Concurrent chemoradiotherapy followed by surgery in locally advanced squamous cell carcinoma of the oesophagus: a single centre experience.

Abstract

Data on combined modality treatment for locally advanced squamous cell carcinoma of the oesophagus involving Asian patients are limited.
A retrospective study of 56 consecutive patients with this condition treated with concurrent chemoradiotherapy followed by surgery in a single tertiary institution in Singapore was performed.
The median overall survival of the entire cohort was 14.1 months [95% confidence interval (CI); range, 8.6 to 19.6 months]. In patients who underwent successful oesophagectomy after chemoradiotherapy (n = 17), the median survival was 27.8 months compared to 9.8 months for those who did not have surgery (n = 39) (P = 0.046, log-rank test). The median time to first relapse for the entire cohort was 16.1 months (95% CI, 7.7 to 24.5 months). The time to first relapse was 23.9 months in the subgroup of patients with successful surgery and 12.1 months in the group which did not (P = 0.147, log-rank test). The high proportion of patients who were medically unfit for surgery or declined surgery may have conferred a selection bias.
Concurrent chemoradiotherapy followed by surgery is feasible in selected patients. The benefit of adding of surgery to chemoradiotherapy is still controversial and we await the results of randomised controlled trials comparing chemoradiotherapy with surgery versus chemoradiotherapy alone.

Full text

Page 1

June 2005, Vol. 34 No. 5
369
Combination Treatment Oesophageal Carcinoma—NS Wong et al
Concurrent Chemoradiotherapy followed by Surgery in Locally Advanced
Squamous Cell Carcinoma of the Oesophagus: A Single Centre Experience
NS Wong,1MBBS, M Med (Int Med), MRCP, KF Foo,1MBBS, M Med (Int Med), MRCP, D Poon,1MBBS, M Med (Int Med), MRCP,
SS Leong,1MBBS, M Med (Int Med), MRCP, WK Wong,2MBBS, FRCS, FAMS, HS Chan,2MBBS, FRCS, FAMS, KC Soo,3MBBS, FRACS, FAMS,
SP Yap,4MBBS, MRCP, FRCR, J Wee,4MBBS, FRCR, FAMS, YB Cheung,5PhD, EH Tan,1MBBS, MRCP, FAMS
Introduction
Carcinoma of the oesophagus is a relatively uncommon
malignancy in Singapore and incidence rates have been
declining since 1968. A total of 506 cases were diagnosed
from 1993 to 1997. The age-standardised rate for the same
period was 5.8 per 100,000.1 The predominant histologic
type is squamous cell carcinoma, although there has been
a shift towards adenocarcinoma histology in recent years in
Western populations. The majority of patients present late
in the course of disease, as a result of which long-term
survival is poor. While concurrent chemoradiotherapy has
been defined as the standard of care in unresectable non-
metastatic disease,2 controversy still exists in the setting of
locally advanced but surgically resectable carcinoma of the
oesophagus.
The 5-year survival rate ranges from 5% to 18% in
several series for patients with locoregionally advanced
(stage III) carcinoma of the oesophagus treated with surgery
as the sole modality.3-5 Attempts to improve these dismal
results have generated numerous randomised trials studying
the respective roles of preoperative radiotherapy,
preoperative chemotherapy, preoperative combined
chemoradiation, as well as postoperative radiotherapy.
The results of 3 randomised trials comparing concurrent
1Department of Medical Oncology
3Director’s Office
4Department of Therapeutic Radiology
5Biostatistics Unit, Division of Clinical Trials and Epidemiology
National Cancer Centre, Singapore
2Department of General Surgery
Singapore General Hospital, Singapore
Address for Reprints: Dr Kian-Fong Foo, Department of Medical Oncology, National Cancer Centre Singapore, 11 Hospital Drive, Singapore 169610.
Email: dmofkf@nccs.com.sg
Abstract
Introduction: Data on combined modality treatment for locally advanced squamous cell
carcinoma of the oesophagus involving Asian patients are limited. Materials and Methods: A
retrospective study of 56 consecutive patients with this condition treated with concurrent
chemoradiotherapy followed by surgery in a single tertiary institution in Singapore was
performed. Results: The median overall survival of the entire cohort was 14.1 months [95%
confidence interval (CI); range, 8.6 to 19.6 months]. In patients who underwent successful
oesophagectomy after chemoradiotherapy (n = 17), the median survival was 27.8 months
compared to 9.8 months for those who did not have surgery (n = 39) (P = 0.046, log-rank test).
The median time to first relapse for the entire cohort was 16.1 months (95% CI, 7.7 to 24.5
months). The time to first relapse was 23.9 months in the subgroup of patients with successful
surgery and 12.1 months in the group which did not (P = 0.147, log-rank test). The high
proportion of patients who were medically unfit for surgery or declined surgery may have
conferred a selection bias. Conclusion: Concurrent chemoradiotherapy followed by surgery is
feasible in selected patients. The benefit of adding of surgery to chemoradiotherapy is still
controversial and we await the results of randomised controlled trials comparing
chemoradiotherapy with surgery versus chemoradiotherapy alone.
Ann Acad Med Singapore 2005;34:369-75
Key words: Asian, Oesophagectomy, Retrospective study, Singapore
Original Article

Page 2

370
Annals Academy of Medicine
Combination Treatment Oesophageal Carcinoma—NS Wong et al
chemoradiotherapy followed by oesophagectomy versus
oesophagectomy alone have been published. Walsh and
associates6 treated 113 patients suffering from
adenocarcinoma of the oesophagus with 2 cycles of
preoperative 5-fluorouracil and cisplatin concurrent with
radiotherapy and reported a significant improvement in
median survival (16 months versus 11 months) (P = 0.01)
and 3-year survival (32% versus 6%; P = 0.01) compared
to surgery alone.
In a study by Urba et al,7 100 patients were randomised
to preoperative cisplatin, vinblastine and 5-fluorouracil
concurrent with radiotherapy. The predominant histology
in this study was adenocarcinoma. A complete pathologic
response was observed in 28% of patients. With a median
follow-up of 8.2 years, the median survival was similar in
both treatments (16.9 months versus 17.6 months for
multimodal therapy and surgery respectively), and the
improvement in 3-year survival (30% versus 16%) did not
reach statistical significance.
A European study8 included 282 patients with stage I or
II squamous cell carcinoma of the oesophagus who were
randomly assigned surgery alone or surgery preceded by
chemoradiotherapy. Surgery was performed 2 to 4 weeks
after the completion of the preoperative therapy. After a
median follow-up of 55 months, preoperative treatment
was associated with a higher frequency of curative resection,
a significantly longer disease-free survival and time to
local failure, and a lower rate of cancer-related deaths.
However, there was no improvement in median survival
(18.6 months in both groups). This study has been criticised
because of inadequate radiation dose-fractionation schedule,
and the use of single-agent cisplatin rather than multi-agent
chemotherapy. While the results of neoadjuvant
chemoradiotherapy followed by oesophagectomy are well
reported in the Western literature, there are limited data
regarding the outcomes of similar treatment regimens in
Asian populations.
Between 1995 and 1999, the role of chemoradiotherapy
followed by oesophagectomy was assessed at the Singapore
General Hospital to address this issue. This report presented
the results of such a treatment strategy for the entire cohort
and also the separate results for patients who had
chemoradiotherapy alone or chemoradiotherapy followed
by surgery.
Materials and Methods
A retrospective study of concurrent chemoradiotherapy
followed by oesophagectomy in 56 consecutive patients
with locally advanced carcinoma of the oesophagus treated
between September 1995 and September 1999 at the
Singapore General Hospital, a tertiary care hospital in
Singapore, was performed. Patients eligible for concurrent
chemoradiotherapy during this period fulfilled the following
criteria: age >21, histologically confirmed T3-4 N0-1 M0
(1983 American Joint Committee on Cancer Staging)
squamous cell carcinoma of the oesophagus, performance
status <2 (WHO criteria) with adequate renal, bone marrow
and liver functions. Patients were not eligible if there was
evidence of metastasis, tracheo-oesophageal fistula, or
carcinoma of the cervical oesophagus. All patients
underwent physical examination, chest X-ray, computed
tomography (CT) scans of the chest and abdomen,
oesophagoduodenoscopy, barium swallow, bronchoscopy
and electrocardiogram prior to commencement of treatment.
Endoscopic ultrasound was performed whenever possible.
Radiotherapy
The sequence of therapy is summarised in Figure 1.
Radiotherapy was administered to a dose of 50 Gy (10 Gy
per week in 5 fractions). Radiotherapy was begun on the
first day of the chemotherapy. All patients were planned
using 3D conformal techniques. The planned target volumes
extended 5 cm beyond the longitudinal margins of the
tumour defined by the endoscopic and radiologic findings,
1.5 cm beyond the radial margins. The radiation treatments
were given in 2 phases with the initial Phase I delivered as
parallel and opposed pair of fields. This was treated to a
total dose of 40 Gy in 20 fractions of 2 Gy per fraction. The
Phase II treatment was planned as a 3 fields technique
aiming to spare the radiation dose to the underlying spinal
cord. This would be delivered to an additional dose of 10
Gy in 5 fractions. All patients were treated daily except on
weekends. If patients were deemed to have unresectable
disease after reassessment or there were positive margins
and/or viable tumour seen in the resected specimens in
Fig. 1. Treatment flowchart.
??????????
??????????
????????????
?????????????????
????????????????????
???????
???????
???????
??????? ???????
???????
???????
??? ???
??????? ???
?
?
?????????
??????????????????????????????????
?????????????????????

Page 3

June 2005, Vol. 34 No. 5
371
Combination Treatment Oesophageal Carcinoma—NS Wong et al
those who had surgery, they would receive an additional
14 Gy in 7 fractions to the Phase II treatment volume.
Chemotherapy
Chemotherapy was started on the same day as
radiotherapy. Cisplatin 20 mg/m2 was given with intravenous
hydration over 6 hours on days 1 to 4, with mannitol and
frusemide diuresis on day 1 while 5-fluorouracil 1gm/m2
was given over 6 hours on days 1 to 4. Two courses of
chemotherapy were given during radiotherapy at 4-week
intervals. Three weeks after the completion of chemo-
radiotherapy, a repeat evaluation was performed with CT
chest, barium meal and oesophagoduodenoscopy. Patients
deemed operable underwent transthoracic or transhiatal
oesophagectomy. Patients with inoperable disease were
given an additional cycle of the same chemotherapy
concomitant with 14 Gy of radiotherapy in 7 fractions.
Patients with operable disease underwent transthoracic or
transhiatal oesophagectomy. If resection margin was
positive or viable tumour was present in the resection
specimen, patients were also treated with an additional
cycle of the same chemotherapy concomitant with 14 Gy of
radiotherapy in 7 fractions.
Clinical complete response was defined as the absence of
visible tumour on CT scan, barium meal, oesophago-
duodenoscopy and repeat biopsy. Pathological complete
response was defined as the absence of viable tumour in the
resection specimen on histopathologic examination.
Fig. 2. Overall survival for entire cohort.
Fig. 3. Overall survival by treatment.
Fig. 4. Time to relapse for whole cohort.
Fig. 5. Time to first relapse by treatment.
???
???
???
???
???
???
????????????????
??????
?????????????????
????????
?
?? ??
??
?????
???
???
???
???
???
???
????????????????
??????
?
????
??
?????
?????????????????
??????????????????????????????????
?????????????????????????
???
???
???
???
???
???
??????????????????????????
??????
?
?? ???? ??
?????????????????
????????
???
???
???
???
???
???
??????????????????????????
??????
? ???? ?? ??
?????????????????
??????????????????????????????????
?????????

Page 4

372
Annals Academy of Medicine
Combination Treatment Oesophageal Carcinoma—NS Wong et al
Overall survival was calculated from time of diagnosis
until time of death. Time to first relapse was calculated
from time of diagnosis until the first documented relapse.
In the event that death of relapse did not occur the observation
was censored. Time to local relapse was calculated from
time of diagnosis to time of documented locoregional
recurrence, progression from clinical partial response or
local progression while on chemoradiotherapy.
Statistical Analysis
Data analysis was performed using SPSS (Release 11.5).
Survival curves for relapse-free and overall survival were
calculated using the Kaplan-Meier method and compared
using the log-rank test. Statistical significance was shown
by P <0.05.
Pretreatment characteristics of the patients are summarised
in Table 1. The median age of the patients was 66 years
(range, 43 to 84). There was a male preponderance and
nearly all patients were Chinese. Ninety per cent of patients
had performance status 0 to 1. Fifty-eight per cent of the
tumours were well to moderately differentiated and 18%
were poorly differentiated. Twenty-six per cent of the
tumours were located in the upper third of the thoracic
oesophagus while 49% and 25% were located in the middle
and lower third respectively. Eighty-one per cent of tumours
were staged as T3 and 18% as T4. Sixty-three per cent of
patients were node-negative and 37% were node-positive.
A total of 119 cycles of chemotherapy concurrent with
radiotherapy were administered to 56 patients. The median
number of cycles per patient was 2 (range, 1 to 3). Three
patients required interruption of radiotherapy for more
than 1 week and 2 patients had interruption for less than 1
week.
The modalities of treatment received are summarised in
Table 2. Seventeen patients (30%) underwent oesopha-
gectomy after completion of chemoradiotherapy. Oesopha-
gectomy was abandoned in 3 patients (5%) due to the
presence of unresectable disease intraoperatively. Thirty-
six patients (64%) received only chemoradiotherapy, of
which 15 (42%) declined surgery while 21 (58%) where
unfit for surgery. Most of these 21 patients where unfit due
to poor baseline lung or cardiac function, although 7
patients in this group did not have surgery due to clinical
disease progression.
Fifty-four patients were assessable for toxicity and the
results are summarised in Table 3. No toxic deaths from
chemoradiotherapy occurred. Ten patients (18%) developed
grade 3 oesophagitis. Seventeen patients (32%) developed
grade 3 or 4 neutropaenia, of whom 5 (29%) developed
neutropenic fever. Four patients (7%) developed tracheo-
oesophageal fistula, which were treated with oesophageal
stents. There were 2 deaths after oesophagectomy. One
patient had subcutaneous emphysema and pneumonia
and passed away 37 days after surgery while the other
patient developed pneumonia and passed away 15 days
after surgery.
The response to chemoradiotherapy is summarised in
Table 4. For the entire cohort, 25 patients (45%) had
Table 2. Treatment Received
Modality No. (%)
Chemoradiotherapy followed by oesophagectomy
Chemoradiotherapy followed by attempted
oesophagectomy
Chemoradiotherapy only
17 (30)
3 (5)
36 (64)
Table 1. Patient Characteristics
Characteristics All patients
(n = 56)
No. (%)
CRT
(n = 39)
No. (%)
CRT + S
(n = 17)
No. (%)
Age (y)
Median
Range
666863
43-8443-84 47-75
Sex
Male
Female
44 (79)
12 (21)
30 (77)
9 (18)
14 (82)
3 (18)
Race
Chinese
Indian
55 (98)
1 (2)
38 (97)
1 (3)
17 (100)
0
Differentiation
Well
Moderately
Poorly
Unknown
4 (7)
28 (50)
10 (18)
14 (25)
3 (8)
18 (46)
7 (18)
11 (28)
1 (6)
10 (59)
3 (18)
3 (18)
Site
Upper
Middle
Lower
14 (25)
28 (50)
14 (25)
13 (33)
17 (44)
9 (23)
1 (6)
11 (65)
5 (29)
Performance status
0
1
2
3
Unknown
1(2)
50 (89)
0
1 (2)
4 (7)
1(3)
35 (90)
0
1 (3)
2 (5)
0
15 (88)
0
0
2 (12)
Tumour stage
T3
T4
Unknown
45 (80)
10 (18)
1 (2)
30 (77)
8 (21)
1 (3)
15 (88)
2 (12)
0
Nodal stage
N0
N1
Endoscopic ultrasound
36 (64)
20 (36)
15 (27)
27 (69)
12 (31)
11(28)
9 (53)
8 (47)
4 (31)
CRT: chemoradiotherapy; S: successful oesophagectomy

Page 5

June 2005, Vol. 34 No. 5
373
Combination Treatment Oesophageal Carcinoma—NS Wong et al
persistent disease and 19 (34%) had clinical complete
response to induction chemoradiotherapy. Seven (13%)
patients progressed while on treatment, mostly from
progressive local disease although 1 patient had distant
metastasis while undergoing chemoradiotherapy. Seventeen
out of 56 patients eventually had successful oesopha-
gectomy. In this select group of patients, 7 (41%) had
residual tumour after induction chemoradiotherapy while
10 (59%) had pathologic complete response. Out of the 9
clinical complete responders to chemoradiotherapy who
proceeded to have surgery, only 2 (22%) were found to
have residual tumour in the oesophagectomy specimen.
The overall survival of the entire group is shown in
Figure 2. The median overall survival was 14.1 months
[95% confidence interval (CI), 8.6 to 19.6 months]. A log-
rank test showed significant difference between patients
with and without successful oesophagectomy (P = 0.046)
(Fig. 3). In patients who underwent successful
oesophagectomy after chemoradiotherapy, the median
survival was 27.8 months, compared to 9.8 months for
those who did not have oesophagectomy. In patients who
had successful oesophagectomy after concurrent
chemoradiotherapy, the log-rank test did not show any
significant difference in terms of survival between patients
Table 3. Chemoradiotherapy Toxicity Graded by Common
Toxicity Criteria (Version 2.0)
ToxicityGrade 0
No. (%)
Grade 1
No. (%)
Grade 2 Grade 3 Grade 4
No. (%) No. (%) No. (%)
Anaemia
Neutropaenia
Oesophagitis
Sepsis
Thrombocytopaenia
TEF
Vomiting
20 (37)
19 (35)
13 (24)
39 (72)
34 (63)
52 (93)
35 (65)
22 (41)
9 (17)
16 (30)
1 (2)
15 (28)
-
9 (17)
11 (20)
9 (17)
15 (28)
4 (7)
5 (9)
-
8 (15)
1 (2)
10 (19)
10 (19)
9 (17)
0
4 (7)
2 (4)
0
7 (13)
0
1 (2)
0
0
0
n = 54 (2 patients not evaluable for haematological toxicity due to missing
data)
TEF: tracheo-oesophageal fistula
Table 4. Response to Chemoradiotherapy
Response All patients
(n = 56)
No. (%)
CRT
(n = 39)
No. (%)
CRT+S
Clinical assessment
Path assessment
Persistent disease
Clinical CR
Path PR
Path CR
PD
No data
25 (45)
19 (34)
-
-
7 (13)
5 (9)
17 (44)
10 (26)
-
-
7 (18)
5 (13)
8 (47)
9 (53)
-
-
0
0
-
-
7 (41)
10 (59)
0
0
CR: complete response; CRT: chemoradiotherapy; Path: pathologic; PD:
progressive disease; PR: partial response; S: successful oesophagectomy
No. (%) No. (%)
Table 5. Pattern of First Relapse
Site of first relapse All patients
(n = 56)
No. (%)
CRT
(n = 39)
No. (%)
CRT+S
(n = 17)
No. (%)
No relapse
Locoregional
Distant
Locoregional/distant
Missing data
23 (41)
11 (20)
13 (23)
6 (11)
3 (5)
17 (44)
10 (26)
6 (15)
3 (8)
3 (8)
6 (35)
1 (6)
7 (41)
3 (18)
0
CRT: chemoradiotherapy; S: successful oesophagectomy
who had pathologic complete response and those with
residual disease (P = 0.760). The overall survival for
patients with pathological complete response was 37.5
months, compared to 27.3 months for patients with residual
disease.
Three patients were excluded from the time to first
relapse analysis due to missing data. Table 5 summarises
the pattern of first relapse. The median time to first relapse
for the entire cohort was 16.1 months (95% CI, 7.7 to 24.5
months) (Fig. 4). A log-rank test failed to show any
significant difference between patients with and without
successful oesophagectomy (P = 0.147) (Fig. 5). In patients
who underwent successful oesophagectomy after
chemoradiotherapy, the time to relapse was 23.9 months,
compared to 12.1 months for those who did not have
oesophagectomy. In patients who had successful
oesophagectomy after concurrent chemoradiotherapy, the
log-rank test did not show any significant difference between
patients who had pathologic complete response and those
with residual disease (P = 0.241). The time to relapse for
patients with pathological complete response was 54.3
months, compared to 14.8 months for patients with residual
disease.
The median time to distant relapse was 27 months (95%
CI, 8.6 to 45.3 months). A log-rank test did not show any
significant difference between patients with and without
successful oesophagectomy (P = 0.742). In patients who
underwent successful oesophagectomy after chemo-
radiotherapy, the time to distant relapse was 27.0 months,
compared to 38.0 months for those who did not have
oesophagectomy. In patients who had successful oesopha-
gectomy after concurrent chemoradiotherapy, the log-rank
test did not show any significant difference between patients
who had pathologic complete response and those with
residual disease (P = 0.489). The time to distant relapse for
patients with pathological complete response was 66.5
months, compared to 23.9 months for patients with residual
disease.
The median time to local relapse for the entire cohort was
58.2 months. A log-rank test did not show any significant
difference between patients with and without successful