase, inhibiting synthesis of thymidine,
DNA, and RNA. Increasing understanding of the molecular
pharmacology of fluorouracil has led to the development of
strategies to increase its efficacy.
The first strategy to be tested was coadministration with the
immunostimulatory, antihelminthic drug levamisole, but despite
promising early results, recent trials have not convincingly
shown significant improvements in outcome compared with
fluorouracil alone. In addition, no persuasive mechanism for the
assumed synergism between fluorouracil and levamisole has
been found.
In contrast, addition of folinic acid increases and prolongs
the inhibition of the target enzyme (thymidylate synthase) and
seems to confer improved clinical outcome compared with
fluorouracil alone in advanced disease and when used in
adjuvant therapy.
The side effects of chemotherapy based on fluorouracil vary
according to the regimen (most commonly given as bolus
intravenously daily for 5 days every 4 weeks or bolus weekly).
They include nausea, vomiting, an increased susceptibility to
infection, oral mucositis, diarrhoea, desquamation of the palms
and soles, and, rarely, cardiac and neurological toxic effects
Established benefits of fluorouracil
based adjuvant chemotherapy
Early adjuvant trials were retrospective and underpowered and
failed to show any therapeutic benefit with respect to recurrence
rate or survival. In 1990, however, the results of the intergroup
trial were published. In this study 318 patients with stage B
colorectal malignancy were randomised for surgical treatment
alone or surgery followed by fluorouracil plus levamisole. In
addition, 929 patients with stage C malignancy received surgery
alone, surgery plus levamisole, or surgery plus fluorouracil and
levamisole. For these patients there was a 33% reduction in the
odds of death and a 41% decrease in recurrence among those
treated with fluorouracil plus levamisole compared with surgery
alone or surgery plus levamisole.
In contrast with levamisole, combining folinic acid with
fluorouracil is pharmacologically rational, and documented
benefit in advanced disease led to the logical extension of this
combination into adjuvant therapy. Three large randomised
adjuvant phase III trials produced confirmatory evidence of
improved, disease-free survival at three years and improved
overall survival in patients treated with fluorouracil plus folinic
acid, with a 25-30% decrease in the odds of dying from colon
cancer (or an absolute improvement in survival of 5-6%
compared with controls).
Recently a meta-analysis of updated individual data from all
unconfounded randomised studies of adjuvant chemotherapy
(including the above three trials) has been undertaken
(Colorectal Cancer Collaborative Group, unpublished). Overall,
there was a 6-7% absolute improvement in survival with
chemotherapy compared with surgery alone (SD 2.3, P = 0.01).
The analysis advised that on current evidence the combination
of fluorouracil plus folinic acid should be accepted as
“standard” adjuvant chemotherapy for patients with Dukes’s
type C colon cancer.
Controversies in adjuvant therapy
Despite convincing evidence that adjuvant chemotherapy
improves disease-free survival and overall survival in Dukes’s
type C colon cancer (an estimated six deaths prevented for 100
patients treated), several controversies surrounding the
application of this form of treatment still exist.
Length of treatment and optimal dose of fluorouracil plus
folinic acid
Lengthy adjuvant treatment has adverse effects on patients’
quality of life as well as financial implications. A recent North
American study, however, has shown that six months’ treatment
is as effective as 12 months’.
Determining the optimal dose is important: high dose
folinic acid is 10 times as expensive as low dose. This issue has
been addressed in the “certain” arm of the United Kingdom
Co-ordinating Committee on Cancer Research’s QUASAR
(“quick and simple and reliable”) trial (patients with Dukes’s
type C colon cancer). The trial uses the principle of
randomising according to certain or uncertain indication: if, for
a particular subgroup of patients the worth in receiving some
form of adjuvant chemotherapy is definitely established from
published randomised controlled trials (for example, patients
with Dukes’s type C colon cancer) then these patients are
randomised to the certain indication arm (with a choice of
different drugs and regimens); if, however, no definitive
evidence exists of worth in a particular subgroup (for example,
in patients with Dukes’s type B colon cancer or with rectal
cancer) then the patients are randomised into the uncertain
indication arm (chemotherapy v no chemotherapy). The results
from QUASAR’s certain arm show that neither high dose
folinic acid nor levamisole contribute to improved survvial.
Role of adjuvant chemotherapy in lower risk groups
Inadequate data exist on the effect of chemotherapy in stage B
colon cancer. The proportional reduction in annual risk is
probably similar for stage B and stage C patients. If the
proportional reductions in mortality are similar, the absolute
benefits in terms of five year survival would be somewhat
smaller for stage B patients than for stage C patients because of
lower risk of recurrence (perhaps two to three lives saved per
100 patients treated).
Patients with stage B cancers who have prognostic
indicators that suggest a high risk of recurrence (for example,
perforation, vascular invasion, poor differentiation) might
benefit proportionately more than patients with stage B cancer
without high risk indicators and these variables might define a
subgroup of patients who might merit adjuvant chemotherapy.
Little evidence exists, however, on the prognostic predictability
of these various features.
Use of adjuvant therapy in rectal cancer
Insufficient evidence exists to support the routine use of
systemic chemotherapy in either Dukes’s type B or type C rectal
cancer. Anatomical constraints make the rectum less accessible
to the surgeon, so it is much more difficult to achieve wide
excision of the tumour, and about 50% of recurrences are in the
pelvis itself rather than at distant sites. This means that locally
directed radiotherapy is a useful adjuvant weapon, and this has
been assessed for rectal cancer both before and after surgery.
In the largest trial of preoperative radiotherapy (the Swedish
rectal cancer trial), radiotherapy produced a 61% decrease in
local recurrence and an improvement in overall survival (58% v
48%) compared with surgery alone. Radiotherapy after surgery
seems to be less effective, even at higher doses, possibly because
of rapid repopulation of tumour cells after surgery or relative
hypoxia around the healing wound.
Only one trial, the Uppsala trial in Sweden, has directly
compared radiotherapy before and after surgery. Despite a
higher dose after surgery, a significant reduction occurred in
local recurrence rates among patients treated before surgery
(12% v 21%, P < 0.02).
Animal studies have suggested that fluorouracil may prime
the tumour cells and increase the cytotoxic effect of subsequent
radiotherapy. Some clinical data support the role of
chemoradiotherapy combinations in rectal cancer, but further
clinical evidence of benefit needs to be provided before this
treatment could be considered for routine use. The uncertain
arm of the QUASAR trial will help to resolve this issue.
Role of portal venous infusional therapy
Fluorouracil is an S phase specific drug, and yet its active
metabolites have a half life of about 10 minutes, which limits its
target, when given as a bolus, to the small fraction of cells in the
S phase at the time of administration. Infusional therapy can
therefore affect a greater proportion of cells. In addition, the
most common site for micrometastases after resection of a
colorectal tumour is the liver. In contrast with macroscopically
identifiable metastases of advanced disease, which derive their
blood supply from the hepatic artery, these micrometastases are
thought to be supplied by the portal vein. Therefore delivering
chemotherapy via the portal vein should provide high
concentrations of the drug at the most vulnerable site and lead
to substantial first pass metabolism, which should attenuate any
systemic toxicity. The established regimen for portal fluorouracil
in adjuvant therapy is a course of 5-7 days starting immediately
after surgery. A meta-analysis of 10 randomised trials showed a
4.7% improvement in absolute survival with portal venous
infusion therapy compared with surgery alone; however, the
confidence intervals were wide and the statistical benefit is not
robust. Indeed the AXIS trial, the largest single trial of portal
venous infusion to date, randomising 4000 patients after
surgery either to the infusion therapy or to observation alone at
five years, suggests no significant differences in overall survival.
Future role of adjuvant therapy
The use of adjuvant therapy in colorectal cancer over the past
40 years has centred on fluorouracil, alone and in combination,
and on the fine tuning of regimen and route of administration.
Current trials are considering new drugs (eg irinotecan and
oxaliplatin) and their sequencing, as well as innovative
techniques, such as immunotherapy and gene therapy. These
techniques will be considered in detail later in the series. Gene
therapy and immunotherapy are likely to function optimally,
however, when cellular load is low, blood supply is good, and
small clusters of cells are surrounded by effectors of the
immune system; these therapies may therefore be most suitable
as adjuvant therapy rather than for use in advanced disease.
All cytotoxic agents are rigorously tested and applied in
advanced disease before being used in adjuvant therapy. New
agents that are now entering adjuvant trials will be fully
described in the next article in this series.
