close

Вход

Забыли?

вход по аккаунту

?

Clinicopathologic features of small pancreatic adenocarconoma A collective study

код для вставкиСкачать
968
Adjuvant Preoperative Radiotherapy in Patients with
Rectal Carcinoma
Adverse Effects during Long Term Follow-Up of Two Randomized Trials
Torbjorn Holm, M.D.'
Toom Singnomklao, M.s.*
Lars-Erik Rutqvist, M.D., Ph.D.2
Bjorn Cedermark, M.D., P m . '
' Department of Surgery, Karolinska Hospital,
Stockholm, Sweden.
' The
Oncology Center, Karolinska Hospital,
Stockholm, Sweden.
BACKGROUND. Adjuvant preoperative radiotherapy of patients with primary rectal
carcinoma improves local control and survival, but also may increase the risk of
early postoperative morbidity and mortality. I n addition, the possible late adverse
effects of this treatment are largely unknown.
METHODS. The present study was based on 1027 curatively operated patients included in 2 prospective randomized trials of preoperative radiotherapy for rectal
carcinoma patients (Stockholm I and Stockholm I1 Trials). The goal was to assess
whether long term intercurrent morbidity and mortality were increased in patients
allocated to the preoperative treatment. A computerized linkage of the randomized
patients to a population-based registry of the Stockholm County Council was used
to study hospital admissions for six groups of a priori defined diseases, putatively
related to late adverse effects of the radiation.
RESULTS. Preoperative radiotherapy significantly increased the incidence of venous
thromboeiubolism ( P = 0.01), femoral neck and pelvic fractures (P = 0.031, intestinal obstruction (P = 0.021, and postoperative fistulas ( P = 0.01). For arterial disease
and genitourinary tract diseases, no difference in risk was found between irradiated
and non irradiated patients. Radiotherapy significantly reduced rectal carcinoma
deaths in both trials and also improved overall survival in the Stockholm I1 trial.
The late intercurrent mortality was similar in irradiated and nonirradiated patients.
CONCLUSIONS. Although high dose, short term, preoperative radiotherapy can improve outcome after surgery for rectal carcinoma, there also may be an increased
risk for long term morbidity. Refinement of the radiotherapy technique and a
more accurate selection of patients suitable for the treatment will probably further
improve the results, at least in regard to treatment-related complications. Cancer
1996; 78:968-76. 8 1996 Afnericniz Cancer SocLeiy.
KEYWORDS: rectal carcinoma, preoperative radiotherapy, adverse effects, morbidity.
H
Supported by research grants from: The Cancer
Society in Stockholm, The Swedish Cancer Society and Jerzy och Eva Cederbaums Minervafond.
Address for reprints. Torbjorn Holm, M.D., Department of Surgery, Karolinska Hospital, S-171
76 Stockholm, Sweden.
Received February 5, 1996; revision received
May 1, 1996; accepted May 24, 1996.
6 1996 American Cancer Society
igh dose, short term, preoperative radiotherapy improves local
control'-' and survival"' in patients with primary rectal carcinorm. I n two randomized trials, the early postoperative mortality
rate was significantly increased in irradiated patients compared with
nonirradiated contro1s,2.3and in most such trials the early postoperative morbidity increased, mainly due to a higher incidence of wound
sepsis in the irradiated g r ~ u p . ' - ~ .Late
~ , ' effects of pelvic irradiation
in gynecologic and prostate carcinoma are described in standard textbooks"m9; however, the possible late adverse effects of preoperative
radiotherapy in rectal carcinoma are not known. Reported radiotherapy trials have used varying total doses, target volumes, fractionation,
and portal arrangements and the risk of late adverse effects may be
related to the radiotherapy technique used.
Preoperative Irradiation in Rectal Carcinoma/Holm et al.
The current study was based on 1027 curatively
operated patients with rectal carcinoma included in 2
randomizcd radiotherapy trials initiated by the Stockholm Rectal Cancer Study Group. 'I'he aim was to assess if high dose, short term, preoperative radiotherapy in rectal carcinoma causes increased intercurrent
morbidity and mortality during long term follow up.
PATIENTS AND METHODS
In 1980, the Stockholm Rectal Cancer Study Group
initiated a prospective, randomized trial on preoperative high dose, short term, radiotherapy in operable
rectal adenocarcinoma (Stockholm I Trial). Patients
living in the County of Stockholm, in the city of
Malmii, or on the Island of Gotland who had a biopsy
proven, primary resectable, rectal adenocarcinoma
were eligible. Patients were randomized to radiotherapy followed by surgery (RT+) or to surgery alone
( H I ' ). Those allocated to preoperative radiotherapy
received a total dose of 25 Gray (Gy) in 5 fractions
during 1 week and surgery was performed within 1
week thereafter. IJntil February 1987, the trial included
849 patients. During follow-up, a significant reduction
in the local recurrence rate was observed in the radiotherapy group, but there was also an increased postoperative mortality (within 30 days) in irradiated patients
(8%) compared with nonirradiated patients (2%).This
increased postoperative mortality was mainly the result of a higher risk of cardiovascular deaths in elderly
patients after radiotherapy. Details and results of the
study were published previously.".'"
Concurrently, another Swedish radiotherapy trial
compared pre- with postoperative radiotherapy in primary rectal carcinoma. The preoperative irradiation
regimen was similar to that in the Stockholm I Trial,
but the target volume was smaller and a three- or fourportal technique was used. N o significant increase in
postoperative mortality in irradiated patients was reported from this trial."
I n the light of these results, the Stockholm Rectal
Cancer Study Group initiated a new trial in March
1987 (Stockholm I1 Trial). In this study, only patients
living in the County of Stockholm and in the Island of
Gotland were included. The protocol was identical
with that of the first trial, except that a four-portal
technique was used, the target volume was reduced,
and patients aged older than 80 years were excluded.
The Stockholm 11 Trial closed in May 1993, when 557
patients had been included. I n a preliminary report
from this trial,'' the local recurrence rate was reduced
and the survival improved after preoperative radiotherapy, and there was no significant increase in postoperative mortality in irradiated patients compared
with nonirradiated patients (2% vs. 1%).
969
Details of the Radiotherapy Techniques
In both trials, the patients were treated in the supine
position with a fractionation dose of 5 Gy over 5 or 7
days (treatment interruption occurred during the
weekend). The Stockholm I Trial used a two-field technique to a large volume including the anus, rectum,
perirectal tissues, perineum, and regional lymph node
stations (including the inguinal lymph nodes, the obturator foramina, and the paravertebral lymph nodes)
up to the level of the second lumbar vertebra (Fig
l a ) ' 2 Cobalt-60 units or 8-12 megavolt (MV) linear
accelerators were used and the dose was prescribed
to the patient's central axis.
In the Stockholm I1 trial, a four-field box technique
was used and the treatment volume was reduced cranially (mid-IA) (Fig. Ib). The dose was prescribed to
the isocenter and 6-21 MV linear accelerators were
used.
Design of the Present Study
The Stockholm County Council registers basic demographic information on all county residents in a computerized register. The registration is based on the use
of an identification number that is unique to all persons living in Sweden. The data in this register are
supplemented with information regarding admissions
for inpatient care from nearly all hospitals in the region. Only one small, private hospital and a few long
term care centers do not report admissions, but none
of these institutions have an emergency unit and most
have a geriatric profile. It is estimated that approximately 95% of all hospital admissions in the county are
included in the County Council register.':' The register
includes information about time and duration of the
hospital stay, as well as the main discharge diagnosis
according to the International Classification of Iliseases (ICD) of the World Health Organization.
Of the 1406 patients included in the two Stockholm trials, 1241 were Stockholm County residents. To
avoid confounding from morbidity related to residual
rectal carcinoma only those 1027 patients who underwent curative surgery were included in the current
study. Surgery was defined as curative if the surgeon
reported a complete resection of the primary tumor
with regional lymph nodes and no distant metastases,
and the pathologist reported tumor free margins of
the specimen. Curative surgery was performed in 500
irradiated (RT+) and 527 nonirradiated ( I U - ) patients. In the irradiated group, 283 patients were irradiated with a 2-field technique and 217 with a 4-field
technique.
'The curatively operated patients were matched to
the County Council register and information was obtained on all hospital admissions for each individual
970
CANCER September 1,1996 / Volume 78 / Number 5
FIGURE 1. (a) Irradiated volume in the Stockholm I Trial, using a two-field technique. (b) Irradiated volume in the Stockholm II Trial, using
a four-field “box” technique.
patient from the date of randomization until January
1995.
All hospital admissions after the date of randomization were analyzed and the total number of admissions and number of days spent in the hospital were
recorded, irrespective of discharge diagnoses.
Late morbidity was analyzed only for diagnoses
related to groups of diseases that previously have been
reported as complications after radiotherapy to the
pelvis. The rationale for this was to diminish potential
problems related to “mass significance”. These a priori defined groups of diseases were: venous thromboembolism, diseases of the arteries, femoral neck and
pelvic fractures, intestinal obstruction, fistulas (including anal fistulas), and genitourinary tract diseases.
The analysis of hospital admissions was performed
using the three-digit code for specific diagnoses in
ICD-8, which was used until January 1987, and ICD-
Preoperative Irradiation in Rectal CarcinomalHolm et al.
971
TABLE 1
Analyzed Diagnoses in ICD 8 and ICD 9 with Corresponding Groups of Diseases
Analyzed diagnoses
Group of disease
ICD8
ICD9
Venous t h r o i n b o e i n b o l l s m
Arlerial disease
Femoral neck and pelvic fractrues
Inlestinal obstruction
Fistulas
Genitourinary tract disease
450,451,452,453
440, 443, 444, 445
808,820
560
565, 566. 599.03, 596.00, 629.80, 998.60
600, 601, 602, 612-629
415B, 451, 452,453
440,444, 557
808, 820
560
565, 566, 5968, 596C, 599B, 619, 99%
GOO, 601, 602, 614-629
IC1). International Classification of Diseases of the World Health Organization
-" I
E 8
Stockholm I RT+
0
Stockholm I RT-
#
P
.-...* ....-...* ..._
*____ *...*....(,
x....
6-
-
StockholmIRT~
.
Slockholm I RT.... .... Stockholm II RT+
-0-
.x"
Stockholm II RT+
Stockholm II RT-
YedI
At nsk
Stockholm1 RTi 283
RT- 289
203
226
158
168
123
125
Siu.kholmllRT4. 217
172
171
106
47
US
34
RT- 238
106
103
61
68
FIGURE 2. Median in-hospital stay (days) each year after randomization
according to allocated treatment. RT: radiotherapy.
9, which was used thereafter. The analyzed diagnoses
with corresponding groups of diseases are shown in
Table 1.
Only the first admission for any diagnosis within
a group of diseases was included. The rationale for
analyzing only the first hospital admission was the fact
that if a patient is transferred from one department to
another, this is registered as two separate admissions,
usually with the same discharge diagnosis.
Only hospital admissions before the first local or
distant rectal carcinoma recurrence were included and
all patient data were analyzed according to the allocated treatment, regardless of whether the patient actually received that treatment. Twenty-one patients
(13 irradiated and 8 nonirradiated) emigrated from the
county during the period studied. Data on these patients were censored at the date of emigration.
Because the follow-up time in the second Stockholm trial was shorter than in the first trial, the cumulative incidence at 5 years was also estimated for irradiated and nonirradiated patients in both trials for
comparative purposes.
Cause specific mortality was analyzed using the
officially recorded underlying causes of death, available from the Swedish National Central Bureau of Statistics.
Statistical Methods
Cumulative incidences of the first admission due to a
diagnosis within the studied groups of diseases were
calculated using actuarial methods. Comparisons of
distributions in irradiated and nonirradiated patients
were made using the log rank test.
RESULTS
The median follow-up times in the Stockholm I and
Stockholm I1 trials were 140 months (range, 95-178
months) and 62 months (range, 19-95 months) respectively.
In both trials, irradiated patients had a slightly
longer hospital stay during the first year after randomization, but after that period the median time spent
in hospital, due to any cause, was similar in irradiated
and nonirradiated patients (Fig 2).
972
CANCER September 1, 1996 / Volume 78 / Number 5
TABLE 2
Number of First Hospital Admissions, According to Allocated Treatment, in Each Group of Studied Diseases at 5 Years of Follow-Up in the 2
Stockholm Trials
Stockholm I trial
RT t
n 283
Venous thromboembolism
Arterial disease
Femoral neck and pelvic fractures
Intestinal obstruction
Postuperative fistulas
Stockholm I1 trial
RT t
RT -
RR
(95% CI)
n = 217
n = 238
RR
(95% CI)
1.5
0.8
2.5
(0.7-3.2)
(0.3-2.7)
(1-5.4)
(0.9-2.81
(0.8-4.1)
7
2
1
20
2
2.3
2
3.4
1.G
15
4
4
20
9
RT-
n
= 289
I1
6
5
20
8
16
5
13
32
14
1.8
(1-6.4)
(0.4-9.8)
(0.6-19.7)
(0.6-2)
(1.2-12.3)
1.1
3.8
RTt: radiotherapy followed b\ iarp,en'; RT-: w r p m alone; RR: relative risk CI: confidence i n t e n d
.........
....
...........
....
Slockholm I RT+
!
Slockholm I RTStockholm I1KTt
I
u$
Sla-kholm II RT-
0
I
2
4
......
42
_.=.............*........
-..Q-..
Stockholm I RT-
SIDckhoIm IIKT-
.......
.......9............ 0 ............ ............*............
-.
Ywri
5
I
2
3
RT+ 283
RT- 289
221
233
185
159
157
139
189
135
120
Slocwlohn RT+ 217
KT. 238
186
193
IS5
122
h5
141
103
92
74
5
Stockholm 1 RT+
----*---Stockholm II RT+
0
YCm
A1 nsk
l
k
-c
0
4
Alnsk
Stockholml RT+ 283
RT- 289
224
232
Sl~ckholmll RT+ 21J
RT- 238
187
192
190
165
190
159
155
I23
103
142
146
137
122
134
94
67
74
54
Slockholrnl
128
54
FIGURE 4. Cumulative incidence of first hospital admission due to arte-
FIGURE 5. Cumulative incidence of first hospital admission due to femo-
rial disease, according to allocated treatment in the two Stockholm trials.
RT: radiotherapy.
ral neck and pelvic fractures, according to allocated treatment in the two
Stockholm trials. RT: radiotherapy.
Venous Thromboembolism
The risk of venous thromboembolism was significantly
increased in the RT+ group during follow-up. Thirtyeight irradiated (7.5%)and 19 nonirradiated (3.6%)patients were admitted with this diagnosis, (relative risk
[RR] 2 [95%confidence interval (CI). 1.2-3.41; P = 0.01)
The increased risk was established during the first
4 months after surgery and after that period the curves
were parallel, indicating a similar risk in both groups
(Fig 3). When the cumulative incidence of venous
thromboembolism at 5 years after surgery was analyzed separately for the 2 trials, irradiated patients in
both trials had a higher estimated RR but the increase
was not statistically significant (Table 2).
of this event later than 5 years after randomization
revealed a first admission with arterial disease in 9 of
195 patients in the RT+ group compared with 2 of 172
patients in the RT- group ( P = 0.05).
Diseases of the Arteries
The cumulative incidence of first admissions related to
diseases of the arteries was not significantly different
between irradiated and nonirradiated patients during
follow-up in either of the trials. The cumulative incidence of arterial disease at 5 years is shown in Figure
4 and the relative risks in Table 2. However, an analysis
Femoral Neck or Pelvic Fractures
Twenty-seven irradiated (5.3%) and 13 nonirradiated
(2.4%) patients were admitted with a femoral neck or
pelvic fracture during follow-up. This difference was
statistically significant (RR 2 [95% CI, 1.1-3.71; P =
0.03). The increased cumulative incidence in the RT+
group was observed mainly during the first 3 years
after randomization but after that period the risk was
similar in both groups (Fig. 5). The estimated RR of
femoral neck or pelvic fractures at 5 years was increased in the RT+ group in both trials, but when the
trials were analyzed separately, the difference between
irradiated and nonirradiated patients was not statistically significant (Table 2).
Intestinal Obstruction
The risk of intestinal obstruction was significantly
higher in irradiated patients. Sixty-seven irradiated
Preoperative Irradiation in Rectal Carcinoma/Holm et al.
973
years was higher in irradiated patients in both trials,
but statistically significant only in the second trial.
74
.79
144
IIb
8Y
MI
!?3
95
hR
4Y
FIGURE 6. Curnulalive incidence of first hospital admission due to intestinal obstruction according to allocated treatment in the two Stockholm
trials RT radiotherapy
I0
1
__._.__..
_c
SiakholrnIRTt
Siakholrn IRT-
__..-
SlwMolm I1 RT+
....*.... Scakholm I1 RT-
............ ............
........... __._______._,,
~
YtAE
A,
I
”*
I
3
S~~rkholml RT+ 2W7
RT- 289
21Y
22R
IS5
187
IN)
SlockholmU RT+ 217
187
IYZ
l5U
I!I
142
in1
RT- 238
IS4
4
5
119
134
128
119
YI
74
(IS
54
FIGURE 7. Cumulative incidence of first hospital admission due to fistulas, according to allocated treatment in in the two Stockholm trials. RT:
radiotherapy.
(13.3%) and 45 nonirradiated (8.5‘rO)patients were admitted with this complication during the entire followup period (RH 1.6 195% CI, 1.1-2.21; P = 0.02). As
shown in Figure 6, the cumulative incidence was similar in both groups during the first 2 years after randomization, but after that period the incidence increased
in irradiated patients in the first Stockholm trial. In
the second Stockholm trial, the cumulative incidence
was similar in RT+ and RT- patients during 5 years
of follow-up. Table 2 presents the relative risk for intestinal obstruction separately for the two trials.
Fistulas
The risk of postoperative fistulas was significantly
higher after preoperative irradiation; 24 patients
(4.8%) in the RT+ group had this complication versus
10 patients (1.9%) in the RT-group (RH 2.3 [95% CI,
1.2-4.61; P = 0.01). The cumulative incidence curves
for the R T + and RT- groups in the two Stockholm
trials are shown in Figure 7. The estimated RR at 5
Diseases Related to the Genitourinary Tract
The number of first events related to diseases of the
bladder and urethra were similarly distributed irrespective of preoperative radiotherapy in both trials
and the cumulative incidence during follow-up was
similar (data not shown).
With regard to diagnoses related to bynecologic
disease, the cumulative incidence of these events were
also similar in irradiated and nonirradiatcd patients
in both trials (data not shown).
Long Term lntercurrent Mortality
The cause specific mortality according to allocated
treatment in the two Stockholm trials is shown in Table 3.
Preoperative radiotherapy significantly reduced
the incidence of death due to rectal carcinoma during
the entire follow-up period (1’ = 0.0008). This benefit
was observed in both trials. In the second Stockholm
trial, radiotherapy improved overall survival.
The incidence of cardiovascular deaths after radiotherapy was significantly increased in the first trial
(RR 4.4 195% C1, 2.1-9.51; I-’ = 0.0001) and in the second trial (RIi 3.9 195% CI, 1.1-1 1.91; I-’ = 0.02). This
increased risk in the irradiated group was present only
during the first year after randomization and mainly
during the first 3 months, postoperatively. Mortality
due to other intercurrent diseases (mainly infectious
complications) was increased after radiotherapy during the first year after randomization in the first trial,
but not in the second.
DISCUSSION
Adjuvant high dose, short term, preoperative radiotherapy has been shown to reduce the local recurrence
rate after surgery for rectal carcinoma in several randomized trials’-.”and results of recent studies demonstrate that the overall survival can he significantly improved in irradiated
However, preoperative irradiation in rectal carcinoma has also been shown to increase morbidity’ :’,‘ and mortality.3 in the early postoperative period. Information regarding the possible long term
intercurrent morbidity and mortality caused by this
treatment is scarce. TO our knowledge, no data on long
term, noncancer-related morbidity has been published from any of the major trials on adjuvant preoperative radiotherapy in rectal carcinoma.
In Sweden, the use of an identification number
that is unique to all persons living in the country
makes it possible to trace individuals and decreases
974
CANCER September 1,1996 / Volume 78 / Number 5
TABLE 3
Cause Specific Mortality According to Allocated Treatment in the Two Stockholm Trials
Stockholm I trial (No. of deaths
in allocated treatment groups)”
> 12 mo after
randomization
5 12 mo after
randomization
RTt
Stockholm I1 trial (No. of deaths
in allocated treatment groupslh
RT t
RT -
RT-
c 12 mo after
randomization
RT t
RT -
> 12 mo after
randomization
RT t
~~~
Cardiovascular disease
Other iimrcurrent disease
Rectal carcinoma
Other malignancies
Total
24
19
11
2
56
4
31
5
19
1
18
85
8
29
142
32
19
9
2
I
2
1
I
115
9
22
38
8
0
0
174
20
0
26
46
RT
-
~
5
2
58
2
67
RTT: iadiotlierapy followed by surgey; RT-: surgery alone.
” Stockholm 1 [rial: Irradiated patielits (RTTI n = 283; nonirradiatztl p a t i e m [RT-I n = 289.
“Stockholm II trial: Irradiated patients IRl’tl n = 217; nonirradiated patients (RT-)
11=
238.
the risk of losing patients to follow-up. The computerized register on all hospital admissions for people living in Stockholm County provides an opportunity for
a comprehensive long term follow-up of patients, although no data on morbidity not necessitating in-hospita1 care are available.
In this study, data on intercurrent morbidity was
obtained from the above mentioned register, which is
estimated to have a completeness of approximately
9570, only 2% of the data on patients were censored
because they moved from the county during the period studied.
It seems reasonable to assume that any misclassification of diagnoses in the register was unrelated to
the allocated treatment (i.e., preoperative radiotherapy). Such nondifferential misclassification may have
impaired the statistical precision but did not introduce
bias. A prolonged postoperative hospital stay for patients given preoperative radiotherapy has been reported previously.’ In our study, irradiated patients
were also hospitalized for a slightly longer period during the first year after surgery. However, overall inhospital care, in terms of total number of days spent
in the hospital and total number of hospital admissions was similar in irradiated and nonirradiated patients during the follow-up period. The preoperative
radiotherapy reduced the rate of local recurrence to
approximately 50%,3which implies that irradiated patients were in the hospital more often due to noncancer-related morbidity.
Venous thromboembolic complications after radiotherapy have been described in case reports,14but
it has not been a well established complication. Goldberg et aL2 reported an increased risk of postoperative
thromboenibolic complications within 30 days in pre-
operatively irradiated patients with rectal carcinoma.
In a previous study, we found an increased mortality
due to early postoperative cardiovascular complications in irradiated patients, especially in those with a
preoperative history of cardiovascular disease.” In the
current study, the risk of thromboembolic complications was increased during the 4 months after surgery
in irradiated patients in both Stockholm trials, irrespective of radiation field technique. Radiation causes
a diminished plasminogen activator activity in endothelial cells’6and it is possible that preoperative irradiation affects vascular endothelium, resulting in coagulation disturbances with an increased incidence of
thromboembolic complications. However, such a
mechanism remains tentative and warrants further
studies.
Arterial lesions as a result of radiation injury are
well known8 and are late complications. The chronic
changes in the arteries may be angiographically indistinguishable from atheroscler~sis.~~
The cumulative
incidence rates of first admissions with a diagnosis of
arterial disease that we found were not statistically
different between irradiated and nonirradiated patients during the follow-up period, but it may be that
the number of patients at risk after 5 years was not
sufficiently large to detect such a difference.
Fractures after therapeutic irradiation have been
described in the literature.8,’8,’’After irradiation, the
bone becomes hypocellular and osteoblastic damage
leads to decreased matrix formation and disturbance
of bone mineralization. Fracture can then occur
through the atrophic weak bone in weight-bearing
areas, such as the femoral neck and pelvis.*’ In a recent
report by Grisby et al.,“’the cumulative actuarial incidence of femoral neck fracture after groin irradiation
Preoperative Irradiation in Rectal Carcinorna/Holrn et al.
was 1 1 "/n and 15'%at 5 and 10 years, respectively. Pelvic
insufficiency fractures after radiotherapy to the pelvis
have also been described and may be difficult to diagIlose~:'l
..J1
Our study demonstrates an increased risk of
fractures. necessitating in-hospital care, in irradiated
patients during the first 3 years after radiotherapy.
After that time, the increase in cumulative incidence
was similar in irradiated and nonirradiated patients
and was probably attributable to the increasing age of
the patients. 130th Stockholm trials showed an increased risk of femoral neck and pelvic fractures after
radiotherapy, but the difference in risk between irradiated and nonirradiated patients was not statistically
significant in the second trial. The inclusion of older
patients in the first Stockholm trial probably explains
why the cumulative incidence of fractures in the control group was higher in the first than in the second
trial, and it is likely that radiotherapy increases the
absolute risk of fractures more in older patients with
osteoporosis.
Late radial ion intestinal injury after radiotherapy
to the abdomen and pelvis is also well described in the
literature'" and several studies report late morbidity in
the form of intestinal obstruction after postoperative
radiotherapy in patients with rectal carcinoma." I Iowever, the frequency of late small bowel toxicity after
preoperative irradiation in rectal carcinoma has not
been studied in detail. In a study from the Uppsala
group,"' it was reported that the frequency of small
bowel obstruction was 6%)after surgery alone, 5% after
preoperative radiotherapy plus surgery, and 1 1% after
surgery plus postoperative radiotherapy. These results
must be interpreted with caution however, because
the analysis was not based on patients randomly allocated to the three treatment groups; patients in the
surgery alone group were those allocated to postoperative radiotherapy who did not receive this treatment.
Our data show that the risk of bowel obstruction was
higher after preoperative radiotherapy plus surgery
compared with surgery alone in patients with rectal
carcinoma. Bowel obstruction induced by radiation
seems to he a relatively late event, because the risk in
irradiated patients increased aftcr 2 years. Its increased incidence in irradiated patients was present
in the Stockholm I trial but not in the Stockholm I1
trial, a difference that is likely to be explained by the
larger volume, including more small bowel, in the first
Stockholm trial.
The current study also showed an increased risk of
postoperat ive fistulas after preoperative radiotherapy.
This finding was present during the first 6 years after
therapy and was present in both Stockholm trials. The
complication has heen described previouslf,26 and is
probably due to an increased frequency of infections
975
and development of fibrosis after irradiation and surgery. The incidence of fistulas in the control group was
lower in the second trial than in the first trial. This is
difficult to explain hut may in part be due to changes
in surgical techniques with time, such as number of
abdominoperineal excisions, use of perineal drains,
and better control of infections. The mean age of the
patients was higher in the first trial, a fact that may
also have contributed to this observation.
With regard to urinary tract disease and gynecological disease, we found no differences in irradiated
and nonirradiated patients. This was somewhat unexpected because these organs were within the treatment fields. One reason could be that symptoms from
late complications induced by radiation in these organs are relatively mild and usually do not demand in
hospital care. Another reason could be that radiation
in this dose level does not significantly increase the
rate of these complications.
One problem with this study is that the follow-up
for the Stockholm I1 Trial is relatively short for detection of long term complications. The comparison between t h e two trials of the cumulative incidence of
complications at 5 years is an attempt to overcome
this problem, but it must be appreciated (hat the incidence of long term morbidity may be underestimated
in the second Stockholm trial. In addition, an analysis
of morbidity leading to in-hospital care probably underestimates the risk of late side effects from radiotherapy because less severe complications (e.g., pelvic
insufficiency fractures and irradiation cystitis) are
treated on an outpatient basis and arc thus not recorded in the County Council register.
The amount and nature of the radiotherapy-induced complications were dependent on the irradiated volume andlor on the treatment field technique,
with less pronounced differences between irradiated
and nonirradiated patients in the Stockholm [I trial,
which used four portals and a smaller treatment volume. Because individual dose planning was not performed, dose inhomogeneities could not be analyzed
for the individual patients. Ilowever, the risk that regions within the target receive an overdose is higher
in the two-field technique with lower energies than
with the four-field technique with higher energies. I n
addition. the single fraction dose of 5 Gy in these trials
is quite high and it is known that with higher doses
per fraction the risk of late toxicity increases." Another
fractionalion with lower doses per fraction may reduce
the risk of late complications.
The reduced risk of rectal carcinoma deaths and
the increased risk of early cardiovascular deaths and
infectious complications after radiotherapy has been
described in previous reports from the Stockholm Rec-
976
CANCER September 1,1996 / Volume 78 / Number 5
tal Cancer Study Group.".".'"." This study showed that
the late intercurrent mortality (i.e., mortality after 1
year) was similar irrespective of adjuvant treatment.
In conclusion, although high dose, short term, adjuvant preoperative radiotherapy in patients with rectal carcinoma improves local control and decreases
death from the disease, there is also an increased long
term postoperative morbidity. The treatment volume,
dose per fraction, and total dose probably determines
the severity of this morbidity. Refinement of the radiotherapy technique may be needed to further improve
the results of this treatment, at least in regard to late
treatment-related complications. Further studies on
different, patient-related. risk factors associated with
an increased incidence of late complications are also
desirable. Identification of such risk factors may enable the exclusion of preoperative radiotherapy in high
risk patients, a decision that of course must be related
to the stage of the disease and to the risk of rectal
carcinoma recurrence in each patient.
REFERENCES
1.
2.
Gerard A, Buyse M, Nordlinger €3. Loygue J, Pene F, Kempf
P, et al. Preoperative radiotherapy as adjuvant treatment in
rectal cancer. Final results of a randomized study of the
European Organization for Research and Treatment of Cancer (EORTC).Ar7n Surg 1988;208(5):606- 14.
Goldberg PA, Nicholls RJ, Porter N H , Love S, Grimsey JE.
Long-term results of a randomised trial of short-course lowdose adjuvant pre-operative radiotherapy for rectal cancer:
reduction in local treatment failure. Eur J Cancer 1994;
30A(11):1602-6.
3.
Stockholm Rectal Cancer Study Group. Preoperative shortterm radiation therapy in operable rectal carcinoma. Cancer
1990;66:49-55.
Swedish Rectal Cancer Trial. Local recurrence rate in a randomized multicentre trial of preoperative radiotherapy compared with operation alone in resectable rectal carcinoma.
E I L ~S1Ll
J .g 1996; 162~397-402.
5. Cedermark B, SRCS Group. Randomized study on preoperative radiation therapy in rectal carcinoma. Ann Surg Oncol
In press.
(5. Swedish Rectal Cancer Trial. Increased 5-years survival after
preoperative irradiation in rectal cancer (abstract).Int Colorectal Dis; 11(3):128.
7. Swedish Rectal Cancer Trial. Initial report from a Swedish
multicentre study examining the role of preoperative irradiation in the treatment of patients with resectable rectal carcinoma. Br 1 Surg 1993;80:1333-6.
8. Davies GC. Early and late complications of oncological treatment. In: Veronesi U, editor. Surgical oncology. A European
handbook. Berlin: Springer-Verlag, 1989;348-62.
9. van der Kogel A, Ang K. Complications related to radiotherapy. In: Peckham M, editor. Oxford textbook of oncology.
Oxford: Oxford University Press, 19952295-306.
10. Stockholm Rectal Cancer Study Group. Short-term radiotherapy for adenocarcinoma of the rectum. An interim anal-
ysis of a randomized multicenter trial. A m J Cliia Oncd
1987; 10:369-75.
11. Pihlman L, Glimelius B. Pre- or postoperative radiotherapy
in rectal and rectosigmoid carcinoma: report from a randomized multicentre trial. Airiz Surg 1990;211(2):187-95.
12. Kligerman M. Radiation therapy for rectal carcinoma. Semirz
O!ICO/1976;3(4):407- 13.
13. Fornander T. Long-term adjuvant tamoxifen in postmenopausal breast cancer. Adverse effects among patients included in a randomized clinical trial [thesisl. Stockholm,
Sweden: Karolinska Institute. 1990.
14. Schreiber DP, Kapp DS. Axillary-subclavian vein thrombosis
following combination chemotherapy and radiation therapy
in lymphoma. bit J Radiat OrrcoI Biol Phys 1986; 12(3):3915.
15. Holm T, Rutqvist L-E, Johansson H, Cedermark B. Postoper-
ative mortality in rectal cancer, treated with or without preoperative radiotherapy: causes and risk factors. Br J Sirrg
1996; 881964-8.
16. Astedt B, Bergentz SE, Svanberg L. Effect of irradiation on
the plasminogen activator content in rat vessels. E.xperientia
1974;30:1466-7.
17. Bergqvist D, Jonsson K, Nilsson M. Treatment of arterial
lesions after radiation therapy. Surg Gyizecol Obsret 1987;
165:116-20.
18. Grigsby PW, Roberts HL, Perez CA. Femoral neck fracture
following groin irradiation. 11zt J Radial Oncol Biol Phys
1995; 32( 1):63-7.
19. Parker R, Berry H. Late effects of therapeutic irradiation on
the skeleton and bone marrow. Cnracer 1976;37:1162-71.
20. Ergun H, Howland WJ. Postirradiation atrophy of mature
bone. CRC Crit Rev Diugir Imaging 1980; lan:225-43.
21. Blomlie V, Lien H, lversen T. Winderen M, Tvera K.Radiation
22.
4.
23.
24.
25.
26.
27.
induced insufficiency fractures of the sacrum: evaluation
with MR imaging. Radiology 1993; 188241-4.
Abe H, Nakamura M, Takahashi S, Maruoka S, Ogawa Y,
Sakamoto K. Radiation induced insufficiency fractures of
the pelvis: evaluation with """'Tc-metylene diphosphonate
scintigraphy. A m J Roeizrg 1992; 158:599-602.
Coia LR, Myersori RJ, Tepper JE. Late effects of radiation
therapy on the gastrointestinal tract. I n t J Radial Ot2col Biol
P h y ~1995; 3 1(5) :1213-36.
Mak AC, Rich TA, Schultheiss TE, Kavanagh B, Ota DM,
Romsdahl MM. Late complications of postoperative radiation therapy for cancer of the rectum and rectosigmoid. Iizt
J Radial Oncol Riol Phys 1994; 28(3):597-603.
Frykholm G, Glimelius B. Pihlman L. Preoperative or postoperative irradiation in adenocarcinoma of the rectum: final
treatment results of a randomised trial and an evaluation of
late secondary effects. Dis Colori Rectzma 1993;36(6):564-72.
Eifel P. Levenback C, Wharton J, Oswald M. Time course
and incidence of late complications in patients treated with
radiation therapy for FIG0 Stage IB carcinoma of the uterine
cervix. Int J Rndint Orrcol Biol Phys 1995;32(5):1289- 1300.
Fletcher GH. Hypofractionation: lessons from coniplications. Radiotker Orrcol 1991;20:10-5.
Документ
Категория
Без категории
Просмотров
8
Размер файла
787 Кб
Теги
features, stud, small, pancreaticum, adenocarconoma, clinicopathologic, collection
1/--страниц
Пожаловаться на содержимое документа