close

Вход

Забыли?

вход по аккаунту

?

906

код для вставкиСкачать
Significance of Surgical Margin on the Prognosis of
Patients with Ewing’s Sarcoma
A Report from the Cooperative Ewing ’s Sarcoma Study
Toshifumi Ozaki, M.D.’
Axel Hillmann, M.D.’
Christiane Hoffmann, M.D?
Christian Rube, M.D?
Sebastian Btasius, M . D . ~
Jiirgen Dunst, M.O?
Herbert Jiirgens, M.D?
Winfried Winkelmann, M.o.’
’
Department of Orthopaedics, Westfalische
Wilhelms-University, Munster, Germany.
Department of Pediatric Hematology and Oncology, Westfalische Wilhelms-University, Munster, Germany.
Department of Radiation Oncology, Westfalische Wilhelms-University, Munster, Germany.
Department of Pathology, Westfalische Wilhelms-University, Munster, Germany.
Department of Radiotherapy, Martin Luther University Halle-Wittenberg, Halle, Germany.
Presented in part at the Joint Meeting of the
EMSOS [European Musculo-Skeletal Oncology
Society) and AMSTS [American Musculo-Skeletal Tumor Society), Florence, Italy, May 8-9,
1995.
Supported in part by a grant from Deutsche
Krebshilfe (grant # M43/92/Ju2), EC BlOMEDl
(grant # BMHl-CT92-1341).
The authors thank Mrs. S. Jabar for the statistical analysis and her help in preparing the article.
Address for reprints: Winfried Winkelmann,
M.D., Department of Orthopaedics, Westfalische Wilhelms-University, Albert-SchweitzerStr. 33, 48129 Munster, Germany.
Received March 11,1996; accepted April 26,1996.
0 1996 American Cancer Society
BACKGROUND. There is little information reg,arding an adequate surgical margin
for local control of Ewing’s sarcoma.
METHODS. Two hundred and forty-four patients (PTS) with Ewing’s sarcoma who
were registered in the Cooperative Ewing’s Sarcoma Studies underwent surgical
treatment. Ninety-four PTS underwent definirive surgery (surgery alone), 131 PTS
received postoperative irradiation, and 19 PTS received preoperative irradiation.
The surgical margins were distributed as follows: radical, 29 PTS; wide, 148 PTS;
marginal, 39 PTS; and intralesional, 28 PTS. The impact of the surgical margin on
the treatment outcome of PTS was analyzed statistically.
RESULTS. The local or combined (local recurrence and systemic metastasis) relapse
rate after surgery with or without irradiation was significantly lower compared
with that after definitive irradiation (irradiation alone) (7% vs. 31%, P < 0.0001).
The local or combined relapse rate after complete resection (radical or wide margin) with or without irradiation was less compared with that after incomplete
resection (marginal or intralesional margin) with or without irradiation (5% vs.
12% P = 0.0455). The local or combined relapse rate did not greatly decreased
after irradiation after incomplete surgery (from 14% to 12%). In both groups of
good (viable tumor cells <lo%) and poor (viable cells 210%) histologic response,
the difference in systemic or combined relapse rate between patients undergoing
complete and incomplete surgery was not significant. The 10-year overall survival
of the PTS for each of the margins was distributed as follows: radical, 58%;wide,
65%; marginal, 61%; and intralesional, 71% (,P = not significant).
CONCLUSIONS. Surgery in patients with Ewing’s sarcoma adds to the safety of local
control. Under the current treatment regimen with intensive chemotherapy and
irradiation, complete resection of the tumor appears capable of decreasing the
risk of local recurrence. Cancer 1996; 78:892-900.0 1996American Cancer Society.
KEYWORDS bone, neoplasms, malignant, Ewing’s sarcoma, surgery, therapy, surgical margin.
E
wing’s sarcoma is a malignant bone tumor comprised of small
round cells. The expression of neural markers distinguishes conventional Ewing’s sarcoma from malignant primitive neuroectoderma1 tumor (PNET).’ The treatment outcome of Ewing’s sarcoma has
rapidly improved with the increasing availability of adjuvant theraWith the CESS 81 protocol, the first Cooperative Ewing’s Sarcoma Study (CESS) was initiated in 1981 by the German Society of
Pediatric Oncology and Hematology (GPOHI4and followed consecutively by other trials. To date, now, more than 1000 patients (PTS)
have been treated according to the CESS trials.
Although the impact of surgery for PTS with Ewing’s sarcoma
Surgical Margin of Ewing's Sarcoma lOzaki et al.
.
,
0
10
20
30
40
.
.
.
.
.
50
60
70
80
90
.
100 ($6)
FIGURE 1. Distribution of surgical margin. Three patients whose tumor
volume was not clear were excluded. Numbers in each of the values
indicate the number of patients.
is sometimes controversial, surgery seems to increase
the local control rate and survival rate of PTS with
Ewing's ~ a r c o r n a . ~ Because
' ~ - ' ~ of the higher local relapse rate after irradiation alone, surgery is regarded
as an important modality for local treatment in the
CESS trials. In recent years, the surgical procedures
have tended to be more conservative to acquire better
function with salvaged limbs compared with amputation and hemipelvectomy. A low local relapse rate was
reported even after surgery with a marginal or intralesional margin.2 However, an adequate surgical margin
to prevent local relapse has never been satisfactorily
addressed. Moreover, the effect of postoperative irradiation for the local control after surgery with incomplete resection remains unclear. The relationship between histologic response and optimal surgical margin
is also obscure. In this study, the influence of surgery
and surgical margins on the local control and overall
survival of PTS with Ewing's sarcoma were investigated. The significance of irradiation after surgery and
the correlation between treatment outcome and surgical margins according to the histologic response of
preoperative treatment were also analyzed.
PATIENTS AND METHODS
Characteristics and Chemotherapy
Protocol PTS registered for the consecutive CESS 81,
86, and 91P trials were investigated. Requirement for
classification as protocol PTS were: absence of visible
metastases at diagnosis, initiation of treatment within
3 weeks after biopsy, neither pretreatment nor primary
local treatment, and the enrollment occurring within
6 weeks after the start of therapy. The age for protocol
PTS was 25 years or younger in CESS 81 and 86 and
35 years or younger in CESS 91P. The chemotherapy
for the CESS 81 trial was comprised of the four-drug
regimen vincristine, actinomycin-D, cyclophosphamide, and doxorubicin (VACA).4The CESS 86P trial
893
was the pilot study of CESS 86. In the CESS 86 trial,
PTS with standard risk (small tumors in the proximal
or distal region) received the VACA regimen and those
with high risk (tumors in the central region or of large
volume) received the VAIA regimen, in which ifosfamide replaced cyclophosphamide. The CESS 86P and
CESS 86 trials were regarded as the same protocol. In
the CESS 91P trial, a pilot phase of the ongoing European Intergroup Cooperative Ewing's Sarcoma Study
(EICESS) 92 trial, standard risk PTS (same definition
as CESS 86) received the VACA regimen and high risk
PTS received either VAIA alone or with additional etoposide as a fifth agent (EVAIA) by randomization. Theses protocols have been published in detail elsewhere.*,5The regular period of chemotherapy prior to
definitive local treatment was 18 weeks in PTS treated
according to the CESS 81 trial, 9 weeks in CESS 86,
and 12 weeks in CESS 91P.
Of 255 protocol PTS without primary metastases
who underwent surgery, 244 PTS for whom information regarding the surgical margin was available were
reviewed for this study. Fifty-nine, 23, 126, and 36 PTS
were treated according to the CESS 81, 86P, 86, and
CESS 91P trials, respectively (Fig. 1).Their ages ranged
between 2 and 25 years, with a median age of 14 years.
The male to female ratio was 150 to 94 (1.6 to 1).
According to the histologic diagnosis, 214 PTS had Ewing's sarcoma and 30 had PNET. All belonged to Enneking's Stage IIB.I3 In this study, the local control
with surgical treatment according to each surgical
margin is emphasized. Although there are slight differences among the protocols, all PTS were analyzed together.
Ninety-five tumors originated in the central region: pelvis, 42 tumors; rib, 26 tumors; scapula, 17
tumors; clavicle, 6 tumors; spine, 2 tumors; sternum,
1 tumor; and skull, one tumor. Seventy-three tumors
originated in the proximal region: femur, 57 tumors;
and humerus, 16 tumors. Seventy-six tumors were located in the distal region: fibula, 33 tumors; tibia, 29
tumors; foot, 9 tumors; ulna, 4 tumors and radius, 1
tumor. Tumor volume was evaluated according to the
method described by Gobel et aL1*A tumor volume
of 100 mL or higher was defined as a large tumor (150
PTS), and that below 100 mL as a small tumor (91
PTS). In three PTS, the information about tumor size
was not available. The follow-up period ranged from
4.7 to 159.1 months (median, 61.1 months). In surviving PTS, the shortest follow-up time was 24.4 months.
Modalities of Local Treatment
Of these 244 PTS, 94 underwent surgery alone (definitive surgery). One hundred thirty one PTS underwent
894
CANCER August 15,1996 / Volume 78 / Number 4
TABLE 1
Local or Combined Failures According to Surgical Margin and Type of Local Treatment
Adequate
Radical
Wide
Inadequate
Marginal
Intralesional
Surgery (%I
Surgery t postop RT (%)
Preop RT t surgery (%)
Total (%I
0/27 (0)
2/53 (4)
0/2 (0)
-
6/80 (8)
0115 (0)
0129 (0)
81148 (5)
1/ 10 (10)
1/4 (25)
2/26 (8)
4/23 (17)
013 (0)
Oil (0)
3/39 (8)
5/28 (181
postop: postoperative; R T radiation therapy; preop: preoperative.
surgery with postoperative irradiation. Nineteen PTS
received preoperative irradiation followed by surgery.
Surgery
All surgical procedures were classified according to
Enneking’s criteria.I3 The information regarding the
surgical margin was reported to the CESS trial office by
the surgeons and pathologists who carefully examined
the resected specimen. Definitions were as follows:
radical: the whole tumor-bearing compartment is removed en bloc; wide: the tumor and its pseudocapsule
were removed en bloc surrounded by healthy tissue
within the tumor-bearing compartment; marginal: the
tumor was removed en bloc, but the line of resection
ran through the pseudocapsule or reactive area of the
tumor, and microscopic residual disease was likely;
and intralesional: the tumor was opened during surgery, the surgical field was contaminated, and microscopic or macroscopic residual disease remained.
Radical and wide margins were termed adequate and
marginal and intralesional margins were termed inadequate. Of 29 PTS who underwent surgery with a radical margin, 27 PTS underwent definitive surgery and
2 PTS received postoperative irradiation (Table 1). Of
148 PTS who underwent surgery with a wide margin,
53 PTS underwent definitive surgery, 80 PTS received
postoperative irradiation, and 15 PTS received preoperative irradiation. Of 39 PTS with a marginal margin,
10 PTS underwent definitive surgery, 26 PTS received
postoperative irradiation, and 3 PTS received preoperative irradiation. Of 28 PTS with an intralesional margin, 4 PTS underwent definitive surgery, 23 PTS received postoperative irradiation, and 1 patient received preoperative irradiation.
Irradiation
Doses of irradiation ranged between 22.4 and 76 Gray
(Gy) (median; 46 Gy). The recommended protocol
dose of postoperative irradiation was 36 Gy in CESS
81 and 44.8 Gy in CESS 86.Is In CESS 91P, preoperative
irradiation was recommended for PTS with a less than
50% reduction of evaluable soft tissue mass after 2
courses of chemotherapy. The dose of the preoperative
irradiation in 19 PTS ranged from 36 to 63.2 Gy (median, 51.8 Gy).
Histologic Response
After surgery, the surgical specimen was examined histologically and classified into 6 grades according to
the criteria of regression by Salzer-Kuntschik et a1.,16
Grade 1: no viable tumor cells detectable; Grade 2: 1
focus of the viable cells or a tumor island smaller than
0.5 cm remained; Grade 3: less than 10% of viable
tumor cells remained; Grade 4: 10-50% of the viable
tumor cells remained; Grade 5: greater than 50% of
the viable tumor cells remained; and Grade 6: no histologic response to chemotherapy. The breakpoint between good and poor response by definition was located between Grades 3 and 4. Hence, Grades 1, 2,
and 3 were termed good, and Grades 4, 5 , and 6 were
termed poor responses.
Relapses
Relapses were classified in three types: local (only local
recurrence), systemic (only metastases), and combined (local recurrence and systemic metastases), at
the time of first relapse.
Statistical Analysis
Statistical significance of the differences between proportions was evaluated by the chi-square test with
Fisher’s correction with the criterion of P < 0.05.’’
The cumulative probability of relapse free survival was
calculated by the Kaplan-Meier method.18 Univariate
log rank analysis was used to compare the c u r ~ e s . ’ ~
TREATMENT AND RESULTS
Treatment Modalities and Failures
First, the general impact of surgery on treatment outcome was evaluated, and thereafter the influence of
Surgical Margin of Ewing's Sarcoma /Ozaki et al.
RAD
895
Radical
n=29
n=lO2
0P
n=94
OP+RAD
n=l31
Local Failure
Marginal
Local Failure
Combined Failure
n=39
u
Combined Failure
R A D+ 0 P
n=l9
Systemic Failure
Remission
Systemic Failure
Remission
0
FIGURE 2. Relapse pattern according to the modalities of the treatment.
Relapses were classified as local, combined (local and systemic), and
systemic. Numbers in each of the values indicate the number of the patients.
the different surgical margins was analyzed. One hundred two PTS who received irradiation alone (definitive irradiation) were compared with 244 PTS who underwent surgery. Of the 102 PTS who received definitive irradiation as alocal treatment, 15 (15%), 17 (17%),
and 15 (15%) PTS developed local, combined, and systemic relapses, respectively (Fig. 2). Of 94 PTS who
underwent definitive surgery without additional irradiation, 3 (3%), 1 (l%),and 27 (29%) PTS developed
local, combined, and systemic relapses, respectively.
Of 1.31 PTS who received surgery and postoperative
irradiation, 7 (5%),5 (4%),and 35 (27%)PTS developed
local, combined and systemic failures, respectively. Of
19 PTS who received preoperative irradiation and surgery, 5 (26%) PTS developed systemic failures.
Impact of Surgery on Local Failure
The local relapse rate of PTS from the above mentioned 3 groups with surgery (10 of 244) was significantly lower compared with that of the PTS with definitive irradiation (15 of 102) (4%vs. 15%; P = 0.0011).
The combined relapse rate of PTS of the 3 groups including surgery (6 of 244) was also significantly lower
compared with that of the PTS with definitive irradiation (17 of 102) (2% vs. 17%; P < 0.0001). If local and
combined relapses were analyzed together with respect to local control, the local or combined relapse
rate of the 3 groups including surgery (16 of 244) was
significantly lower compared with that of the PTS with
definitive irradiation (32 of 102) (7% vs. 31%; P <
20
40
60
80
100 (%)
FIGURE 3. Relapse pattern according to the surgical margin. Relapses
were classified as local, combined (local and systemic), and systemic.
Numbers in each of the values indicate the number of the Patients.
diation (67 of 244) was significantly higher than that
of the group with definitive irradiation (15 of 102) (27%
vs. 15%;P = 0.0123). If systemic and combined failures
were analyzed together with respect to systemic metastases, the difference of systemic or combined failure
between the 3 groups including surgery (73 of 244;
30%) and PTS with definitive irradiation (31%)was not
significant.
Impact of Treatment Modalities on Overall Survival
The 10-year overall survival of the 244 PTS who underwent surgery with or without irradiation was significantly superior to that of the 102 PTS with definitive
irradiation (65% vs. 50%; P = 0.0088). The 10-year
overall survival of the PTS who underwent definitive
irradiation definitive surgery, postoperative irradiation, and preoperative irradiation were 50% 60%, 68%,
and 79%, respectively ( P = 0.0457). If the I0-year overall survival of 19 PTS who received preoperative irradiation was separated from that of the 225 PTS without
preoperative irradiation, 79% and 65% of the 10-year
survival were obtained ( P = not significant).
0.0001).
Factors Influencing the Surgical Margin
Tumors that originated in the central region were resected with inadequate surgical margins more often
than those in the proximal or distal region (44 of 95,
46% vs. 23 of 149, 15%; P < 0.0001) (Fig. 1). There
were no significant differences of the distribution of
the surgical margins (adequate or inadequate) between or among the variables of sex, tumor size, protocol, treatment with or without preoperative irradiation, and histologic response.
Impact of Surgery on Systemic Failure
Conversely, the systemic failure rate of the PTS of the
3 groups that underwent surgery with or without irra-
Surgical Margins and Relapse Pattern
Of 29 PTS who underwent surgery with a radical margin, no PTS developed local or combined failures;
896
CANCER August 15,1996 / Volume 78 I Number 4
“t
40
20
0
- ___
-_-
- Marginal (n=39)
Radical (n=29)
1
I
0
40
80
120
160
(mon.)
FIGURE 4. Overall survival according to the type of the surgical margin. The differences among the
surgical margins were not significant.
however, 11 PTS (38%) developed systemic failures
(Fig. 3). Of 148 PTS who underwent surgery with a
wide margin, 7 (5%), 1 (1%),and 38 (26%) PTS developed local, combined, and systemic relapses, respectively. Of 39 PTS who underwent surgery with a marginal margin, l (3%),2 (5%),and 11 (28%) PTS developed local, combined, and systemic relapses,
respectively. Of 28 PTS who underwent surgery with
, 7 (25%)
an intralesional margin, 2 (7%),3 ( l l % )and
PTS developed local, combined, and systemic relapses, respectively. The local or combined failure
rate had a tendency to increase as the resection line
got closer to the tumor. There was no significant
difference in the local relapse rate after surgery between adequate (radical and wide) (7 of 177; 4%)
and inadequate (marginal and intralesional) margins (3 of 67; 4%). However, the combined relapse
rate after surgery with adequate surgical margins
(1 of 177) was significantly lower compared with
that after surgery with inadequate surgical margins
(5/67) (1% vs. 7%; P = 0.0066). If local and combined relapses are analyzed together with respect
to local control, the local or combined relapse rate
after surgery with adequate surgical margins (8 of
177) was significantly lower compared with that
after surgery with inadequate margins (8/671 (5%
vs. 12%; P = 0.0455). There was no significant difference in the systemic failure rate between adequate
(49 of 177; 28%) and inadequate (18 of 67; 27%)
margins. Even if systemic or combined relapses
were analyzed together with respect of systemic
metastasis, the difference in systemic or combined
failure rate between adequate (50 of 177; 28%) and
inadequate margins (23 of 67; 34%) was not significant.
Surgical Margin and Overall Survival
If all 244 patients were analyzed together, the 10-year
overall survival of PTS with a radical margin was 58%
(12 of 29 failed); wide margin, 65% (41 of 148 failed);
marginal margin, 61% (14 of 39 failed); and intralesional margin, 71% (8 of 28 failed) (Fig. 4). There was
no significant difference among these four groups. The
10-year overall survivals of 225 PTS who did not receive preoperative irradiation were 58%with a radical
margin (12 of 29 failed), 66% with a wide margin (38
of 133 failed), 61% with a marginal margin (13 of 36
failed), and 70% with an intralesional margin (8 of 27
failed) (P-not significant). The 10-year overall survivals
of 19 PTS who received preoperative irradiation were
as follows; wide: 80% (3 of 15 failed); marginal: 67%
(1 of 3 failed); and intralesional: 100% (0 of 1 failed)
(P-not significant).
The survival rate after relapse of 16 PTS who developed local or combined relapses was poor (27%).
Analysis of Local or Combined Failures According to the
Surgical Margin and Type of Local Treatment
The frequency of local failures in each surgical margin
was analyzed according to the difference of the treatment modalities. In PTS who underwent surgery with
or without postoperative irradiation, local or combined relapse rates increased according to the transition of the surgical margins from radical to intrale-
Surgical Margin of Ewing’s Sarcoma /Ozaki et al.
TABLE 2
Local or Combined Failures according to Tumor Location
and Surgical Margin
Adequate
Central
Proximal
Distal
= 0.0077). Tumors that relapsed locally after surgery
with adequate margins were distributed as follows:
three in the rib, one in the clavicle, one in the pelvis,
one in the scapula, one in the femur, and one in the
fibula.
Inadequate
R (%)
W (%I
M (%)
IL(%)
Total (%)
010 (0)
017 (0)
0122 (01
6/51 (12)
1/56 (18)
1/41 (2)
1122 (51
118 (13)
119 (Ill
5/22 (23)
012 (01
014 (01
12/95 (13)
2/73 (31
2/76 (3)
H: radical; Cz‘: wide: M: marginal: I 1 inlralesional
TABLE 3
Failures ,4ccording to Histologic Response and Adequacy
of Surgical Margin
Good (1-3)
Adequate
Inadequate
Poor (4-6)
Adequate
lnadeauate
Local and
systemic (%)
No.
Local (%I
121
39
5 (4)
2 (5)
0 (0)
47
27
2 (4)
0 I01
897
Systemic (%I
Total (%I
4 (10)
26 (22)
6 1151
31 (26)
12 (31)
1 (2)
1 (41
22 (471
12 1441
25 (53)
13 1481
sional (Table 1). In PTS who underwent surgery with
adequate margins, the local or combined relapse rate
did not decrease after postoperative irradiation (6 of
82; 7%) compared with that after definitive surgery
(2 of 80; 3%). In PTS who underwent surgery with
inadequate margins, the local or combined relapse
rate decreased after postoperative irradiation (6 of 49;
12%)compared with that after definitive surgery, but
the difference was not significant (2 of 14; 14%). No
local relapses developed in PTS who underwent preoperative irradiation.
Analysis of Local or Combined Failures According to the
Surgical Margin and Tumor Location
The influence of surgical margin on local control may
change with the tumor location. The local or combined relapse rate of the tumors in the central region
(12 of 95) was higher than that of tumors in the proximal or distal region (4 of 149) (13%vs. 3%;P = 0.0031)
(Table 2). After surgery with inadequate surgical margins, the local or combined relapse rate of the tumors
in the central region was 14% (6 of 44) and that of the
tumors in the proximal or distal region was 9% (2 of
23) (P-not significant). After surgery with adequate
margins the local or combined relapse rate of the tumors in the central region was 12% (6 of 51) and that
in the proximal or distal region was 2% (2 of 126) (P
Histologic Response and Relapse Pattern
The local failure rate for PTS with a good histologic
response was 4% (7 of 160) and that for PTS with poor
response was 3% (2 of 74) (Table 3). The combined
failure rate for PTS with good response was 3% (4 of
160) and that for PTS with poor response was 3% (2
of 74). The local or combined failure rate for PTS with
good histologic response was 7% (11 of 160) and that
for PTS with poor response was 5% (4 of 74) (P-not
significant). In PTS with good response, the local or
combined relapse rate after surgery with inadequate
margins (6 of 39) was significantly higher than with
adequate margins (5 of 121) (15% vs. 4%; P = 0.0258).
In the poor response group, the local or combined
relapse rate after adequate surgical margins (3 of 47)
was not lower than that after surgery with inadequate
margins (1 of 27) (6% vs. 4%; P-not significant).
The systemic failure rate for PTS with good response was 20% (32 of 160) and that for PTS with poor
response was 46% (34 of 74) ( P < 0.0001). The systemic
or combined failure rate for PTS with good response
was 23% (36 of 160) and that for PTS with poor response was 49% (36 of 74) (P = 0.0001). In both groups
with good and poor response, the difference of systemic or combined relapse rate between adequate and
inadequate surgical margins was not significant.
DISCUSSION
The development of the surgical techniques and limbsparing surgery have made surgical treatment an attractive choice for local treatment of Ewing’s sarcoma.
In a large number of PTS, a good histologic response is
obtained after preoperative chemotherapy; however,
viable tumor cells often remain in the resected specim e n ~Thus,
. ~ PTS who underwent surgical excision of
the primary tumor had a better local control and survival than those who received definitive irradiation.’,6-10 In 1980, Marcove and Rosen12 reported the
excellent local control rate of PTS with Ewing’s sarcoma who underwent en bloc surgical resection (clear
margin) with postoperative irradiation. Although surgery is currently the most important modality of local
treatment for PTS with Ewing’s sarcoma, there is little
information about the treatment results for the different surgical margins in PTS with Ewing’s sarcoma.
Most of the local relapses are imputed to an inadequate excision that leaves viable tumor cells behind.
Conversely, local relapses do not necessarily occur
898
CANCER August 15, 1996 I Volume 78 I Number 4
after surgery with an inadequate margin.2 Under the
current treatment with intensive chemotherapy and
irradiation, a high local control rate may be expected
even after inadequate surgical margins. However, the
systemic failure rate may be increased after surgery
with an inadequate margin because of the possibility
of intravascular dissemination of the viable tumor cells
during surgery. However, this hypothesis has never
been proven. In this reported series, the combined or
systemic failure rate after irradiation is 31% compared
with 30% after surgery, 28% after surgery with radical
or wide (adequate) margins, and 34% after surgery
with marginal or intralesional (inadequate) margins.
The local or combined failure rate of PTS who
underwent surgery (7%) was significantly lower compared with that of PTS with definitive irradiation (31%)
(Fig. 2). PTS who received definitive irradiation might
include a considerable number of the PTS with inoperable lesions, in contrast to those who underwent surgery. In the surgically inaccessible tumors, the local
control rate naturally becomes poor. The high local or
combined failure rate in PTS who received definitive
irradiation would lead to the worst prognosis. Apart
from the inoperable lesions, definitive irradiation
would be the treatment of choice in PTS for whom
surgical resection of the tumor with adequate margins
is considered either to be difficult or believed to result
in severe functional deficit.15Even after chemotherapy
and 45-54 Gy preoperative irradiation, viable tumor
cells remained in approximately 50% of the PTS.20If
tumor is surgically accessible, surgery will add to the
safety of local control.
In general, tumor size and site are thought to affect the distribution of the acquired surgical margin.
The effect of preoperative treatment may also closely
influence the acquired surgical margins because PTS
who undergo surgical resection tend to show good
response to preoperative treatment.6 In this study, primary tumor site affected the acquired surgical margin;
inadequate surgical margins were obtained in the PTS
with tumor in the central region (46%) more often
than in the proximal or distal regions (15%) (Fig. 1).
Tumors in the central area are close to important organs. In the clavicle, rib, or spine, tumors are likely to
infiltrate into the neurovascular bundles, the vertebral
bodies, or the spinal cord, respectively. Because of the
close distance to essential organs in the pelvic area,
acquiring adequate margins becomes difficult. Due to
the three-dimensional anatomy of the pelvis," the
evaluation of the exact tumor extent is also complicating.
In a report from Mayo Clinic, PTS who underwent
complete excision of the tumor did not develop local
relapses and survival was significantlyimproved. How-
ever, the survival rate of PTS with debulking surgery
was not significantly different from that of PTS who
underwent no ~urgery.'~'~
Conversely, in the report
from the Intergroup Ewing's Sarcoma Study (IESS)
group, no significant differences were noted in local
failure rates in PTS with pelvic tumors treated with
either complete or incomplete resection followed by
irradiation." In PTS divided into groups based on each
surgical margin, local or combined relapse rates increased from radical to intralesional margins (Fig. 3).
The statistical difference of the local or combined relapse rate was noted between adequate (4%) and inadequate margins (12%).The PTS who developed local
relapses had a poor prognosis, so it is advisable to
choose adequate surgical margins to prevent local or
combined failures. Systemic metastasis does not seem
to be affected by the adequacy of surgical margins.
Although similar margins were acquired, the local
failure rate was different according to tumor location.
In this study, after surgery in central tumors (13%),
local relapses developed more often than in proximal
or distal tumors (3%)(Table I). This tendency was also
noted after surgery with adequate margins. Three of
six central tumors that relapsed locally after surgery
with adequate margins originated in the ribs. The evaluation of surgical margins in rib tumors is difficult
because of the pleural dissemination or implantation
of the tumor cells. The criteria of the surgical margin
used in this study would not be suitable for evaluation
of the surgical adequacy of the chest wall tumors.
If the tumor is violated or the surgical margin is
contaminated, it is generally believed that postoperative irradiation becomes necessary." In this series,
there was a tendency for the local or combined relapse
rate after intralesional surgery to decrease from 25%
to 17% with the addition of postoperative irradiation
(P-not significant) (Table 2). After marginal resection,
the local or combined relapse rate decreased from 10%
to 8% after postoperative irradiation. The effect of
postoperative irradial.ion for preventing local relapses
after surgery with an inadequate margin was not satisfactory, so acquiring adequate margins becomes more
important for the local control than additional postoperative irradiation afi er surgery with inadequate margins. In PTS who underwent surgery with a wide margin, the local or combined relapse rate did not decrease after postoperative irradiation. The tumor
locations of six PTS who had local relapse after surgery
with a wide margin and postoperative irradiation were
distributed as follow!;: three in the rib, one in the scapula, one in the clavicle, and one in the pelvis. This
result would also be affected by centrally located tumors, especially by the rib tumors as mentioned
above.
Surgical Margin of Ewing’s Sarcoma lOzaki et al.
PTS with a histologic good response tended to
have better relapse free survival than those with poor
respo~ise.~
This result is attributed to the low relapse
rates in PTS with a good histologic response (Table 3).
Even if good response was observed after preoperative
treatment, the local or combined relapse rate after
surgery with inadequate margins was significantly
higher than that after surgery with adequate margins.
In PTS with poor histologic response, a contradicted
trend between adequacy of the surgical margin and
the local or combined relapse was noted. This might
be because the number of PTS is fairly small and the
prognosis of the PTS with poor response was poor so
that survival was not long enough to develop local
relapses. From these results, at least, surgery with inadequate margins, even after good histologic response, does not become pertinent for local control of
Ewing’s sarcoma. Indeed the PTS with poor response
suffered systemic or combined relapses more often
than those with good response, but surgical margin
did not influence systemic or combined relapse rate
in either the good or poor response groups.
Although it was difficult to determine the relationship between surgical margin and overall survival, the
inadequate margins led to local and combined failure
more often than the adequate margins. The number
of PTS who developed local relapses is not large
enough to show a definitive influence on overall survival. But most PTS who developed local or combined
relapses had a poor prognosis compared with those
without relapses. From these results, there are no
proper reasons to justify an inadequate surgical margin. Although adequate margins should be the objective for good local control, inadequate margins sometimes cannot be avoided because the other options
would be amputation or procedures leading to significant restrictions of the extremital functions. Furthermore, in some PTS, acquiring an adequate margin
is difficult because of anatomic tumor locations. For
PTS who suffered local relapses, further limb-salvage
surgery becomes difficult and the risk of rapid systemic dissemination is high. If PTS undergo surgery
with inadequate margins, additional therapy to prevent local relapses should be considered. Options include amputation, additional wider resection, brachytherapy, and postoperative irradiati~n.’~
In recent
years, brachytherapy has been introduced to CESS trials, mf the margin is anticipated to be inadequate or
actually contaminated. No severe complications and
no local relapses have been reported in a series of 22
PTS after bra~hytherapy.~~
A prospective randomized
trial on soft tissue sarcoma also demonstrated improvement of the local control rate even though the
overall survival did not i m p r ~ v e . ’ ~Brachytherapy
~‘~
899
may be the current method of choice for local control
after an inadequate surgery because it can increase
the local control without additional surgical invasion.
High dose irradiation to the lower extremities led
to unacceptable functional morbidity, so primary amputation was enco~raged.’~
Because preoperative irradiation did not raise the complication rate of surgery
in Ewing’s sarcoma,28with the initiation of CESS 91P,
preoperative irradiation was incorporated in the treatment schedule to further sterilize the tumor-bearing
compartment before surgery. Resection of the involved bone after preoperative irradiation may not
only lessen the risk of postirradiation sarcoma,6’21
but
may also be useful for preventing postirradiative
pathologic fractures and local relapses. In 19 PTS who
received preoperative irradiation, none developed local relapses, even after limited surgery. The number
of PTS who received preoperative irradiation was still
quite small; nevertheless, good local control by the
preoperative irradiation could be expected from this
preliminary result.
For further improvement of the overall survival
in PTS with Ewing’s sarcoma, not only local but also
systemic failures must be controlled more effectively.
The elevation of the chemotherapeutic dose intensity
under the current supportive care to avoid side effects
and the selection of antitumor drugs for Ewing’s sarcoma are relatively limited. Myeloablative therapy
with high dose anticancer agents was reported to improve the prognosis of PTS with multifocal primary
disease or multiple relapse^.'^ However, this treatment
is currently not justified for all PTS with Ewing’s sarcoma without primary metastasis but is restricted to
high risk subgroups. The influence of cytokine regulation of Ewing’s sarcoma cells in vitro has recently been
dem~nstrated.~’
Although these results are still in
vitro, the response of Ewing’s sarcoma cells to cytokines might lead to a new modality of systemic treatment.
In conclusion, the local or combined relapse rate
after surgery with radical or wide margin was significantly lower compared with that after surgery with
marginal or intralesional margins. The local or combined relapse rate did not significantly decrease after
irradiation after inadequate surgery. The difference in
10-year overall survival of the PTS for each of the margins was also not significant; however, PTS who developed local or combined relapses had a very poor prognosis. Therefore, acquiring adequate surgical margins
appears to remain beneficial.
REFERENCES
1.
Roessner A, Jurgens H. Round cell tumors of bone. Path01
Res Pract 1993;189:1111-36.
900
2.
3.
4.
5.
6.
7.
8.
9.
10
11
12
13.
14.
15.
16.
17.
CANCER August 15,1996 / Volume 78 / Number 4
Bacci G, Toni A, Avella M, Manrfini M, Sudanese A, Ciaroni
D, et al. Long-term results in 144 localized Ewing’s sarcoma
patients treated with combined therapy. Cancer 1989;
63:1477-86.
Jurgens HF. Ewing’s sarcoma and peripheral primitive neuroectodermal tumor. Curr Opin Oncol 1994;6:391-6.
Jurgens H, Exner U, Gadner H, Harms D, Michaelis J, Sauer
R, et al. Multidisciplinary treatment of primary Ewing’s sarcoma of bone. A 6-year experience of European cooperative
trial. Cancer 1988;61:23-31.
Jurgens H, Treuner J, Miinkler K, Gobel U. Ifosfamide in
pediatric malignancies. Semin Oncol 1989; 16:46-50.
Sailer SL, Harmon DC, Mankin HJ, Truman IT, Suit HD.
Ewing’s sarcoma: surgical resection as a prognostic factor.
Int J Radiat Oncol Biol Phys 1988;15:43-52.
Wilkins RM, Pritchard DJ, Burgert EO Jr, Unni KK. Ewing’s
sarcoma of bone. Experience with 140 patients. Cancer
1986;5812551-5.
Rosen G, Caparros B, Nirenberg A, Marcove RC, Huvos AG,
Kosloff C, et al. Ewing’s sarcoma: ten-year experience with
adjuvant chemotherapy. Cancer 1980;47:2204- 13.
Burgert EO, Nesbit ME, Garnsey IA. Multimodal therapy for
the nonpelvic localized Ewing’s sarcoma of bone: intergroup
study IESS-11. J Clin Oncol 1990;8:1514-24.
Neff JR. Nonmetastatic Ewing’s sarcoma of bone: the role
of surgical therapy. Clin Orthop 1986;204:lll-7.
Pritchard DJ, Dahlin DC, Dauphine RT, Taylor WF, Beabout
JW. Ewing‘s sarcoma. A clinicopathological and statistical
analysis of patients surviving five years or longer. J Bone
Joint Surg Am 1975;57:10-6.
Marcove RC, Rosen G. Radical en bloc excision of Ewing’s
sarcoma. Clin Orthor, 1980;153:86-91.
Enneking WF, Spanler SS, Goodman MA. A system for the
surgical staging of musculoskeletal sarcoma. Clin Ortkop
1980; 153:106-21.
Gobel V, Jurgens H, Etspuler G, Kemperdick H, Jungblut RM,
Steinen U, et al. Prognostic significance of tumor volume in
localized Ewing’s sarcoma of bone in children and adolescents. J Cancer Res Clin Oncol 1987; 113:187-91.
Dunst J, Burgers JMV, Hawliczek R, Rurten R, Winkelmann
W, Salzer-Kuntschik M, et al. Radiation therapy as local
treatment in Ewing’s sarcoma. Cancer 1991;67:2818-25.
Salzer-Kuntschik M, Delling D, Beron G, Sigmund R. Morphological grades of regression on osteosarcoma after polychemotherapy: Study COSS 80. J Cancer Res Clin Oncol
1983;106121-4.
Fleiss JL. Statistical methods for rates and proportions. 2nd
edition. New York: John Wiley and Sons, 1981:14-22.
18. Kaplan EI, Meier P. Non-parametric estimation for incomplete observations. J A m Stat Assoc 1958;52457-81.
19. Pet0 R, Peto J. Asymptotically efficient rank in variant test
procedures. J R Stat Soc 1972; 135A185-98.
20. Jurgens H, Hoffmann CH, Braun-Munzinger G, Blasius S,
Salzer-Kuntschik M, Winkelmann W, et al. Histological response following preoperative chemoradiotherapy: the
CESS experience [abstract]. Joint Meeting EMSOS-AMSTS,
Florence, Italy, May 8-9, 1995:63.
21. Frassica FJ, Frassica DA, Pritchard DJ, Schomberg PL, Wold
LE, Sim FH. Ewing’s sarcoma of the pelvis. CLinicopathological features and treatment. J Bone Joint Surg A m 1993;
75~1457-65.
22. Pritchard DJ. Indications for surgical treatment of localized
Ewing’s sarcoma of bone. Clin Orthop 1980 153:39-43.
23. Evans RG, Nesbit ME, Gehan Eh, Garnsey LA, Burgert 0,
Vir:tti TJ, et al. Multimodal therapy for the management of
localized Ewing’s sarcoma of pelvic and sacral bones: a report from the second intergroup study. J Clin Oncol
19!3 1; 9: 1173-80.
24. Tanabe KK, Pollock RE, Ellis LM, Murphy A, Sherman N,
Romsdahl M. Influence of surgical margins on outcome in
patients with preoperatively irradiated extremity soft tissue
sarcomas. Cancer 1993;i3:1652-9.
25. Hillmann A, Rubee CH, Rod1 R, Lindner N, Hoffmann CH,
Schuck A, et al. Intraoperative brachytherapy of Ewing’ssarcoma [abstract] Joint Meeting EMSOS-AMSTS, Florence, Italy, May 8-9, 1995:94.
26. Brennan MF, Casper ES, Harrison LB, Shiu MH, Gaynor J,
Hajdu S. The role of multimodality therapy in soft-tissue
sarcoma. Ann Surg 1991;214:328-38.
27. Lewis RJ, Marcove RC, Rosen G. Ewing’s sarcoma-functioiial effects of radiatio’n therapy. J Bone Joint Surg A m
1977;59:325-31.
28. Hillmann A, Rube CH, Lindner N, Rod1 R, Hoffmann C, Jurgens H, et al. Does preoperative radiation therapy (RT) of
Ewing’s sarcoma increase the complication rate? A review
of the cases at the bone tumor center in Munster [abstract].
European Musculo-skeletal Oncology Society, 7th Meeting,
Amsterdam, October 6-7, 1994.
29. Burdach S, Jurgens H, Peters C, Nurnberger W, Mauz-Korholz C, Korholz D, et al. Myeloablative radiochemotherapy
and hematopoietic stem-cell rescue in poor-prognosis Ewing’s sarcoma. J Clin Oncol 1993;11:1482-8.
30. van Valen F, Winkelmann W, Burdach S, Gobel U, Jurgens H.
Interferon y and tumor necrosis factor cy induce a synergistic
antiproliferative response in human Ewing’s satrcoma cells
in vitro. J Cancer Res Clin Oncol 1993;119:615-21.
Документ
Категория
Без категории
Просмотров
6
Размер файла
917 Кб
Теги
906
1/--страниц
Пожаловаться на содержимое документа