Int J Gynecol Cancer 2002, 12, 158170

Pathologic complete remission after preoperative

intracavitary radiotherapy of cervical cancer stage Ib
and IIa is a strong prognostic factor for long-term
survival: analysis of the Radiumhemmet
data 19891991
C. BESKOW,* A.-K. GREN-CRONQVIST, F. GRANATH, B. FRANKENDAL* &
R. LEWENSOHN
Departments of *Gynecologic Oncology and Hospital Physics, Radiumhemmet, Karolinska Hospital, and Unit of
Medical Radiation Biology, Cancer Centre Karolinska, Karolinska Hospital, and Department of Medical Epidemiology,
Karolinska Institutet, Stockholm, Sweden

Abstract. Beskow C, gren-Cronqvist A-K, Granath F, Frankendal B,

Lewensohn R. Pathologic complete remission after preoperative intracavitary radiotherapy of cervical cancer stage Ib and IIa is a strong prognostic
factor for long-term survival: analysis of the Radiumhemmet data 1989
1991. Int J Gynecol Cancer 2002;12:158170.
The purpose of this study was to evaluate the treatment results of preoperative brachytherapy and the prognostic value of pathologic complete remission after preoperative intracavitary irradiation in patients
with stage Ib and IIa cervical carcinoma in relation to recurrence rate and
survival. The clinical records of 185 patients with stage Ib (129 patients)
and IIa (56 patients) cervical carcinoma, consecutively admitted to Radiumhemmet from January 1989 to December 1991 were reviewed. The
median follow-up time was 71 months. In 121 patients the treatment
consisted of uterovaginal intracavitary irradiation, according to the
Stockholm technique, followed by surgery. Tumor remission assessed in
the surgical specimen was classified as pathologic complete remission
(pCR) if no microscopic tumor was found or incomplete pathologic remission (non-pCR) if microscopic residual tumor was found. Postoperative external beam radiation was added to cases with metastases in pelvic nodes or residual tumor in the resection margins. The diseasespecific 5-year survival was 87% and 75% for stage Ib and IIa,
respectively, for the patient population treated with preoperative intracavitary radiotherapy and surgery. After intracavitary radiation, 79% of
the patients obtained pCR of the primary tumor. Five-year survival in
those with pCR was 95%, compared with 46% in those with non-pCR (P
< 0.0001). Patients with pCR and no lymph node metastases had a 98%
5-year survival as compared to a 5-year survival of 64% in patients with
non-pCR and node negativity (P < 0.0001). Locoregional relapses were
diagnosed in 2% of the patients with pCR compared to 54% in patients
with non-pCR (P < 0.0001). Multivariate analysis revealed non-pCR (RR
= 6.42) and node positivity (RR = 4.59) as nonfavorable factors for surAddress correspondence and reprint requests to: Rolf Lewensohn, MD, Unit of Medical Radiobiology, Cancer Center Karolinska R8:00,
Karolinska Hospital, S-171-76 Stockholm, Sweden. E-mail: Rolf.Lewensohn@onkpat.ki.se.
2002 IGCS

Cervical cancer and remission after radiotherapy

159

vival, while tumor size was not found to be of independent significance

for survival. Pathologic complete remission after intracavitary irradiation is a strong favorable prognostic factor in node-negative patients.
The combination of preoperative intracavitary radiotherapy and surgery
results in a high cure rate and aids in identifying patients at risk for
relapse who might be subject to adjuvant therapy.
KEYWORDS: cervical carcinoma, pathologic complete remission, preoperative intracavitary radiotherapy, survival.

Successful treatment of early stage cervical cancer

(stage Ib and IIa) requires control of the primary tumor and eradication of metastases when present in
regional nodes. Stage Ib and IIa carcinoma of the cervix can be cured by radical surgery, by radiotherapy,
or by a combination of these two modalities. Five-year
survival rates of 6690% for stage Ib and 6788% for
stage IIa have been reported for primary surgery or
radiotherapy alone(1,2). However, a reliable evaluation
of the results from studies comparing primary surgery
with radiotherapy alone requires that the patient
populations do not differ regarding selection of patients as well as the choice of adjuvant treatment. The
few randomized studies(3,4) comparing treatment with
primary surgery to radiotherapy alone do not give a
conclusive answer to the question of what should be
regarded as optimal treatment of these patients.
With radical surgery, pathologic findings such as
lymph node metastases and residual tumor in resection margins may be assessed, resulting in the identification of at-risk patients who could possibly benefit from adjuvant treatment. Combined treatment
with preoperative intracavitary radiotherapy (ICRT)
and surgery offers the possibility to study the effect of
radiotherapy on the primary tumor. The prognostic
value of pathologic complete remission after preoperative ICRT for survival in patients with early stage
cervical carcinoma has not been clearly defined. Some
authors claim that complete remission of the tumor
after preoperative ICRT is a favorable prognostic factor(57), while others found that eradication of the primary tumor has borderline or no significant impact on
survival(8,9). We are not aware of any randomized
study comparing preoperative brachytherapy and surgery with primary surgery. There is a randomized
study comparing radiotherapy alone versus brachytherapy plus surgery in patients with cervical carcinoma stage IIa and IIb, where residual tumor after
brachytherapy was not prognostic of death from recurrent disease(10).
We herein report the results of a retrospective
analysis of the treatment of stage Ib and IIa cervical

carcinoma using intracavitary irradiation followed by

surgery. The main purpose was to evaluate the prognostic significance of pathologic complete remission
after intracavitary radiotherapy as related to recurrence rates and survival.

Materials and methods

Patient population
The records of 185 patients with primary cervical carcinoma, consecutively referred to and treated at the
Department of Gynecologic Oncology, Radiumhemmet, from January 1989 to December 1991 were studied. Age distribution showed the two-peak distribution, which is well known according to other Swedish
reports on cervical cancer(11). Multiple biopsies were
taken from the cervix and fractionated curettage was
performed on all patients for diagnosis. The histologic
material from patients diagnosed at other clinics was
retrieved for review at the Department of Pathology,
Radiumhemmet, at the time of clinical staging of the
patient. Squamous cell carcinoma was diagnosed in
139 cases (75%), adenocarcinoma in 36 cases (19%),
and adenosquamous carcinoma in 7 cases (4%). Two
percent had other histologies; clear cell carcinoma in
two cases and in one case the tumor had neuroendocrine differentiation. There were 10% well-differentiated, 49% moderately differentiated, and 37% poorly
differentiated tumors. One patient had an undifferentiated tumor. In seven cases (4%), the pathologist was
unable to classify the grade of histologic differentiation.
All patients were subject to follow-up at the Department of Gynecologic Oncology, Radiumhemmet, during the first 5 years after treatment. After that the follow-up was carried out by the referring gynecologist.
Median time of follow-up was 71 months (range 48
110 months), except for one patient who was lost to
follow-up after 11 months. Clinical staging of the tumor according to the International Federation of Gynecology and Obstetrics (FIGO)(12) was carried out un 2002 IGCS, International Journal of Gynecological Cancer 12, 158170

160

Beskow et al.

der general anesthesia or under spinal anesthesia by

two gynecologic oncologists. Staging investigations
included a chest X-ray, cystoscopy, and intravenous
pyelogram.
There were 129 patients in FIGO stage Ib and 56
patients in FIGO stage IIa. For the analysis of the relation between tumor size and response to radiotherapy, as well as survival, the patients were divided
into three groups according to primary tumor size:
small tumor, <2 cm, medium tumor, >2 cm<4cm, and
large tumor, >4 cm.
Of 185 patients, 44 patients were treated with radiotherapy only, due to medical conditions excluding
them from surgery. Twelve patients were treated with
either surgery only or a combination of primary surgery and adjuvant radiotherapy, due to pregnancy,
fallopian tube abscess, benign cystoma of the ovary,
and myoma of the uterus. In all, 129 patients underwent radiotherapy followed by surgery. Of these, six
patients received a combination of external beam radiation and intracavitary radiotherapy preoperatively
due to difficulties in localizing the cervical canal. In
123 patients the treatment consisted of two uterovaginal intracavitary treatments with a 3-week interval,
followed by surgery consisting of radical hysterectomy and pelvic lymph adenectomy 4 weeks later. In
two patients, total hysterectomy was not performed,
although sampling of lymph nodes was carried out. In
these cases it was not possible to evaluate the degree
of pathologic remission of the primary tumor. Consequently the degree of pathologic remission after ICRT
was possible to evaluate in 121 patients.
Tumor remission after intracavitary treatment was
diagnosed in the surgical specimen and was classified
as pathologic complete remission (pCR) if no microscopic tumor was found, or incomplete pathologic remission (non-pCR) if a microscopic residual tumor
was found. Microscopic residual tumor in the surgical
specimen was not considered as an indication for postoperative external beam radiotherapy (EBRT). Only if
there was tumor growth at the resection margin was
postoperative EBRT given to a pelvic field with a prescribed dose of 45 Gy. Four patients received postoperative EBRT due to tumor growth at the resection
margins.
Pelvic lymph node resection was performed in 134
patients, of which 25 were found to have lymph node
metastases. Twenty-three of these patients received
postoperative EBRT including a pelvic and a paraaortic field, with a prescribed dose of 45 Gy and 40 Gy,
respectively. Two patients received pelvic radiation
only. The postoperative EBRT started within 45
weeks after surgery.
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

Recurrences in the pelvis were classified as locoregional. Recurrences outside the pelvis were considered as distant relapses. Cases with both locoregional
and distant relapse were grouped together with cases
with locoregional relapses. The degree of clinical remission was assessed by clinical examination 6 weeks
after the end of treatment and biopsies were performed to confirm clinical signs of persistent disease.
Persistent disease was found in 10 of 185 cases and
was classified as locoregional recurrence. Among
these, five patients were treated according to the standard treatment, including two fractions of brachytherapy followed by WertheimMeig surgery. In two
patients, hysterectomy was not performed after
brachytherapy, although lymph node resection was
performed. One patient received a combination of
EBRT and brachytherapy preoperatively and two patients were not suitable for surgery and were treated
with radiotherapy only.
Intracavitary treatment
During 19891991 the brachytherapy technique at the
Department of Gynecologic Oncology, Radiumhemmet, gradually changed from a manual technique with
radium applicators to a remote after-loading technique with cesium-137. Patients were treated in accordance with the modified individualized Stockholm
technique(13), which included two uterovaginal intracavitary treatments with a 3-week interval, followed
by either surgery or external beam radiation 4 weeks
later.
Forty-five of 121 patients received brachytherapy
using the manual radium technique. The manual technique included a combination of an intrauterine tube
(4370 mg Ra) and a vaginal applicator (5070 mg Ra)
not fixed to each other. The dose rate in the bladder
and rectum, measured by a gammameter, served as
the basis for determining the treatment time and thus
the total given dose. The dosage was quoted in milligram-hours of radium (mghRa)(14). The treatment consisted of two fractions giving a total uterovaginal
mean dose of 6500 mghRa (48007900 mghRa). The
dose rate at point A was estimated retrospectively and
varied between 1.0 and 1.23 Gy/hr (mean 1.1 Gy/hr).
Seventy-six of 121 patients received brachytherapy
using the remote after-loading technique. The afterloading technique was gradually introduced during
the years 19891990, using a circular-shaped vaginal
applicator which was fixed to the uterine applicator in
the Selectron system(14). The dose rate at point A, defined according to the Manchester method(15), varied
between 1.20 and 1.45 Gy/hr (mean 1.35 Gy/hr).

Cervical cancer and remission after radiotherapy

Computerized dosimetry was performed for each

patient, generating isodose distributions in three
planes. The dose prescriptions referred to point A, as
defined by the International Committee on Radiation
Units and Measurement (ICRU)(16). When the afterloading technique was introduced the dose prescriptions were calculated to mimic the doses given by the
manual technique. The ordered dose to point A for
each fraction varied between 20 and 22.5 Gy, giving a
total dose of 4045 Gy.

External beam radiotherapy

Treatment with external beam irradiation was effected
by linear accelerators (621 MV), based on computerized dose planning. The upper limit of the para-aortic
field reached the space between thoracic vertebras
XXI. Anterior-posterior fields were used with a daily
fraction of 1.6 Gy, given 5 days/wk. The pelvic field
included the internal and external iliac nodes and the
lower common iliac nodes up to the level of the space
between L4 and L5. All patients undergoing irradiation following intracavitary therapy had a central
shield placed in accordance to the intracavitary applicators, with a width of 4 cm and with different shielding power depending on the dose already given to the
bladder and rectum by brachytherapy.

161

Moderate (grade 2): Symptoms resulting in intermittent or persistent interference with normal activity
and/or requiring investigation such as recto- or cystoscopy.
Major (grades 3 and 4): Symptoms that affect the performance status of the patient and that require surgery or invasive procedures. Grade 4 was defined as
treatment-related death.
Fistulas to the vagina were not counted as a complication if histopathologic examination showed recurrent disease in the fistula area.
Statistical analysis
Survival was analyzed by life-table methods. Comparison of survival between groups was carried out by
log-rank test. Survival analysis with covariates was
performed by the Cox proportional hazards regression model, including the following variables: clinical
tumor stage, lymph node metastases, pathologic complete remission, age, histologic type, histologic grade,
technique of brachytherapy, and tumor size. Survival
time was calculated from the date of clinical staging to
death or last time of follow-up. Survival curves show
disease-specific survival if no other information is
given. For analysis of proportions (e.g., relapse) Fishers exact test was used.

Results
Surgery
Radical hysterectomy, bilateral salpingo-oophorectomy, and pelvic lymphadenectomy were performed
on patients after intracavitary irradiation, according to
the class III WertheimMeigs procedure (17) . The
lymph node dissection included the external iliac, hypogastric, and obturator lymph nodes. The parametria
and the uterosacral ligaments were resected at the pelvic wall. One to 2 cm of the upper part of the vagina
were excised in most cases. A greater portion of the
vagina was resected if there were clinical signs of tumor growth at the resection margins.

Complications
Our review of patient records enabled us to register
late complications and classify them according to the
glossary of Chassagne et al.(18) Late side effects were
defined as complications persistent or occurring more
than 3 months after the end of radiotherapy. We did
not include minor complications (grade 1) in this
analysis.

In this retrospective analysis we studied records from

patients with cervical cancer consecutively admitted
to Radiumhemmet during the years 19891991. The
analysis is based on a total of 185 patients, 129 patients
in stage Ib and 56 patients in stage IIa. We found that
the overall 5-year survival rate of patients with stage
Ib disease was 82% and with stage IIa, 66%. The disease-specific 5-year survival rate was 87% and 73% for
stage Ib and IIa, respectively (Fig. 1). In the group of
patients treated with preoperative ICRT and surgery
we found a disease-specific 5-year survival rate of 87%
in stage Ib and 75% in stage IIa (Fig. 2); the overall
5-year survival rate was 86% and 75% for stages Ib and
IIa, respectively.
When analyzing survival in relation to histologic
type and grade in 185 patients, we found no difference
in survival between adenocarcinoma and squamous
cell carcinoma, with a 5-year survival rate for patients
with squamous cell carcinoma of 77% and 80% for
patients with adenocarcinoma. There was a trend for
better 5-year survival in well- and moderately differentiated cases compared with low-differentiated
cases, but this difference was not statistically significant.
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

162

Beskow et al.

Fig. 1. Disease-specific survival of 185

patients FIGO stage Ib and IIa cervical
cancer treated at Radiumhemmet during
19891991.

Among the 134 patients who underwent surgery,

including lymph node dissection, positive pelvic
nodes were found in 25 patients (19%), with 12 of 104
patients in stage Ib (12%) and 13 of 30 patients in stage
IIa (43%). Five-year survival in node-negative patients
was 92%, compared to 38% in node-positive patients
(P < 0.0001). Node positivity was a prognostic factor
irrespective of stage.
We were interested in understanding the impact of
tumor remission after preoperative brachytherapy on
survival. Therefore we analyzed whether or not the
degree of radiation-induced histopathologic remission

in the surgical specimen had a prognostic impact on

survival. Table 1 shows tumor characteristics in relation to pCR. Pathologic complete remission after intracavitary treatment was found in 79% of patients.
Patients with pCR showed a 95% 5-year survival rate
compared to a 46% 5-year survival rate in patients
with non-pCR (P < 0.0001) (Fig. 3).
Tumor size is a known prognostic factor in early
stage cervical cancer. When analyzing survival in relation to primary tumor size, we found 5-year survivals of 94%, 88%, and 73% for small, medium, and large
tumors, respectively (Fig. 4). The number of patients

Fig. 2. Survival of 121 patients stage Ib

and IIa treated with preoperative
intracavitary radiotherapy.
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

Cervical cancer and remission after radiotherapy

Table 1.

Tumor characteristics in relation to tumor remission

No. of patients
Tumor
characteristics
Stage IB
Stage IIA
Node negative
Node positive
Histologic type
SCC
Adenocarcinoma
Others
Grade of
differentiation
Highmoderate
Low
Unknown
Tumor size
Small
Medium size
Large

Pathological
complete
remission

Nonpathological
complete
remission

Total
no. of
patients

95 (79%)

26 (21%)

121

79 (81%)
16 (67%)
86 (86%)
9 (43%)

18 (19%)
8 (33%)
14 (14%)
12 (57%)

97
24
100
21

68 (81%)
22 (76%)
5 (63%)

16 (19%)
7 (24%)
3 (37%)

84
29
8

60 (81)
29 (73%)
6 (86%)

14 (19%)
11 (27%)
1 (14%)

74
40
7

30 (97%)
37 (88%)
28 (58%)

1 (3%)
5 (12%)
20 (42%)

31
42
48

presenting with different tumor sizes and the frequency of pCR for the different groups are shown in
Table 1. Five-year survival in relation to pCR and primary tumor size is shown in Table 2. In patients with
pCR, tumor size had no significant impact on survival.
Patients with large tumors and pCR had a 5-year survival rate of 96% compared to 40% in patients with
non-pCR (P < 0.0001).
As noted above, node positivity is a negative prognostic factor for survival. We therefore separately ana-

163

lyzed the degree of the primary tumor remission in

node-negative and node-positive patients. In nodenegative patients we found 5-year survivals of 97%,
92%, and 88% in patients with small, medium, and
large tumors, respectively. Pathologic complete remission was found in 86% of node-negative patients compared to 43% in node-positive patients (Table 1) For
node-negative patients with pCR (86 patients), the
5-year survival rate was 98% compared to a 64%
5-year survival rate in patients with non-pCR (14 patients) (P < 0.0001) (Fig. 5). The 5-year survival in relation to pCR and primary tumor size in nodenegative cases is shown in Table 3. In patients with
large tumors we found a 5-year survival of 100% in
cases with pCR compared to 56% in cases with nonpCR (P < 0.0001).
Twenty-one of 121 patients had positive lymph
nodes, 12 patients in stage Ib (12%) and 9 patients in
stage IIa (38%). Patients with positive pelvic nodes
found at surgery received postoperative EBRT. For
node-positive cases we found a trend for better survival of cases with pCR of the primary tumor compared with non-pCR cases. In patients with pCR and
positive pelvic nodes (9 patients), the 5-year survival
rate was 67% compared to 25% in patients with nonpCR and positive pelvic nodes (12 patients) (Fig. 6).
This result was, however, not statistically significant
(P = 0.11). When relating survival to primary tumor
size in node-positive cases we found 5-year survivals
of 50%, 33%, and 44% for small, medium, and large
tumors, respectively. Five-year survival in relation to
pCR and primary tumor size in node-positive patients

Fig. 3. Impact of pathologic complete

remission on survival of patients stage Ib
and IIa treated with preoperative
intracavitary radiotherapy. pCR =
pathologic complete remission; non-pCR =
nonpathologic complete remission.
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

164

Beskow et al.

Fig. 4. Survival in relation to tumor size of

121 patients stage Ib and IIa treated with
preoperative intracavitary radiotherapy.

Table 2. Five-year survival in relation to pathologic complete

remission and primary tumor size
pCR

Small tumors
Medium tumors
Large tumors

Non-pCR

No. of
cases

5-year
survival

No. of
cases

5-year
survival

30
37
28

28/30 (93%)
34/37 (92%)
27/28 (96%)

1
5
20

1/1 (100%)
3/5 (60%)
8/20 (40%)

is shown in Table 4. Considering the impact of pCR as

well as node positivity, different prognostic groups
may be based on these variables (Fig. 7).
Since pCR seems to be important for a favorable

outcome, we were interested in understanding whether differences in histology could predict the degree of
remission. The distribution of histologic type and
grade of tumor differentiation in relation to degree of
remission is shown in Table 1. Histologic type and
grade of differentiation did not predict pCR or nonpCR in this series. The question was also addressed
whether the technique of brachytherapy (i.e., manual
technique versus remote after-loading technique) had
any impact on survival. The kind of brachytherapy
technique used did not have an impact on the frequency of pCR. Likewise, there was no significant dif-

Fig. 5. Impact of pathologic complete

remission on survival of patients with
pelvic lymph node metastases treated
without preoperative intracavitary
radiotherapy. pCR = pathologic complete
remission; non-pCR = nonpathologic
complete remission.
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

Cervical cancer and remission after radiotherapy

Table 3. Five-year survival in relation to pathologic complete

remission and primary tumor size in node-negative cases
pCR

Small tumors
Medium tumors
Large tumors

Table 4. Five-year survival in relation to pathologic complete

remission and primary tumor size in node-positive cases

Non-pCR

No. of
cases

5-year
survival

No. of
cases

5-year
survival

28
35
23

27/28 (96%)
33/35 (94%)
23/23 (100%)

1
4
9

1/1 (100%)
3/4 (75%)
5/9 (56%)

ference in mean dose of the groups of patients obtaining pCR and non-pCR (Table 5).
A multivariate analysis was performed to evaluate
several prognostic factors for survival (Table 6). The
variables that were analyzed were clinical stage (IIa
versus Ib), pelvic lymph node metastases, non-pCR,
patient age (assessed by grouping patients into 10year cohorts), tumor histology (adenocarcinoma versus squamous cell carcinoma), tumor grade (high and
medium versus low), technique of brachytherapy
(manual versus after-loading), and tumor size (medium size and large tumors versus small tumors). The
only variables that were found to be of independent
significance for decreased survival were non-pCR (P =
0.0015) and the presence of pelvic lymph node metastases (P = 0.0065), with relative risks (RR) of 6.42 and
4.59, respectively. Large tumor size was not found to
be an independent variable for decreased survival in
this series (RR = 1.23).
Recurrences
A total of 21 recurrences were found among 121 patients treated with preoperative brachytherapy. Six-

165

pCR

Small tumors
Medium tumors
Large tumors

Non-pCR

No. of
cases

5-year
survival

No. of
cases

5-year
survival

2
2
5

1/2 (50%)
1/2 (50%)
4/5 (80%)

0
1
11

0/1 (0%)
3/11 (27%)

teen patients showed primary locoregional failures

and five patients developed distant metastases. Table
7 shows the relapse pattern in relation to pathologic
tumor remission. We found two locoregional recurrences in 95 patients with pCR (2%) compared to 14
patients with locoregional recurrence out of 26 patients with non-pCR (54%) (P < 0.0001).
Complications
A total of 12 patients among 121 patients (10%) treated
with intracavitary irradiation followed by surgery developed late complications. All were classified as
grade 2 except for one treatment-related death. There
were no vaginal fistulas. Four of 25 patients that received postoperative EBRT developed side effects
compared to 8 of 96 patients that were treated with
brachytherapy and surgery only. The type of complication in relation to treatment is shown in Table 8. The
patient who died was a 51-year-old who underwent
two intracavitary insertions, and pelvic lymph node
metastases were found during surgery as well as a
second primary malignancy of the ovaries. She re-

Fig. 6. Impact of pathologic complete

remission on survival of 21 patients with
pelvic lymph node metastases treated with
preoperative intracavitary radiotherapy,
surgery, and postoperative external beam
radiotherapy. pCR = pathologic complete
remission. Non-pCR = non-pathologic
complete remission.
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

166

Beskow et al.

Fig. 7. Survival in relation to the degree of

pathologic complete remission and pelvic
lymph node metastases in patients with
preoperative intracavity radiotherapy. pCR
= pathologic complete remission. Non-pCR
= non-pathologic complete remission. () =
lymph node negative. (+) = lymph node
positive.

Table 5.

Technique of brachytherapy and tumor remission

No. of patients
Manual technique
Mean dose (mghRa)
Remote afterloading
Mean dose (Gy)
point A

Pathologic
complete
remission

Nonpathologic
complete
remission

Total
no. of
patients

95
36 (80%)
6200 1100
59 (78%)

26
9 (20%)
6700 650
17 (22%)

121
45

43.8 2.4

43.0 2.3

76

Table 6. Multivariate analysis of prognostic factors in patients

treated with preoperative brachytherapy
Factor
FIGO stage
Node positivity
Non-pCR
Age
Histologic type
Grade of differentiation
Technique of brachytherapy
Tumor size
Medium
Large

Risk
ratio

Confidence
limits

Table 7.

Relapse pattern in relation to tumor remission

Total
No
Locoregional Distant
no. of
recurrence recurrence
metastases patients

Pathologic
complete
remission
89 (94%)
Nonpathologic
complete
remission
11 (42%)
Total
100 (83%)

2 (2%)

4 (4%)

95

14 (54%)
16 (13%)

1 (4%)
5 (4%)

26
121

Table 8. Distribution of late complications in relation to treatment type

0.76
4.59
6.42
1.00
0.58
1.03
1.32

0.202.54
1.5414.2
2.0322.5
0.951.06
0.181.66
0.352.80
0.434.69

0.66
0.0065
0.0015
0.87
0.32
0.96
0.64

1.08
1.23

0.198.30
0.229.54

0.94
0.82

ceived postoperative EBRT and additional chemotherapy. Ten weeks after the end of EBRT a laparotomy was performed due to bowel obstruction and the
patient died 2 weeks later due to bowel perforation.

Discussion
The present retrospective analysis of the response of
preoperative intracavitary radiotherapy of cervical
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

ICRTa +
surgery
(n = 96)
Bladder
Rectum
Bowel
Total no. of complications

5
2
1
8 (8%)

ICRT +
surgery +
EBRTb
(n = 25)
1
5
6 (24%)

Total
(n = 121)
6
2
6
14c (12%)

Intracavitary radiotherapy.
External beam radiotherapy.
c
Two patients developed complications from two organs.
b

cancer stages Ib and IIa addresses the importance of

primary radiotherapy response to the outcome of patients admitted to Radiumhemmet during the period
19891991. First, we conclude that our treatment results after preoperative intracavitary radiotherapy
compares well with other treatment techniques. Second, we found that pCR after preoperative intracavitary radiotherapy is a significant variable for survival.
Looking at our total material of consecutively ad-

Cervical cancer and remission after radiotherapy

mitted patients during the study period, our results of

5-year disease-specific survival rates of 87% and 73%
in stage Ib and IIa, respectively, compare well with
treatment results from other centers. We have not
found studies reporting treatment results of consecutive patient populations, and thus a direct comparison
with other materials is not possible. Comparing specifically the patients treated with preoperative ICRT
and surgery, our results of a 5-year disease-specific
survival rate of 85% in stage Ib and 75% in stage IIa
compares well with the results reported using similar
treatment techniques(1922).
The main focus of the present study was to understand whether or not there is an impact of pCR of the
primary tumor concerning the outcome of the patients
treated by preoperative ICRT. Even though it is well
known that lymph node metastases confer a poorer
prognosis, our multivariate analysis showed that pCR
was a stronger factor for survival than lymph node
metastases. Thus we found a strong correlation between pCR of the primary tumor and survival (P <
0.0001). It is important to note that this correlation is
not simply related to tumor size (see below).
There was a high frequency of pCR (79%) in our
series which is of the same order as found by others(68,21,23). There are, however, reports on lower degrees of tumor sterilization(5,9,24,25). The reason for
these differences remains obscure at present. Factors
such as the selection of patients, differences in tumor
size or stages of disease, involvement of lymph nodes,
as well as differences in histology, radiation dose, and
technique of brachytherapy should be considered
when comparing results obtained in different studies.
The time interval between the end of radiation and
surgery may influence the rate of observed pCR. The
rate of eradicated cancer seen in specimens after primary irradiation became larger with a longer time interval to surgery(24). In our material, no important
time variation existed and thus no difference in pCR
due to this factor could be studied.
One factor that might have to be considered when
analyzing what governs the rate of pCR is the delivered dose rate. During the period our patients were
admitted to Radiumhemmet the technique of brachytherapy changed from the manual technique, with a
mean dose rate of 1.1 Gy/hr, to the remote afterloading technique, with a mean dose rate of 1.35 Gy/
hr. We did not find a significant difference in the rate
of pCR between the two techniques. Further, multivariate analysis did not reveal any significant impact
for survival as a function of the technique. Of interest,
however, is that Lambin et al.(24) report on a prospective randomized trial comparing two dose rates (0.38

167

Gy/hr versus 0.73 Gy/hr) in patients with cervical

carcinoma treated with preoperative brachytherapy.
They observed an inverse dose rate effect for mediumsize tumors (4059 mm), with significantly more sterilizations observed in stage II (proximal) patients in
the lower dose rate group. Thus differences in dose
rate may have an impact on tumor response. This was
not, however, the case in our material.
The importance of pCR for survival has been addressed by a number of authors reporting on preoperative ICRT. Dealing with several of these reports it
was not possible to conclude whether pCR is a statistically significant favorable prognostic factor(5,7,21,25).
There are, however, some reports that provide statistics for such comparisons. Mundt et al.(9) report no
significant difference in disease-free survival rates between patients with or without residual disease in the
pathologic specimen after preoperative ICRT in a
study including stage Ib and IIa cervical cancer with
tumors larger than 2 cm. This is in accordance with the
result reported by Calais et al.(8), who concluded that
the pathologic status of the cervix was not a prognostic factor in a study of patients with tumors less than
4 cm. Sundfor et al.(10), who performed a randomized
trial comparing radical radiotherapy versus brachytherapy plus surgery in patients with cervical cancer
stage IIa and IIb, report that residual tumor in the
surgical specimen after preoperative brachytherapy
was not prognostic for death from recurrent cancer. In
contrast, Gerdin et al.(6) report that residual disease in
the hysterectomy specimen after brachytherapy is a
statistically significant predictive factor for recurrence
as well as survival. Our present study adds information on this issue, since we found a strong correlation
between pCR and survival (P < 0.0001).
Several reports on treatment results in early stage
cervical carcinoma in which patients were treated with
radiotherapy alone show that tumor size seems to be
an important factor for survival(2628). Of interest,
however, is that Grigsby et al.(29) found no statistically
significant effect of pretreatment cervical tumor size
on survival in patients with stage Ib and IIa cervical
cancer treated with preoperative radiation, including
intracavitary and external radiotherapy. Our results
using a multivariate analysis is supportive of Grigsby
et al.s findings. In patients with pCR, tumor size had
no significant impact on survival. While pCR of large
tumors confers an excellent prognosis, the presence of
node metastases will, however, impair the likelihood
of good long-term survival.
Since the majority of patients with stage Ib and IIa
cervical cancer present with node-negative disease, it
is important to identify tumor-related risk factors for
2002 IGCS, International Journal of Gynecological Cancer 12, 158170

168

Beskow et al.

recurrence and decreased survival in this group of

patients. Several reports have identified variables such
as capillary lymphatic space invasion (CLS), deep stromal invasion (DSI), and large tumor size (LTS) as independent risk factors for recurrence and mortality(3033). Attempts have been made with different
treatment regimens to increase survival rates in this
prognostically unfavorable group. Keys et al.(34) reported results of a randomized trial comparing chemotherapy, radiation, and hysterectomy with radiation and hysterectomy for patients with bulky stage Ib
cervical carcinoma. Only patients with no radiographic evidence of lymphadenopathy on computerized scanning or lymphangiography were eligible for
the study. The 3-year survival rates were 74% in the
group given radiotherapy alone and 83% in the chemoradiation group, which was a statistically significant difference. Sedlis et al.(35) report the results of a
randomized trial of pelvic radiation therapy versus no
further therapy in node-negative patients with stage
Ib cervical carcinoma treated with primary radical surgery. Cases with a combination of risk factors including CLS, DSI, and LTS were selected for the study. A
significant reduction in recurrence was found in the
radiotherapy group, with a 2-year recurrence-free survival of 88% compared to the 2-year survival of 79% in
patients with no additional therapy. In the present
study we found a highly significant reduction of locoregional recurrences in patients with pCR compared
to non-pCR patients. When using preoperative
brachytherapy, CLS, DSI, and adjacent tissues at risk
for invasion are targeted, leading to optimization of
tumor sterilization of adjacent tissue. This in turn is
also important for the survival difference seen between cases with pCR and non-pCR. The difference in
5-year survival among node-negative patients with
large tumors and non-pCR compared with patients of
the same group but with pCR (56% compared to
100%) points out that intrinsic resistance factors of the
tumor have a major importance for therapy outcome.
What influence histologic subtype has on radioresponsiveness and prognosis is still unclear. There are
several reports on similar survival rates of adenocarcinomas and squamous cell carcinomas (SCCs)(6,3638).
However, others have reported a better survival
rate for patients with SCC compared to patients
with adenocarcinoma or adenosquamous carcinoma(31,33,3944). These studies, however, do not answer the question of what importance histology has
on primary tumor remission. Our data do not support
any difference in radiosensitivity, measured as the degree of pCR, between SCC and adenocarcinoma. Similarly we did not find any significant difference in sur 2002 IGCS, International Journal of Gynecological Cancer 12, 158170

vival between patients with SCC compared with

adenocarcinoma.
The aim of cancer treatment is to obtain a high cure
rate at the cost of a low complication rate. In early
stage cervical carcinoma, most patients have a high
probability for long-term survival and it is important
that their quality of life is not impaired by complications originating from the treatment given. In patients
treated with primary surgery, additional postoperative EBRT plays an important role in the total complication rate. Landoni et al.(3) demonstrate in a randomized study of radical surgery versus radiotherapy
alone in stage Ib and IIa cervical cancer a total frequency of 28% grade 23 morbidity in the primary
surgery group, including cases with adjuvant external
radiotherapy, as compared to 12% in the radiotherapy
group. He also reported that 64% (108 patients) of the
cases in the primary surgery group received postoperative EBRT, of which 58% (63 patients) were affected
by other risk factors than positive nodes. In our material, 21% of the patients treated with preoperative
ICRT and surgery received postoperative EBRT. Only
four of those cases (16%) were due to other causes
than node positivity. The use of preoperative ICRT
may reduce the need for adjuvant EBRT and thus lowers the risk of late complications due to postoperative
EBRT. When discussing preservation of vaginal function, this is of importance, together with the possibility
to perform a less radical resection of the vagina after
preoperative ICRT.
It has been reported that treatment with full-course
radiotherapy followed by surgery results in a higher
degree of serious complications than irradiation
alone(45,46). In contrast, Perez et al.(47) report no difference in morbidity in patients with stage Ib, IIa, and
limited IIb treated with irradiation alone or combined
with surgery. According to the literature, the risk of
major late complications after preoperative brachytherapy show a great variation. A review of 20 reports
concerning treatment results after preoperative
brachytherapy was performed by Mundt et al.(9). The
risk of developing fistulas varied between 0 and 5.2%.
Kjorstad et al.(48) concluded in a retrospective analysis
of 612 patients treated with preoperative ICRT and
radical surgery that no significant increase in complications could be attributed to the use of preoperative
ICRT. The brachytherapy technique employed and radiation dose, as well as the radiation schedule and the
type of surgical procedure performed, may be important in determining the morbidity of combined
therapy.
In our series there were no major complications
such as fistulas in patients treated with ICRT and sur-

Cervical cancer and remission after radiotherapy

gery, including cases with additional postoperative

EBRT. This might be explained by the Stockholm
technique, which is based on individualization. Another factor of importance for the low frequency of
complications in our material is probably the time relation between ICRT, surgery, and EBRT. Our hypothesis is that the 3-week interval between the two fractions of ICRT as well as the 4-week interval between
ICRT and surgery and the 4- to 5-week interval between surgery and additional EBRT will provide time
for normal tissue to repair.

10

Conclusion
Based on the result from this retrospective analysis we
conclude that pathologic complete remission after preoperative intracavitary radiotherapy is a strong indicator for long-term survival in patients with stage Ib
and IIa cervical cancer. Patients with pCR have a significantly lower frequency of local recurrences as compared to patients with non-pCR. Brachytherapy according to the Stockholm method results in a low rate
of moderate and major late complications. The combination of preoperative brachytherapy and surgery
offers a clinical opportunity to identify patients with
poor survival (non-pCR) and other factors, such as
lymph node metastases, which may aid in the selection of patients for adjuvant therapy.

11

12

13

14

15

16

Acknowledgments

17

This study was supported by a grant from the Stockholm Cancer Society (project no. 00:168).

18

References

19

1 Benedet J, Odicino F, Maisonneuve P et al. Annual report

on the results of treatment in gynecological cancer, vol.
23. J Epidemiol Biostat 1998;3:534.
2 Hoskin WJ, Perez CA, Young RC. Principles and Practice of
Gynecologic Oncology, 2nd ed. Philadelphia: LippincottRaven, 1997.
3 Landoni F, Maneo A, Colombo A et al. Randomised
study of radical surgery versus radiotherapy for stage
Ib-IIa cervical cancer. Lancet 1997;350:53540.
4 Newton M. Radical hysterectomy or radiotherapy for
stage I cervical cancer. A prospective comparison with 5
and 10 year follow-up. Am J Obstet Gynecol 1975;123:
5359.
5 Bonar LD. Results of radical surgical procedures after
radiation for treatment of invasive carcinoma of the uterine cervix in a private praxis. Am J Obstet Gynecol
1980;136:10068.
6 Gerdin E, Cnattingius S, Johnson P, Pettersson B. Prognostic factors and relapse patterns in early-stage cervical

20

21

22

23

24

169

carcinoma after brachytherapy and radical hysterectomy. Gynecol Oncol 1994;53:3149.

Pilleron JP, Durand JC, Lenoble JC. Carcinoma of the
uterine cervix, stages I and II, treated by radiation
therapy and extensive surgery (1000 cases). Cancer
1972;29:5936.
Calais G, Le Floch O, Chauvet B, Reynand-Bougnoux A,
Bougnoux P. Carcinoma of the uterine cervix stage IB
and early stage II. Prognostic value of the histological
tumor regression after initial brachytherapy. Int J Radiat
Oncol Biol Phys 1989;17:12315.
Mundt AJ, Waggoner S, Herbst A, Rotmensch J. Preoperative intracavitary brachytherapy in early-stage cervical carcinoma. Am J Clin Oncol 1999;22:737.
Sundfor K, Trope CG, Kjorstad KE. Radical radiotherapy
versus brachytherapy plus surgery in carcinoma of the
cervix 2A and 2B. Acta Oncol 1996;35(suppl 8):99107.
Swedish Cancer Registry, Centre for Epidemiology. Cancer incidence in Sweden 1991. Stockholm: National
Board of Health and Welfare, 1994.
Pettersson F, ed. Annual Report on the Results of
Treatment in Gynecological Cancer, vol. 20. Stockholm:
Radiumhemmet, Karolinska Sjukhuset, 1988.
Kottmeier HL. Surgical and radiation treatment of carcinoma of the uterine cervix. Acta Obstet Gynecol Scand
1964;43(suppl 2):148.
Wahlstam R. History of brachytherapy in Sweden including the origins of the Stockholm & Heyman techniques. Brachyther J 1992;6:7981.
Tod M, Meredith WJ. Treatment of cancer of the cervix
uteria revised Manchester method. Br J Radiol
1953;26:2527.
International Committee on Radiation Units and Measurement. Dose and volume specifications for reporting
intracavitary therapy in gynecology. ICRU report no. 38.
Bethesda, MD: ICRU, 1985.
Piver MS, Rutledge F, Smith JP. Five classes of extended
hysterectomy for women with cervical cancer. Obstet Gynecol 1974;44:26572.
Chassagne D, Sismondi P, Horiot JC et al. A glossary for
reporting complications of treatment in gynecological
cancers. Radiother Oncol 1993;26:195202.
Gerbaulet AL, Kunkler IH, Kerr GR et al. Combined radiotherapy and surgery: local control and complications
in early carcinoma of the uterine cervixthe Villejuif
experience, 19751984. Radiother Oncol 1992;23:6673.
Iversen T, Kjorstad KE, Martimbeau PW. Treatment results in carcinoma of the cervix stage IB in a total population. Gynecol Oncol 1982;14:15.
Rampone JF, Klem V, Kolstad P. Combined treatment of
stage IB carcinoma of the cervix. Obstet Gynecol
1973;41:1637.
Timmer PR, Aalders JG, Bouma J. Radical surgery after
preoperative intracavitary radiotherapy for stage IB and
IIA carcinoma of the uterine cervix. Gynecol Oncol
1984;18:20612.
Einhorn N, Patek E, Sjberg B. Outcome of different
treatment modalities in cervix carcinoma stage IB and
IIA. Cancer 1985;55:94955.
Lambin P, Gerbaulet A, Kramar A et al. A comparison of
early effects with two dose rates in brachytherapy of
cervix carcinoma in a prospective randomised trial. Eur J
Cancer 1994;30A:31220.

2002 IGCS, International Journal of Gynecological Cancer 12, 158170

170

Beskow et al.

25 Marziale P, Atlante G, Le Pera V, Mariano T, Pozzi M,

Iacovelli A. Combined radiation and surgical treatment
of stages IB and IIA and B carcinoma of the cervix. Gynecol Oncol 1981;11:17583.
26 Eifel PJ, Morris M, Wharton JT, Oswald MJ. The influence of tumor size and morphology on the outcome of
patients with FIGO stage IB squamous cell carcinoma of
the uterine cervix. Int J Radiat Oncol Biol Phys 1994;29:9
16.
27 Kapp KS, Stuecklschweiger GF, Kapp DS, Poschauko J,
Pickel H, Hackl A. Prognostic factors in patients with
carcinoma of the uterine cervix treated with external
beam irradiation and Ir-192 high-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 1998;42:53140.
28 Perez CA, Grigsby PW, Chao KSC, Mutch DG, Lockett
MA. Tumor size, irradiation dose, and long-term outcome of carcinoma of uterine cervix. Int J Radiat Oncol
Biol Phys 1998;41:30717.
29 Grigsby PW, Perez CA, Chao KSC et al. Lack of effect of
tumor size on the prognosis of carcinoma of the uterine
cervix stage IB and IIA treated with preoperative irradiation and surgery. Int J Radiat Oncol Biol Phys
1999;45:64551.
30 Delgado G, Bundy B, Zaino R, Sevin BU, Creasman W,
Major F. Prospective surgical pathological study of disease-free interval in patients with stage IB squamous cell
carcinoma of the cervix: a Gynecologic Oncology Group
study. Gynecol Oncol 1990;38:3527.
31 Fuller AF Jr, Elliot N, Kosloff C, Hoskins WJ, Lewis JL Jr.
Determinants of increased risk for recurrence in patients
undergoing radical hysterectomy for stage IB and IIA
carcinoma of the cervix. Gynecol Oncol 1989;33:349.
32 Sevin BU, Lu Y, Bloch D, Nadji M, Koechli O, Averette H.
Surgically defined prognostic parameters in patients
with early cervical carcinomaa multivariate survival
tree analysis. Cancer 1996;78:143846.
33 Samlal RAK, van der Velden J, Ten Kate FJW, Schilthuis
MS, Hart AA, Lammes FB. Surgical pathologic factors
that predict recurrence in stage IB and IIA cervical carcinoma patients with negative pelvic lymph nodes. Cancer 1997;80:123440.
34 Keys H, Bundy B, Stehman F et al. Cisplatin, radiation
and adjuvant hysterectomy compared with radiation
and adjuvant hysterectomy for bulky stage IB cervical
carcinoma. N Engl J Med 1999;340:115461.
35 Sedlis A, Bundy B, Rotman M, Lentz S, Muderspach L,
Zaino R. A randomized trial of pelvic radiation therapy
versus no further therapy in selected patients with stage
IB carcinoma of the cervix after radical hysterectomy and
pelvic lymphadenectomy: a Gynecologic Oncology
Group study. Gynecol Oncol 1999;73:17783.
36 Grigsby PW, Perez CA, Kuske RR et al. Adenocarcinoma

2002 IGCS, International Journal of Gynecological Cancer 12, 158170

37

38

39

40

41

42

43

44

45

46

47

48

of the cervix: lack of evidence for a poor prognosis. Radiother Oncol 1988;12:28996.
Kenter GG, Ansink AC, Heintz APM, Delemarre J, Aartsen EJ, Hart AA. Low stage invasive carcinoma of the
uterine cervix stage IIIa morphological prognostic factors. Eur J Surg Oncol 1988;14:18792.
Kilgore LC, Soong SJ, Gore H, Shingleton HM, Hatch
KD, Partridge EE. Analysis of prognostic features in adenocarcinoma of the cervix. Gynecol Oncol 1988;31:137
48.
Artman LE, Hoskins WJ, Bibro MC et al. Radical hysterectomy and pelvic lymphadenectomy for stage Ib carcinoma of the cervix: 21 years experience. Gynecol Oncol
1987;28:813.
Eifel PJ, Burke TW, Morris M, Smith TL. Adenocarcinoma as an independent risk factor for disease recurrence in patients with stage Ib cervical carcinoma. Gynecol Oncol 1995;59:3844.
Hopkins MP, Morley GW. A comparison of adenocarcinoma and squamous cell carcinoma of the cervix. Obstet
Gynecol 1991;77:9127.
Kamura T, Tsukamoto N, Tsuruchi N et al. Multivariate
analysis of the histopathological prognostic factors of
cervical cancer in patients undergoing radical hysterectomy. Cancer 1992;69:1816.
Morley GW, Seski JC. Radical pelvic surgery versus radiation therapy for stage I carcinoma of the cervix (exclusive of microinvasion). Am J Obstet Gynecol 1976;126:
78594.
Samlal RA, van der Velden J, Schilthuis MS, Gonzalez D,
Ten Kate FJ, Hart AA. Identification of high-risk groups
among node-positive patients with stage Ib and IIa cervical carcinoma. Gynecol Oncol 1997;64:4637.
Eifel PJ, Levenback C, Wharton JT, Oswald MJ. Time
course and incidence of late complications in patients
treated with radiation therapy for FIGO stage IB carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys
1995;32:1289300.
Weems DH, Mendenhall WM, Bova FJ, Marcus RB Jr,
Morgan LS, Million RR. Carcinoma of the intact uterine
cervix, stage IB-IIA-B, 6 cm in diameter: irradiation
alone vs preoperative irradiation and surgery. Int J Radiat
Oncol Biol Phys 1985;11:19114.
Perez CA, Grigsby PW, Camel HM, Galakatos AE,
Mutch D, Lockett MA. Irradiation alone or combined
with surgery in stage IB, IIA and IIB carcinoma of the
uterine cervix: update of a nonrandomized comparison.
Int J Radiat Oncol Biol Phys 1995;31:70316.
Kjorstad KE, Martimbeau PW, Iversen T. Stage IB carcinoma of the cervix, the Norwegian Radium Hospital:
results and complications. Gynecol Oncol 1983;15:427.

Accepted for publication November 15, 2001.