- Email: [email protected]
External beam and intraoperative electron irradiation for locally advanced soft tissue sarcomas
Inl J Radmrrun Oncology BIO/ Phys.. Vol. Printed in the U.S.A. All nghts reserved.
0360-3016/93 $6.00 + .OO Copyright (El 1993 Pergamon Press Ltd.
25, PP. 647-656
??Phase I/II Clinical Trials EXTERNAL BEAM AND INTRAOPERATIVE ELECTRON IRRADIATION FOR LOCALLY ADVANCED SOFT TISSUE SARCOMAS LEONARD L. GUNDERSON, M.D., M.S.,’ DAVID M. NAGORNEY, M.D.,’ DONALD C. MCILRATH, M.D.,* JENNIFER M. FIECK,~ HARRY S. WIEAND, PH.D.,~ ALVARO MARTINEZ, M.D.,’ DOUGLAS J. PRITCHARD, M.D.,4 FRANK SIM, M.D.,4 JAMES A. MARTENSON, M.D.,’ JOHN H. EDMONSON, M.D.5 AND JOHN H. DONOHUE, M.D.* Radiation Oncology’. Surgery’, Cancer Center Statistic$, Orthopedics4 and Medical Oncology’ Mayo Clinic and Mayo Medical School, Rochester, MN Purpose: Intraoperative irradiation with electrons was used in conjunction with external beam irradiation and maximal surgical resection in 20 patients with locally advanced soft tissue sarcomas or desmoids. This manuscript presents results with regard to tolerance of treatment and its impact on tumor control and survival. Methods and Materials: Ten patients presented with previously untreated primary sarcomas and 10 at the time of local recurrence (two had recurrent desmoid tumors). Tumor location was retroperitoneal in 19 and in the low anterior neck in one. A partial or gross total resection was performed prior to the external beam or intraoperative component of irradiation in every patient, but all had positive resection margins. Patients received 4500-6000 cGy of fractionated, external beam irradiation and an IORT dose of 1000-2000 cGy. Chemotherapy was given only at the time of disease progression. Results: Fourteen of 20 patients (70%) were alive; 11 (55%) were free of disease (4/10 primary, 7/10 recurrent), but 1 required hemipelvectomy for salvage. Progression within the intraoperative irradiation field was documented in only 1 patient (5%) and within the external beam field in 3/20 (15%). Blood born distant metastasis occurred in 5 patients (25%) and peritoneal seeding in 1 (5%). The distant failure incidence by grade was l/8 (13%) for Grades 1, 2 and 5/12 (42%) for Grades 3, 4. Only 1 patient (5%) developed a 2 severe neuropathy, and small bowel obstruction requiring exploration also occurred in a single patient. Conclusion: In view of acceptable tolerance and the high current rate of local tumor control, in spite of incomplete surgical resections, further evaluation of intraoperative irradiation as a component of treatment is indicated for locally advanced primary and recurrent soft tissue sarcomas. Intraoperative irradiation, Soft tissue sarcomas, Sarcoma irradiation, Locally advanced lesions.
INTRODUCTION
would be required to provide a reasonable probability of local control of tumor (28, 29). Many radiation oncologists, therefore, have recommended observation instead of adjuvant external irradiation since such dose levels ex-
Although surgery alone for retroperitoneal sarcomas is associated with a high incidence of local recurrence ( 1, 2, 15, 16, 21, 32, 34), the literature contains little data on the use of adjuvant irradiation as a component of treatment. Since retroperitoneal sarcomas often produce only minimal symptoms until they reach extensive size, surgical resections are often marginal, and microscopically involved resection margins are not infrequent. With the latter scenario, external irradiation doses of 6000-7000 cGy
ceed gastrointestinal tolerance (small intestine and stomach, *large bowel dependent on volume) (3, 5, 7, 8). In an attempt to decrease local recurrence and improve survival with an acceptable risk of complications, our institution has combined intraoperative irradiation (IORT), using an electron beam, with external beam irradiation and maximal surgical resection for locally advanced ret-
Presented in part at the American Society of Therapeutic Radiology and Oncology, New Orleans, LA, 9-14 October 1988.
the preparation of the manuscript, and to the radiation oncology
physicists and technologists, operating room nurses, and many
Reprint requests to: Leonard L. Gunderson, M.D., M.S., Chairman, 55905.
Radiation
Oncology,
Mayo Clinic, Rochester,
additional physicians who were involved in the treatment patients. Supported in part by NC1 Contract CM-27528. Accepted for publication 17 September 1992.
MN
Acknowledgements-The Chambers
authors appreciate the efforts of Julie and the Mayo Clinic Typing Service for assistance in 647
of the
648
I. J. Radiation Oncology 0 Biology 0 Physics
Volume 25, Number 4. 1993
roperitoneal and trunk sarcomas. The intent of this paper is to present our experience with the combined treatment approach in a group of 20 patients who presented with either primary or locally recurrent lesions.
METHODS
IORT was scheduled but not given in an additional 12 patients with retroperitoneal sarcomas during this time period. Three patients were excluded prior to exploration; radiographic evidence of locally extensive disease in 2 patients, and 1 patient was against exploration. At exploratory laparotomy, 5 patients were excluded because of disease factors; peritoneal seeding in 2 patients, and locally extensive disease that could neither be resected nor adequately treated with IORT in 3 patients. Total resection was accomplished in 4 additional patients. IORT was not given because of negative margins in 2 patients, 1 patient was against reoperation for IORT (microscopic residual), and 1 had myocardial infarction after resection and reoperation for IORT was not medically indicated. General criterion for patient selection and preoperative evaluation have been outlined previously (9, 11, 12). All patients have a staging work-up including history, physical examination, and diagnostic radiographs to determine the extent of local and regional disease and to exclude distant disease. Patients with distant metastases do not usually receive IORT in our institution, because expected life span
AND MATERIALS
Patient selection and work-up Twenty adult patients without previous irradiation for their locally advanced primary or recurrent soft tissue sarcoma of the retroperitoneum or trunk were treated with curative intent over a 6-year period with a regimen that included intraoperative irradiation (IORT) with electrons, have had minimum follow-up of 15 months, and form the basis of this report. One additional patient with a locally recurrent retroperitoneal sarcoma was excluded from all analyses because a full component of external irradiation could not be delivered due to previous external irradiation to this region after resection of a primary liposarcoma in the right hemiabdomen. One of the 20 patients included in this report had previously received 3000 cGy for a seminoma, necessitating restricted external beam fields during irradiation of a right paravertebral sarcoma. Patients who received IORT for extremity sarcomas are not included in this analysis.
would not be adequate to evaluate treatment related effectiveness or tolerance. In view of the high incidence of lung metastases in patients with high-grade soft tissue sar(CT) of the chest is roucomas, computed tomography tinely obtained at the time of initial evaluation in addition
Table 1. Tumor characteristics
Disease category
No. patients
Primary, resected-residual Primary, unresectable for cure (disease fixation) Recurrent, resected-residual Recurrent, unresectable for cure Total
7 3* 7 3* 20
Histological types and grade
Gr 112
Gr 314
Chondrosarcoma Hemangiopericytoma Leiomyosarcoma Liposarcoma Malignant fibrous histiocytoma Neurofibrosarcoma Synovial cell sarcoma Sarcoma-not otherwise specified Total soft tissue sarcomas Desmoid (recurrent)
3 1 0 1 0 1 -
0 0 4 1 2 3 1
0 6
1 12
Maximum tumor dimension before resection (cm) o-5 5.1-10 10.1-15 > 15 Total * All six had resection for a seminoma.
after 4500-5000
Primary 4 4 2 0 10 cGy. In 1 of 6, external fields were restricted,
Recurrent 4 3 3 0 10
Total 3 1
4 2 2 4 1 1 18 2 Total 8 7 5 0 20
since 3000 cGy had been previously
delivered
Intraoperative and external RT, sarcoma 0 L. L.
to a CT and/or magnetic resonance imaging (MRI) of the site of the primary or recurrent lesion. Tumor characteristics are shown in Table 1. At time of presentation for irradiation, 10 patients had previously untreated primary lesions and 10 had localized recurrences ( 14 presented after an attempt at surgical resection and six before resection). All histological grades and tumor types of soft tissue sarcomas were eligible and recurrent desmoids were also included (Table 1). Histological Grade of 1 to 4 was assessed by Broder’s classification. Maximum tumor dimension was obtained from the preoperative CT scans and resected specimens. Treatment factors External irradiation. External beam irradiation methods have been fairly consistent in this group of patients. Doses of 4500-6040 cGy were delivered in 180 cGy fractions, 5 days per week over 5-61 weeks with linear accelerators using 10 MV photons ( 18 of 20 received external
irradiation
Completed planned external irradiation Method of irradiation Postop* (4500-6040 cGy/25-33 fractions) Low dose preop + postopt Preop (4500-5200 cGy/26-29 fractions) IORT energy and dose Electron energy (MeV) 9 12 15 18 IORT dose (cGy) 1000-1250 1500 1750 2000 Total fields/patients+ Location
factors No.
W)
20*
(100)
13
(65) (5)
1 6 Primary
649
et al.
doses I 5580 cGy). For pelvic lesions, treatments were given with 4 field shaped external beam techniques similar to those used for locally advanced colorectal cancers (6, 7, 10). With extrapelvic lesions, unresected or residual disease plus 5 cm margins was included to 4500 cGy with parallel opposed fields; reduced fields with 2-3 cm margins were treated to 5000 to 5500 cGy. As noted in Table 2, 20 of 20 patients (100%) completed all planned external irradiation. Five of twenty patients were referred to Mayo for consideration of resection and IORT after external radiation was instituted or completed at other institutions. In 2 of the 5 patients, the external fields were felt to be either marginal or inadequate because of narrow tumor-free margins beyond gross disease (in 1 patient this was due to inaccurate reconstruction of tumor volume from a preirradiation CT; in the second patient fields were designed with intentional narrow margins because of the previous delivery of 3000 cGy for a primary seminoma).
Table 2. Irradiation External
GUNDERSON
(30) Recurrent
Total
5 3 3 1
2 4 2 4
I 3 1 12/10
2 8 0 2 12110
2 1
4 (5)” 2
6 (7)$ 3
2** 60-W
1** 2**
0 1” 0
0 1
3 8 1 1 1
1
7 I 5 5 9 11 1 3 24120
of IORT boost field
Extrapelvic-
patients
Para-aortic or paravertebral Iliac fossa Pelvic+- 10 patients Presacrum Sidewall Anterior (Bladder) Inguinal femoral region Neck- 1 patient * 4500125 -4;5040-5580128-3
1 -7;5760/32
1
- 1;6040/33 - 1.
+ Preop 75013, Postop 4500125. * IORT Fields: Abutting- 1 (1500 cGy, res(m), 12 MeV; 2000 cGy, res(g), 18 MeV); Field within field- 1; P two separate fields-2 (one patient with two separate areas of recurrence in right paravertebral region, another with two separate areas of residual pelvic sidewall and left inguinal femoral region after resection of primary disease. +* Presacrum and pelvic sidewall within single field in one patient each group-counted in both locations.
in the left
650
1. J. Radiation Oncology 0 Biology 0 Physics
Surgery. A partial or gross total resection was performed in our institution in each patient prior to the external beam or IORT component of irradiation. All had positive resection margins. Small vascular clips were placed at the margins of residual disease to help define and document IORT fields. While titanium clips produce less CT artifact, they sometimes cannot be located on lateral simulator films when designing external irradiation fields. One patient had an initial attempt at resection at an outside institution but was referred to Mayo for IORT because of residual disease within a previous field of external irradiation given for a primary seminoma. Additional tumor was resected at Mayo prior to IORT. Extent of resection before or after external irradiation and amount of residual disease by group is as follows: (a) primary-complete or gross total resection but with microscopic residual [i-es(m)] - 5, partial resection with gross residual [res(g)] - 5; (b) recurrent-res(m) - 6, res(g) 4. Planned high-dose preoperative irradiation (4500-5200 cGy in 180 cGy fractions) was delivered before an attempt at surgical resection in 6 patients. Subsequent gross total (4 patients) or subtotal resection (2 patients) was performed. Resection of adjacent organs or major vessels was performed in 12 of the 20 patients (5 of 10 primary and 7 of 10 recurrent) which helped achieve a gross total resection in 9 of the 12 patients. The remaining 3 patients had gross residual in a pelvic sidewall or presacrum. Six additional patients without extended resection had gross residual: pelvic sidewall or presacrum-3 patients, aorta or vena cava- 2 patients, brachial plexus- 1 patient. IORT. Decisions regarding the appropriateness of an IORT supplement were made jointly by the radiation oncologist and surgeon. When high-dose preoperative irradiation preceded an attempt at resection, the patient was restaged 3f-4 weeks after completion of irradiation with studies including a CT scan of the lungs as well as a CT of the primary or locally recurrent lesion. If distant metastasis was not diagnosed, resection was undertaken in a hospital operating room. The radiation oncologist joined the surgeon at that procedure to help determine whether a subsequent IORT boost was feasible and to make decisions regarding target volume as well as size and shape of the IORT cylinder. If surgical exploration preceded external irradiation and residual or unresectable disease remained after attempts at resection, a similar team decision was made with regard to IORT. The patient had a secondary reexploration in an operating room in the Radiation Oncology Department (9, 12), preferably in 1-2 days, but occasionally as late as 7 days. The area at risk was reexposed, and the patient was transferred within a sterile corridor to the linear accelerator room where the IORT boost was delivered. As discussed previously, IORT is used in our institution as a “boost” dose to supplement external radiation (9, 11, 12). The IORT dose is calculated at the 90% isodose and is usually dependent on the amount of disease remaining after attempts at resection: microscopic residual
Volume 25, Number 4. 1993
(res[mJ)-1000-1250 cGy, gross residual (res[g]) 2 2 cm-l 500 cGy, unresected or res(g) 2 2 cm-l 7502000 cGy. The size and shape of the lucite cylinders (cones) used for IORT are dependent on location of the IORT field with the intent of including tumor or tumor bed with at least a 1 cm margin (i.e., 5 cm tumor necessitates 2 7 cm cylinder). For all pelvic locations, circular cones with 15” or 30” bevels were utilized to approximate the anatomy of the presacrum, pelvic sidewall or anterior pelvis. With the 30” bevel cones, the depth of the isodose curves is more shallow at the heel end of the cone than the toe end (20). The radiation oncologist must be aware of these differences in determining placement of the treatment cone relative to tumor bed or residual tumor. For extrapelvic lesions, rectangular and eliptical cones with flat or 20” bevel ends were most frequently used. IORT characteristics are shown in Table 2. Electron energy varied from 9 to 18 MeV with the lower energies of 9 and 12 MeV used after gross total resection or with residual disease of minimal thickness (8 of 12 fields in patients who presented with primary disease). The 15 and I8 MeV energies and doses of 1500-2000 cGy were used more commonly in patients with recurrent disease. Twenty-four IORT fields were treated in the 20 patients ( I9 of 20 patients had lesions in the retroperitoneum and the other had a recurrent desmoid in the low neck). Abutting fields were used in 1 patient, two separate areas were treated in 2. In I other patient, a shrinking field IORT technique was used to deliver 1000 cGy with a large cone to both gross residual tumor and an area of gross total removal but narrow or microscopically positive margins, and a smaller cone was used to encompass the gross residual tumor for an additional 500 cGy (total dose within second IORT field of 1500 cGy). Potential dose-limiting organs within an IORT field included rectum (2 patients), ureter (3 patients), bladder (1 patient), peripheral nerve, spinal cord, and vessels. Peripheral nerve and vessels are usually within paravertebral, pelvic sidewall, and presacral fields. In para-aortic fields, vessels are always present and spinal cord and peripheral nerve are often present (dependent on width of fields-if field width exceeds vertebral body width, will have nerve roots deep to the field within the psoas muscle). When spinal cord may be in the IORT field, the dose contribution from IORT is limited to I 500 cGy by proper choice of electron beam energy. In such patients, the external beam dose to spinal cord is limited to 3600-3960 cGy in 180 cGy fractions with the aid of multiple field techniques. Follow-up
Most patients returned for scheduled follow-up at 3month intervals for 2 years, 6-month intervals for 3 years, then yearly. Both patterns of failure and treatment-related complications were coded prospectively using criterion developed by the NC1 IORT contract group of which we were a member (30). CT scans of the abdomen and pelvis
Intraoperative and external RT. sarcoma 0 L. L. GUNDERSON et al.
651
Table 3. Survival and disease status Survival (mo)
No.
Status Primary Alive (5) NED With disease Dead (5) With disease Recurrent Alive (9) NED With disease Dead (1) With disease
4 1
24, 26. 32, 33 1st (DM)
5
9, 11, 35, 37; (DM), 18 (RF)
7 (l)* 2*
15, 16, 18.5. 19, 45, 45 (CF + LF)*, 69 16 (LF + PS); 64.5 (LF)
1
40 (RF)
* Third patient had CF + LF-salvaged with hemipelvectomy, NED 34 months later. CF = Central failure or disease progression within IORT field; LF = Local failure in external beam field: RF = Regional failure outside irradiation field, DM = Distant metastases; PS = Peritoneal seeding.
with contrast and CT scans of the chest were performed at 6-month intervals. Disease progression within the IORT
Sarcoma IORT--Mayo SURVIVAL--ALL DEATHS
field (marked by surgical clips) was defined as central failure (CF) and within the external beam field as local failure (LF). Distant failures (DF) were separated into blood born
distant metastases (DM) and peritoneal seeding (PS). At time of progression, solitary areas of failure were resected, if feasible, and other treatment was individualized using chemotherapy, irradiation or combined modalities. Chemotherapy was never used as initial treatment, but only at time of disease progression. 20
Statistical analyses Complete follow-up data were available on all patients for 2 15 months from initiation of treatment for this malignancy (i.e., for patients who presented with recurrence, duration was measured from initiation of treatment for the recurrent disease). Overall survival time was calculated from initiation of treatment for this malignancy (surgery or start of preoperative irradiation) to the date of the most recent follow-up or to time of death using the Kaplan and Meier method (14). Disease-free survival was measured from initiation of treatment to the date of last follow-up or to time of disease progression. RESULTS Survival and disease status Survival and disease status are shown in Table 3. Six of 20 patients died (9, 11, 18, 35, 374, and 40 months)four with metastatic disease. Of the 10 patients who presented with primary disease, 5 (50%) were alive and 4 (40%) were disease-free (24, 26, 32, and 33 months). Of the 10 who presented at time of recurrence, 9 (90%) were alive with 7 currently free of disease (15, 16, 18.5, 19,45, 45, and 69 months). One of 7 required surgical salvage with hemipelvectomy and is disease-free 34 months after that procedure. Of the 6 patients with preoperative irradiation prior to an attempt at resection, 4 have died of
Ahveat intelval
Prlmaq Recurrent Totat
O-
18
6 5 11
1 5 6
0 2 2
I
J
1
2
3
4
5
I,”
1
0
I
,
0 2 2
Years (a)
DISEASE-FREE
SURVIVAL
Total
2
I
40
Primary (n = 10)
cz *O -Primary - Recurrent I
0’ 0
6 10 I
6 5 I
1
I 2
1 5 I
I 3
Years (b) Fig. 1. Kaplan Meier survival curves for locally advanced, pri-
mary and recurrent soft tissue sarcoma and recurrent desmoid treated with external beam irradiation, surgical resection and IORT. (a) Overall survival, all deaths included (primary, recurrent, total group). (b) Disease free survival.
652
1. J. Radiation Oncology 0 Biology 0 Physics
disease progression, 3 with distant metastases (each had a high-grade lesion). Overall and disease-free survival curves are shown in Figure 1. Actuarial survival at 2.5 years (Fig. la) was 83%, 69% in patients who presented with primary disease and 100% in those presenting with recurrence. Actuarial 5 year survival for the entire group was 48.5%. When patients who died without evidence of disease were censored, 2.5 year survival was 89% for the total group and 78 and lOO%, respectively, for primary versus recurrent disease presentation (curves not shown). With regard to diseasefree survival, patients who presented for treatment at the time of a localized recurrence have done at least as well as those with primary lesions (Fig. lb).
Volume 25,
Number
4, 1993
Impact of histological grade on the incidence of distant failure and total failures (any disease progression) is shown in Table 7. Distant failures appeared to be more frequent with high grade lesions (5/12 or 42% vs. l/8 or 13%).
DISTANT METASTASIS-FREE SURVIVAL ____-___-_______-___----_ Recurrent (n = 10)
Total group
8 c”-
60 -
‘5 ‘2
40
Primary (n = 10)
-
cz
Incidence and patterns offailure
Incidence and patterns of failure were evaluated in detail (Fig. 2 and Table 4). Disease progression occurred in 10 of 20 patients or 50% (primary 6/ 10, recurrent 4/ 10). Central failure (CF) in the IORT boost was documented in only one patient (5% of patients at risk or 10% of failures) and local failures (LF) in the external beam field in three (O/ 10 primary; 3/ 10 recurrent-30%). Hemipelvectomy for salvage was performed in 1 patient with both CF and LF who is currently free of disease. Distant metastases occurred in 5 patients (5/ 10 primary-50%; O/ 10 recurrent). None were amenable to resection. Peritoneal seeding was documented in only 1 patient (5%) but this estimate may be falsely low because of a low incidence of reoperations or autopsies. Ascites on serial CT’s or physical exam has not been observed, however. Disease control curves are shown in Figure 2. Patterns of failure differed markedly between primary and recurrent disease patients, but part of that difference may be due to histological factors. The incidence of both total and distant failure was higher in primary versus recurrent disease presentation (Table 4). All blood born distant metastases were documented in the primary disease group but 4 of 5 patients had high grade lesions. All three failures within the irradiation field were in the group who presented at the time of recurrence. However, if the 2 patients with marginal external fields are censored, then the local failure risk in the recurrent group becomes I of 8 or 12.5%. The impact of resection on local failure (within external irradiation field) and total failure is shown in Table 5. A trend for a decrease in the total failure incidence with gross total resection versus partial resection is suggested only if the 2 patients with marginal external fields are not included in the analysis. An expanded analysis is shown in Table 6 with regard to the 5 patients with tumor progression within or at the margin of irradiation fields. One regional and one local failure may have been prevented with larger fields (optimal fields could not be used in the patient with subsequent regional failure in view of previous external irradiation for seminoma and the patient with local failure also had limited peritoneal seeding found at exploration).
20r-
6 10
Primary Recurrent
1
0
6 5
1 5
2
3
Years (a)
ABDOMINAL CONTROL 100
1
1
Primary (n = 10)
60 8 !F ‘5 .-
60
-
2
40 -
Recurrent (n = 10)
cz 2o
-
Primary Recurrent
0 10
I
I
0
I
I
1
0
1 5
6 5
I
I 3
2
Years (b)
LOCAL CONTROL Primary (n = 10)
Total
Recurrent (n = 10)
2o 0
. 0
Primary Recurrent
I
6 10 I
1 5
6 5
I
I
1
2
I
I 3
Years (c) Fig. 2. Disease control curves. (a) Local control (in radiation field-external or IORT)-one patient required salvage with hemipelvectomy. (b) Abdominal control. (c) Distant control (free of blood born or peritoneal metastases).
lntraoperative and external RT, sarcoma 0 L. L. GUNDERSON
653
et al.
Table 4. Incidence and patterns of failure Patterns of failure (any component) CF
LF
DF
RF
Failures Site
No. %
No. (%)
No. (%)
No. (%)
No. (%)
Total group Primary Recurrent
10/20 (50) 6/lO (60) 4/ 10 (40)
1 (5) 0 1 (lo)+
3 (15) 0 3 (30)+
6 (30)* 5 (50) I (lo)*
2 (10) 1 (10) 1 (10)
See Table 3 for CF, LF, and RF, DF-distant failure either hematogenous or peritoneal (W) Are of total group at risk. * PS in conjunction with LF (PS component in 5% at risk). + CF + LF was salvaaed with hemipelvectomv. both external + IORT fields were marginal: a second recurrent patient with LF also had marginal external beam fields (EW). _
Complications Significant complications occurred in 6 patients. The IORT surgical procedure or IORT electron boost were
implicated as a potential causative factor in 5 of the 6 (nerve- 1, wound-2, urologic- 1, fibrosis- 1). Severe (Grade 3 or 4) symptomatic or objective neuropathy occurred in 1 patient (5%) who was treated with abutting IORT fields. This was a Grade 3 motor neuropathy which resulted in foot drop but caused no pain. At the time of analysis this patient was free of disease within the IORT fields but had local failure in the opposite side of the pelvis. Another patient had a Grade 2 neuropathy requiring occasional pain medications. Small bowel obstruction (SBO), related to surgery and external irradiation, occurred in 2 patients. This resolved with conservative measures during hospitalization in 1 patient (5%), but reoperation was required in the other (5% of 20 at risk). DISCUSSION When surgery is the sole treatment modality for retroperitoneal sarcomas, the incidence of subsequent local
Table 5. Impact of resection on local and total failure Failure patternamount of residual Local failure Microscopic residual (gross total removal) Gross residual Totals Any failure Microscopic residual Gross residual Totals
Primary No.
Recurrent
Total
No. %
No. %
O/5 O/5 O/IO
2/6* 114 3/10 (30)
2/11 (18)+ l/9 (11) 3/20 (15)
315 315 6/10 (60)
2/6 214 4/10 (40)
5/l 1 (45)+ 5/9 (56) 10/20 (50)
* l-marginal external field due to previous 3000 cGy for seminoma (non-Mayo); l-marginal IORT field (Mayo). + Becomes O/9 LF and 3/9 (33%) any failure for res(m) if the two patients with marginal external fields are censored for this analysis.
recurrence has been as high as 70-90% ( 1, 2, 15, 16, 2 1, 32, 34). If gross total resection can be accomplished, the risk of subsequent local recurrence in patients who presented with primary tumors was 60% (37/63) in a surgical series from Mayo (2) and 37% (15/41) from Memorial (13). The latter incidence is falsely low as only patterns of initial failure were reported. When patients presented at the time of recurrence and had a gross resection of tumor, Memorial reported a subsequent local recurrence incidence of 59% (23/39) as initial site of failure. Neither the Mayo nor Memorial series, however, represent surgery alone data. In the Mayo series, 18 of the 63 with complete resection received adjuvant therapy (irradiation - 15, chemotherapy - 1, both -2). At Memorial, 18 of 4 1 patients with complete resection of primary lesions received postoperative therapy, but no information was given concerning added treatment in the 39 with complete resection of locally recurrent lesions. Survival at 5 years in surgery alone series has varied from 25-50% after complete resection and from 5-20% with subtotal resection. External beam irradiation has been combined with surgical resection with or without chemotherapy for locally advanced soft tissue sarcomas of the trunk and extremities. Although local control has appeared to improve when external irradiation is combined with surgical resection for retroperitoneal sarcomas (4, 15, 3 1, 34) the dose of irradiation that can be delivered safely is much lower than with extremity sarcomas (22. 29) in view of dose limiting structures including small intestine, stomach, liver, kidney, and spinal cord. In published series from the NCI, both acute and chronic small bowel morbidity were excessive in the treatment arms using moderate to high doses of external irradiation with no IORT (4, 18). The National Cancer Institute (NCI) conducted a randomized trial from 1980- 1985 in which 35 patients with retroperitoneal sarcomas were randomized to receive external irradiation + IORT ( 18). All patients had primary lesions, and none had received prior irradiation or chemotherapy. All had gross total resection, but most had microscopic residual. Patients randomized to external alone received 3500 to 4000 cGy to an extended field over 4-5 weeks and an additional 1500 cGy over 2 weeks to a
I. J. Radiation Oncology 0 Biology 0 Physics
654
Volume 25. Number 4, 1993
Table 6. Expansion of central (CF), local (LF) and regional failures* Radiation factors Presentation
Original size (cm) amount residual
histology
External (cGy/fractions)
IORT (size/dose/energy)
Failure* pattern
Comment
D. M. Ret R iliac fossa (2nd) Gr 4 leiomyosarc
8.5 x 8 x 5 res (m)
5.0, 30” 1500,9E
CF-LF
5040128
S. S. Ret R retroperit (1st) Gr 3 liposarc
11 x 1.5 res (m)
8 x 15,20” 1500, 12E
LF-PS
4554121
S. G. Ret L pelvis (2nd) Gr 2 chondrosarc C.H. Ret L pelvis (3rd) Gr 2 chondrosarc J. P. Pri R retroperit Gr 3 leiomyosarc
10.5 x 9 X 8.5 1) res (g) 2) res (m)
(I)* 5.5, 2000 18E
LF
5340/30
(2)* 6.0, 1500 12E
10X8x7 res (g)
8.0, 15” 2000, 18E
RF
5200126
R inguinal DOD 40 mo
10 x 5 x 3.5 res (m)
7 x 12, 20” 1900, 9E
RF
4500125
Ext RT (EW) marginal 2” previous ext. RT for seminoma; DOD 18mo
IORT field marginal NED 34 mo after hemipelvectomy (45 mo total) External RT marginal (EW) Alive 16 mo R pelvis_4440/25Fx Alive 64.5 mo
* See Table 2 for definition of failure patterns. Ret = Recurrent; Pri = Primary; Sarc = Sarcoma; R = Right, L = left; Res (m) = Microscopic residual; Res (g) = Gross residual; E = Electron energy in MeV; NED = No evidence of disease; DOD = Dead of disease; Fx = fractions; Ext RT = External irradiation; EW = Elsewhere. + Abutting fields with maximum 0.5 cm overlap.
reduced field. The IORT group received 3500-4000 cGy in 4-5 weeks to an extended field and an IORT dose of 2000 cGy to abutting fields (2-6 abutting fields) plus intravenous misonidazole. Some apparent advantages for IORT patients were seen in the NC1 trial with a minimum follow-up of 15 months. Treatment-related small bowel complications were substantially lower in the IORT group. Severe acute enteritis occurred in 12 of 20 versus 0 of 15 patients (p < 0.05), chronic enteritis in 7 of 20 versus 2 of 15 (p < 0.05), and fistulae in 6 of 20 versus 0 of 15 (p < 0.05). Actuarial disease control within irradiation fields at 4-5 years was 78% with IORT plus external irradiation versus 30% with external irradiation alone. In view of small patient numbers and limited follow-up, these differences did not reach
statistical significance in the initial publication (18). In an updated analysis, local/regional failure in the tumor bed or abdomen occurred in 6 of 15 IORT (40%) versus 15 of 20 non-IORT patients (75%) which achieved borderline statistical significance at p = 0.08 (24). In the updated analysis, gastrointestinal complications were again statistically less in the IORT patients at 2/ 15 versus 9/20 patients (p < 0.05). With marginal resection of retroperitoneal sarcomas or proven microscopic residual disease, doses of 6000-7000 cGy would usually be necessary in order to achieve local control with a frequency of 80-90% (22, 28, 29). Such doses could not be safely achieved with external beam approaches for retroperitoneal sarcomas. Even with external doses to only 5500 cGy in the NC1 series, acute
Table 7. Impact of histological grade on distant and total failure Distant failure (DF)
Grade
112 314
Total failures
Total
With local control
No. (%)
No. (%)
l/8* (13) 5/12 (42)
l/7 4/10
6/20’ (30)
5/17* (28)
Totals
(14) (40) Totals Sarcoma Desmoid
Primary
Recurrent
l/2 518
214 214
6/10 -
o/2
* Includes two patients with desmoid - O/2 DF. + Total failure for low grade lesions 3/8 (37.5%) if desmoids combined with Gr 1, 2. * DF = DM (5) + PS (1).
418
No. (%) 3/6 (50)+ 7/12 (58) lo/18 (56) o/2+
lntraoperative and external RT, sarcoma 0 L. L. GUNDERSON
and chronic small bowel complications were excessive. Although improvements in survival were not seen in this randomized trial, it is appropriate to consider the combined IORT external beam approach as preferable treatment to external beam alone in view of nonexistent chronic small bowel risks and improved local tumor control. In the present series, similar optimistic results were seen regarding tumor control within irradiation fields with the same minimum follow-up of 15 months as in the NC1 series. In patients who presented with primary disease, no local failures had occurred within either the external beam or IORT field at the time of this analysis. Local failure had occurred in 3 of 10 patients in the group who presented at the time of local recurrence. Our analysis has shown no definite relationship between the amount of residual remaining after maximal resection and subsequent disease progression within irradiation fields. However, the IORT dose in our institution is based on the amount of residual disease remaining after maximal surgical resection (1000-l 250 cGy with 5 microscopic residual, 1500-2000 cGy for gross residual or unresected disease). Doses less than 1500 cGy are rarely used for recurrent lesions because of the greater potential for tumor bed hypoxia from multiple operations. Failures within or at the margin of external or IORT irradiation fields were probably related to inadequate fields in 3 patients, 2 received the external component of irradiation elsewhere (Table 3). In 1 patient external fields were marginal because of a previous course of irradiation for seminoma and in the second were due to inaccurate reconstruction of tumor volumes from computed tomography. As with extremity sarcomas, it would be reasonable to have initial field margins 5-7 cm beyond unresected or residual tumor or clipped areas of tumor adherance (i.e., achieve those margins relative to potential microscopic spread along muscle, nerve, and vessel). It may be safer to treat these extended margins to 4500 cGy in 180 cGy fractions with preoperative rather than postoperative irradiation. Preoperative irradiation may reduce the incidence of severe complications because the tumor is helping to displace the dose-limiting small intestine. After resection of large retroperitoneal sarcomas, small bowel may become immobile because of adherence to denuded surfaces thus increasing the risk of acute or chronic treatment-related enteritis. Small bowel morbidity was acceptable in the current series with an incidence of
et al.
655
small bowel obstruction requiring reoperation of only 5% (l/20). With regard to IORT field design, the entire tumor bed is not intentionally treated in our institution. As previously noted, that is treated with fractionated external beam techniques to 4500-5040 cGy, and IORT doses of lOOO2000 cGy are used as a boost to areas of positive or narrow resection margins. If unresected or gross residual exists after maximal surgical resection, it must be encompassable within a single IORT cylinder. Abutting IORT fields are considered only in the context of a gross total resection. Blood-born metastases occurred with higher frequency with primary vs recurrent disease presentation in our se-
ries. This difference was likely due to more high-grade lesions (i.e., 8 Grade 3 or 4 primary lesions versus 4 in the recurrent group). The incidence of distant metastases was 13% in grade 1 and 2 lesions (l/S) versus 42% with Grade 3 and 4 (5/ 12). We will be attempting to integrate chemotherapy regimens in future IORT treatment strategies for high-grade lesions. With regard to survival, we were intrigued that patients who presented at the time of recurrence have done as well as those who had primary lesions. This finding may be related in part to a lower incidence of distant metastases (only 4 of 10 patients in the recurrent group had highgrade lesions). However, it is encouraging that aggressive approaches seem effective and appropriate for carefully selected patients with localized recurrence in addition to those with locally advanced primary disease. The aggressive techniques used in this series of patients appear promising. All patients in this series had either proven microscopic residual disease after gross total resection of locally advanced trunk sarcomas or gross residual. Most would have been expected to recur locally and die of their malignancy with only a surgical approach. We are encouraged by the excellent disease control within irradiation fields, the acceptable tolerance with regard to small bowel and peripheral nerve (17, 19, 23-27), and the favorable early survival. Further evaluation of IORT combined with external irradiation and maximal resection seems indicated for locally advanced primary and recurrent retroperitoneal sarcomas on the basis of this and other IORT series (18, 33). Attempts dt integration of chemotherapy in a protocol setting are probably indicated for grade 3 and 4 lesions in view of metastatic risks of 40% or greater. While randomized trials would be of interest, patient numbers may be too small to allow this to occur.
REFERENCES 1. Cody, H. S.; Turnbull, A. D.; Former, J. G.; Hadju, S. I. The continuing challenge of retroperitoneal sarcomas. Cancer 47:2147-2152;1981. 2. Dalton, R. R.; Donahue, J. H.; Mucha, P.; Van Heerden, J. A.; Reiman, H. M.; Chen, S. Management of retroperitoneal sarcomas. Surgery 106:725-733;1989. 3. Gallagher, M. J.; Brereton, H. D.; Restock, R. A.; Zero, J. M.; Zekoski, D. A.; Poyss, L. F.; Richter, M. P.; Kligermann, M. M. A prospective study of treatment techniques to minimize the volume of pelvic small bowel with reduction
of acute and late side effects associated with pelvic irradiation. Int. J. Radiat. Oncol. Biol. Phys. 12:1565-1573;1986. 4. Glenn, J.; Sindelar, W. F.; Kinsella, T.; Glatstein, E.; Tepper, J.; Costa, J.; Baker, A.: Sugarbaker, P.; Brennan, M. F.; Seipp, C.; Wesley, R.; Young, R. C.; Rosenberg, S. A. Results of multimodality therapy of resectable soft tissue sarcomas of the retroperitoneum. Surgery 97:316-324;1984. 5. Green, N.; ha, G.; Smith, W. R. Measures to minimize small intestine injury in the irradiated pelvis. Cancer 35: 1633-1640:1975.
656
I.
J. Radiation Oncology 0 Biology0 Physics
6. Gunderson, L. L.; Cohen, A. M.; Dosoretz, D. E.; Shipley, W. V.; Hedberg, S. A.; Wood, W. C.; Rodkey, G. V.; Suit, H. D. Residual, unresectable, or recurrent colorectal cancer: External beam irradiation and intraoperative electron beam boost + resection. Int. J. Radiat. Oncol. Biol. Phys. 9: 15971606;1983. 7. Gunderson, L. L.; Cohen, A. M.; Welch, C. W. Residual, inoperable, or recurrent colorectal cancer: Surgical radiotherapy interaction. Am. J. Surg. 1395 18-525; 1980. 8. Gunderson, L. L.; Martenson, J. A. Gastrointestinal tract radiation tolerance. Front. Radiat. Ther. Oncol. 23:277298;1989. 9. Gunderson, L. L.; Martin, J. K.; Earle, J. D.; Byer, D.; Voss, M.; Fieck, J.; Kvols, L.; Rorie, D.; Martinez, A.; Nagomey. D. M.; O’Connell, M. J.; Weber, F. Intraoperative and external beam irradiation & resection: Mayo pilot experience. Mayo Clin. Proc. 59:691-699;1984. 10. Gunderson, L. L.; Russell, A. H.; Llewellyn, H. T.; Dopke, K. P.; Tepper, J. E. Treatment planning for colorectal cancer: Radiation and surgical techniques and value of small bowel films. Int. J. Radiat. Oncol. Biol. Phys. 11:1379-1393;1985. 11. Gunderson, L. L.; Shipley, W. U.; Suit, H. D.; Epp, E. R.: Nardi, G.; Wood, W.; Cohen, A.; Nelson, J.; Battit, G.; Biggs, P. J.; Russell, A.; Rockett, A,; Clark, D. Intraoperative irradiation: A pilot study combining external beam irradiation with “boost” dose intraoperative electrons. Cancer 49:22592266; 1982. 12. Gunderson, L. L.; Tepper, J. E.; Biggs, P. J.; Goldson, A.: Martin, J. K.; McCullough, E. C.; Rich, T. A.; Shipley, W. U.; Sindelar, W. F.; Wood, W. C. Intraoperative +- external beam irradiation. Curr. Probl. Cancer 7: l-69; 1983. 13. Jaques, D. P.; Coit, D. G.; Hajdu, S. I.; Brennan, M. D. Management of primary and recurrent soft tissue sarcoma of the retroperitoneum. Ann. Surg. 2 12:5 l-59: 1990. 14. Kaplan, E. L.; Meier, P. Nonparametric estimation for incomplete observations. J. Am. Stat. Assoc. 53:457-48 1;1958. 15. Karakousis, C. P.; Velez, A. F.; Einrich, L. J. Management of retroperitoneal sarcomas and patient survival. Am. J. Surg. 150:376-380; 1985. 16. Kinne, D. W.; Chu, F. C. H.; Huvos, A. G.; Yagoda, A.; Fortner, J. G. Treatment of primary and recurrent retroperitoneal liposarcoma: Twenty-five year experience at Memorial Hospital. Cancer 3 1:53-64; 1973. 17. Kinsella, T. J.; Sindelar, W. F.; DeLuca, A. M.; Pezeshkpair, G.; Smith, R.; Maher, M.; Terrill, R.; Miller, R.; Mixon, A.; Harwell, J. F.; Rosenberg, S. A.; Glatstein, E. Tolerance of peripheral nerve to intraoperative radiotherapy (IORT): Clinical and experimental studies. Int. J. Oncol. Biol. Phys. 11:1579-1585;1985. 18. Kinsella, T. J.; Sindelar, W. F.; Lack, E.; Glatstein, E.; Rosenberg, S. A. Preliminary results of a randomized study of adjuvant radiation therapy in resectable adult retroperitoneal soft tissue sarcomas. J. Clin. Oncol. 6: 18-25; 1988. 19. LeCouteur, R. A.; Gillette, E. L.; Powers, E. L.; Child, G.; McChesney, S. L.; Ingram, J. T. Peripheral neuropathies following experimental intraoperative radiation therapy (IORT). Int. J. Radiat. Oncol. Biol. Phys. 17:583-590;1989.
Volume 25. Number 4. 1993 20. McCullough, E. C.; Anderson, J. A. The dosimetric properties of an applicator system for intraoperative electron beam therapy utilizing a clinic 18 accelerator. Med. Phys. 9:261-268;1982. 2 I. McGrath, P. C.; Neifeld, J. P.: Lawrence, W.: DeMay, R. M.; Kay, S.; Horsley, J. S.; Parker, G. A. Improved survival following complete excision of retroperitoneal sarcomas. Ann. Surg. 200:200-204:1984. 22. Schray, M. D.; Gunderson, L. L.; Sim, F. H.: Pritchard, D. J.; Shives, T. C.; Yeakel, P. D. Soft tissue sarcomas. Integration of bradytherapy, resection, and irradiation. Cancer 66:45 l-456; 1990. 23. Shaw, E. G.; Gunderson, L. L.; Martin, J. K.; Beart, R. W.; Nagorney, D. M.; Podratz, K. C. Peripheral nerve and ureteral tolerance to intraoperative radiation therapy. Clinical and dose response analysis. Radiother. Oncol. 18:247255:1990. 24. Sindelar, W. F.; Kinsella, T. J.; Chen, P. W.; DeLaney, T. F.; Tepper, J. E.; Rosenberg, S. A.; Glatstein. E. Intraoperative radiotherapy in sarcomas of the retroperitoneum: Final results of a prospectively randomized clinical trial. (in press). 25. Sindelar, W. F.; Kinsella, T. J.; Hoekstra, P., et al. Treatment complications in intraoperative radiotherapy. Int. J. Oncol. Biol. Phys. 11:117; 1985. 26. Sindelar, W. F.; Kinsella, T.; Tepper, J.; Travis, E. L.; Rosenberg, S. A.; Glatstein, E. Experimental and clinical studies with intraoperative radiotherapy. Surg. Gyn. Obstet. 157: 205-219;1983. 27. Sindelar, W. F.; Tepper, J.: Travis, E. L.; Terrill. R. Tolerance of retroperitoneal structures to intraoperative irradiation. Ann. Surg. 196:601-608; 1982. 28. Suit, H. D. Radiation biology: A basis for radiotherapy. In: Fletcher, G. H., ed. Textbook of radiotherapy, 2nd edition. Philadelphia: Lea and Febiger; 1973: 75- 12 1. 29. Suit, H. D.; Mankin, H. J.; Wood, W. C.; Proppe, K. H. Preoperative. intraoperative, and postoperative radiation in the treatment of primary soft tissue sarcoma. Cancer 55: 2254-2267: 1985. 30. Tepper, J. E.; Gunderson, L. L.; Orlow, E.; Cohen, A. M.: Hedberg, S. E.; Shipley, W. U.; Blitzer, P. H.; Rich, T. Complications of intraoperative radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 10: 183 1- 1839; 1984. 31. Tepper. J. E.; Suit, H. D.; Wood, W. C.; Proppe, K. H.; Harmon, D.; McNulty, P. Radiation therapy of retroperitoneal soft tissue sarcomas. Int. J. Radiat. Oncol. Biol. Phys. 10:825-830:1984. 32. Wile, A. G.: Evans, H. L.; Romsdahl, M. Leiomyosarcoma of soft tissue: A clinicopathologic study. Cancer 48: 10221032;1981. 33. Willett, C. G.; Suit, H. D.; Tepper, J. E.; Mankin, H. J.; Conrery, K.; Rosenberg, A. L.; Wood, W. C. Intraoperative electron beam radiation for retroperitoneal soft tissue sarcoma. Cancer 68:278-283; 199 1. 34. Wist, E.; Sotheim. 0. P.; Jacobsen, A. B.; Blom, P. Primary retroperitoneal sarcoma. Acta. Radio]. Oncol. 24:305310:1985.
0360-3016/93 $6.00 + .OO Copyright (El 1993 Pergamon Press Ltd.
25, PP. 647-656
??Phase I/II Clinical Trials EXTERNAL BEAM AND INTRAOPERATIVE ELECTRON IRRADIATION FOR LOCALLY ADVANCED SOFT TISSUE SARCOMAS LEONARD L. GUNDERSON, M.D., M.S.,’ DAVID M. NAGORNEY, M.D.,’ DONALD C. MCILRATH, M.D.,* JENNIFER M. FIECK,~ HARRY S. WIEAND, PH.D.,~ ALVARO MARTINEZ, M.D.,’ DOUGLAS J. PRITCHARD, M.D.,4 FRANK SIM, M.D.,4 JAMES A. MARTENSON, M.D.,’ JOHN H. EDMONSON, M.D.5 AND JOHN H. DONOHUE, M.D.* Radiation Oncology’. Surgery’, Cancer Center Statistic$, Orthopedics4 and Medical Oncology’ Mayo Clinic and Mayo Medical School, Rochester, MN Purpose: Intraoperative irradiation with electrons was used in conjunction with external beam irradiation and maximal surgical resection in 20 patients with locally advanced soft tissue sarcomas or desmoids. This manuscript presents results with regard to tolerance of treatment and its impact on tumor control and survival. Methods and Materials: Ten patients presented with previously untreated primary sarcomas and 10 at the time of local recurrence (two had recurrent desmoid tumors). Tumor location was retroperitoneal in 19 and in the low anterior neck in one. A partial or gross total resection was performed prior to the external beam or intraoperative component of irradiation in every patient, but all had positive resection margins. Patients received 4500-6000 cGy of fractionated, external beam irradiation and an IORT dose of 1000-2000 cGy. Chemotherapy was given only at the time of disease progression. Results: Fourteen of 20 patients (70%) were alive; 11 (55%) were free of disease (4/10 primary, 7/10 recurrent), but 1 required hemipelvectomy for salvage. Progression within the intraoperative irradiation field was documented in only 1 patient (5%) and within the external beam field in 3/20 (15%). Blood born distant metastasis occurred in 5 patients (25%) and peritoneal seeding in 1 (5%). The distant failure incidence by grade was l/8 (13%) for Grades 1, 2 and 5/12 (42%) for Grades 3, 4. Only 1 patient (5%) developed a 2 severe neuropathy, and small bowel obstruction requiring exploration also occurred in a single patient. Conclusion: In view of acceptable tolerance and the high current rate of local tumor control, in spite of incomplete surgical resections, further evaluation of intraoperative irradiation as a component of treatment is indicated for locally advanced primary and recurrent soft tissue sarcomas. Intraoperative irradiation, Soft tissue sarcomas, Sarcoma irradiation, Locally advanced lesions.
INTRODUCTION
would be required to provide a reasonable probability of local control of tumor (28, 29). Many radiation oncologists, therefore, have recommended observation instead of adjuvant external irradiation since such dose levels ex-
Although surgery alone for retroperitoneal sarcomas is associated with a high incidence of local recurrence ( 1, 2, 15, 16, 21, 32, 34), the literature contains little data on the use of adjuvant irradiation as a component of treatment. Since retroperitoneal sarcomas often produce only minimal symptoms until they reach extensive size, surgical resections are often marginal, and microscopically involved resection margins are not infrequent. With the latter scenario, external irradiation doses of 6000-7000 cGy
ceed gastrointestinal tolerance (small intestine and stomach, *large bowel dependent on volume) (3, 5, 7, 8). In an attempt to decrease local recurrence and improve survival with an acceptable risk of complications, our institution has combined intraoperative irradiation (IORT), using an electron beam, with external beam irradiation and maximal surgical resection for locally advanced ret-
Presented in part at the American Society of Therapeutic Radiology and Oncology, New Orleans, LA, 9-14 October 1988.
the preparation of the manuscript, and to the radiation oncology
physicists and technologists, operating room nurses, and many
Reprint requests to: Leonard L. Gunderson, M.D., M.S., Chairman, 55905.
Radiation
Oncology,
Mayo Clinic, Rochester,
additional physicians who were involved in the treatment patients. Supported in part by NC1 Contract CM-27528. Accepted for publication 17 September 1992.
MN
Acknowledgements-The Chambers
authors appreciate the efforts of Julie and the Mayo Clinic Typing Service for assistance in 647
of the
648
I. J. Radiation Oncology 0 Biology 0 Physics
Volume 25, Number 4. 1993
roperitoneal and trunk sarcomas. The intent of this paper is to present our experience with the combined treatment approach in a group of 20 patients who presented with either primary or locally recurrent lesions.
METHODS
IORT was scheduled but not given in an additional 12 patients with retroperitoneal sarcomas during this time period. Three patients were excluded prior to exploration; radiographic evidence of locally extensive disease in 2 patients, and 1 patient was against exploration. At exploratory laparotomy, 5 patients were excluded because of disease factors; peritoneal seeding in 2 patients, and locally extensive disease that could neither be resected nor adequately treated with IORT in 3 patients. Total resection was accomplished in 4 additional patients. IORT was not given because of negative margins in 2 patients, 1 patient was against reoperation for IORT (microscopic residual), and 1 had myocardial infarction after resection and reoperation for IORT was not medically indicated. General criterion for patient selection and preoperative evaluation have been outlined previously (9, 11, 12). All patients have a staging work-up including history, physical examination, and diagnostic radiographs to determine the extent of local and regional disease and to exclude distant disease. Patients with distant metastases do not usually receive IORT in our institution, because expected life span
AND MATERIALS
Patient selection and work-up Twenty adult patients without previous irradiation for their locally advanced primary or recurrent soft tissue sarcoma of the retroperitoneum or trunk were treated with curative intent over a 6-year period with a regimen that included intraoperative irradiation (IORT) with electrons, have had minimum follow-up of 15 months, and form the basis of this report. One additional patient with a locally recurrent retroperitoneal sarcoma was excluded from all analyses because a full component of external irradiation could not be delivered due to previous external irradiation to this region after resection of a primary liposarcoma in the right hemiabdomen. One of the 20 patients included in this report had previously received 3000 cGy for a seminoma, necessitating restricted external beam fields during irradiation of a right paravertebral sarcoma. Patients who received IORT for extremity sarcomas are not included in this analysis.
would not be adequate to evaluate treatment related effectiveness or tolerance. In view of the high incidence of lung metastases in patients with high-grade soft tissue sar(CT) of the chest is roucomas, computed tomography tinely obtained at the time of initial evaluation in addition
Table 1. Tumor characteristics
Disease category
No. patients
Primary, resected-residual Primary, unresectable for cure (disease fixation) Recurrent, resected-residual Recurrent, unresectable for cure Total
7 3* 7 3* 20
Histological types and grade
Gr 112
Gr 314
Chondrosarcoma Hemangiopericytoma Leiomyosarcoma Liposarcoma Malignant fibrous histiocytoma Neurofibrosarcoma Synovial cell sarcoma Sarcoma-not otherwise specified Total soft tissue sarcomas Desmoid (recurrent)
3 1 0 1 0 1 -
0 0 4 1 2 3 1
0 6
1 12
Maximum tumor dimension before resection (cm) o-5 5.1-10 10.1-15 > 15 Total * All six had resection for a seminoma.
after 4500-5000
Primary 4 4 2 0 10 cGy. In 1 of 6, external fields were restricted,
Recurrent 4 3 3 0 10
Total 3 1
4 2 2 4 1 1 18 2 Total 8 7 5 0 20
since 3000 cGy had been previously
delivered
Intraoperative and external RT, sarcoma 0 L. L.
to a CT and/or magnetic resonance imaging (MRI) of the site of the primary or recurrent lesion. Tumor characteristics are shown in Table 1. At time of presentation for irradiation, 10 patients had previously untreated primary lesions and 10 had localized recurrences ( 14 presented after an attempt at surgical resection and six before resection). All histological grades and tumor types of soft tissue sarcomas were eligible and recurrent desmoids were also included (Table 1). Histological Grade of 1 to 4 was assessed by Broder’s classification. Maximum tumor dimension was obtained from the preoperative CT scans and resected specimens. Treatment factors External irradiation. External beam irradiation methods have been fairly consistent in this group of patients. Doses of 4500-6040 cGy were delivered in 180 cGy fractions, 5 days per week over 5-61 weeks with linear accelerators using 10 MV photons ( 18 of 20 received external
irradiation
Completed planned external irradiation Method of irradiation Postop* (4500-6040 cGy/25-33 fractions) Low dose preop + postopt Preop (4500-5200 cGy/26-29 fractions) IORT energy and dose Electron energy (MeV) 9 12 15 18 IORT dose (cGy) 1000-1250 1500 1750 2000 Total fields/patients+ Location
factors No.
W)
20*
(100)
13
(65) (5)
1 6 Primary
649
et al.
doses I 5580 cGy). For pelvic lesions, treatments were given with 4 field shaped external beam techniques similar to those used for locally advanced colorectal cancers (6, 7, 10). With extrapelvic lesions, unresected or residual disease plus 5 cm margins was included to 4500 cGy with parallel opposed fields; reduced fields with 2-3 cm margins were treated to 5000 to 5500 cGy. As noted in Table 2, 20 of 20 patients (100%) completed all planned external irradiation. Five of twenty patients were referred to Mayo for consideration of resection and IORT after external radiation was instituted or completed at other institutions. In 2 of the 5 patients, the external fields were felt to be either marginal or inadequate because of narrow tumor-free margins beyond gross disease (in 1 patient this was due to inaccurate reconstruction of tumor volume from a preirradiation CT; in the second patient fields were designed with intentional narrow margins because of the previous delivery of 3000 cGy for a primary seminoma).
Table 2. Irradiation External
GUNDERSON
(30) Recurrent
Total
5 3 3 1
2 4 2 4
I 3 1 12/10
2 8 0 2 12110
2 1
4 (5)” 2
6 (7)$ 3
2** 60-W
1** 2**
0 1” 0
0 1
3 8 1 1 1
1
7 I 5 5 9 11 1 3 24120
of IORT boost field
Extrapelvic-
patients
Para-aortic or paravertebral Iliac fossa Pelvic+- 10 patients Presacrum Sidewall Anterior (Bladder) Inguinal femoral region Neck- 1 patient * 4500125 -4;5040-5580128-3
1 -7;5760/32
1
- 1;6040/33 - 1.
+ Preop 75013, Postop 4500125. * IORT Fields: Abutting- 1 (1500 cGy, res(m), 12 MeV; 2000 cGy, res(g), 18 MeV); Field within field- 1; P two separate fields-2 (one patient with two separate areas of recurrence in right paravertebral region, another with two separate areas of residual pelvic sidewall and left inguinal femoral region after resection of primary disease. +* Presacrum and pelvic sidewall within single field in one patient each group-counted in both locations.
in the left
650
1. J. Radiation Oncology 0 Biology 0 Physics
Surgery. A partial or gross total resection was performed in our institution in each patient prior to the external beam or IORT component of irradiation. All had positive resection margins. Small vascular clips were placed at the margins of residual disease to help define and document IORT fields. While titanium clips produce less CT artifact, they sometimes cannot be located on lateral simulator films when designing external irradiation fields. One patient had an initial attempt at resection at an outside institution but was referred to Mayo for IORT because of residual disease within a previous field of external irradiation given for a primary seminoma. Additional tumor was resected at Mayo prior to IORT. Extent of resection before or after external irradiation and amount of residual disease by group is as follows: (a) primary-complete or gross total resection but with microscopic residual [i-es(m)] - 5, partial resection with gross residual [res(g)] - 5; (b) recurrent-res(m) - 6, res(g) 4. Planned high-dose preoperative irradiation (4500-5200 cGy in 180 cGy fractions) was delivered before an attempt at surgical resection in 6 patients. Subsequent gross total (4 patients) or subtotal resection (2 patients) was performed. Resection of adjacent organs or major vessels was performed in 12 of the 20 patients (5 of 10 primary and 7 of 10 recurrent) which helped achieve a gross total resection in 9 of the 12 patients. The remaining 3 patients had gross residual in a pelvic sidewall or presacrum. Six additional patients without extended resection had gross residual: pelvic sidewall or presacrum-3 patients, aorta or vena cava- 2 patients, brachial plexus- 1 patient. IORT. Decisions regarding the appropriateness of an IORT supplement were made jointly by the radiation oncologist and surgeon. When high-dose preoperative irradiation preceded an attempt at resection, the patient was restaged 3f-4 weeks after completion of irradiation with studies including a CT scan of the lungs as well as a CT of the primary or locally recurrent lesion. If distant metastasis was not diagnosed, resection was undertaken in a hospital operating room. The radiation oncologist joined the surgeon at that procedure to help determine whether a subsequent IORT boost was feasible and to make decisions regarding target volume as well as size and shape of the IORT cylinder. If surgical exploration preceded external irradiation and residual or unresectable disease remained after attempts at resection, a similar team decision was made with regard to IORT. The patient had a secondary reexploration in an operating room in the Radiation Oncology Department (9, 12), preferably in 1-2 days, but occasionally as late as 7 days. The area at risk was reexposed, and the patient was transferred within a sterile corridor to the linear accelerator room where the IORT boost was delivered. As discussed previously, IORT is used in our institution as a “boost” dose to supplement external radiation (9, 11, 12). The IORT dose is calculated at the 90% isodose and is usually dependent on the amount of disease remaining after attempts at resection: microscopic residual
Volume 25, Number 4. 1993
(res[mJ)-1000-1250 cGy, gross residual (res[g]) 2 2 cm-l 500 cGy, unresected or res(g) 2 2 cm-l 7502000 cGy. The size and shape of the lucite cylinders (cones) used for IORT are dependent on location of the IORT field with the intent of including tumor or tumor bed with at least a 1 cm margin (i.e., 5 cm tumor necessitates 2 7 cm cylinder). For all pelvic locations, circular cones with 15” or 30” bevels were utilized to approximate the anatomy of the presacrum, pelvic sidewall or anterior pelvis. With the 30” bevel cones, the depth of the isodose curves is more shallow at the heel end of the cone than the toe end (20). The radiation oncologist must be aware of these differences in determining placement of the treatment cone relative to tumor bed or residual tumor. For extrapelvic lesions, rectangular and eliptical cones with flat or 20” bevel ends were most frequently used. IORT characteristics are shown in Table 2. Electron energy varied from 9 to 18 MeV with the lower energies of 9 and 12 MeV used after gross total resection or with residual disease of minimal thickness (8 of 12 fields in patients who presented with primary disease). The 15 and I8 MeV energies and doses of 1500-2000 cGy were used more commonly in patients with recurrent disease. Twenty-four IORT fields were treated in the 20 patients ( I9 of 20 patients had lesions in the retroperitoneum and the other had a recurrent desmoid in the low neck). Abutting fields were used in 1 patient, two separate areas were treated in 2. In I other patient, a shrinking field IORT technique was used to deliver 1000 cGy with a large cone to both gross residual tumor and an area of gross total removal but narrow or microscopically positive margins, and a smaller cone was used to encompass the gross residual tumor for an additional 500 cGy (total dose within second IORT field of 1500 cGy). Potential dose-limiting organs within an IORT field included rectum (2 patients), ureter (3 patients), bladder (1 patient), peripheral nerve, spinal cord, and vessels. Peripheral nerve and vessels are usually within paravertebral, pelvic sidewall, and presacral fields. In para-aortic fields, vessels are always present and spinal cord and peripheral nerve are often present (dependent on width of fields-if field width exceeds vertebral body width, will have nerve roots deep to the field within the psoas muscle). When spinal cord may be in the IORT field, the dose contribution from IORT is limited to I 500 cGy by proper choice of electron beam energy. In such patients, the external beam dose to spinal cord is limited to 3600-3960 cGy in 180 cGy fractions with the aid of multiple field techniques. Follow-up
Most patients returned for scheduled follow-up at 3month intervals for 2 years, 6-month intervals for 3 years, then yearly. Both patterns of failure and treatment-related complications were coded prospectively using criterion developed by the NC1 IORT contract group of which we were a member (30). CT scans of the abdomen and pelvis
Intraoperative and external RT. sarcoma 0 L. L. GUNDERSON et al.
651
Table 3. Survival and disease status Survival (mo)
No.
Status Primary Alive (5) NED With disease Dead (5) With disease Recurrent Alive (9) NED With disease Dead (1) With disease
4 1
24, 26. 32, 33 1st (DM)
5
9, 11, 35, 37; (DM), 18 (RF)
7 (l)* 2*
15, 16, 18.5. 19, 45, 45 (CF + LF)*, 69 16 (LF + PS); 64.5 (LF)
1
40 (RF)
* Third patient had CF + LF-salvaged with hemipelvectomy, NED 34 months later. CF = Central failure or disease progression within IORT field; LF = Local failure in external beam field: RF = Regional failure outside irradiation field, DM = Distant metastases; PS = Peritoneal seeding.
with contrast and CT scans of the chest were performed at 6-month intervals. Disease progression within the IORT
Sarcoma IORT--Mayo SURVIVAL--ALL DEATHS
field (marked by surgical clips) was defined as central failure (CF) and within the external beam field as local failure (LF). Distant failures (DF) were separated into blood born
distant metastases (DM) and peritoneal seeding (PS). At time of progression, solitary areas of failure were resected, if feasible, and other treatment was individualized using chemotherapy, irradiation or combined modalities. Chemotherapy was never used as initial treatment, but only at time of disease progression. 20
Statistical analyses Complete follow-up data were available on all patients for 2 15 months from initiation of treatment for this malignancy (i.e., for patients who presented with recurrence, duration was measured from initiation of treatment for the recurrent disease). Overall survival time was calculated from initiation of treatment for this malignancy (surgery or start of preoperative irradiation) to the date of the most recent follow-up or to time of death using the Kaplan and Meier method (14). Disease-free survival was measured from initiation of treatment to the date of last follow-up or to time of disease progression. RESULTS Survival and disease status Survival and disease status are shown in Table 3. Six of 20 patients died (9, 11, 18, 35, 374, and 40 months)four with metastatic disease. Of the 10 patients who presented with primary disease, 5 (50%) were alive and 4 (40%) were disease-free (24, 26, 32, and 33 months). Of the 10 who presented at time of recurrence, 9 (90%) were alive with 7 currently free of disease (15, 16, 18.5, 19,45, 45, and 69 months). One of 7 required surgical salvage with hemipelvectomy and is disease-free 34 months after that procedure. Of the 6 patients with preoperative irradiation prior to an attempt at resection, 4 have died of
Ahveat intelval
Prlmaq Recurrent Totat
O-
18
6 5 11
1 5 6
0 2 2
I
J
1
2
3
4
5
I,”
1
0
I
,
0 2 2
Years (a)
DISEASE-FREE
SURVIVAL
Total
2
I
40
Primary (n = 10)
cz *O -Primary - Recurrent I
0’ 0
6 10 I
6 5 I
1
I 2
1 5 I
I 3
Years (b) Fig. 1. Kaplan Meier survival curves for locally advanced, pri-
mary and recurrent soft tissue sarcoma and recurrent desmoid treated with external beam irradiation, surgical resection and IORT. (a) Overall survival, all deaths included (primary, recurrent, total group). (b) Disease free survival.
652
1. J. Radiation Oncology 0 Biology 0 Physics
disease progression, 3 with distant metastases (each had a high-grade lesion). Overall and disease-free survival curves are shown in Figure 1. Actuarial survival at 2.5 years (Fig. la) was 83%, 69% in patients who presented with primary disease and 100% in those presenting with recurrence. Actuarial 5 year survival for the entire group was 48.5%. When patients who died without evidence of disease were censored, 2.5 year survival was 89% for the total group and 78 and lOO%, respectively, for primary versus recurrent disease presentation (curves not shown). With regard to diseasefree survival, patients who presented for treatment at the time of a localized recurrence have done at least as well as those with primary lesions (Fig. lb).
Volume 25,
Number
4, 1993
Impact of histological grade on the incidence of distant failure and total failures (any disease progression) is shown in Table 7. Distant failures appeared to be more frequent with high grade lesions (5/12 or 42% vs. l/8 or 13%).
DISTANT METASTASIS-FREE SURVIVAL ____-___-_______-___----_ Recurrent (n = 10)
Total group
8 c”-
60 -
‘5 ‘2
40
Primary (n = 10)
-
cz
Incidence and patterns offailure
Incidence and patterns of failure were evaluated in detail (Fig. 2 and Table 4). Disease progression occurred in 10 of 20 patients or 50% (primary 6/ 10, recurrent 4/ 10). Central failure (CF) in the IORT boost was documented in only one patient (5% of patients at risk or 10% of failures) and local failures (LF) in the external beam field in three (O/ 10 primary; 3/ 10 recurrent-30%). Hemipelvectomy for salvage was performed in 1 patient with both CF and LF who is currently free of disease. Distant metastases occurred in 5 patients (5/ 10 primary-50%; O/ 10 recurrent). None were amenable to resection. Peritoneal seeding was documented in only 1 patient (5%) but this estimate may be falsely low because of a low incidence of reoperations or autopsies. Ascites on serial CT’s or physical exam has not been observed, however. Disease control curves are shown in Figure 2. Patterns of failure differed markedly between primary and recurrent disease patients, but part of that difference may be due to histological factors. The incidence of both total and distant failure was higher in primary versus recurrent disease presentation (Table 4). All blood born distant metastases were documented in the primary disease group but 4 of 5 patients had high grade lesions. All three failures within the irradiation field were in the group who presented at the time of recurrence. However, if the 2 patients with marginal external fields are censored, then the local failure risk in the recurrent group becomes I of 8 or 12.5%. The impact of resection on local failure (within external irradiation field) and total failure is shown in Table 5. A trend for a decrease in the total failure incidence with gross total resection versus partial resection is suggested only if the 2 patients with marginal external fields are not included in the analysis. An expanded analysis is shown in Table 6 with regard to the 5 patients with tumor progression within or at the margin of irradiation fields. One regional and one local failure may have been prevented with larger fields (optimal fields could not be used in the patient with subsequent regional failure in view of previous external irradiation for seminoma and the patient with local failure also had limited peritoneal seeding found at exploration).
20r-
6 10
Primary Recurrent
1
0
6 5
1 5
2
3
Years (a)
ABDOMINAL CONTROL 100
1
1
Primary (n = 10)
60 8 !F ‘5 .-
60
-
2
40 -
Recurrent (n = 10)
cz 2o
-
Primary Recurrent
0 10
I
I
0
I
I
1
0
1 5
6 5
I
I 3
2
Years (b)
LOCAL CONTROL Primary (n = 10)
Total
Recurrent (n = 10)
2o 0
. 0
Primary Recurrent
I
6 10 I
1 5
6 5
I
I
1
2
I
I 3
Years (c) Fig. 2. Disease control curves. (a) Local control (in radiation field-external or IORT)-one patient required salvage with hemipelvectomy. (b) Abdominal control. (c) Distant control (free of blood born or peritoneal metastases).
lntraoperative and external RT, sarcoma 0 L. L. GUNDERSON
653
et al.
Table 4. Incidence and patterns of failure Patterns of failure (any component) CF
LF
DF
RF
Failures Site
No. %
No. (%)
No. (%)
No. (%)
No. (%)
Total group Primary Recurrent
10/20 (50) 6/lO (60) 4/ 10 (40)
1 (5) 0 1 (lo)+
3 (15) 0 3 (30)+
6 (30)* 5 (50) I (lo)*
2 (10) 1 (10) 1 (10)
See Table 3 for CF, LF, and RF, DF-distant failure either hematogenous or peritoneal (W) Are of total group at risk. * PS in conjunction with LF (PS component in 5% at risk). + CF + LF was salvaaed with hemipelvectomv. both external + IORT fields were marginal: a second recurrent patient with LF also had marginal external beam fields (EW). _
Complications Significant complications occurred in 6 patients. The IORT surgical procedure or IORT electron boost were
implicated as a potential causative factor in 5 of the 6 (nerve- 1, wound-2, urologic- 1, fibrosis- 1). Severe (Grade 3 or 4) symptomatic or objective neuropathy occurred in 1 patient (5%) who was treated with abutting IORT fields. This was a Grade 3 motor neuropathy which resulted in foot drop but caused no pain. At the time of analysis this patient was free of disease within the IORT fields but had local failure in the opposite side of the pelvis. Another patient had a Grade 2 neuropathy requiring occasional pain medications. Small bowel obstruction (SBO), related to surgery and external irradiation, occurred in 2 patients. This resolved with conservative measures during hospitalization in 1 patient (5%), but reoperation was required in the other (5% of 20 at risk). DISCUSSION When surgery is the sole treatment modality for retroperitoneal sarcomas, the incidence of subsequent local
Table 5. Impact of resection on local and total failure Failure patternamount of residual Local failure Microscopic residual (gross total removal) Gross residual Totals Any failure Microscopic residual Gross residual Totals
Primary No.
Recurrent
Total
No. %
No. %
O/5 O/5 O/IO
2/6* 114 3/10 (30)
2/11 (18)+ l/9 (11) 3/20 (15)
315 315 6/10 (60)
2/6 214 4/10 (40)
5/l 1 (45)+ 5/9 (56) 10/20 (50)
* l-marginal external field due to previous 3000 cGy for seminoma (non-Mayo); l-marginal IORT field (Mayo). + Becomes O/9 LF and 3/9 (33%) any failure for res(m) if the two patients with marginal external fields are censored for this analysis.
recurrence has been as high as 70-90% ( 1, 2, 15, 16, 2 1, 32, 34). If gross total resection can be accomplished, the risk of subsequent local recurrence in patients who presented with primary tumors was 60% (37/63) in a surgical series from Mayo (2) and 37% (15/41) from Memorial (13). The latter incidence is falsely low as only patterns of initial failure were reported. When patients presented at the time of recurrence and had a gross resection of tumor, Memorial reported a subsequent local recurrence incidence of 59% (23/39) as initial site of failure. Neither the Mayo nor Memorial series, however, represent surgery alone data. In the Mayo series, 18 of the 63 with complete resection received adjuvant therapy (irradiation - 15, chemotherapy - 1, both -2). At Memorial, 18 of 4 1 patients with complete resection of primary lesions received postoperative therapy, but no information was given concerning added treatment in the 39 with complete resection of locally recurrent lesions. Survival at 5 years in surgery alone series has varied from 25-50% after complete resection and from 5-20% with subtotal resection. External beam irradiation has been combined with surgical resection with or without chemotherapy for locally advanced soft tissue sarcomas of the trunk and extremities. Although local control has appeared to improve when external irradiation is combined with surgical resection for retroperitoneal sarcomas (4, 15, 3 1, 34) the dose of irradiation that can be delivered safely is much lower than with extremity sarcomas (22. 29) in view of dose limiting structures including small intestine, stomach, liver, kidney, and spinal cord. In published series from the NCI, both acute and chronic small bowel morbidity were excessive in the treatment arms using moderate to high doses of external irradiation with no IORT (4, 18). The National Cancer Institute (NCI) conducted a randomized trial from 1980- 1985 in which 35 patients with retroperitoneal sarcomas were randomized to receive external irradiation + IORT ( 18). All patients had primary lesions, and none had received prior irradiation or chemotherapy. All had gross total resection, but most had microscopic residual. Patients randomized to external alone received 3500 to 4000 cGy to an extended field over 4-5 weeks and an additional 1500 cGy over 2 weeks to a
I. J. Radiation Oncology 0 Biology 0 Physics
654
Volume 25. Number 4, 1993
Table 6. Expansion of central (CF), local (LF) and regional failures* Radiation factors Presentation
Original size (cm) amount residual
histology
External (cGy/fractions)
IORT (size/dose/energy)
Failure* pattern
Comment
D. M. Ret R iliac fossa (2nd) Gr 4 leiomyosarc
8.5 x 8 x 5 res (m)
5.0, 30” 1500,9E
CF-LF
5040128
S. S. Ret R retroperit (1st) Gr 3 liposarc
11 x 1.5 res (m)
8 x 15,20” 1500, 12E
LF-PS
4554121
S. G. Ret L pelvis (2nd) Gr 2 chondrosarc C.H. Ret L pelvis (3rd) Gr 2 chondrosarc J. P. Pri R retroperit Gr 3 leiomyosarc
10.5 x 9 X 8.5 1) res (g) 2) res (m)
(I)* 5.5, 2000 18E
LF
5340/30
(2)* 6.0, 1500 12E
10X8x7 res (g)
8.0, 15” 2000, 18E
RF
5200126
R inguinal DOD 40 mo
10 x 5 x 3.5 res (m)
7 x 12, 20” 1900, 9E
RF
4500125
Ext RT (EW) marginal 2” previous ext. RT for seminoma; DOD 18mo
IORT field marginal NED 34 mo after hemipelvectomy (45 mo total) External RT marginal (EW) Alive 16 mo R pelvis_4440/25Fx Alive 64.5 mo
* See Table 2 for definition of failure patterns. Ret = Recurrent; Pri = Primary; Sarc = Sarcoma; R = Right, L = left; Res (m) = Microscopic residual; Res (g) = Gross residual; E = Electron energy in MeV; NED = No evidence of disease; DOD = Dead of disease; Fx = fractions; Ext RT = External irradiation; EW = Elsewhere. + Abutting fields with maximum 0.5 cm overlap.
reduced field. The IORT group received 3500-4000 cGy in 4-5 weeks to an extended field and an IORT dose of 2000 cGy to abutting fields (2-6 abutting fields) plus intravenous misonidazole. Some apparent advantages for IORT patients were seen in the NC1 trial with a minimum follow-up of 15 months. Treatment-related small bowel complications were substantially lower in the IORT group. Severe acute enteritis occurred in 12 of 20 versus 0 of 15 patients (p < 0.05), chronic enteritis in 7 of 20 versus 2 of 15 (p < 0.05), and fistulae in 6 of 20 versus 0 of 15 (p < 0.05). Actuarial disease control within irradiation fields at 4-5 years was 78% with IORT plus external irradiation versus 30% with external irradiation alone. In view of small patient numbers and limited follow-up, these differences did not reach
statistical significance in the initial publication (18). In an updated analysis, local/regional failure in the tumor bed or abdomen occurred in 6 of 15 IORT (40%) versus 15 of 20 non-IORT patients (75%) which achieved borderline statistical significance at p = 0.08 (24). In the updated analysis, gastrointestinal complications were again statistically less in the IORT patients at 2/ 15 versus 9/20 patients (p < 0.05). With marginal resection of retroperitoneal sarcomas or proven microscopic residual disease, doses of 6000-7000 cGy would usually be necessary in order to achieve local control with a frequency of 80-90% (22, 28, 29). Such doses could not be safely achieved with external beam approaches for retroperitoneal sarcomas. Even with external doses to only 5500 cGy in the NC1 series, acute
Table 7. Impact of histological grade on distant and total failure Distant failure (DF)
Grade
112 314
Total failures
Total
With local control
No. (%)
No. (%)
l/8* (13) 5/12 (42)
l/7 4/10
6/20’ (30)
5/17* (28)
Totals
(14) (40) Totals Sarcoma Desmoid
Primary
Recurrent
l/2 518
214 214
6/10 -
o/2
* Includes two patients with desmoid - O/2 DF. + Total failure for low grade lesions 3/8 (37.5%) if desmoids combined with Gr 1, 2. * DF = DM (5) + PS (1).
418
No. (%) 3/6 (50)+ 7/12 (58) lo/18 (56) o/2+
lntraoperative and external RT, sarcoma 0 L. L. GUNDERSON
and chronic small bowel complications were excessive. Although improvements in survival were not seen in this randomized trial, it is appropriate to consider the combined IORT external beam approach as preferable treatment to external beam alone in view of nonexistent chronic small bowel risks and improved local tumor control. In the present series, similar optimistic results were seen regarding tumor control within irradiation fields with the same minimum follow-up of 15 months as in the NC1 series. In patients who presented with primary disease, no local failures had occurred within either the external beam or IORT field at the time of this analysis. Local failure had occurred in 3 of 10 patients in the group who presented at the time of local recurrence. Our analysis has shown no definite relationship between the amount of residual remaining after maximal resection and subsequent disease progression within irradiation fields. However, the IORT dose in our institution is based on the amount of residual disease remaining after maximal surgical resection (1000-l 250 cGy with 5 microscopic residual, 1500-2000 cGy for gross residual or unresected disease). Doses less than 1500 cGy are rarely used for recurrent lesions because of the greater potential for tumor bed hypoxia from multiple operations. Failures within or at the margin of external or IORT irradiation fields were probably related to inadequate fields in 3 patients, 2 received the external component of irradiation elsewhere (Table 3). In 1 patient external fields were marginal because of a previous course of irradiation for seminoma and in the second were due to inaccurate reconstruction of tumor volumes from computed tomography. As with extremity sarcomas, it would be reasonable to have initial field margins 5-7 cm beyond unresected or residual tumor or clipped areas of tumor adherance (i.e., achieve those margins relative to potential microscopic spread along muscle, nerve, and vessel). It may be safer to treat these extended margins to 4500 cGy in 180 cGy fractions with preoperative rather than postoperative irradiation. Preoperative irradiation may reduce the incidence of severe complications because the tumor is helping to displace the dose-limiting small intestine. After resection of large retroperitoneal sarcomas, small bowel may become immobile because of adherence to denuded surfaces thus increasing the risk of acute or chronic treatment-related enteritis. Small bowel morbidity was acceptable in the current series with an incidence of
et al.
655
small bowel obstruction requiring reoperation of only 5% (l/20). With regard to IORT field design, the entire tumor bed is not intentionally treated in our institution. As previously noted, that is treated with fractionated external beam techniques to 4500-5040 cGy, and IORT doses of lOOO2000 cGy are used as a boost to areas of positive or narrow resection margins. If unresected or gross residual exists after maximal surgical resection, it must be encompassable within a single IORT cylinder. Abutting IORT fields are considered only in the context of a gross total resection. Blood-born metastases occurred with higher frequency with primary vs recurrent disease presentation in our se-
ries. This difference was likely due to more high-grade lesions (i.e., 8 Grade 3 or 4 primary lesions versus 4 in the recurrent group). The incidence of distant metastases was 13% in grade 1 and 2 lesions (l/S) versus 42% with Grade 3 and 4 (5/ 12). We will be attempting to integrate chemotherapy regimens in future IORT treatment strategies for high-grade lesions. With regard to survival, we were intrigued that patients who presented at the time of recurrence have done as well as those who had primary lesions. This finding may be related in part to a lower incidence of distant metastases (only 4 of 10 patients in the recurrent group had highgrade lesions). However, it is encouraging that aggressive approaches seem effective and appropriate for carefully selected patients with localized recurrence in addition to those with locally advanced primary disease. The aggressive techniques used in this series of patients appear promising. All patients in this series had either proven microscopic residual disease after gross total resection of locally advanced trunk sarcomas or gross residual. Most would have been expected to recur locally and die of their malignancy with only a surgical approach. We are encouraged by the excellent disease control within irradiation fields, the acceptable tolerance with regard to small bowel and peripheral nerve (17, 19, 23-27), and the favorable early survival. Further evaluation of IORT combined with external irradiation and maximal resection seems indicated for locally advanced primary and recurrent retroperitoneal sarcomas on the basis of this and other IORT series (18, 33). Attempts dt integration of chemotherapy in a protocol setting are probably indicated for grade 3 and 4 lesions in view of metastatic risks of 40% or greater. While randomized trials would be of interest, patient numbers may be too small to allow this to occur.
REFERENCES 1. Cody, H. S.; Turnbull, A. D.; Former, J. G.; Hadju, S. I. The continuing challenge of retroperitoneal sarcomas. Cancer 47:2147-2152;1981. 2. Dalton, R. R.; Donahue, J. H.; Mucha, P.; Van Heerden, J. A.; Reiman, H. M.; Chen, S. Management of retroperitoneal sarcomas. Surgery 106:725-733;1989. 3. Gallagher, M. J.; Brereton, H. D.; Restock, R. A.; Zero, J. M.; Zekoski, D. A.; Poyss, L. F.; Richter, M. P.; Kligermann, M. M. A prospective study of treatment techniques to minimize the volume of pelvic small bowel with reduction
of acute and late side effects associated with pelvic irradiation. Int. J. Radiat. Oncol. Biol. Phys. 12:1565-1573;1986. 4. Glenn, J.; Sindelar, W. F.; Kinsella, T.; Glatstein, E.; Tepper, J.; Costa, J.; Baker, A.: Sugarbaker, P.; Brennan, M. F.; Seipp, C.; Wesley, R.; Young, R. C.; Rosenberg, S. A. Results of multimodality therapy of resectable soft tissue sarcomas of the retroperitoneum. Surgery 97:316-324;1984. 5. Green, N.; ha, G.; Smith, W. R. Measures to minimize small intestine injury in the irradiated pelvis. Cancer 35: 1633-1640:1975.
656
I.
J. Radiation Oncology 0 Biology0 Physics
6. Gunderson, L. L.; Cohen, A. M.; Dosoretz, D. E.; Shipley, W. V.; Hedberg, S. A.; Wood, W. C.; Rodkey, G. V.; Suit, H. D. Residual, unresectable, or recurrent colorectal cancer: External beam irradiation and intraoperative electron beam boost + resection. Int. J. Radiat. Oncol. Biol. Phys. 9: 15971606;1983. 7. Gunderson, L. L.; Cohen, A. M.; Welch, C. W. Residual, inoperable, or recurrent colorectal cancer: Surgical radiotherapy interaction. Am. J. Surg. 1395 18-525; 1980. 8. Gunderson, L. L.; Martenson, J. A. Gastrointestinal tract radiation tolerance. Front. Radiat. Ther. Oncol. 23:277298;1989. 9. Gunderson, L. L.; Martin, J. K.; Earle, J. D.; Byer, D.; Voss, M.; Fieck, J.; Kvols, L.; Rorie, D.; Martinez, A.; Nagomey. D. M.; O’Connell, M. J.; Weber, F. Intraoperative and external beam irradiation & resection: Mayo pilot experience. Mayo Clin. Proc. 59:691-699;1984. 10. Gunderson, L. L.; Russell, A. H.; Llewellyn, H. T.; Dopke, K. P.; Tepper, J. E. Treatment planning for colorectal cancer: Radiation and surgical techniques and value of small bowel films. Int. J. Radiat. Oncol. Biol. Phys. 11:1379-1393;1985. 11. Gunderson, L. L.; Shipley, W. U.; Suit, H. D.; Epp, E. R.: Nardi, G.; Wood, W.; Cohen, A.; Nelson, J.; Battit, G.; Biggs, P. J.; Russell, A.; Rockett, A,; Clark, D. Intraoperative irradiation: A pilot study combining external beam irradiation with “boost” dose intraoperative electrons. Cancer 49:22592266; 1982. 12. Gunderson, L. L.; Tepper, J. E.; Biggs, P. J.; Goldson, A.: Martin, J. K.; McCullough, E. C.; Rich, T. A.; Shipley, W. U.; Sindelar, W. F.; Wood, W. C. Intraoperative +- external beam irradiation. Curr. Probl. Cancer 7: l-69; 1983. 13. Jaques, D. P.; Coit, D. G.; Hajdu, S. I.; Brennan, M. D. Management of primary and recurrent soft tissue sarcoma of the retroperitoneum. Ann. Surg. 2 12:5 l-59: 1990. 14. Kaplan, E. L.; Meier, P. Nonparametric estimation for incomplete observations. J. Am. Stat. Assoc. 53:457-48 1;1958. 15. Karakousis, C. P.; Velez, A. F.; Einrich, L. J. Management of retroperitoneal sarcomas and patient survival. Am. J. Surg. 150:376-380; 1985. 16. Kinne, D. W.; Chu, F. C. H.; Huvos, A. G.; Yagoda, A.; Fortner, J. G. Treatment of primary and recurrent retroperitoneal liposarcoma: Twenty-five year experience at Memorial Hospital. Cancer 3 1:53-64; 1973. 17. Kinsella, T. J.; Sindelar, W. F.; DeLuca, A. M.; Pezeshkpair, G.; Smith, R.; Maher, M.; Terrill, R.; Miller, R.; Mixon, A.; Harwell, J. F.; Rosenberg, S. A.; Glatstein, E. Tolerance of peripheral nerve to intraoperative radiotherapy (IORT): Clinical and experimental studies. Int. J. Oncol. Biol. Phys. 11:1579-1585;1985. 18. Kinsella, T. J.; Sindelar, W. F.; Lack, E.; Glatstein, E.; Rosenberg, S. A. Preliminary results of a randomized study of adjuvant radiation therapy in resectable adult retroperitoneal soft tissue sarcomas. J. Clin. Oncol. 6: 18-25; 1988. 19. LeCouteur, R. A.; Gillette, E. L.; Powers, E. L.; Child, G.; McChesney, S. L.; Ingram, J. T. Peripheral neuropathies following experimental intraoperative radiation therapy (IORT). Int. J. Radiat. Oncol. Biol. Phys. 17:583-590;1989.
Volume 25. Number 4. 1993 20. McCullough, E. C.; Anderson, J. A. The dosimetric properties of an applicator system for intraoperative electron beam therapy utilizing a clinic 18 accelerator. Med. Phys. 9:261-268;1982. 2 I. McGrath, P. C.; Neifeld, J. P.: Lawrence, W.: DeMay, R. M.; Kay, S.; Horsley, J. S.; Parker, G. A. Improved survival following complete excision of retroperitoneal sarcomas. Ann. Surg. 200:200-204:1984. 22. Schray, M. D.; Gunderson, L. L.; Sim, F. H.: Pritchard, D. J.; Shives, T. C.; Yeakel, P. D. Soft tissue sarcomas. Integration of bradytherapy, resection, and irradiation. Cancer 66:45 l-456; 1990. 23. Shaw, E. G.; Gunderson, L. L.; Martin, J. K.; Beart, R. W.; Nagorney, D. M.; Podratz, K. C. Peripheral nerve and ureteral tolerance to intraoperative radiation therapy. Clinical and dose response analysis. Radiother. Oncol. 18:247255:1990. 24. Sindelar, W. F.; Kinsella, T. J.; Chen, P. W.; DeLaney, T. F.; Tepper, J. E.; Rosenberg, S. A.; Glatstein. E. Intraoperative radiotherapy in sarcomas of the retroperitoneum: Final results of a prospectively randomized clinical trial. (in press). 25. Sindelar, W. F.; Kinsella, T. J.; Hoekstra, P., et al. Treatment complications in intraoperative radiotherapy. Int. J. Oncol. Biol. Phys. 11:117; 1985. 26. Sindelar, W. F.; Kinsella, T.; Tepper, J.; Travis, E. L.; Rosenberg, S. A.; Glatstein, E. Experimental and clinical studies with intraoperative radiotherapy. Surg. Gyn. Obstet. 157: 205-219;1983. 27. Sindelar, W. F.; Tepper, J.: Travis, E. L.; Terrill. R. Tolerance of retroperitoneal structures to intraoperative irradiation. Ann. Surg. 196:601-608; 1982. 28. Suit, H. D. Radiation biology: A basis for radiotherapy. In: Fletcher, G. H., ed. Textbook of radiotherapy, 2nd edition. Philadelphia: Lea and Febiger; 1973: 75- 12 1. 29. Suit, H. D.; Mankin, H. J.; Wood, W. C.; Proppe, K. H. Preoperative. intraoperative, and postoperative radiation in the treatment of primary soft tissue sarcoma. Cancer 55: 2254-2267: 1985. 30. Tepper, J. E.; Gunderson, L. L.; Orlow, E.; Cohen, A. M.: Hedberg, S. E.; Shipley, W. U.; Blitzer, P. H.; Rich, T. Complications of intraoperative radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 10: 183 1- 1839; 1984. 31. Tepper. J. E.; Suit, H. D.; Wood, W. C.; Proppe, K. H.; Harmon, D.; McNulty, P. Radiation therapy of retroperitoneal soft tissue sarcomas. Int. J. Radiat. Oncol. Biol. Phys. 10:825-830:1984. 32. Wile, A. G.: Evans, H. L.; Romsdahl, M. Leiomyosarcoma of soft tissue: A clinicopathologic study. Cancer 48: 10221032;1981. 33. Willett, C. G.; Suit, H. D.; Tepper, J. E.; Mankin, H. J.; Conrery, K.; Rosenberg, A. L.; Wood, W. C. Intraoperative electron beam radiation for retroperitoneal soft tissue sarcoma. Cancer 68:278-283; 199 1. 34. Wist, E.; Sotheim. 0. P.; Jacobsen, A. B.; Blom, P. Primary retroperitoneal sarcoma. Acta. Radio]. Oncol. 24:305310:1985.