Annals of Oncology 8: 575-581, 1997.

© 1997 Kluwer Academic Publishers. Printed in (he Netherlands.

Original article

Preliminary results of a phase II study of high-dose radiation therapy and

neoadjuvant plus concomitant 5-fluorouracil with CDDP chemotherapy for
patients with anal canal cancer: A French cooperative study*

D. Peiffert,1 J.-F. Seitz,2 P. Rougier,3 E. Francois,4 F. Cvitkovic,5 X. Mirabel,6 S. Nasca,7

M. Ducreux,3 J. M. Hannoun-Levi, 2 A. Lusinchi,3 E. Debrigode,8 T. Conroy,1 J. P. Pignon 3 &
J.P.Gerard 9
'Centre Alexis Vautnn, Nancy; 2Institut J. Paoli I. Calmetle, Marseille; ^Institut Gustave Roussy, Villejuif; * Centre Antome Lacassagne, Nice;
5
Centre Rene Huguenin, St Cloud; ^Centre Oscar Lambret, Lille; 1 Centre Jean Godinot, Reims, ^Centre Vald'Aurelle, Montpellier; 9H6pital Lyon
Sud, Lyon, France and the Digestive Tumors Group of the French 'Federation Nationale des Centres de Lutte Contre le Cancer'

Summary lism on D4. The remaining 29 received the entire treatment,

Background: Chemotherapy (5-fluorouracil-mitomycin Q con-
comitant with radiotherapy (RT) increases local control and
colostomy-free survival in advanced anal canal carcinomas
(ACC). The purpose of this prospective trial was to analyse
the toxicity of and response to an induction chemotherapy
combining 5-fluorouracil (5-FU) and CDDP administered
concomitantly with irradiation.
Patients and methods; Thirty patients (24 F/6 M, mean age
60, range 38-74) with an advanced ACC > 40 mm and/or with
node involvement were prospectively treated (1 Tl, 16 T2, 8 T3,
5 T4,10 Nl, 1 N2, 8 N3) from November 1994 to January 1996.
Two induction and two concomitant cycles of 5-FU (800 mg/
m2 Dl^t infusion) and CDDP (80 mg/i.v./m2 at Dl) were
delivered. RT consisted of 45 Gy (1.8 Gy/fr, 5 fr/w) on pelvis
± inguinal nodes or 30 Gy (3 Gy/fr, 4 fr/w) by direct perineal
field. A boost (15-20 Gy) was delivered six weeks later.
Results: Toxicity one patient died of a pulmonary embo- study, radiation therapy
with reduced 5-FU doses in 11 patients because of acute tox-
icity. The RT boost was delayed for one patient (aplasia). In
109 cycles, 3 grade 4 and 17 grade 3 toxicities were observed;
there were no toxic deaths. Tumor response: the complete
response (CR) and partial response (PR) rates were, respec-
tively, 11% and 61% after induction chemotherapy, 59% and
31% after concomitant radiochemotherapy and 96% and 0%
two months after completion of the treatment. No tumor
progression was observed.
Conclusion: the treatment was well tolerated and there was
good compliance. After induction chemotherapy, most of the
patients were in PR, with some even in CR. After completion
of the treatment all but one were in CR. The tumor response
and the long term results of 50 patients will be analysed before
initiation of a randomised trial is considered.
Key words: anal cancer, cisplatin, 5-fluorouracil, phase II

Introduction tases are frequent, with 40% of cancer deaths caused by

disease outside the pelvis in the UKCCCR trial [1].
The prognosis of anal canal carcinomas remains poor. There is a need to increase the therapeutic ratio of the
On the basis of the results of recent randomised trials, combined treatments by dose intensification, employing
concomitant radiotherapy and chemotherapy combin- the spatial and additive cooperation of both chemo-
ing 5-fluorouracil (5-FU) and mitomycin C (MMC) therapy and radiotherapy.
must be considered the standard treatment for locally Cisplatin (CDDP) containing regimens (usually with
advanced anal carcinoma [1, 2], The objective role of 5-FU) have been used for anal cancer, with an objective
MMC in the improvement of the results has been proven response rate of 55% in recurrent or metastatic disease
despite its related late complications [3]. However, in [4-6] and of 82% in neoadjuvant schemes [7].
locally advanced tumours both the two-year local recur- When CDDP replaced MMC in concomitant treat-
rence rate of 25% and colostomy rate of 30% remain ments of head and neck tumours it seemed to cause less
high after combined treatment. The complete response toxicity. Preliminary results in the anal canal [8-10]
rate after conservative treatment for locally advanced confirmed the good tolerance of this schedule, with an
tumours is 83% (biopsy proven) in the RTOG trial [3] overall response rate of 95%.
and 80% in the EORTC trial [2]. Metachronous metas- CDDP with 5-FU concomitant with a salvage irradia-
tion has yielded complete histologic responses in 12 of
' Presented as an Oral Communication at the 21st European Society of 22 patients with positive biopsies after initial 5-FU-
Medical Oncology, October 1996, Vienna. MMC combined treatment [3].
576
The impact of induction chemotherapy with 5-FU—
CDDP on pharyngo-laryngeal conservative treatment
[11, 12] has recently been proven: organ preservation
was achieved by radiation therapy in good responders,
the surgical salvage was possible under good conditions
in non-responders, and distant metastases were statisti-
cally delayed, and perhaps averted.
To obtain the best therapeutic ratio in the anal canal
without risk of tumour progression in poor responders,
we designed an original scheme combining an induction
chemotherapy with a concomitant radiochemotherapy.
An induction chemotherapy aims to obtain a partial or
complete response so that the irradiation can be started
on a less infiltrative tumour, to obtain a higher local
control and a lower risk of complications. Its role in
distant metastasis has still to be evaluated. The aim of
the concomitant chemotherapy was to increase the local
control. The choice of a 5-FU-CDDP combined sched-
ule was based on the lower toxicity of CDDP, given
concomitantly with irradiation, compared with MMC.
The split course scheme for irradiation was maintained,
based on the results of Cummings [13] and Papillon [14],
and on the impact of a boost on a subclinical residual
disease [3].
The objective of this study was to analyse in a multi-
centric trial the toxicity and the tumour response of an
induction and concomitant 5-FU-CDDP chemother-
apy for locally advanced anal canal carcinomas.
Patients and methods
The phase II study was approved by the Nancy University Ethics
Committee (Comite de Protection des Personnes se pretant a la
Recherche Biomedicale), and all of the patients gave their informed
consent. Eligibility criteria included the presence of a biopsy proven
squamous cell carcinoma of the anal canal (International Union
Against Cancer UICC 1987 classification [15], margin primaries ex-
cluded), WHO performance status 0-1, no distant metastasis, tumour
size greater than 40 millimetres and/or involved nodes, age 75 years or
less, seric creatinine =£ 130 mmol/1, no contra-indication to 5-FU, no
other tumour and no previous treatment of the anal canal tumour.
Pretreatment evaluations included history and physical examina-
tion, with documentation of measurable disease and performance
status, anorectal functional status, complete blood count, ECG, chest
radiography, abdominoperineal CT-Scan, endorectal ultrasonography.
Treatment
The treatment scheme (Figure 1) combined four cycles of chemo-
therapy, the first two as induction chemotherapy, the third and fourth
cycles administered concomitantly with the first course of irradiation,
and an irradiation boost after a one to two months interval. The
patients received one cycle of chemotherapy every 28 days consisting
of 5-FU, 800 mg/ra 2 per day by continuous infusion on day 1 through
day 4, and of CDDP, 80 mg/m 2 intravenously on day 1 in a one-hour
infusion, CDDP given with the usual pre- and posthydration proto-
col (at least 4 litres during the day of CDDP infusion). During the
cycles delivered concomitantly, the irradiation was started one to two
hours after injection of the CDDP. Reduction of the 5-FU dose was
prescribed in instances of neutropenia, thrombopenia, diarrhea, or
mucositis (WHO scores), as follows: 25%, 50% and 100% for grade 2,
grade 3 and grade 4, respectively. The CDDP was not prescribed when
Cycle «1 Cycle #2 Cycle #3 Cycle # 4
Dl
t
D28
t
D56
t
D84
Gap Boost
6 weeks
BT
: continuous 5-Fluorouradl infusion during 4 days: 800 mg / m 2 / day
: d> platinum: IV bolus on 1 hour before 5 FU on D»y 1:80 mg / m 2 / d«y
;-EBI \ : External Beam Irradiation (see text)
Pelvic fields: 45 Gy / 1.8 Gy per fraction / 5frper week
or perineal field : 36 Gy / 3 Gy perfr/ 4frper week
: EBI: 2 Gy per fr / 5frper w 20Gy if Pirtiil response
15 Gy if Complete response
: 192 IT interstitial brachytherapy boost
Figure 1. Study scheme.
seric creatinine was > 130 mmol/1, if the increase of the basal value
was higher than 50%, or in instances of grade 2-4 neurologic toxicity.
The fourth cycle of chemotherapy was not prescribed when the irradia-
tion was to be delivered within three weeks (perinea! technique). The
usual recommendations for antiemetic therapy before the infusions
were adapted for the individual local practices or institutions. The
routine use of hematopoietic growth factors or antibiotics was not
planned.
The radiotherapy was delivered with megavoltage therapy units.
According to clinical presentation, the recommended target volume
was the anorectal region and the pelvic nodal areas, including the
unilateral or bilateral inguinal nodes if involved Depending on local
practice, the small-volume technique of irradiation (with 6 MV X-rays
or Cobalt gamma-rays) could be used, not including the pelvic nodes
(perineal technique) described by Papillon [14]. The boost was limited
to the initially involved areas.
The techniques used for the first course of irradiation were: 1) A
box technique (nine patients) with four orthogonal fields (Antero
posterior (AP), Postero-anterior (PA) and 2 Lat fields). The superior
border of the fields was L5-S1, the lower border included the penanal
region, and the lateral borders the pelvic rim, with blocks shielding the
small bowel. The dose delivered on the isocenter was 45 Gy, in 25
fractions and five weeks ; 2) Large AP-PA fields (13 patients) including
the inguinal nodes with larger lateral borders including the inguinal
areas (45 Gy in 25 fractions and five weeks). The clinically involved
inguinal fields were boosted till 60-65 Gy; 3) A perineal technique
(seven patients) in tumours without node involvement, including only
the anorectal area and perirectal nodes. The dose was delivered in the
canal, prescribed at 5 cm depth on the 95% isodose, 3 Gy per fraction,
four fractions per week, in three weeks.
After an interval of four to six weeks (mean = 5 weeks, range =
3-10 weeks), the tumour bed was boosted by external beam irradiation
(X-rays or electrons) or Paris System 192-Iridium interstitial brachy-
therapy, and 20 Gy or 15 Gy was delivered in instances of partial
response or complete response, respectively.
Response assessment
Response and toxicity were graded according to the WHO criteria [15]
All of the patients were scheduled for two induction cycles and the one
or two cycle(s) of chemotherapy concomitant with the first course of
irradiation in order to be eligible for response. The decision of a
definitive conservative treatment was taken after the interval, depend-
ing on the response and the possibility of obtaining a good functional
result.
Toxicity was assessed after each cycle of chemotherapy, one month
after the end of the first course of the radiotherapy (before the boost or
 577
the surgery) and two months after the end of the treatment. The results dermitis, one grade 2 neutropenia; 50% dose reduction:
of the response were determined after the two cycles of induction
one grade 2 and two grade 3 diarrhea). The total dose of
chemotherapy, after the end of the first course of radiotherapy (before
the boost) and two months after the end of the treatment. CDDP could be prescribed to all of the patients.
One cycle was delayed for three weeks in one patient
following a febrile aplasia. Another patient with a grade
4 neutropenia at the first cycle received granulocyte
Results colony-stimulating factor instead of a dose reduction,
presented no further hematologic toxicity, and could
Patient characteristics then be given the full doses.
Of the 22 patients who received the fourth cycle of
From November 1994 to January 1996, 30 patients, 24 chemotherapy nine (41%) had dose reductions, and 13
females and six males with a mean age of 60 years received the total dose.
(range 38-74), were included. According to the 1987
UICC classification [16], there were 1 Tl, 16 T2, 8 T3, 5
T4,10 NO, 11 Nl, 1 N2, 8 N3 (Table 1). The mean tumour
size was 42 mm (range 15-75), with an invasion of the
circumference lower or equal to one-third in 12 patients, 30pts INCLUSION
between one- and two-thirds in 14 patients, and larger
than two-thirds in four patients. All of the patients were
followed and evaluated for toxicity and response.
30pts Cycle #1 1 death (at Day 4 from pulmonary
Of the 30 patients included, one died of a pulmonary embolism)
embolism during the first cycle of induction chemother-
apy on day 4. The remaining 29 completed the treatment
\t
(Figure 2): 25 received the definitive conservative sched-
29pts Cycle # 2
ule, one received the schedule followed by an abdomino-
perineal resection for no response, and three received
the induction and concomitant chemotherapy and the
first course of radiotherapy, but were surgically salvaged
for no change or partial response. 29pts Cycle # 3

Toxicity during the chemotherapy

Cycle # 4
In general, the toxicity of the treatment was considered
tolerable; the results are described in Table 2. 3pts
A total number of 109 cycles of chemotherapy were Abdominoperineal resection
delivered. None of the seven patients treated with the lpt
direct perineal field technique received the fourth cycle 26pts BOOST
4pts
of chemotherapy.
The full prescribed dose of the first cycle was deliv-
25pts
ered for all of the patients. A reduction of the 5-FU dose
was prescribed and delivered for the second and sub- 2 month follow-up 29pts
sequent cycles in two patients (25% dose reduction for
grades 2 and 4 mucositis), for the third and subsequent Figure .?. Treatment sch
cycles in two patients (25% dose reduction: two grade 3
neutropenia), for the fourth cycle in seven patients (25%
dose reduction: two grade 2 diarrheas, one grade 3 Table 2. Acute toxicity.

Cycle # 1 Cycle # 2 Cycle # 3 Cycle # 4

Table 1. Distribution of the tumors according TNM classification.
gr gr gi" gr gr gr gr gr gr gr
Tl T2 «; 40 T2 > -10 T3 T4 2 3 4 2 3 4 2 3 4 2 3 4
mm mm • -\ A • ft f\
Neutropenia u 1 2 0 4 0 i 0 u
NO 0 0 7 3 1 11 Thrombo- u
Nl 1 2 2 3 2 10 penia 0 0 0 0 0 0 0 0 i 0 0 0
N2 0 1 0 0 0 1 Nausea 5 2 0 6 0 0 4 2 0 3 0 0
N3 0 2 2 2 2 8 Diarrhea 1 0 0 2 0 0 5 2 0 4 l 0
1 5 11 8 5" 30 Cutaneous 0 0 0 1 0 0 1 4 0 1 3 0
i 1 1 f\ 1 A
MUCOSltlS i 1 1 I U y) 1 1 U
* Two patients with T size ^ 40 mm classified T4 for a rectovaginal
u u u
wall involvement. * Oral cavitv nnuco sitis.
578
Toxicity during the irradiation
Of the 29 patients who were irradiated, 26 could be
given the prescribed dose in the planned time.
The box technique, used in nine patients, delivered 45
Gy in 25 fractions in 36 days (range 33-57 days). In only
one patient was there a protraction of over 38 days
because of a delay of three weeks due to a grade 3
dermitis, but was able to receive the prescribed dose in
57 days.
The AP-PA technique, used in 13 patients, delivered
45 Gy in 25 fractions (range 24—26) in 36 days (range
33-48).
The perineal technique used in seven patients deliv-
ered 35 Gy at mid-canal (range 29-37) in 8 to 14
fractions and 14 to 21 days. It was interrupted at the
eighth fraction in one patient because of acute toxicity
(aplasia), and there was an 80-day interval before the
boost.
The interval between the end of the first course and
the boost was on average 39 days (range 21-80 days),
less than two months for all but one patient. The overall
time of irradiation was an average of 77 days (range Si-
ll 2), statistically less for the patients treated by the
perineal technique (66 days) and/or brachytherapy boost
(69 days) than by the other techniques (P = 0.02).
The boost was delivered by a brachytherapy in 17
patients, by a box technique in four patients or a direct
perineal field in five patients. The dose delivered was 20
Gy in 16 patients, 15 Gy in seven patients and 25 Gy in
three patients.
None of the patients who received dose reductions
during the induction chemotherapy experienced toxicity
during the irradiation.
Tumour response
One patient presenting with a large infiltrative tumour
and a stenosis painful enough to preclude digital exam
could not be objectively evaluated during the treatment,
but only at the two-month follow-up exam. This patient
was considered to have shown no response and was
scheduled for surgical salvage; however, several negative
biopsies, an absence of symptoms and no change in the
CT scan encouraged us to continue the follow-up, and
after one year, this patient had no sign of progression.
The other 28 patients were evaluated.
None of the tumours progressed during the treat-
ment. The tumour response rates are detailed in Table 3,
with 72% response (complete and partial) after the two
cycles of induction chemotherapy, 90% after the con-
comitant chemotherapy and the first course of radio-
therapy, and 96% complete response at the end of the
treatment.
Surgical salvage
The four surgically-salvaged patients were treated by
complete abdomino-perineal resection without morbidity.
Table 3. Tumor response after the successive phases of the treatment.
After # CR (%) PR (%) NC EV
Eval
Induction chemotherapy 28 3(11%) 17(61%) 8" 0
Concomitant radiochemo-
therapy 29 17(59%) 9(31%) 3' 0
Boost 26 25(96%) 0 1* 0
Initial surgery (local con-
trol) 29 28(97%) 0 1° 0
Abbreviations: Eval = number of patients evaluable; CR = complete
response; PR = partial response; NC = no change; EV = tumour
progression.
* One patient not clinically evaluable: in CR after one year follow-up.
Three of them, 1 T4 Nl, 1 T3 NO and 1 T2 Nl, were
operated on after poor tumour regression (two no
change and one partial regression) one month after the
concomitant treatment and were not boosted.
The fourth (T4 NO) presented an apparent complete
response at the end of the treatment but relapsed two
months after a brachytherapy boost (20 Gy).
All had residual tumours on their surgical samples
and three had nodal involvement. One had only small
residual ilots on the surgical sample.
Discussion
The choice of a chemotherapy combining 5-FU and
CDDP was based on previous experimental trials on
anal canal and on other epidermoid carcinomas [4-12].
The best agents to use and the best way to combine
chemotherapy with irradiation remain debatable.
Multi-agent treatment has been more effective than
single-agent [3]. The administration of 5-FU in a four-
day infusion was chosen because of its good tolerance in
previous anal canal series [13, 14].
Continuous infusion has been proposed in anorectal
tumours to increase the additive effect. However, it
increased the acute toxicity when delivered seven days
per week at a dose of 300 mg/m2/day, achieving 40%
temporary discontinuation of 5-FU infusion for acute
toxicity [17]. Afive-days-per-weekschedule yielded low-
er toxicity. With respect to tumour control, a benefit has
been demonstrated for continuous infusion compared
to a three-day infusion in postoperative concomitant
radiochemotherapy for rectal tumour [18], but this
benefit was only significant in distant metastases and
overall survival, while no improvement in local control
was observed.
The impact of the total dose of chemotherapy was
considered rather than the duration of the infusion. In
our study this dose effect was obtained by the induction
chemotherapy.
Furthermore, the irradiation fields of the conserva-
tive treatment of anal canal tumours include the whole
canal and the margin. The acute toxicity of a continuous
concomitant treatment is potentially high, and can lead

to a prolongation of the irradiation's overall duration.
Nevertheless, this continuous scheme will be explored
by an ongoing trial of the EORTC.
As second drug, we chose cisplatin because of its
cytotoxic activity in the treatment of squamous cell
cancers, especially when combined with 5-FU [19, 20].
The mode of delivering it in a single bolus at each cycle
in our study was based on animal studies [21, 22], and
the feasibility in other protocols [19]. Only one study
used CDDP in a continuous infusion, with no objective
benefit [17].
The doses of 5-FU and CDDP were chosen to ensure
good tolerance in these often elderly patients. The 5-FU
dose of the concomitant treatment was between the ones
used by Cummings [13] (1000 mg/m2 during five days),
and by Papillon [14] (600 mg/m2 during four days), and
was similar to the dose used by Nigro [23] (1000 mg/m2
during four days). The dose of CDDP was lower than the
one usually used by the other authors [4, 7-10, 24] for
the same reason of age but was similar to the doses used
in concomitant chemotherapy for other epidermoid tu-
mours, e.g., oesophageal cancers [19]. The same scheme
and same doses were chosen for the induction and
concomitant chemotherapy to avert prescription errors.
Tolerance of this experimental scheme was usually
good. One patient died of a pulmonary embolism during
the first cycle of chemotherapy which is considered as an
intercurrent event. All the other patients could complete
the whole treatment.
The dose reduction prescribed in the occurrence of
side effects during the two induction cycles, made it
possible to prevent any toxicity for these patients during
the concomitant treatment, and to avoid an interruption
of the irradiation.
Tolerance of the concomitant treatment was worse,
especially cutaneously and digestively: in fact, only 59%
of the patients could be given the full dose of chemo-
therapy during the fourth cycle.
It is possible that the technique of irradiation has an
impact on acute toxicity, especially for the perineal
technique, but it has no influence on the dose or dura-
tion of the treatment. Nevertheless, this latter technique
should be avoided on radiobiological grounds because
of potential late toxicity.
The response rate at the end of the treatment was
high and comparable to those obtained by other series
using induction 5-FU-CDDP (Gerard, 80% [9] and
Brunet, 90% [7]) or concomitant 5-FU-MMC treatment
series, (90% in Papillon's series with stages III and IV
tumours, 90% in Cumming's series with all stages, and
80% in the EORTC trial). Moreover, it must be empha-
sised that included in our trial were only patients with
locally advanced tumours excluding anal margin, which
are considered to be of better prognosis [25]. The high
response rate after the first two cycles of induction
chemotherapy, and the absence of tumour progression,
confirm that the patients were not penalised as far as
local control is concerned, even with these lower doses
of chemotherapy compared with the doses usually used
579
in other tumours. The high chemosensitivity of epider-
moid anal canal tumours makes these results possible.
Other schedules with a split course using the same
chemotherapy but lower doses of irradiation (two
courses of 20 Gy in 10 fractions) showed good tolerance
of the 5-FU-CDDP-radiotherapy association, but did
not obtain good local control [24]. This confirms the
impact of the radiation dose on the local control, pre-
viously described [13, 26].
Other drugs such as carboplatin have also been used
with 5-FU in concomitant treatment with complete
regression in all of the patients [27], but only in small
series.
In head and neck cancer, the impact of induction
chemotherapy on survival, although lower than for a
concomitant single agent treatment (OR: 1/1.77), is
positive in the literature-based meta-analysis of Munro
[28], with an odds ratio (OR) of 1/1.2, The impact on the
time to distant metastasis at three years was proven on
larynx cancers (surgery arm: 36%, chemotherapy arm:
25%, relative risk: 0.58) and the possibility of a surgical
salvage was retained [12]. The high chemosensitivity of
anal tumours could augment this benefit for distant
metastases. The efficiency of a multiagent vs. a single
agent chemotherapy has already been cited [26] and
proven [3].
The major flaw in induction chemotherapy is the
related possibility of tumour progression in non-re-
sponders with its potential for becoming inaccessible
to a surgical salvage. None of the patients in our series
had significant disease progression during the induction
chemotherapy. The four abdominoperineal resections
were performed without incident. The small residual
disease observed in the sample from one patient could
perhaps have been controlled by a boost. Even if a
selection of radioresistant cell lines could be suggested,
the concomitant radiochemotherapy makes it possible
to achieve complete responses at the same rate as in
other series.
These preliminary results are encouraging but must
be confirmed by a longer follow-up to quantify the late
toxicity, the long term local control, the sphincter con-
servation, the survival and the occurrence of distant
metastases.
Induction chemotherapy containing CDDP is fea-
sible and well tolerated in these usually elderly patients.
Objective responses that it induces when used as sole
treatment could reflect a potential effect on transient
cells and on microscopic disease outside the pelvis. This
justifies our position with respect to an escalation of
chemotherapy doses for the purpose of reducing the risk
of distant metastasis. This chemotherapy is more easily
delivered as induction than as adjuvant treatment, espe-
cially after combined radiochemotherapy schedules, be-
cause of the duration of the treatment, the acute radio-
dermitis, a potential surgical salvage, or general perfor-
mance status.
If the tolerance and response rates are confirmed for
the next 20 planned patients, as well as the long term
580
results after a minimum of two years follow-up, the high
rate of inclusion for these selected cases (two patients
per month) should make it possible to design a prospec-
tive comparative study, in a larger cooperative group.
The following modifications suggested for further
study are dose reductions for the third and fourth cycles
of chemotherapy to lower the acute toxicity and allow a
shorter interval between the courses of irradiation, then
delivering the irradiation in a shorter time as recently
recommended [29, 30]. The dose of 5-FU for the third
cycle should be 600 mg/m2 D1-D4. The fourth cycle of
chemotherapy may perhaps be deleted as it has not
proven useful and is difficult to deliver. The interval
between the two courses of irradiation may be reduced
to two to four weeks, especially with the non-accelerated
techniques (box technique or AP-PA large fields), and
the brachytherapy boost helps reduce the overall dura-
tion of treatment.
Conclusion
This prospective study showed the feasibility of a neo-
adjuvant and concomitant 5-FU-CDDP chemotherapy
and conservative radiotherapy for locally advanced epi-
dermoid carcinomas of the anal canal. The compliance
and tolerance were good, with few major toxicities.
Most of the patients were in partial response and a few
of them in complete response after the first two cycles of
induction chemotherapy. The complete response rate
was high after the end of the first course of irradiation
and all patients could be considered in complete re-
sponse at the end of the treatment. The long-term results
will be analysed in the series of 50 patients who will be
included before initiation of a prospective randomised
trial comparing this scheme in one arm, after minor
modifications, to a reference arm.
Acknowledgement
Supported by a grant of the Ligue Nationale Contre le
Cancer.
References
1. UKCCCR Anal Cancer Trial Working Party. Epidermoid anal
cancer: Results from the UKCCCR randomised trial of radio-
therapy alone versus radiotherapy, 5-fluorouracil, and mitomycin.
Lancet 1996; 348: 1049-54
2. Roelofsen F, Bosset JF, Eschwege F et al. Concomitant radio-
therapy and chemotherapy superior to radiotherapy alone in the
treatment of locally advanced anal cancer. Results of a phase III
randomized trial of the EORTC radiotherapy and gastrointestinal
cooperative group. Proc Am Soc Clin Oncol 1995; 14: 194 (Abstr).
3. Flam M, John M, Pajak TH et al. Role of mitomycin in combina-
tion with fluorouracil and radiotherapy, and of salvage chemo-
radiation in the definitive nonsurgical treatment of epidermoid
carcinoma of the anal canal: Results of a phase III randomized
intergTOup study. J Clin Oncol 1996; 14: 2527-39.
4. Ajani JA, Carrasco CH, Jakson DE et al. Combination of cispla-
tin plus fluoro-pyrimidine chemotherapy effective against liver
metastases from carcinoma of the anal canal. Am J Med 1989;
87: 221-4.
5 Mahjoubi M, Sadek H, Fran£ois E et al. Epidermoid anal canal
carcinoma (EACC): Activity of cisplatin (P) and continuous
5-fluorouracil (5-FU) in metastatic (M) and/or local recurrent
(LR) disease. Proc Am Soc Clin Oncol 1990; 9: 114 (Abstr).
6. Salem PA, Habboubi N, Anaissie E et al. Effectiveness of cisplatin
in the treatment of anal squamous cell carcinoma. Cancer Treat
Rep 1985; 69: 891-3.
7. Brunet R, Sadek H, Vignoud J et al. Cisplatin (P) and 5-fluoro-
uracil (5-FU) for the neoadjuvant treatment (Tt) of epidermoid
anal canal carcinoma (EACC) Abstract. Proc Am Soc Clin Oncol
1990,9: 104 (Abstr).
8 Diaz E, Young K, Dc La Rosa E. Cisplatin (CDDP) and 5-fluoro-
uracil (5-FU) plus simultaneous radiotherapy (RT) for the treat-
ment of epidermoid carcinoma of the anal region (ECHR). Proc
Am Soc Clin Oncol 1993, 12: 190 (Abstr).
9 Gerard JP, Romestaing P, Mahe M et al. Cancer du canal anal:
Role de l'association 5-FU-cisplatinum. LyonChir 1991; 87: 74-6.
10. Martenson JA, Lipsitz SR, Wagner H et al. Initial results of a
phase II trial of high-dose radiation therapy, 5-fluorouracil, and
cisplatin for patients with anal cancer (E4292): An Eastern
Cooperative Oncology Group Study. Int J Radiat Oncol Biol
Phys 1996; 35: 745-9.
11. Department of Veterans Affairs Cancer Study Group. Induction
chemotherapy plus induction compared with surgery plus radia-
tion in patients with advanced laryngeal cancers. N Engl J Med
1991; 324: 1065-90.
12. Lefebvre JL, Chevalier D, Luboinski B et al. Larynx preservation
in pyriform sinus cancer: Preliminary results of a European
Organization for Research and Treatment of Cancer phase III
trial. J Natl Cancer Inst 1996; 88: 890-9;
13. Cummings BJ, Keane TJ, O'Sulhvan B et al. Epidermoid anal
cancer: Treatment by radiation alone or by radiation and 5-fluo-
rouracil with and without mitomycin C Int J Radiat Oncol Biol
Phys 1991; 21: 1115-25.
14. Papillon J, Montbarbon JF. Epidermoid carcinoma of the anal
canal. A series of 276 cases. Dis Colon Rectum 1987; 30: 324-33.
15. Word Health Organisation. WHO Handbook for Reporting Re-
sults of Cancer Treatment Geneva: WHO 1979
16 International Union Against Cancer. In Hermaneck P, Sobin LM
(eds): TNM Classification of Malignant Tumors, 4th edition.
Springer-Verlag 1987; 50-2.
17. Hughes LL, Rich TA, Delclos SL et al. Radiotherapy for anal
cancer: Experience from 1979 to 1987. Int J Radiat Oncol Biol
Phys 1989; 17: 1153-60.
18. O'Connel MJ, Martenson JA, Wieand HS et al. Improving adju-
vant therapy for rectal cancer by combining protracted infusion
fluorouracil with radiation therapy after curative surgery. N Engl
J Med 1994; 331: 502-7.
19. Herskovic A, Martz K, Al-Sarraf M et al. Combined chemo-
therapy and radiotherapy compared with radiotherapy alone in
patients with cancer of oesophagus. N Engl J Med 1992; 326:
1593-8.
20. Jacobs C, Lyman G, Velez-Garcia E et al. A phase HI randomized
study comparing cisplatin and fluorouracil as single agents and in
combination for advanced squamous cell carcinoma of the head
and neck. J Clin Oncol 1992; 10: 257-63.
21. Steel GG. Principles for the combination of radiotherapy and
chemotherapy, In Horwich A, Arnold E (eds): Combined Radio-
therapy and Chemotherapy in Clinical Oncology. London 1992;
14-22.
22. Lilieveld P, Scoles MA, Brown JM et al. The effect of treatment in
fractionated schedules with the combination of X-irradiation and
6 cytotic drugs on the RIF. One tumor and normal mouse skin.
Int J Radiat Oncol Biol Phys 1985; 11: 1911-21.
23. Nigro NO, Vaitkevicius VK, Considiner BJ. Combined treatment
for cancer of the anal canal. A preliminary report. Dis Colon
Rectum 1974; 17- 354-6.
 581
24. Seitz JF, Giovannini M, Padaut-Cesana J et al. Traitement des 29. Allal AS, Mermillod B, Kurtz JM et al. Impact of therapeutic
carcinomes epidermoldes du canal anal (CECA) par chimio- factors on local control in T2-T3 anal carcinoma treated by
therapie (5-FU-CDDP) et radiotherapie concomitante. Resultats radiation or chemotherapy. Int J Radiat Oncol Biol Phys 1996,
chez 15 patients. Gastroenterol Clin Biol 1992; 16 A48 (Abstr). 36: 297 (Abstr).
25. Jensen SL, Hagen K, Harling H. Long-term prognosis after 30. Constantinou EC, Daly W, Fung CTet al. Dose-time considera-
radical treatment for squamous cell carcinoma of the anal canal tions in the treatment of anal cancer. Int J Radiat Oncol Biol Phys
and anal margin. Dis Colon Rectum 1988; 31: 273-8. 1996; 36: 295 (Abstr).
26. RichTA, Ajani JA, Morrison WH et al. Chemoradiation therapy
for anal cancer: Radiation plus continuous infusion of 5-fluoro- Received 20 January 1997; accepted 22 May 1997.
uracil with or without cisplatin. Radiat Oncol 1993; 27: 207-15
27 Svensson C, Kaigas M, Goldmann S. Induction chemotherapy
with carboplatin and 5-fluorouracil in combination with radio- Correspondence to:
therapy in locoregional advanced epidermoid carcinoma of the D. Peiffert, MD
anus. Preliminary results. Int J Colored Dis 1992; 7: 122—4. Centre Alexis Vautnn
28 Munro AJ. An overview of randomised controlled trials of adju- Avenue de Bourgogne, Brabois
vant chemotherapy in head and neck cancer. Br J Cancer 1995; 71: 54511 Vandoeuvre-les-Nancy
83-91. France