The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

PREOPERATIVE RADIOTHERAPY COMBINED WITH

TOTAL MESORECTAL EXCISION FOR RESECTABLE RECTAL CANCER

ELLEN KAPITEIJN, M.D., CORRIE A.M. MARIJNEN, M.D., IRIS D. NAGTEGAAL, M.D., HEIN PUTTER, PH.D.,
WILLEM H. STEUP, M.D., PH.D., THEO WIGGERS, M.D., PH.D., HARM J.T. RUTTEN, M.D., PH.D.,
LARS PAHLMAN, M.D., PH.D., BENGT GLIMELIUS, M.D., PH.D., J. HAN J.M. VAN KRIEKEN, M.D., PH.D.,
JAN W.H. LEER, M.D., PH.D., AND CORNELIS J.H. VAN DE VELDE, M.D., PH.D.,
FOR THE DUTCH COLORECTAL CANCER GROUP*

ABSTRACT preoperative radiotherapy also improved the rate of

Background Short-term preoperative radiotherapy survival at five years.7 A recent meta-analysis8 conclud-
and total mesorectal excision have each been shown ed that the combination of preoperative radiotherapy
to improve local control of disease in patients with and surgery, as compared with surgery alone, signif-
resectable rectal cancer. We conducted a multicenter, icantly improved overall survival and cancer-specific
randomized trial to determine whether the addition survival.
of preoperative radiotherapy increases the benefit of The recognition that involvement of the circum-
total mesorectal excision. ferential margin by tumor cells is important in local
Methods We randomly assigned 1861 patients with recurrences has led to the general use of total mes-
resectable rectal cancer either to preoperative radio-
orectal excision,9-13 in which the entire mesorectum is
therapy (5 Gy on each of five days) followed by total
mesorectal excision (924 patients) or to total meso- enveloped and resected by precise, sharp dissection.
rectal excision alone (937 patients). The trial was con- Improvements in local control with this technique
ducted with the use of standardization and quality- have been shown, mainly in retrospective series.9-12,14
control measures to ensure the consistency of the In previous studies of radiotherapy for rectal cancer,
radiotherapy, surgery, and pathological techniques. surgery was not standardized. Since surgical technique
Results Of the 1861 patients randomly assigned is a key factor in the success of tumor control,15-17
to one of the two treatment groups, 1805 were eligi- standardization and quality control with respect to
ble to participate. The overall rate of survival at two surgery are indispensable for evaluating the effects of
years among the eligible patients was 82.0 percent adjuvant therapy. Optimal quality must also include
in the group assigned to both radiotherapy and sur- the use of standardized methods of pathological ex-
gery and 81.8 percent in the group assigned to sur-
gery alone (P=0.84). Among the 1748 patients who un-
amination.18 A prospective, randomized trial was or-
derwent a macroscopically complete local resection, ganized by the Dutch Colorectal Cancer Group to
the rate of local recurrence at two years was 5.3 per- investigate the efficacy of preoperative radiotherapy
cent. The rate of local recurrence at two years was in combination with standardized total mesorectal
2.4 percent in the radiotherapy-plus-surgery group excision in patients with rectal cancer.19 In this arti-
and 8.2 percent in the surgery-only group (P<0.001). cle, we present the results of the trial after a median
Conclusions Short-term preoperative radiotherapy follow-up of two years.
reduces the risk of local recurrence in patients with
rectal cancer who undergo a standardized total mes- METHODS
orectal excision. (N Engl J Med 2001;345:638-46.) Eligibility, Randomization, and Sample Size
Copyright 2001 Massachusetts Medical Society.
Patients were enrolled between January 1996 and December
1999. To be eligible, patients had to have histologically confirmed
adenocarcinoma of the rectum, without evidence of distant me-
tastases, and the inferior margin of the tumor had to be located
not farther than 15 cm from the anal verge and below the level

L
OCAL recurrence is a serious problem in of S12. Patients with fixed tumors or tumors that were treated
the treatment of rectal cancer, since it causes
disabling symptoms and is difficult to treat.1,2
There is a high incidence of local recur- From the Departments of Surgery (E.K., C.J.H.V.), Clinical Oncology
rence (15 to 45 percent) after conventional surgery, (C.A.M.M.), and Medical Statistics (H.P.), Leiden University Medical
Center, Leiden; the Departments of Pathology (I.D.N., J.H.J.M.K.) and
in which blunt dissection of the rectal fascia often Radiotherapy (J.W.H.L.), University Medical Center St. Radboud, Nijmegen;
fails to remove all the tissue that may bear tumor.3-5 the Department of Surgery, Leyenburg Hospital, The Hague (W.H.S.); the
Department of Surgery, University Hospital Groningen, Groningen (T.W.);
In an attempt to improve local control and sur- and the Department of Surgery, Catharina Hospital, Eindhoven (H.J.T.R.)
vival after conventional surgery, radiotherapy has been all in the Netherlands; and the Departments of Surgery (L.P.) and On-
given. The only randomized trial that compared pre- cology (B.G.), Akademiska Sjukhuset, Uppsala, Sweden. Address reprint
requests to Dr. van de Velde at the Department of Surgery K6-R, Leiden
operative and postoperative radiotherapy showed the University Medical Center, P.O. Box 9600, 2300 RC Leiden, the Nether-
superiority of preoperative radiotherapy for local con- lands, or at [email protected].
trol.6 The Swedish Rectal Cancer Trial found that *Other participating investigators are listed in the Appendix.

638 N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
PREOPERATIVE RADIOTHERAPY COMBINED WITH TOTAL MESOREC TAL EXCISION FOR RESEC TABLE REC TAL CANCER

by local (transanal) resection were excluded. Patients with previous on the basis of the number of eligible patients who had macroscop-
or coexisting cancer and those who had previously undergone large- ically complete local resection without distant metastasis. Analyses
bowel surgery, chemotherapy, or radiotherapy of the pelvis were of postoperative morbidity and mortality were based on the total
also excluded. number of eligible patients who underwent resection.
After informed consent had been obtained, we randomly as-
signed the patients to treatment with preoperative radiation (5 Gy RESULTS
on each of five days) followed by total mesorectal excision or to Patients
total mesorectal excision alone. Randomization was performed at
the central trial office and was based on permuted blocks of six, A total of 1861 patients were randomly assigned to
with stratification according to center and the expected type of one of the two treatment groups. There were 1530
operation (low anterior resection or abdominoperineal resection). patients from 84 Dutch hospitals, 228 from 13 Swed-
The trial was approved by the medical ethics committees of all the
participating hospitals. The trial design and the calculation of the ish hospitals, and 103 from 11 other European and
sample size have been described in detail elsewhere.19 Canadian centers. Of these 1861 patients, a total of
56 were found to be ineligible before randomization,
Follow-up including 4 patients for whom there was no infor-
Clinical evaluation every three months during the first year af- mation on eligibility. Our analysis therefore included
ter surgery and yearly thereafter for at least two more years was 1805 eligible patients. Of these, 1653 patients had a
mandatory and included yearly liver imaging and endoscopy. Local
recurrence was defined as evidence of a tumor within the lesser pel- curative resection. Of the remaining 152 patients, 57
vis or the perineal wound. Distant recurrence was defined as evi- did not undergo a macroscopically complete local re-
dence of a tumor in any other area. Recurrence at the colostomy section, and 95 were found to have distant metastasis
site or in the inguinal region was also classified as distant recurrence. at surgery (Table 1). The characteristics of the 1805
Quality Control patients who were eligible for the study and the fea-
In the Netherlands, participating surgeons attended workshops
tures of their tumors were similar in the two treat-
and symposiums, saw instructional videotapes, and were monitored ment groups (Table 2). In 28 patients (2 percent), no
by specially trained instructor surgeons. At each hospital, the first tumor was found in the resected specimen, despite a
five total mesorectal excisions were supervised by an instructor preoperative biopsy that showed an adenocarcinoma.
surgeon.19 Pathologists were trained to identify lateral spread of
tumor according to the protocol of Quirke et al.18 The results of Protocol Violations
histopathological examination of the specimens were reviewed by
a panel of supervising pathologists and a quality manager. 20 Pa- Patients with major or minor protocol violations,
tients eligibility and treatment and the details of follow-up were or both, were included in all the analyses.
checked by study coordinators. Local and distant recurrences were
Major Violations
confirmed radiologically or histologically and checked by a radi-
ation oncologist. Of the 897 eligible patients assigned to undergo
In Sweden, the technique of total mesorectal excision was in-
troduced on a national basis several years ago,12,13 as was the pro-
radiotherapy before total mesorectal excision, 29 did
tocol of Quirke et al.18 The European Organization for Research not receive preoperative radiotherapy for the follow-
and Treatment of Cancer participated in this trial under protocol ing reasons: known metastases (8 patients), carcinoma
40971. Visits to other participating hospitals and specialists were in situ (1), sigmoid carcinoma (3), a second cancer
made before the start of the trial to ensure the quality of treatment (1), withdrawal of informed consent (11), and phys-
at those sites. For logistic reasons, no quality control with respect
to radiotherapy, surgery, or pathological examination was performed ical limitations that made radiotherapy impossible (5).
outside the Netherlands during the trial. Long-term preoperative radiotherapy was given to sev-
en patients for locally advanced tumors. One patient
Statistical Analysis was unable to tolerate surgery and was treated with
Case-report forms were sent to the central trial office, where in- long-term radiotherapy alone. Preoperative radiother-
formation on the forms was entered into a data base and analyzed apy was discontinued in 14 patients, mainly because
with SPSS statistical software (version 9.0 for Windows, SPSS,
Chicago). Chi-square tests were used to compare proportions. of neurotoxicity.
MannWhitney tests were used to compare quantitative and or- Of the 908 eligible patients assigned to total mes-
dinal variables. Univariate analyses of survival were carried out by orectal excision alone, 3 patients withdrew their in-
the KaplanMeier method, and the evaluation of differences be- formed consent and requested radiotherapy (5 Gy on
tween the two groups was performed with the log-rank test. The
Cox proportional-hazards model was used to calculate hazard ra-
each of five days), and 8 patients had advanced local
tios and 95 percent confidence intervals in the univariate and mul- tumors for which long-term preoperative radiother-
tivariate analyses. A two-sided P value of 0.05 or less was considered apy was given.
to indicate statistical significance. Postoperative adjuvant therapy was not allowed in
The starting point for the analyses of survival and recurrence patients who had microscopically tumor-free margins
was the day of surgery. Data on patients who were alive or free of
recurrence were censored at the time of the last follow-up. The without spillage of tumor cells during the operation.
analysis of overall survival was performed on an intention-to-treat Of 1759 eligible patients with available information
basis and thus included all the eligible patients. The rate of local re- on margins and tumor spillage, 1351 (77 percent) had
currence was calculated on the basis of the number of eligible pa- tumor-free margins without tumor spillage. Eighty-
tients who underwent a macroscopically complete local resection.
The rate of distant recurrence was calculated on the basis of the
five of these patients (38 in the group assigned to ra-
number of eligible patients who did not have distant metastasis at diotherapy and surgery and 47 in the group assigned
the time of surgery. The overall rate of recurrence was calculated to surgery alone) received adjuvant therapy (chemo-

N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org 639

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

TABLE 1. CHARACTERISTICS OF THE ELIGIBLE AND INELIGIBLE PATIENTS AND RATES OF

MACROSCOPICALLY COMPLETE LOCAL RESECTION, ACCORDING TO TREATMENT GROUP.*

ALL
VARIABLE PATIENTS TREATMENT GROUP
RADIOTHERAPY SURGERY
PLUS SURGERY ALONE

no. (%)

Randomly assigned to treatment 1861 924 937

Ineligible for participation 56 27 29
No adenocarcinoma 8 5 3
Fixed tumor 2 0 2
Tumor treated by transanal resection 2 2 0
Tumor >15 cm from anal verge 5 4 1
Previous cancer 21 8 13
Coexisting cancer 11 4 7
Previous large-bowel surgery, pelvic radiotherapy, 3 2 1
or chemotherapy
No information on eligibility 4 2 2
Eligible for participation 1805 (97) 897 (97) 908 (97)
Incomplete local resection
Without distant metastases 31 10 21
With distant metastases 26 14 12
Complete local resection 1748 (94) 873 (94) 875 (93)
With distant metastases 95 47 48
Without distant metastases (curative) 1653 (89) 826 (89) 827 (88)

*Percentages are based on the total numbers of patients randomly assigned to one of the two treat-
ment groups.

therapy, radiotherapy, or chemoradiotherapy), which

TABLE 2. CHARACTERISTICS OF THE 1805 ELIGIBLE PATIENTS.* was a major protocol violation.
Minor Violations
RADIOTHERAPY

CHARACTERISTIC
PLUS SURGERY
(N=897)
SURGERY ALONE
(N=908)
P
VALUE
Of the 846 eligible patients randomly assigned to
preoperative radiotherapy who received the total dose
Age yr 0.79 of 25 Gy, the interval between the first day of radio-
Median 65 66
Range 2688 2392 therapy and the day of surgery exceeded 10 days in
Sex no. (%) 0.92 110 patients (13 percent). In 127 of the patients (15
Male 573 (64) 578 (64) percent), the upper border of the treatment field was
Female 324 (36) 330 (36)
Distance of tumor from 0.48 at the level of S12 instead of at the promontory,
anal verge no. (%) and in 161 of the patients undergoing an abdomino-
10.115 cm 267 (30) 280 (31)
5.110 cm 384 (43) 364 (40)
perineal resection (19 percent), the perineum was
5 cm 244 (27) 263 (29) not included in the treated volume.
Unknown 2 (<1) 1 (<1)
Type of resection no. (%) 0.12 Postoperative Morbidity and Mortality
None 16 (2) 29 (3)
Low anterior 579 (65) 604 (67) The median interval between randomization and
Abdominoperineal 251 (28) 234 (26) surgery was 21 days in the group assigned to radio-
Hartmann 50 (6) 40 (4)
Unknown 1 (<1) 1 (<1) therapy and surgery and 14 days in the group assigned
TNM stage no. (%) 0.53 to surgery alone. The patients assigned to radiothera-
0 11 (1) 17 (2)
I 265 (30) 244 (27)
py and surgery lost slightly more blood during the op-
II 252 (28) 245 (27) eration than those assigned to surgery alone (median
III 300 (33) 324 (36) loss, 1000 vs. 900 ml; P<0.001), and of the patients
IV 61 (7) 61 (7)
Unknown or no resection 8 (<1) 17 (2) who had an abdominoperineal resection, those as-
signed to radiotherapy had more perineal complica-
*Characteristics were unknown in some cases because not all case-report tions than those assigned to surgery alone (26 per-
forms were received. Because of rounding, not all percentages total 100.
TNM denotes tumornodemetastasis. cent vs. 18 percent, P=0.05). No other significant
A Hartmann resection is a low anterior resection without the construc- differences with respect to postoperative morbidity
tion of an anastomosis. and mortality were found between the two groups.

640 N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
PREOPERATIVE RADIOTHERAPY COMBINED WITH TOTAL MESOREC TAL EXCISION FOR RESEC TABLE REC TAL CANCER

Follow-up Overall Survival

As of February 2001, surviving eligible patients The rate of overall survival at two years was 82.0
without local recurrence had been followed for a me- percent in the group assigned to radiotherapy before
dian of 24.9 months (range, 1.1 to 56.0). Of these pa- surgery and 81.8 percent in the group assigned to
tients, 87 percent were followed for at least one year, surgery alone (P=0.84) (Fig. 1). The hazard ratio
54 percent for at least two years, 24 percent for at for death in the group assigned to surgery alone as
least three years, and 5 percent for at least four years. compared with the group assigned to preoperative
Rates of survival and recurrence are presented here radiotherapy was 1.02 (95 percent confidence inter-
at a follow-up of two years. A reanalysis as of June 1, val, 0.83 to 1.25).
2001, produced essentially the same results for all
Local Recurrence
the major end points of the study.
The rate of local recurrence at two years was 5.3
Events
percent in the population of 1748 patients who un-
As of February 2001, 365 (20 percent) of the derwent a macroscopically complete local resection.
1805 eligible patients had died. Of the 365 deaths, The rates of local recurrence at two years were 2.4
61 occurred postoperatively, 231 were related to rec- percent in the group assigned to radiotherapy before
tal cancer (growth of the primary tumor [in cases of surgery and 8.2 percent in the group assigned to sur-
macroscopically incomplete resection] or recurrence), gery alone (P<0.001) (Fig. 2). According to a univari-
and 70 were not related to rectal cancer. In three pa- ate analysis, the hazard ratio for local recurrence in the
tients, the cause of death was unknown. group assigned to surgery alone as compared with the
Local recurrence occurred in 87 patients. Of these group assigned to preoperative radiotherapy plus sur-
87 patients, 45 (52 percent) had local recurrence gery was 3.42 (95 percent confidence interval, 2.05
alone, 28 (32 percent) had both local and distant re- to 5.71).
currences, and 14 (16 percent) had local recurrence In the univariate analyses, treatment-group assign-
after distant metastasis was found at surgery (in 9 pa- ment (P<0.001), the location of the tumor (distance
tients) or during follow-up (in 5). A total of 227 pa- of the tumor from the anal verge) (P=0.003), and
tients were found to have only distant recurrence. the tumornodemetastasis (TNM) stage (P<0.001)

100

90
Overall Survival (%)

80

Surgery aloneH
(n=908)
70

RadiotherapyH
60 plus surgery (n=897)

50

0
0 1 2 3 4
Years
NO. AT RISK
RadiotherapyH 897H 741H 435H 192H 41H
plus surgeryH H H H H H
Surgery alone 908 744 454 207 42

Figure 1. Rates of Overall Survival in the Population of 1805 Eligible Patients, According to Treatment
Group.
At two years, the rate of overall survival was 82.0 percent in the group assigned to radiotherapy and
surgery and 81.8 percent in the group assigned to surgery alone (P=0.84).

N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org 641

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

30

Rate of Local Recurrence (%)

20

Surgery aloneH
(n=875)

10

RadiotherapyH
plus surgery (n=873)

0
0 1 2 3 4
Years
NO. AT RISK
RadiotherapyH 873H 691H 407H 170H 30H
plus surgeryH H H H H H
Surgery alone 875 688 406 173 37

Figure 2. Rates of Local Recurrence in the Population of 1748 Eligible Patients Who Underwent Mac-
roscopically Complete Local Resection, According to Treatment Group.
At two years, the rate of local recurrence was 2.4 percent in the group assigned to radiotherapy and
surgery and 8.2 percent in the group assigned to surgery alone (P<0.001).

were significant predictors of the risk of local recur-

rence. In the multivariate Cox regression analysis (Ta- TABLE 3. RESULTS OF MULTIVARIATE COX REGRESSION ANALYSIS
ble 3), the treatment-group assignment (P<0.001), OF LOCAL RECURRENCE AMONG THE 1748 ELIGIBLE PATIENTS
WITH A MACROSCOPICALLY COMPLETE LOCAL RESECTION.*
the tumor location (P=0.03), and the TNM stage
(P<0.001) were independent predictors of the risk
of local recurrence, whereas the type of resection VARIABLE HAZARD RATIO (95% CI) P VALUE
(P=0.90) had no independent prognostic value with Treatment group <0.001
respect to this end point. Radiotherapy and surgery 1.00
Univariate subgroup analyses showed that preop- Surgery alone 3.41 (2.055.70)
Distance of tumor from anal verge 0.03
erative radiotherapy reduced the risk of local recur- 10.115 cm 1.00
rence significantly in patients who had tumors with 5.110 cm 2.13 (1.134.01) 0.02
an inferior margin less than or equal to 5 cm (P= 5 cm 2.78 (1.226.31) 0.02
Type of resection 0.90
0.05) or 5.1 to 10 cm (P<0.001) from the anal verge Low anterior 1.00
(Table 4). Radiotherapy had no significant effect on Abdominoperineal 1.15 (0.592.24) 0.68
tumors located 10.1 to 15 cm from the anal verge Hartmann 1.16 (0.423.25) 0.78
TNM stage <0.001
(P=0.17). For TNM stage II and III tumors, pre- I 1.00
operative radiotherapy had a significant beneficial ef- II 3.44 (1.269.39) 0.02
III 9.69 (3.8924.2) <0.001
fect (P=0.01 and P<0.001, respectively), which was IV (distant metastases but complete 16.2 (5.4048.6) <0.001
not observed for TNM stage I and IV tumors (P= local resection)
0.15 and P=0.25, respectively). However, tests for
*A variable was included in the multivariate analysis if its P value in the
interaction among the tumor location, TNM stage, univariate analysis was less than 0.10. Patients with missing data were ex-
and treatment-group assignment in a multivariate cluded from the analysis of local recurrence. Twenty-eight patients without
analysis showed no significant interaction between a tumor (TNM stage 0) were excluded from the multivariate analysis be-
cause they were not at risk for local recurrence. CI denotes confidence in-
tumor location and treatment-group assignment (P= terval and TNM tumornodemetastasis.
0.08) or between the TNM stage and treatment- A Hartmann resection is a low anterior resection without the construc-
group assignment (P=0.61), suggesting that the treat- tion of an anastomosis.

642 N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
PREOPERATIVE RADIOTHERAPY COMBINED WITH TOTAL MESOREC TAL EXCISION FOR RESEC TABLE REC TAL CANCER

TABLE 4. RESULTS OF UNIVARIATE LOG-RANK ANALYSES OF TWO-YEAR RATES

OF LOCAL RECURRENCE AMONG THE 1748 ELIGIBLE PATIENTS
WITH A MACROSCOPICALLY COMPLETE LOCAL RESECTION,
ACCORDING TO SELECTED PROGNOSTIC VARIABLES.*

VARIABLE RADIOTHERAPY PLUS SURGERY SURGERY ALONE P VALUE

NO. OF LOCAL NO. OF LOCAL
PATIENTS RECURRENCE PATIENTS RECURRENCE
AT RISK AT 2 YR AT RISK AT 2 YR

% %

Overall 873 2.4 875 8.2 <0.001

Sex
Male 555 2.5 557 7.2 <0.001
Female 318 2.2 318 9.8 <0.001
Distance of tumor from anal verge
10.115 cm 262 1.3 271 3.8 0.17
5.110 cm 372 1.0 350 10.1 <0.001
5 cm 237 5.8 253 10.0 0.05
Type of resection
Low anterior 577 1.2 603 7.3 <0.001
Abdominoperineal 248 4.9 232 10.1 0.02
Hartmann 47 3.2 39 10.7 0.18
TNM stage
I 265 0.5 244 0.7 0.15
II 251 1.0 241 5.7 0.01
III 298 4.3 324 15.0 <0.001
IV (distant metastases but 47 10.1 48 23.8 0.25
complete local resection)

*Patients with missing data were excluded from the analysis of local recurrence. Twenty-eight pa-
tients without a tumor (TNM stage 0) were excluded from the multivariate analysis because they
were not at risk for local recurrence. In a Cox proportional-hazards analysis of age (as a continuous
variable), the hazard ratio for local recurrence at two years was 0.99 (95 percent confidence interval,
0.95 to 1.04; P=0.77) in the group of 873 patients assigned to radiotherapy and surgery and 1.01
(95 percent confidence interval, 0.99 to 1.04; P=0.21) in the group of 875 patients assigned to sur-
gery alone. TNM denotes tumornodemetastasis.
A Hartmann resection is a low anterior resection without the construction of an anastomosis.

ment effect did not differ among the subgroups an- DISCUSSION
alyzed (data not shown). In this trial, we evaluated the efficacy of short-term
Distant Recurrence
preoperative radiotherapy combined with standard-
ized total mesorectal excision in patients with resec-
The rate of distant recurrence at two years was table rectal cancer. We found that radiotherapy before
14.8 percent in the group assigned to radiotherapy total mesorectal excision can improve local control
and surgery and 16.8 percent in the group assigned of disease.
to surgery alone (P=0.87). The hazard ratio for dis- Reported rates of local control after surgery for
tant recurrence in the surgery-only group as com- rectal cancer vary widely. In studies of conventional,
pared with the radiotherapy-plus-surgery group was nonstandardized surgery, usually with a minimal fol-
1.02 (95 percent confidence interval, 0.80 to 1.30). low-up of five years, rates of local recurrence have
been 15 to 45 percent.3-5 By contrast, surgeons who
Overall Recurrence
specialize in total mesorectal excision report local-
The overall rate of recurrence (the rate of local re- recurrence rates of 7 percent or less.9-11 The low rate
currence and distant recurrence) at two years was of local recurrence in the group assigned to total
16.1 percent in the group assigned to radiotherapy mesorectal excision only in our study (8.2 percent at
and surgery and 20.9 percent in the group assigned two years) demonstrates that similar excellent results
to surgery alone (P=0.09). The hazard ratio for any can be achieved by other surgeons at multiple cen-
recurrence in the surgery-only group as compared ters after they are trained in the procedure.
with the radiotherapy-plus-surgery group was 1.21 We found that preoperative radiotherapy further
(95 percent confidence interval, 0.97 to 1.52). reduced the two-year rate of local recurrence from

N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org 643

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

8.2 percent to 2.4 percent, an indication of the value icantly decrease the risk of local recurrence of resecta-
of preoperative radiotherapy when used in conjunc- ble rectal cancer. This result was achieved in a large,
tion with standardized surgery. In the Swedish Rectal multicenter trial that included extensive instruction
Cancer Trial, the reduction in the rate of local recur- and quality control of the surgical technique. In this
rence at five years from 27 percent in the surgery- large group of patients who underwent standardized
only group to 11 percent in the radiotherapy-plus- surgery, short-term preoperative radiotherapy further
surgery group improved the rate of overall survival reduced the risk of local recurrence.
at this time point from 48 percent in the surgery-
only group to 58 percent in the combined-treatment Supported by grants from the Dutch Cancer Society (CKVO 95-04),
group.7 An effect of preoperative radiotherapy on the Dutch National Health Council (OWG 97/026), and the Swedish Can-
cer Society.
overall survival has not yet been detected in our trial,
probably because of the small number of local recur- APPENDIX
rences and the short follow-up. However, we believe
The following investigators participated in the study: the Netherlands
that a median follow-up time of 24.9 months is suf- Surgeons: A.B. Bijnen and P. de Ruiter, Medisch Centrum Alkmaar; B.
ficient to detect the effect of preoperative radiother- van Ooijen, Algemeen Christelijk Ziekenhuis Eemland, Amersfoort; D. van
apy on local recurrences, 55 to 80 percent of which Geldere and R.P.A. Boom, Ziekenhuis Amstelveen, Amstelveen; R.P. Bleich-
rodt and S. Meyer, Academisch Ziekenhuis Vrije Universiteit, Amsterdam;
occur during the first 2 years after surgery, with the R.M.J.M. Butzelaar, E.P. Steller, W.F. van Tets, and A.C.H. Boissevain, Sint
peak rate at 6 to 12 months.4,21,22 Lucas Andreas Ziekenhuis, Amsterdam; F.J. Sjardin, Bovenij Ziekenhuis,
The beneficial effect of preoperative radiotherapy Amsterdam; J.F.M. Slors, Academisch Medisch Centrum, Amsterdam;
W.H. Bouma and J.G.J. Roussel, Gelre Ziekenhuizen, Apeldoorn; J.H.G.
in our trial was observed for all tumor locations 15 cm Klinkenbijl and E.J. Spillenaar Bilgen, Ziekenhuis Rijnstate, Arnhem; P.M.
or less from the anal verge and for all TNM stages. Kruyt and W.K. de Roos, Stichting Ziekenhuisvoorzieningen Gelderse Val-
lei, Bennekom; E.J.R. Slingenberg and P.D. de Rooij, Sint Ziekenhuis
However, in a univariate subgroup analysis, the effect Lievensberg, Bergen op Zoom; M.A.J.M. Hunfeld, Rode Kruis Zieken-
was not significant in patients who had tumors with huis, Beverwijk; A.L.A. Meersman, Maasziekenhuis, Boxmeer; J.K.S.
an inferior margin more than 10 cm from the anal Nuytinck, Ignatius Ziekenhuis, Breda; R.M.P.H. Crolla, Ziekenhuis de
Baronie, Breda; J. van der Bijl, Atrium Brunssum, Brunssum, and Atrium
verge and in patients who had TNM stage I or IV Heerlen, Heerlen; G.W.M. Tetteroo, Ijsselland Ziekenhuis, Capelle aan de
tumors. Nevertheless, multivariate tests indicated that Ijssel; L.P.S. Stassen and P.W. de Graaf, Reinier de Graaf Groep, Delft;
the treatment effect probably did not differ among W.A.H. Gelderman and F.G.J. Willekens, Bosch Medicentrum, den Bosch;
I.P.T. van Bebber and E.J. Carol, Stichting Carolus-Liduina-Lindelust
subgroups defined according to tumor location, TNM Ziekenhuis, den Bosch; G.W. Kastelein and H. Boutkan, Stichting Juliana
stage, and treatment assignment. Therefore, consider- KinderziekenhuisRode Kruis Ziekenhuis, Den Haag; C. Ulrich and B.C.
de Vries, Medisch Centrum Haaglanden, Den Haag; H.J. Smeets and J.M.
ing the difficulties involved in predicting the location Heslinga, Stichting Bronovo-Nebo, Ziekenhuis Bronovo, Den Haag; P.V.M.
of tumors high above the anal verge and in determin- Pahlplatz, Ziekenhuis Leyenburg, Den Haag; P. Heres and J.A. van Oijen,
ing the TNM stage preoperatively, the decision not to Stichting het van Weel-Bethesda Ziekenhuis, Dirksland; M. van Hillo,
Stichting Talma Sionsberg, Dokkum; R.J. Oostenbroek and K.G. Tan, Al-
irradiate before surgery should be carefully considered. bert Schweitzer Ziekenhuis, Dordrecht; H.C.J. van der Mijle and R. Loo-
Preoperative radiotherapy does not result in down- ijen, Christelijk Ziekenhuis Nij Smellinghe, Drachten; J.J. Jakimowicz, Ca-
staging23 and is therefore not suitable for locally ad- tharina Ziekenhuis, Eindhoven; O.J. Repelaer van Driel and P.H.M.
Reemst, Diaconessenhuis Eindhoven; E.J.T. Luiten and R.F.T.A. Assmann,
vanced tumors. To avoid short-term irradiation of such Sint Annaziekenhuis, Geldrop; C.M. Dijkhuis, Oosterscheldeziekenhuis,
tumors, we advocate accurate preoperative imaging Goes; R.T. Ottow, Het Groene Hart Ziekenhuis, Gouda; J.T.M. Plukker,
Academisch Ziekenhuis Groningen; E.J. Boerma and R. Silvis, Kennemer
(for example, computed tomography or magnetic res- Gasthuis, Haarlem; J.H. Tomee, Stichting Streekziekenhuis Coevorden-
onance imaging). This lack of down-staging explains Hardenberg, Hardenberg; G.J.M. Akkersdijk, Spaarne Ziekenhuis, Heem-
why short-term preoperative radiotherapy has no ef- stede; C.G.B.M. Rupert, de Tjongerschans Ziekenhuis, Heerenveen;
G.J.C.M. Niessen and G. Verspui, Elkerliek Ziekenhuis, Helmond; J.H.
fect on sphincter preservation, which is often an end Kroesen and J.W. Juttmann, Ziekenhuis Hilversum, Hilversum; J.W.D. de
point in conventional trials of long-term radiotherapy. Waard and M.W.C. de Jonge, Westfries Gasthuis, Hoorn; D.B.W. de Roy
Concern has been expressed about the side effects van Zuidewijn and W. Dahmen, Medisch Centrum Leeuwarden; R. Vree,
J.A. Zonnevylle, Diaconessenhuis, Leiden; E. Klein Kranenbarg and
of hypofractionated radiation.24 In the Stockholm I tri- R.A.E.M. Tollenaar, Leids Universitair Medisch Centrum, Leiden; P.A.
al 25 and Imperial Cancer Research Fund trial,26 post- Neijenhuis, S.A. da Costa, and S.K. Adhin, Rijnland Ziekenhuis, Leider-
dorp; F.J. Idenburg, Medisch Centrum Haaglanden, Leidschendam; H.
operative mortality was higher among patients who van der Veen and C.E.A.M. Hoynck van Papendrecht, IJsselmeerzieken-
received radiotherapy than among those who did not. huizen, Lelystad; C.G.M.I. Baeten, M.F. von Meyenfeldt, and G.L. Beets,
In both trials, a suboptimal irradiation technique in- Academisch Ziekenhuis Maastricht; T. Wobbes, Academisch Ziekenhuis,
Nijmegen Sint Radboud, Nijmegen; E.D.M. Bruggink and L.J.A. Strobbe,
creased the treated volume considerably. In the Swed- Canisius-Wilhelmina Ziekenhuis, Nijmegen; O.J. van West and R.A.J.
ish Rectal Cancer Trial, postoperative mortality did Drr, Pasteurziekenhuis, Oosterhout; C.D. van Duyn, Ziekenhuis Bern-
not increase with radiation, provided that radiothera- hoven, Oss; J.W.M. Bol and T.A.A. van den Broek, Waterlandziekenhuis,
Purmerend; J.M.H. Debets and R.J.A. Estourgie, Laurentius Ziekenhuis,
py was optimal.27 In our trial, there was no difference Roermond; H.W.P.M. Kemperman, Ziekenhuis Franciscus, Roosendaal;
in in-hospital mortality between the two groups. In H.F. Veen, W.F. Weidema, and C.J. van Steensel, Ikazia Ziekenhuis, Rot-
terdam; F. Logeman and A.A.E.A. de Smet, Sint Clara Ziekenhuis, Rot-
the Swedish Rectal Cancer Trial, however, there was terdam; A.W.K.S. Marinelli, Daniel den Hoed Kliniek, Rotterdam; J.H.
more incontinence among patients who underwent Driebeek-van Dam, Havenziekenhuis, Rotterdam; W.R. Schouten and
preoperative irradiation and subsequently underwent P.P.L.O. Coene, Academisch Ziekenhuis Rotterdam, Dijkzigt; M.A. Paul,
Zuiderziekenhuis, Rotterdam; J.J. van Bruggen, Schieland Ziekenhuis,
a sphincter-preserving surgery.28 Schiedam; E.J. Mulder, Antonius Ziekenhuis, Sneek; R. den Toom and A.J.
In conclusion, total mesorectal excision can signif- van Beek, Ruwaard van Putten Ziekenhuis, Spijkenisse; S.J. Brenninkmeyer

644 N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
PREOPERATIVE RADIOTHERAPY COMBINED WITH TOTAL MESOREC TAL EXCISION FOR RESEC TABLE REC TAL CANCER

and G.P. Gerritsen, Tweesteden Ziekenhuis, Tilburg; H.J.M. Oostvogel and B. Stinner, Klinikum der Philips-Universitt, Germany; P. Delrio and
and J.A. Roukema, Sint Elisabeth Ziekenhuis, Tilburg; E.B.M. Theunissen, V. Parisi, Istituto Nazionale per lo Studio e la Cura dei Tumori, Italy; S.
Mesos, Utrecht; L.W.M. Janssen and A. Hennipman, Universitair Medisch Pucciarelli, Universita di Padova, Italy; J. Guimaraes dos Santos, Instituto
Centrum, Utrecht; A.J.M. van Wieringen, Mesos, Utrecht; A. Pronk and Portugues de Ontologica do Porto, Portugal; A. Nihlberg and O. Bendt-
P. Leguit, Diakonessenhuis, Utrecht; F.A.A.M. Croiset van Uchelen and sen, Falu Lasarett, Sweden; G. Lindmark, Helsingborgs Lasarett, Sweden;
R.M.H. Roumen, Sint Joseph Ziekenhuis, Veldhoven; C.L.H. van Berlo A. Trnqvist and T. Hallgren, Centralsjukhuset, Sweden; R. Sjdahl and
and J.F.M. Reinders, Sint Maartens Gasthuis, Venlo; C.D.G.W. Verheij, Sint O. Hallbook, University of Linkping, Sweden; M. Bohe and H. Jiborn,
Elisabeth Ziekenhuis, Venray; J.H. ten Thije, Ziekenhuis Walcheren, Vliss- Allmna Sjukhuset, Sweden; E. Nilsson, Lasarettet i Motala, Sweden; H.
ingen; W. van Overhagen and I.H. Oei, Reinier de Graaf Groep, Voorburg; Krook and G. Arbman, Landstinget i stergtland, Sweden; J. Rutegrd,
E.M.G. Leerkotte and J.W.A. van Luijt, Tweesteden Ziekenhuis, Waalwijk; rnskldsvik Hospital, Sweden; B. Sandzn, Ume University Hospital,
H.C.M. Verkooyen and J.A.L. Jansen, Sint Jans-Gasthuis, Weert; J. Merkx Sweden; W. Graf, Akademiska Sjukhuset, Sweden; K. Smedh, Centralhos-
and J.P. Vente, Hofpoort Ziekenhuis, Woerden; H. de Morree, Stichting pital, Sweden; K. Johansson, Vsterviks Sjukhus, Sweden; and R.J. Heald
Oosterscheldeziekenhuizen, Zierikzee; P.J.J. van Rijn, t Lange Land Ziek- and B.J. Moran, North Hampshire Hospital, United Kingdom.
enhuis, Zoetermeer; and W.F. Blom, Albert Schweitzer Ziekenhuis, Zwijn-
drecht; Pathologists: J.P.A. Baak, Medisch Centrum Alkmaar; H. Barrow- REFERENCES
clough, Algemeen Christelijk Ziekenhuis Eemland, Amersfoort; G.J.A.
Offerhaus, Academisch Medisch Centrum, Amsterdam; G. Brutel de la 1. Wiggers T, de Vries MR, Veeze-Kuypers B. Surgery for local recurrence
Riviere, M.L.F. van Velthuysen, Antoni van Leeuwenhoekziekenhuis, Am- of rectal carcinoma. Dis Colon Rectum 1996;39:323-8.
sterdam; B.A. van de Wiel, Sint Lucas Andreas Ziekenhuis, Amsterdam; 2. Holm T, Cedermark B, Rutqvist LE. Local recurrence of rectal adeno-
H.H. Oushoorn, Bovenij Ziekenhuis, Amsterdam; E. Bloemena, Vrije Uni- carcinoma after curative surgery with and without preoperative radiother-
versiteit, Amsterdam; T.A.J.M. Manschot, Gelre Ziekenhuizen, Apeldoorn; apy. Br J Surg 1994;81:452-5.
J.M. Wiersma-van Tilburg, Ziekenhuis Rijnstate, Arnhem; V. Potters, 3. Harnsberger JR, Vernava VM III, Longo WE. Radical abdominopelvic
Stichting Ziekenhuis Lievensberg, Bergen op Zoom; H.V. Stel, Ziekenhuis lymphadenectomy: historic perspective and current role in the surgical
Gooi-Noord, Blaricum; J. Los, Ignatius Ziekenhuis, Breda; G.W. Verdonk, management of rectal cancer. Dis Colon Rectum 1994;37:73-87.
Atrium Brunssum; C. van Krimpen, S.H. Sastrowijoto, and E.M. van der 4. Phillips RK, Hittinger R, Blesovsky L, Fry US, Fielding LP. Local re-
Loo, Stichting Diagnostisch Centrum Stichting Samenwerkende Delftse currence following curative surgery for large bowel cancer. I. The overall
Ziekenhuizen, Delft; H.A. Meijer, Bosch Medicentrum, den Bosch; P. picture. Br J Surg 1984;71:12-6.
Blok, Ziekenhuis Leyenburg, Den Haag; C.J. Tinga, Stichting Bronovo- 5. Kapiteijn E, Marijnen C, Colenbrander AC, et al. Local recurrence in
Nebo, Ziekenhuis Bronovo, Den Haag; E.C.M. Ooms, Medisch Centrum patients with rectal cancer, diagnosed between 1988 and 1992: a popula-
Haaglanden, Den Haag; C.M. Bruijn-van Duinen, Ziekenhuis Leyenburg, tion-based study in the west Netherlands. Eur J Surg Oncol 1998;24:528-
Den Haag; J.W. Steffelaar, Stichting Juliana KinderziekenhuisRode Kruis 35.
Ziekenhuis, Den Haag; P.J. Westenend, Pathologisch Laboratorium voor 6. Frykholm GJ, Glimelius B, Pahlman L. Preoperative or postoperative ir-
Dordrecht en Omstreken, Dordrecht; I.W.N. Tan-Go and H.M. Peters, radiation in adenocarcinoma of the rectum: final treatment results of a ran-
Stichting Pathologische Anatomie en Medische Microbiologie, Eindhoven; domized trial and an evaluation of late secondary effects. Dis Colon Rec-
E.J.M. Ahsmann, Stichting Laboratoria Goudse Ziekenhuizen, Gouda; J.F. tum 1993;36:564-72.
Keuning, Stichting Pathologisch Anatomisch Laboratorium Kennemer- 7. Swedish Rectal Cancer Trial. Improved survival with preoperative radio-
land, Haarlem; K. van Groningen, Spaarne Ziekenhuis, Heemstede; therapy in resectable rectal cancer. N Engl J Med 1997;336:980-7. [Erra-
P.H.M.H. Theunissen, Atrium Heerlen, Heerlen; F.J.J.M. van Merrien- tum, N Engl J Med 1997;336:1539.]
boer, Elkerliek Ziekenhuis, Helmond; G. Freling, Ziekenhuis Bethesda, 8. Camma C, Giunta M, Fiorica F, Pagliaro L, Craxi A, Cottone M. Pre-
Hoogeveen; A.J.K. Grond, Laboratorium voor de Volksgezondheid in operative radiotherapy for resectable rectal cancer: a meta-analysis. JAMA
Friesland, Leeuwarden; M.C.B. Gorsira, Diaconessenhuis, Leiden; J.J. Ca- 2000;284:1008-15.
lame, Rijnland Ziekenhuis, Leiderdorp; E.A. Neefjes-Borst, IJsselmeer- 9. MacFarlane JK, Ryall RD, Heald RJ. Mesorectal excision for rectal can-
ziekenhuizen, Lelystad; J.W. Arends, Academisch Ziekenhuis, Maastricht; cer. Lancet 1993;341:457-60.
A.P. Runsink, Streeklaboratorium Zeeland, Middelburg; C.A. Seldenrijk, 10. Enker WE, Thaler HT, Cranor ML, Polyak T. Total mesorectal exci-
Stichting Sint Antonius Ziekenhuis, Nieuwegein; M. Mravunac, Canisius- sion in the operative treatment of carcinoma of the rectum. J Am Coll Surg
Wilhelmina Ziekenhuis, Nijmegen; W.S. Kwee, Laurentius Ziekenhuis, Ro- 1995;181:335-46.
ermond; H. van Dekken, Daniel den Hoed Kliniek, Rotterdam; J.C. Ver- 11. Aitken RJ. Mesorectal excision for rectal cancer. Br J Surg 1996;83:
haar and N.A.L. van Kaam, Stichting Pathan, Rotterdam; H. van Dekken, 214-6.
Academisch Ziekenhuis Rotterdam, Dijkzigt; R.W.M. Giard, Sint Clara 12. Martling AL, Holm T, Rutqvist LE, Moran BJ, Heald RJ, Cedermark
Ziekenhuis, Rotterdam; H. Beerman, Zuiderziekenhuis, Rotterdam; B. Effect of a surgical training programme on outcome of rectal cancer in
A.A.M. van der Wurff, Sint Elisabeth Ziekenhuis, Tilburg; M.E.I. Schip- the County of Stockholm. Lancet 2000;356:93-6.
per, Universitair Medisch Centrum, Utrecht; H.M. Ruitenberg, Diakones- 13. Dahlberg M, Glimelius B, Pahlman L. Changing strategy for rectal
senhuis, Utrecht; R.F.M. Schapers, Stichting Pathologisch Laboratorium, cancer is associated with improved outcome. Br J Surg 1999;86:379-84.
Venlo; A.P. Willig, Sint Jans-Gasthuis, Weert; and A.G. Balk, Stichting 14. Havenga K, Enker WE, Norstein J, et al. Improved survival and local
Ziekenhuis De Heel, Zaandam; Radiotherapists: E.H.J.M. Rutten, Medisch control after total mesorectal excision or D3 lymphadenectomy in the treat-
Centrum Alkmaar; D. Gonzalez Gonzalez and G. van Tienhoven, Acade- ment of primary rectal cancer: an international analysis of 1411 patients.
misch Medisch Centrum, Amsterdam; B.J. Slotman and J.A. Langendijk, Eur J Surg Oncol 1999;25:368-74.
Academisch Ziekenhuis Vrije Universiteit, Amsterdam; G.M.M. Bartelink 15. Myerson RJ, Michalski JM, King ML, et al. Adjuvant radiation therapy
and B.M.P. Aleman, Antoni van Leeuwenhoekziekenhuis, Amsterdam; for rectal carcinoma: predictors of outcome. Int J Radiat Biol Phys 1995;
A.H. Westenberg, Arnhems Radiotherapeutisch Instituut, Arnhem; J. 32:41-50.
Pomp, Reinier de Graaf Gasthuis, Delft; C.C.E. Koning and R.G.J. 16. McArdle CS, Hole D. Impact of variability among surgeons on post-
Wiggenraad, Medisch Centrum Haaglanden, Den Haag; F.M. Gescher, operative morbidity and mortality and ultimate survival. BMJ 1991;302:
Ziekenhuis Leyenburg, Den Haag; J.J.F.M. Immerzeel and A.C.A. Mak, 1501-5.
Radiotherapeutisch Instituut Stedendriehoek en Omstreken, Deventer; 17. Hermanek P, Wiebelt H, Staimmer D, Riedl S. Prognostic factors of
J.G. Ribot and H. Martijn, Catharina Ziekenhuis, Eindhoven; D.F.M. de rectum carcinoma experience of the German Multicentre Study
Haas-Kock, Stichting Radiotherapeutisch Instituut Limburg, Heerlen; G. SGCRC. Tumori 1995;81:Suppl:60-4.
Botke and A. Slot, Radiotherapeutisch Instituut Friesland, Leeuwarden; 18. Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rec-
E.M. Noordijk, Leids Universitair Medisch Centrum, Leiden; P. Lambin, tal adenocarcinoma due to inadequate surgical resection: histopathological
Academisch Ziekenhuis Maastricht; J. Hoogenhout, Academisch Zieken- study of lateral tumour spread and surgical excision. Lancet 1986;2:996-9.
huis Nijmegen Sint Radboud, Nijmegen; P.C. Levendag and P.E.J. Hans- 19. Kapiteijn E, Kranenbarg EK, Steup WH, et al. Total mesorectal exci-
sens, Daniel den Hoed Kliniek, Rotterdam; G.S.J. Bunnik and K.A.J. de sion (TME) with or without preoperative radiotherapy in the treatment of
Winter, Dokter Bernard Verbeeten Instituut, Tilburg; J.J. Batterman and primary rectal cancer: prospective randomised trial with standard operative
H.K. Wijrdeman, Universitair Medisch Centrum, Utrecht; and J.M. Tabak and histopathological techniques. Eur J Surg 1999;165:410-20.
and M.F.H. Dielwart, Zeeuws Radiotherapeutisch Instituut, Vlissingen. 20. Nagtegaal ID, Kranenbarg EK, Hermans J, van de Velde CJ, van
Other Countries J.C. Pector, Institut Jules Bordet, Belgium; J. van de Krieken JH. Pathology data in the central databases of multicenter ran-
Stadt, Universit Libre de Bruxelles, Hospital Erasme, Belgium; P.T. Phang domized trials need to be based on pathology reports and controlled by
and J.K. MacFarlane, St. Pauls Hospital, Canada; P. Teniere, Hpital trained quality managers. J Clin Oncol 2000;18:1771-9.
Charles Nicolle, France; J.R. Delpero, Institut J. Paoli et I. Calmettes, 21. Carlsson U, Lasson A, Ekelund G. Recurrence rates after curative sur-
France; B. Sastre, Hpital Sainte-Marguerite, France; B. Nordlinger and C. gery for rectal carcinoma, with special reference to their accuracy. Dis Co-
Penna, Centre Hospitalier Universitaire Ambroise Pare, France; B. Gerdes lon Rectum 1987;30:431-4.

N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org 645

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.
 The Ne w E n g l a nd Jo u r n a l o f Me d ic i ne

22. Rao AR, Kagan AR, Chan PM, Gilbert HA, Nussbaum H, Hintz BL. term results of a randomised trial of short-course low-dose adjuvant pre-
Patterns of recurrence following curative resection alone for adenocarcino- operative radiotherapy for rectal cancer: reduction in local treatment fail-
ma of the rectum and sigmoid colon. Cancer 1981;48:1492-5. ure. Eur J Cancer 1994;30A:1602-6.
23. Marijnen CA, Nagtegaal ID, Kranenbarg EK, et al. No downstaging 27. Initial report from a Swedish multicentre study examining the role of
after short-term preoperative radiotherapy in rectal cancer patients. J Clin preoperative irradiation in the treatment of patients with resectable rectal
Oncol 2001;19:1976-84. carcinoma: Swedish Rectal Cancer Trial. Br J Surg 1993;80:1333-6.
24. Fletcher GH. Hypofractionation: lessons from complications. Radio- 28. Dahlberg M, Glimelius B, Graf W, Pahlman L. Preoperative irradiation
ther Oncol 1991;20:10-5. affects functional results after surgery for rectal cancer: results from a ran-
25. Cedermark B, Johansson H, Rutqvist LE, Wilking N. The Stockholm domized study. Dis Colon Rectum 1998;41:543-9.
I trial of preoperative short term radiotherapy in operable rectal carcinoma:
a prospective randomized trial. Cancer 1995;75:2269-75.
26. Goldberg PA, Nicholls RJ, Porter NH, Love S, Grimsey JE. Long- Copyright 2001 Massachusetts Medical Society.

646 N Engl J Med, Vol. 345, No. 9 August 30, 2001 www.nejm.org

The New England Journal of Medicine

Downloaded from nejm.org at Arizona Health Sciences Center and UMC on January 2, 2013. For personal use only. No other uses without permission.
Copyright 2001 Massachusetts Medical Society. All rights reserved.