Pelvic Radiation with Concurrent Chemotherapy Compared with Pelvic and Para-Aortic Radiation for High-Risk Cervical Cancer

Abstract

We compared the effect of radiotherapy to a pelvic and para-aortic field with that of pelvic radiation and concurrent chemotherapy with fluorouracil and cisplatin in women with advanced cervical cancer. Between 1990 and 1997, 403 women with advanced cervical cancer confined to the pelvis (stages IIB through IVA or stage IB or IIa with a tumor diameter of at least 5 cm or involvement of pelvic lymph nodes) were randomly assigned to receive either 45 Gy of radiation to the pelvis and para-aortic lymph nodes or 45 Gy of radiation to the pelvis alone plus two cycles of fluorouracil and cisplatin (days 1 through 5 and days 22 through 26 of radiation). Patients were then to receive one or two applications of low-dose-rate intracavitary radiation, with a third cycle of chemotherapy planned for the second intracavitary procedure in the combined-therapy group.
Of the 403 eligible patients, 193 in each group could be evaluated. The median duration of follow-up was 43 months. Estimated cumulative rates of survival at five years were 73 percent among patients treated with radiotherapy and chemotherapy and 58 percent among patients treated with radiotherapy alone (P=0.004). Cumulative rates of disease-free survival at five years were 67 percent among patients in the combined-therapy group and 40 percent among patients in the radiotherapy group (P<0.001). The rates of both distant metastases (P<0.001) and locoregional recurrences (P<0.001) were significantly higher among patients treated with radiotherapy alone. The seriousness of side effects was similar in the two groups, with a higher rate of reversible hematologic effects in the combined-therapy group.
The addition of chemotherapy with fluorouracil and cisplatin to treatment with external-beam and intracavitary radiation significantly improved survival among women with locally advanced cervical cancer.

Full text

Volume 340 Number 15
·
1137
The New England
Journal
of
Medicine
© Copyright, 1999, by the Massachusetts Medical Society
VOLUME 340
A
PRIL
15, 1999
NUMBER 15
PELVIC RADIATION WITH CONCURRENT CHEMOTHERAPY COMPARED WITH
PELVIC AND PARA-AORTIC RADIATION FOR HIGH-RISK CERVICAL CANCER
M
ITCHELL
M
ORRIS
, M.D., P
ATRICIA
J. E
IFEL
, M.D., J
IANDONG
L
U
, P
H
.D., P
ERRY
W. G
RIGSBY
, M.D.,
C
HARLES
L
EVENBACK
, M.D., R
ANDY
E. S
TEVENS
, M.D., M
ARVIN
R
OTMAN
, M.D., D
AVID
M. G
ERSHENSON
, M.D.,
AND
D
AVID
G. M
UTCH
, M.D.
A
BSTRACT
Background and Methods
We compared the ef-
fect of radiotherapy to a pelvic and para-aortic field
with that of pelvic radiation and concurrent chemo-
therapy with fluorouracil and cisplatin in women
with advanced cervical cancer. Between 1990 and
1997, 403 women with advanced cervical cancer
confined to the pelvis (stages IIB through IVA or
stage IB or IIA with a tumor diameter of at least 5 cm
or involvement of pelvic lymph nodes) were ran-
domly assigned to receive either 45 Gy of radiation
to the pelvis and para-aortic lymph nodes or 45 Gy
of radiation to the pelvis alone plus two cycles of
fluorouracil and cisplatin (days 1 through 5 and days
22 through 26 of radiation). Patients were then to re-
ceive one or two applications of low-dose-rate intra-
cavitary radiation, with a third cycle of chemothera-
py planned for the second intracavitary procedure in
the combined-therapy group.
Results
Of the 403 eligible patients, 193 in each
group could be evaluated. The median duration of
follow-up was 43 months. Estimated cumulative rates
of survival at five years were 73 percent among pa-
tients treated with radiotherapy and chemotherapy
and 58 percent among patients treated with radio-
therapy alone (P=0.004). Cumulative rates of disease-
free survival at five years were 67 percent among pa-
tients in the combined-therapy group and 40 percent
among patients in the radiotherapy group (P<0.001).
The rates of both distant metastases (P<0.001) and
locoregional recurrences (P<0.001) were significant-
ly higher among patients treated with radiotherapy
alone. The seriousness of side effects was similar in
the two groups, with a higher rate of reversible he-
matologic effects in the combined-therapy group.
Conclusions
The addition of chemotherapy with
fluorouracil and cisplatin to treatment with exter-
nal-beam and intracavitary radiation significantly im-
proved survival among women with locally advanced
cervical cancer. (N Engl J Med 1999;340:1137-43.)
©1999, Massachusetts Medical Society.
From the Departments of Gynecologic Oncology (M.M., C.L., D.M.G.)
and Radiation Oncology (P.J.E.), University of Texas M.D. Anderson Can-
cer Center, Houston; the Statistical Unit, Radiation Therapy Oncology
Group, Philadelphia (J.L.); the Mallinckrodt Institute of Radiology
(P.W.G.) and the Division of Gynecologic Oncology (D.G.M.), Washington
University School of Medicine, St. Louis; the Department of Radiation
Oncology, New York University, New York (R.E.S.); and the Department
of Radiation Oncology, State University of New York Health Science Cen-
ter, Brooklyn (M.R.). Address reprint requests to Dr. Morris at the Depart-
ment of Gynecologic Oncology, University of Texas M.D. Anderson Can-
cer Center, 1515 Holcombe Blvd., Box 34, Houston, TX 77030, or at
morris@mdanderson.org.
F the estimated 13,700 women in the
United States in whom invasive cervical
cancer was diagnosed in 1998,
1
nearly
5000 will ultimately die of the disease be-
cause of the inadequacies of current treatment. In
the United States, cervical cancer disproportionately
affects women who are members of minority groups
and women of low socioeconomic status, partly be-
cause such women tend to have insufficient access to
and knowledge of screening programs for cervical
cancer. The nationwide use of such screening pro-
grams has greatly reduced the incidence of invasive
cervical cancer.
Women with early cervical cancer can be success-
fully treated with radical surgery. Those with a large
cervical lesion at presentation or with spread to the
pelvic lymph nodes or other pelvic tissues are usually
treated with a combination of external-beam and in-
tracavitary radiation.
2-6
A previous study reported
improved survival among women with locally ad-
vanced cervical cancer who received prophylactic ra-
diation to the para-aortic nodes.
7
To eradicate micrometastases and sensitize tumor
cells to radiation, several studies have explored the
use of radiotherapy with concomitant chemothera-
py.
8-11
The results of these studies are inconclusive
and have been criticized because they lacked a com-
parison group treated with radiation alone and be-
O
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .

1138
·
April 15, 1999
The New England Journal of Medicine
cause the radiation therapy used may have been de-
ficient according to current standards. In 1990, the
Radiation Therapy Oncology Group (RTOG) began
a randomized clinical trial to compare the effects on
survival of treatment with extended-field radiation
and treatment with pelvic radiotherapy and concur-
rent chemotherapy in patients with cervical cancer.
We report the first results of this study.
METHODS
Patients
We enrolled women of all ages who had stages IIB through
IVA squamous-cell carcinoma, adenocarcinoma, or adenosqua-
mous carcinoma of the cervix according to the staging system of
the International Federation of Gynecology and Obstetrics (Table
1) or stage IB or IIA of one of these cancers with a tumor diam-
eter of at least 5 cm or biopsy-proved metastasis to pelvic lymph
nodes. Women with a Karnofsky performance score of at least 60
and blood counts and serum levels of blood urea nitrogen, cre-
atinine, and bilirubin that were within normal ranges were eligi-
ble for the study. Women were excluded from the study if they
met any of the following criteria: disease outside the pelvic area
or spread to para-aortic lymph nodes; a prior cancer other than
cutaneous basal-cell carcinoma; medical contraindications to che-
motherapy; a rare histologic subtype; and prior hysterectomy or
transperitoneal staging procedure for cervical cancer, pelvic radio-
therapy, or systemic chemotherapy.
A medical history taking and clinical examination were required
before enrollment. The initial evaluation also included chest radi-
ography, cystoscopy, proctoscopy, a complete blood count, and
measurement of liver and renal function. The renal-collecting sys-
tem of each patient was assessed by intravenous pyelography or
contrast computed tomography. Para-aortic lymph nodes were
evaluated by bipedal lymphangiography or retroperitoneal surgi-
cal exploration.
The surveillance committees of the National Cancer Institute
and participating institutions approved this trial. Patients were
required to understand the trial and provide written informed
consent.
Patients who completed the pretreatment evaluation and met
all eligibility criteria were randomly assigned to receive extended-
field radiotherapy or radiotherapy to the pelvic region with con-
current treatment with cisplatin and fluorouracil. The patients in
each treatment group were stratified according to the tumor stage
(IB, IIA, or IIB vs. III or IVA) and the staging method used for
para-aortic lymph nodes (clinical vs. surgical).
Treatment
External-beam radiation was delivered with anteroposterior
and posteroanterior opposed beams of at least 15-MV photons or
the use of four fields (anteroposterior, posteroanterior, and two
lateral fields) of at least 4-MV photons. For patients who were as-
signed to receive radiotherapy and chemotherapy, the treatment
field extended from the space between L4 and L5 to the mid-
pubis or to a line 4 cm below the most distal vaginal or cervical
site of disease. Lateral fields were designed to encompass S3 pos-
teriorly, with a margin of at least 3 cm from the primary cervical
tumor. Custom shielding was designed to treat the pelvic lymph
nodes, with a margin of at least 1 to 1.5 cm. For patients who
were assigned to receive radiotherapy alone, the pelvic and para-
aortic areas were treated as a continuous area, with a superior-
field border at the space between L1 and L2. The radiation dose
was keyed to the central ray at the patient’s midplane (for antero-
posterior–posteroanterior fields) or to the isocenter of the beams.
The total dose to be delivered to pelvic lymph nodes as well as to
para-aortic nodes was 45 Gy, given at a dose of 1.8 Gy per frac-
tion of radiation.
The radioisotopes used for low-dose-rate intracavitary radio-
therapy were cesium-137 or radium-226 in 392 of 402 patients
(98 percent). The first intracavitary treatment was performed be-
fore or during external-beam radiotherapy, and additional exter-
nal-beam therapy was delivered with a midline block. Interstitial
brachytherapy was used only if necessary to increase the dose di-
rected at distal vaginal sites of disease. Brachytherapy was per-
formed within two weeks (preferably less than one week) after the
completion of pelvic radiation, with the goal of keeping the total
duration of treatment under eight weeks when possible.
The protocol specified that all patients receive a total cumula-
tive dose to point A (a reference location 2 cm lateral and 2 cm
superior to the cervical os) of at least 85 Gy. The suggested max-
imal doses to the bladder, the rectum, and the lateral surface of
the vagina were 75, 70, and 130 Gy, respectively.
Within 16 hours after the first radiation fraction was adminis-
tered, patients in the combination-therapy group received the
first cycle of chemotherapy, which consisted of an intravenous in-
fusion of 75 mg of cisplatin per square meter of body-surface area
over a 4-hour period followed by an intravenous infusion of 4000
mg of fluorouracil per square meter over a 96-hour period. This
timing corresponded to days 1 through 5 of radiation therapy.
Two additional cycles of chemotherapy were scheduled at three-
week intervals. One of these was administered at the time of
the second intracavitary insertion. To avoid treatment delays, in-
tracavitary insertions were performed without chemotherapy if a
patient had a granulocyte count of less than 1500 per cubic mil-
limeter and a platelet count of less than 100,000 per cubic milli-
meter.
Follow-up
During treatment, patients were evaluated weekly by clinical as-
sessments, a complete blood count with differential and platelet
counts, and a pelvic examination. Before each cycle of chemo-
therapy, serum levels of creatinine, urea nitrogen, alanine amino-
transferase, alkaline phosphatase, and bilirubin were measured.
Patients had a pelvic examination under anesthesia at the time of
each intracavitary treatment.
Once treatment ended, patients were evaluated every three
*Data were adapted from the International Federation of Gynecology
and Obstetrics.
12
†This category includes all cases in which hydronephrosis is present, un-
less it is known to be due to another cause.
T
ABLE
1.
S
TAGING
OF
C
ARCINOMA
OF
THE
C
ERVIX
.*
S
TAGE
D
EFINITION
I
IA1, IA2
IB1
IB2
Invasive cancer is confined to the cervix.
Invasive cancer is identified microscopically, with a maximal
depth of 5 mm and a maximal width of 7 mm.
Clinically apparent lesions are no greater than 4 cm in
diameter.
Clinically apparent lesions exceed 4 cm in diameter.
II
IIA
IIB
The tumor extends beyond the cervix but not to the pelvic
wall or distal vagina.
Up to two thirds of the vagina is involved, with no obvious
parametrial involvement.
There is obvious parametrial involvement.
III
IIIA
IIIB
The tumor extends to the pelvic wall or involves the lowest
third of the vagina.†
The lowest third of the vagina is involved but not the pelvic
wall.
The pelvic wall is involved, or hydronephrosis or a nonfunc-
tioning kidney is present.
IV
IVA
IVB
The tumor extends beyond the pelvis or involves the bladder
or rectum.
Rectal or bladder mucosa is involved.
There is spread to distant organs.
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .

PELVIC RADIATION PLUS CHEMOTHERAPY VERSUS PELVIC AND PARA-AORTIC RADIATION FOR CERVICAL CANCER
Volume 340 Number 15
·
1139
months for the first two years, every four months during the third
year, every six months during the fourth and fifth years, and then
annually. Disease status and the degree of treatment-related toxic
effects were assessed by history taking, physical examination, and
appropriate laboratory and radiologic tests. Suspected cases of
persistent or recurrent disease were confirmed by a biopsy when-
ever possible.
Toxicity was assessed at the time of each evaluation with use of
the Cooperative Group Common Toxicity Criteria, the Acute Ra-
diation Morbidity Scoring Criteria, and the Late Radiation Mor-
bidity Scoring Scheme of the RTOG and the European Organi-
zation for Research and Treatment of Cancer.
Quality Control
All chemotherapy records were reviewed by a gynecologic on-
cologist to assess compliance with the protocol. Radiotherapy
records, including data concerning external-beam fields, intracav-
itary placement, doses of radiation to tumor and normal tissues,
and other treatment variables, were reviewed by a radiation on-
cologist. Variations were scored as minor, major but acceptable,
or major and unacceptable, if they differed by more than 5, 10, or
20 percent, respectively, from the specified dose of radiation or
duration of treatment. Each institution’s equipment was calibrat-
ed by employees of the Radiological Physics Center in Houston.
Statistical Analysis
Overall survival was the primary end point for the comparison
of the two treatments and was calculated from the date of study
entry until the date of death or the date of the last follow-up visit.
Death from any cause was considered a failure in the analysis. Dis-
ease-free survival was also compared in the two groups and was
calculated from the date of study entry to the date of the first oc-
currence of disease progression, a second diagnosis of cancer, or
death from any cause, or if none of these events occurred, to the
date of the last follow-up visit.
The Kaplan–Meier method was used to calculate both survival
rates.
13
Log-rank tests were used to compare treatments.
14
All pa-
tients who could be assessed were included in the intention-
to-treat analysis. Five-year rates of secondary end points such as
locoregional recurrence, para-aortic recurrence, and distant me-
tastasis were estimated with the use of cumulative-incidence
methods,
15
and treatment effects were tested with use of the Gray
algorithm.
16
The Cox proportional-hazards model was used to es-
timate the hazard ratios.
17
The statistical significance of associa-
tions between treatment assignments and characteristics of the
patients was assessed by chi-square analysis. Acute side effects of
treatment were defined as those that occurred within 60 days af-
ter the completion of radiotherapy, and late effects were defined
as those occurring or persisting more than 60 days after radio-
therapy.
The study was initially designed to be able to detect a reduc-
tion of 33 percent in the annual death rate, with a statistical pow-
er of 80 percent and a two-sided significance level of 0.05. On
the basis of the results of two earlier RTOG trials,
7, 1 1
we predicted
that the five-year survival rate for the control group (radiotherapy
alone) would be 65 percent for patients with stage IB or II dis-
ease and 40 percent for patients with stage III or IVA disease. We
estimated that 40 to 70 percent of the patients would have stage
IB or II disease. A 33 percent reduction in the death rate as a
result of chemotherapy would yield an absolute improvement of
approximately 10 percent in the five-year survival rate. To detect
such a difference, we predicted that we would need to enroll 400
women in the study over a four-year period and then follow them
for an additional four years. We estimated that 199 women would
have died by the time of the initial analysis of treatment.
Interim analyses were scheduled to occur when 50 percent of
the patients had been enrolled and when the enrollment goal was
met. Early reporting of results was permitted if a significant dif-
ference was observed between the two treatment groups. The
nominal significance level required for early reporting was origi-
nally set at P=0.005, but we subsequently adopted a more con-
servative approach,
18
because it allowed the number of deaths
observed to determine the nominal level required for early re-
porting. The results of the interim analyses, in which patients’
treatment assignments were masked, were presented to the data-
monitoring committee. At the interim analysis conducted in July
1998 after the enrollment goal was met, the difference between
the two groups was determined to have met the requirements for
early reporting. For this early report, we updated the results using
all information received and entered at study headquarters by No-
vember 11, 1998.
RESULTS
Characteristics of the Patients
A total of 403 patients were enrolled in the study
between September 1990 and November 1997: 201
were randomly assigned to receive radiotherapy and
concurrent chemotherapy, and 202 were assigned to
receive radiotherapy alone. Fifteen patients (4 per-
*Because of rounding, not all percentages total 100.
†FIGO denotes International Federation of Gynecology and Obstetrics.
‡Only stage IB and IIA tumors were included.
T
ABLE
2.
C
HARACTERISTICS
OF
THE
P
ATIENTS
.*
C
HARACTERISTIC
R
ADIOTHERAPY
AND
C
HEMOTHERAPY
(N=195)
R
ADIOTHERAPY
A
LONE
(N=193)
Median age — yr 47 47
Race or ethnic group — no. (%)
White
Black
Hispanic
Other or unknown
99 (51)
46 (24)
39 (20)
11 (6)
86 (45)
56 (29)
43 (22)
8 (4)
Karnofsky performance score, <90
— no. (%)
29 (15) 38 (20)
Histologic diagnosis — no. (%)
Squamous-cell carcinoma
Adenocarcinoma
Adenosquamous carcinoma
176 (90)
11 (6)
8 (4)
174 (90)
13 (7)
6 (3)
Extent of differentiation of tumor cells
— no. (%)
Well
Moderate
Poor
Unknown
12 (6)
71 (36)
74 (38)
38 (19)
18 (9)
75 (39)
73 (38)
27 (14)
FIGO stage — no. (%)†
IB
IIA
IIB
III
IVA
54 (28)
11 (6)
71 (36)
53 (27)
6 (3)
52 (27)
13 (7)
69 (36)
57 (30)
2 (1)
Staging method for para-aortic lymph
nodes — no. (%)
Lymphangiography
Lymph-node dissection
Both
145 (74)
36 (18)
14 (7)
141 (73)
35 (18)
17 (9)
Pelvic lymph nodes — no. (%)
All negative
External iliac nodes positive, common
iliac nodes negative
Common iliac nodes positive
Unknown
147 (75)
30 (15)
17 (9)
1 (1)
130 (67)
43 (22)
19 (10)
1 (1)
Tumor diameter‡
<5 cm — no. (%)
»5 cm — no. (%)
Median — cm
4 (6)
61 (94)
6
9 (14)
56 (86)
6
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .

1140
·
April 15, 1999
The New England Journal of Medicine
cent) — six in the combined-therapy group and nine
in the radiotherapy group — were subsequently dis-
qualified because of failure to undergo the required
evaluation of para-aortic lymph nodes (eight pa-
tients), the presence of extrapelvic cancer (two), the
presence of a rare histologic subtype (one), the pres-
ence of a stage IB1 tumor with no involvement of
pelvic nodes (one), the absence of pretreatment data
(two), and receipt of chemotherapy before radio-
therapy (one). The 388 remaining patients (195 in
the combined-therapy group and 193 in the radio-
therapy group) form the basis of this analysis.
The characteristics of the two treatment groups are
summarized in Table 2. There were no significant dif-
ferences in these characteristics between the groups.
Treatment and Compliance
Radiotherapy was delivered according to the pro-
tocol or with minor deviations in 83 percent of the
patients in the combined-therapy group and 84 per-
cent of those in the radiotherapy group. Major but
acceptable treatment deviations occurred in another
11 percent and 9 percent of these groups, respec-
tively. Eleven patients (3 percent) did not undergo
brachytherapy: three patients refused, and eight pa-
tients did not undergo it for other reasons. In all,
26 patients did not complete treatment. The medi-
an total duration of radiation was 58 days, with a
mean dose of 89 Gy delivered to point A in each
group.
Of the 195 patients in the combination-therapy
group, 159 (81 percent) completed at least two cy-
cles of chemotherapy, and 133 (68 percent) com-
pleted three cycles. Chemotherapy was discontinued
because of toxic effects in 9 patients, refusal to con-
tinue in the case of 17 patients, diminished perform-
ance status in 4 patients, and other reasons in 4 pa-
tients. Four patients refused or were unable to receive
any chemotherapy but were included in the analysis
according to the intention to treat.
Side Effects
Although moderate (grade 3) and severe (grade
4) side effects occurred more frequently during or
within 60 days after the completion of treatment
with combined therapy than with radiotherapy alone,
these effects were usually self-limited or resolved
with medical management (Table 3). Hematologic
effects were generally moderate.
The late complications of treatment are summa-
rized in Table 4. There were no significant differenc-
es in the seriousness of late effects between the treat-
ment groups.
*A grade of 3 indicates a moderate effect, a grade of 4 a severe effect, and a grade of 5 a fatal effect.
T
ABLE
3.
W
ORST
S
IDE
E
FFECTS
R
EPORTED
DURING
T
REATMENT
OR
WITHIN
60 D
AYS
AFTER
THE
C
OMPLETION
OF
T
REATMENT
.*
S
IDE
E
FFECT
R
ADIOTHERAPY
AND
C
HEMOTHERAPY
(N=195)
R
ADIOTHERAPY
A
LONE
(N=193)
GRADE
3
GRADE
4
GRADE
5
GRADE
3
GRADE
4
GRADE
5
number of patients (percent)
Skin abnormalities 4 1 0 0 1 0
Nausea and vomiting 14 3 0 2 0 0
Bowel or rectal abnormalities 12 5 0 1 0 0
Bladder abnormalities 2 0 0 0 0 0
Hematologic effects 57 16 0 2 0 0
Other 831 00 0
Maximal grade of toxicity 65 (33) 22 (11) 1 (1) 5 (3) 1 (1) 0
Maximal grade of nonhemato-
logic toxicity
15 (8) 4 (2) 1 (1) 3 (2) 1 (1) 0
*A grade of 3 indicates a moderate effect, and a grade of 4 a severe effect.
†No follow-up data were available for two patients.
T
ABLE
4.
W
ORST
S
IDE
E
FFECTS
OF
T
REATMENT
O
CCURRING
OR
P
ERSISTING
M
ORE
T
HAN
60 D
AYS
AFTER
THE
C
OMPLETION
OF
T
REATMENT
.*
S
ITE
OF
S
IDE
EFFECT
RADIOTHERAPY AND
C
HEMOTHERAPY
(N=193)†
R
ADIOTHERAPY ALONE
(N=193)
GRADE 3 GRADE 4 GRADE 3 GRADE 4
number of patients (percent)
Skin or subcutaneous tissue 1 0 0 1
Small bowel 1 4 0 7
Large bowel or rectum 4 13 2 17
Bladder 4 1 1 2
Ureters 1 2 0 2
Other 2 1 1 3
Maximal grade of toxicity 8 (4) 16 (8) 2 (1) 20 (10)
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .

PELVIC RADIATION PLUS CHEMOTHERAPY VERSUS PELVIC AND PARA-AORTIC RADIATION FOR CERVICAL CANCER
Volume 340 Number 15 · 1141
Outcome
The median duration of follow-up was 43 months.
Follow-up data were available for 193 of the 195 pa-
tients in the combined-therapy group and for all 193
patients in the radiotherapy group. Of these, 147 pa-
tients in the combined-therapy group (76 percent)
and 122 patients in the radiotherapy group (63 per-
cent) were alive at the time of the last analysis. In ad-
dition, 13 patients in the combined-therapy group
and 32 patients in the radiotherapy group were alive
but had recurrent cervical cancer. Kaplan–Meier
analysis revealed that overall survival rates were sig-
nificantly better among patients treated with radio-
therapy and chemotherapy than among those treated
with radiotherapy alone (73 percent vs. 58 percent,
P=0.004) (Table 5 and Fig. 1). Disease-free survival
at five years was 67 percent in the combined-therapy
group and 40 percent in the radiotherapy group,
according to Kaplan–Meier analysis (P<0.001) (Fig.
2). The relative likelihood of disease-free survival in
*CI denotes confidence interval, and FIGO International Federation of Gynecology and Obstetrics.
†No follow-up data were available for two patients.
TABLE 5. ESTIMATED OVERALL SURVIVAL RATES AT FIVE YEARS,
ACCORDING TO STRATIFICATION VARIABLES.*
VARIABLE
RADIOTHERAPY AND CHEMOTHERAPY
(N=193)†
RADIOTHERAPY ALONE
(N=193)
P
VALUE
NO. OF
DEATHS/
TOTAL NO.
OF PATIENTS
5-YR
SURVIVAL
RATE
(95% CI)
NO. OF
DEATHS
/
TOTAL NO.
OF PATIENTS
5-YR
SURVIVAL
RATE
(95% CI)
percent percent
All patients 46/193 73 (65.7–80.3) 71/193 58 (49.8–66.2) 0.004
FIGO stage
IB, IIA, or IIB
III or IVA
26/135
20/58
77 (68.4–85.6)
63 (49.7–76.3)
46/134
25/59
58 (48.2–67.8)
57 (42.7–71.3)
0.002
0.44
Method of evaluating para-aortic
lymph nodes
Lymphangiography only
Lymph-node dissection alone or with
lymphangiography
34/143
12/50
74 (66.2–81.8)
70 (53.1–86.9)
50/141
21/52
60 (50.4–69.6)
54 (39.1–68.9)
0.02
0.07
Figure 1. Kaplan–Meier Estimates of Survival among Patients
Assigned to Receive Radiotherapy and Concurrent Chemother-
apy and Those Assigned to Receive Radiotherapy Alone.
Numbers in parentheses are the numbers of patients alive and
included in a follow-up assessment at three and five years.
0
100
060
10
20
30
40
50
60
70
80
90
12 24 36 48
Months
Radiotherapy alone
P=0.004
(78)
(33)
(44)
(95)
Radiotherapy
and chemotherapy
Survival (%)
Figure 2. Kaplan–Meier Estimates of Disease-free Survival
among Patients Assigned to Receive Radiotherapy and Concur-
rent Chemotherapy and Those Assigned to Receive Radiother-
apy Alone.
Numbers in parentheses are the numbers of patients at risk at
three and five years.
0
100
060
10
20
30
40
50
60
70
80
90
12 24 36 48
Months
Radiotherapy alone
P<0.001
(65)
(24)
(43)
(88)
Radiotherapy
and chemotherapy
Disease-free Survival (%)
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .

1142 · April 15, 1999
The New England Journal of Medicine
the combined-therapy group, as compared with the
radiotherapy group, was 0.48 (95 percent confi-
dence interval, 0.35 to 0.66). The rates of distant
relapse were 14 percent in the combined-therapy
group and 33 percent in the radiotherapy group
(P<0.001), with a relative risk of relapse of 0.39
(95 percent confidence interval, 0.24 to 0.63) in
the combined-therapy group. The rate of locore-
gional recurrences was 19 percent in the combined-
therapy group and 35 percent in the radiotherapy
group (P<0.001), with a relative risk of locore-
gional recurrences of 0.47 (95 percent confidence
interval, 0.31 to 0.71) in the combined-therapy
group.
The estimated five-year survival rates according to
various stratification variables are summarized in
Table 5. Approximately 70 percent of the patients
assigned to each group had stage IB, IIA, or IIB dis-
ease. For these patients, overall survival was signifi-
cantly better if they were treated with radiotherapy
and chemotherapy. There was no significant differ-
ence in overall survival between treatment groups
among patients who had stage III or IVA disease, al-
though the study was not designed to have a suffi-
cient number of patients in these subgroups to test
for a statistically significant difference. For patients
with stage III or IVA disease, the five-year disease-
free survival rates were 58 percent in the combined-
therapy group and 38 percent in the radiotherapy
group (P=0.13).
DISCUSSION
We found that the combination of pelvic radiation
and concomitant chemotherapy with cisplatin and
fluorouracil was more effective for locally advanced
cervical cancer than pelvic and para-aortic radiation
alone. The inclusion of chemotherapy substantially
reduced both local and distant recurrences of cervi-
cal cancer, leading to higher overall and disease-free
survival rates. Although chemotherapy increased the
hematologic toxicity, this effect was reversible and
the incidence of late side effects was similar in the
two treatment groups.
We are not the first to investigate the role of con-
comitant chemotherapy in the treatment of locally
advanced cervical cancer. The Gynecologic Oncolo-
gy Group has also studied the effect of radiotherapy
in combination with either hydroxyurea or placebo
in women with stage IIIB or IVA disease.
19
Although
the group reported significant improvements in over-
all and disease-free survival with the addition of hy-
droxyurea therapy, the study has been criticized for
the use of a low dose of radiation and the poor sur-
vival rates in both groups and because more than
half of the 190 patients who were enrolled could
not be evaluated. Despite these criticisms, the re-
sults of that study and those of trials assessing concur-
rent chemotherapy and radiotherapy for other tu-
mors stimulated studies of the effects of radiotherapy
and various combinations of fluorouracil, cisplatin,
mitomycin, carboplatin, and paclitaxel as treatments
for locally advanced cervical cancer.
8,9,20-23
The re-
sults of these trials have been encouraging, but most
clinicians have not found them sufficiently convinc-
ing to justify the inclusion of chemotherapy in the
routine treatment of locally advanced cervical cancer.
Our treatment regimen differed from previous
regimens in several ways. Our protocol emphasized
the importance of delivering a radiation dose of at
least 85 Gy to point A within eight weeks whenever
possible. As a result, the median dose was higher
and the median duration of treatment was shorter
than those reported in other studies. The results of
the Patterns of Care studies and large retrospective
analyses suggest that these features correlate with
survival rates and rates of local control of cervical
cancer.
24,25
We also used a somewhat more aggressive
regimen of chemotherapy than have other groups.
We used a higher dose of cisplatin (75 mg per square
meter) than that used with fluorouracil in the Gyne-
cologic Oncology Group studies, and we also in-
cluded a third cycle of chemotherapy during one of
the intracavitary procedures. The importance of this
third cycle is uncertain, but because close to 25 per-
cent of the total paracentral dose of radiation is de-
livered with each intracavitary procedure, the addi-
tion of concurrent chemotherapy during this time
may be important.
Reports by Rose et al.
26
and Keys et al.
27
in this
issue of the Journal
strengthen the body of evidence
supporting the use of combined therapy in women
with advanced cervical cancer. Future studies will
continue to evaluate cisplatin and fluorouracil as well
as other drugs to determine the most effective doses
and routes of administration. Weekly or daily infu-
sions of cisplatin are well tolerated.
28,29
For patients
who were receiving postoperative radiation for rectal
cancer, a prolonged, continuous infusion of fluoro-
uracil with pelvic radiation was found to be more ef-
fective than short infusions.
30
We do not know
whether the combination of fluorouracil and cisplat-
in is more effective than either drug alone.
The role of prophylactic extended-field radiation
has always been controversial. In our study, the ben-
efit of para-aortic radiation may have been less than
that observed in an earlier RTOG trial,
7
because our
trial required more rigorous staging of para-aortic
lymph nodes and included patients with more ad-
vanced pelvic disease. However, no radiographic test
is currently capable of detecting microscopic para-
aortic disease, and selected patients may still benefit
from prophylactic para-aortic radiation. For patients
with known metastases to the para-aortic lymph
nodes, para-aortic radiation is probably necessary.
Another RTOG study
31
tested the feasibility of com-
bining hyperfractionated extended-field radiation with
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .

PELVIC RADIATION PLUS CHEMOTHERAPY VERSUS PELVIC AND PARA-AORTIC RADIATION FOR CERVICAL CANCER
Volume 340 Number 15 · 1143
the same regimen of cisplatin and fluorouracil that
we used. The acute side effects were severe, possibly
because of the addition of chemotherapy or the al-
tered regimen of fractionation.
We believe there is now sufficient evidence to rec-
ommend that women with locally advanced cervical
cancer confined to the pelvis receive pelvic radiation
concomitantly with treatment with cisplatin and fluor-
ouracil. Future studies are needed to define the op-
timal regimen for these agents and to evaluate other
combinations. The role of extended-field radiation
with chemotherapy must also be defined.
Supported by grants (U10CA21661 and U10CA32115) from the Na-
tional Cancer Institute.
The views expressed in this article are solely those of the authors and do
not necessarily represent the official views of the National Cancer Institute.
REFERENCES
1. Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics, 1998.
CA Cancer J Clin 1998;48:6-29. [Erratum, CA Cancer J Clin 1998;48:
192.]
2. Eifel PJ, Morris M, Wharton JT, Oswald MJ. The influence of tumor
size and morphology on the outcome of patients with FIGO stage IB squa-
mous cell carcinoma of the uterine cervix. Int J Radiat Oncol Biol Phys
1994;29:9-16.
3. Eifel PJ, Morris M, Oswald MJ, Wharton JT, Delclos L. Adenocarcino-
ma of the uterine cervix: prognosis and patterns of failure in 367 cases.
Cancer 1990;65:2507-14.
4. Delgado G, Bundy BN, Fowler WC Jr, et al. A prospective surgical
pathological study of stage I squamous carcinoma of the cervix: a Gyneco-
logic Oncology Group study. Gynecol Oncol 1989;35:314-20.
5. Alvarez RD, Soong SJ, Kinney WK , et al. Identification of prognostic
factors and risk groups in patients found to have nodal metastasis at the
time of radical hysterectomy for early-stage squamous carcinoma of the cer-
vix. Gynecol Oncol 1989;35:130-5.
6. Stehman FB, Bundy BN, DiSaia PJ, Keys HM, Larson JE, Fowler WC.
Carcinoma of the cervix treated with radiation therapy. I. A multi-variate
analysis of prognostic variables in the Gynecologic Oncology Group. Can-
cer 1991;67:2776-85.
7. Rotman M, Pajak TF, Choi K, et al. Prophylactic extended-field irradi-
ation of para-aortic lymph nodes in stages IIB and bulky IB and IIA cer-
vical carcinomas: ten-year treatment results of RTOG 79-20. JAMA 1995;
274:387-93.
8. Thomas G, Dembo A, Fyles A, et al. Concurrent chemoradiation in ad-
vanced cervical cancer. Gynecol Oncol 1990;38:446-51.
9. Grigsby PW, Graham MV, Perez CA, Galakatos AE, Camel HM, Kao
MS. Prospective phase I/II studies of definitive irradiation and chemother-
apy for advanced gynecologic malignancies. Am J Clin Oncol 1996;19:1-6.
10. Stehman FB, Bundy BN, Thomas G, et al. Hydroxyurea versus mi-
sonidazole with radiation in cervical carcinoma: long-term follow-up of a
Gynecologic Oncology Group trial. J Clin Oncol 1993;11:1523-8.
11. Bauer M, Leibel S, Wasserman T, et al. Effect of misonidazole dose on
survival in patients with stage IIIB-IVA squamous cell carcinoma of the
uterine cervix: an RTOG randomized trial. Int J Radiat Oncol Biol Phys
1986;12:1101-3.
12. International Federation of Gynecology and Obstetrics. Staging an-
nouncement: FIGO staging of gynecologic cancers: cervical and vulva. Int
J Gynecol Cancer 1995;5:319.
13. Kaplan EL, Meier P. Nonparametric estimation from incomplete ob-
servations. J Am Stat Assoc 1958;53:457-81.
14. Mantel N. Evaluation of survival data and two new rank order statistics
arising in its consideration. Cancer Chemother Rep 1966;50:163-70.
15. Kalbfleisch JD, Prentice RL. The statistical analysis of failure time da-
ta. New York: John Wiley, 1980:167-9.
16. Gray RJ. A class of K-sample tests for comparing the cumulative inci-
dence of a competing risk. Ann Stat 1988;16:1141-54.
17. Cox DR. Regression models and life-tables. J R Stat Soc [B] 1972;34:
187-220.
18. Kim K , Tsiatis AA. Study duration for clinical trials with survival re-
sponse and early stopping rule. Biometrics 1990;46:81-92.
19. Hreshchyshyn MM, Aron BS, Boronow RC, Franklin EW III, Shin-
gleton HM, Blessing JA. Hydroxyurea or placebo combined with radiation
to treat stages IIIB and IV cervical cancer confined to the pelvis. Int J Ra-
diat Oncol Biol Phys 1979;5:317-22.
20. Thomas G, Dembo A, Beale F, et al. Concurrent radiation, mitomycin
C and 5-fluorouracil in poor prognosis carcinoma of cervix: preliminary
results of a phase I-II study. Int J Radiat Oncol Biol Phys 1984;10:1785-
90.
21. John M, Flam M, Caplan R , et al. Final results of a phase II chemo-
radiation protocol for locally advanced cervical cancer: RTOG 85-15. Gy-
necol Oncol 1996;61:221-6.
22. Corn BW, Micaily B, Dunton CJ, et al. Concomitant irradiation and
dose-escalating carboplatin for locally advanced carcinoma of the uterine
cervix: an updated report. Am J Clin Oncol 1998;21:31-5.
23. Chen MD, Paley PJ, Potish RA, Twiggs LB. Phase I trial of taxol as a
radiation sensitizer with cisplatin in advanced cervical cancer. Gynecol On-
col 1997;67:131-6.
24. Lanciano RM, Won M, Coia LR , Hanks GE. Pretreatment and treat-
ment factors associated with improved outcome in squamous cell carcino-
ma of the uterine cervix: a final report of the 1973 and 1978 Patterns of
Care studies. Int J Radiat Oncol Biol Phys 1991;20:667-76.
25. Fyles A, Keane TJ, Barton M, Simm J. The effect of treatment dura-
tion in the local control of cervix cancer. Radiother Oncol 1992;25:273-9.
26. Rose PG, Bundy BN, Watkins EB, et al. Concurrent cisplatin-based
radiotherapy and chemotherapy for locally advanced cervical cancer.
N Engl J Med 1999;340:1144-53.
27. Keys HM, Bundy BN, Stehman FB, et al. Cisplatin, radiation, and ad-
juvant hysterectomy compared with radiation and adjuvant hysterectomy
for bulky stage IB cervical carcinoma. N Engl J Med 1999;340:1154-61.
28. Malfetano J, Keys H, Kredentser D, Cunningham M, Kotlove D,
Weiss L. Weekly cisplatin and radical radiation therapy for advanced, recur-
rent, and poor prognosis cervical carcinoma. Cancer 1993;71:3703-6.
29. Fields AL, Anderson PS, Goldberg GL, et al. Mature results of a phase
II trial of concomitant cisplatin/pelvic radiotherapy for locally advanced
squamous cell carcinoma of the cervix. Gynecol Oncol 1996;61:416-22.
30. O’Connell MJ, Martenson JA, Wieand HS, et al. Improving adjuvant
therapy for rectal cancer by combining protracted-infusion fluorouracil
with radiation therapy after curative surgery. N Engl J Med 1994;331:502-
7.
31. Grigsby PW, Lu JD, Mutch DG, Kim RY, Eifel PJ. Twice-daily frac-
tionation of external irradiation with brachytherapy and chemotherapy in
carcinoma of the cervix with positive para-aortic lymph nodes: phase II
study of the Radiation Therapy Oncology Group 92-10. Int J Radiat On-
col Biol Phys 1998;41:817-22.
Copyright © 1999 Massachusetts Medical Society. All rights reserved.
Downloaded from www.nejm.org by MR KEVIN PATRICK on July 19, 2005 .