Inf. J Rndiarion Onco/og~~ Bid. Phys., Vol. Printed m the U.S.A. All rights reserved.
18, pp.
1019-1025
0360-3016/90 $3.00 + .W Copyright IC) 1990 Pergamon Pres5 plc
??Original Contribution
REDUCTION
OF SPINAL IRRADIATION
METASTASES AFTER PREEMPTIVE IN PROSTATIC CANCER
IRVING D. KAPLAN, M.D.,’ RICCARDO VALDAGNI, M.D.,’ RICHARD S. Cox, PH.D.’ AND MALCOLM A. BAGSHAW, M.D.’ ‘Departmentof Radiation Oncology, StanfordUniversitySchool of Medicine, Stanford,CA 94305; and *Departmentof Radiation Oncology, Centro S. Pio X, via F. Nava, 3 1, 20 159 Milano, Italy For this study, 136 patients treatedat Stanford University Hospita1 for prostatic cancer between 1971 and 1980 were selected for review. The patients had received no prior therapy, and had no evidente of bone metastases at time of radiation treatment based on radiographic studies and bone scan. Of this group, 71 patients received extended-field irradiation (paraaortic and pelvic fields), and 65 patients received pelvic irradiation. The pelvic field was treated to 50 Gy and the paraaortic field received 45 Gy to 60 Gy. Al1 patients subsequently underwent routine follow-up examinations and studies at Stanford University Hospital: 1,513 follow-up X rays, bone stans, and CTstans were analyzed for site-specific recurrence. The follow-up ranged from 14 months to 16 yrs from the time of initial treatment, with a mean follow-up of 7 yrs. Lower extremities and ribs were found to be the most common sites of bone metastases. Irradiation of the lumbar spine to a dose of 35 to 60 Gy, coincidental to irradiation of the paraaortic lymph nodes prevented or delayed the development of lumbar spine metastases. The potential mechanism and clinical implications are discussed. Prostatic cancer, Bone metastases.
INTRODUCI’ION
that metastatic deposits would not develop in irradiated bone.
Radiation therapy has been shown to control primary prostatic neoplasms and to palliate metastatic sites of diseases ( 1,2, 7, 15, 18, 19, 20). Batson postulated that prostatic cancer spreads via a venous plexus that drains the prostate and communicates directly with the paravertebml and intravertebral veins, thus explaining the high incidence of vertebral metastases (3). In the mid 197Os, one of the authors of the present study (Bagshaw) observed that when the lumbar spine received irradiation incidental to paraaortic lymph node treatment, subsequent bone stans were often devoid of increased uptake in the irr& diated field, while other bones showed multiple sites of metastatic disease (Figs. la, b; 2a, b; 3a, b). In other conditions, such as in breast cancer, incidental bone irradiation has been shown to decrease the subsequent development of in-field bone metastases (9). This phenomenon could be explained by the lethal irradiation of subclinical disease-early eradication of the “seed” (5, 11). Alternatively, clinically detectable metastases might be delayed by the effect of radiation on the tumor bed (4, 14)-in other words, modification of the “soil” by radiation so
METHODS
AND MATERIALS
1. Patient characteristics A group of one hundred thirty-six patients treated between 197 1 and 1980 were reviewed for this study. None had received prior treatment. At the initiation of radiation therapy, none of the patients had detectable bone metastases based on radiographic studies and bone stans. The Stanford TNM staging system was used (20). Al1 initial biopsies were graded using the Broders system as modified by Kempson and Levine (13); in one hundred eleven patients, the histopathology was also scored by the Gleason method (6) (Table 1). Al1 patients were followed at Stanford. Of this group, 7 1 patients received paraaortic irradiation which coincidentally included the lumbar spine, in addition to pelvic and prostatic irradiation (extendedfield irradiation). Sixty-five patients were irradiated to the pelvis and prostate only (pelvic irradiation). Many of these patients had laparotomy-proven paraaortic adenopathy and received extended-field irradiation. Because of an ex-
Reprint request to: Irving D. Kaplan, M.D. authorswish to thankJoyce Ramback, MargeKeskin, Pat Bird, and Jody Kaplan for their assistance, coopexation,encouragement,and patience.
Acknowledgements-The
1019
Accepted for publication 16 November 1989.
1.J. RadiationOncology0 Biology0 Physics
1020
May 1990,Volume18, Number5
Fig. 1. Radiation portals of a patient who subsequently developed extensive metastasis to bone disease.
isting protocol, patients with laparotomy-proven pelvic nodes were randomized to receive either extended-field or pelvic irradiation. Lymphangiography was performed on 120 of the 136 patients. Histologie confirmation of lymph node status by biopsy or lymphadenectomy was obtained in 69 patients (Table 2). 2. Radiation doses The patients were irradiated at 4 MV with a linear accelerator.* In the pelvic irradiation technique which has been previously described ( 1, 2, 18) 26 Gy was delivered by four fields to the pelvis, including the lymphatics, the seminal vesicles, and the prostate. Next, the prostatic region was irradiated for 20 Gy (prostatic boost) using a right and left lateral 120-degree moving beam. Then, an additiona124 Gy was delivered to the pelvis by the original four fields. Figure 4 shows a typical treatment plan for the pelvic fields. The in-field pelvic bone received 30-50 Gy. The paraaortic region was treated from L 1 through L4 by a four-field technique. Lateral obliques were used in
* Varian Clinac IV.
addition to anterior and posterior portals to minimize the dose to the kidneys and spinal cord. A skin gap was calculated between the pelvic and paraaortic fields to abut fields without producing an overlap. A typical treatment plan is shown in Figure 4. The prescribed dose to the paraaortic region ranged from 45 Gy-60 Gy. With the exception of two patients who received 20 and 44 Gy, 54 patients received 45-50 Gy, and 15 patients received 50.560 Gy. Typically, the incidental dose to the vertebral bodies ranged from 35-50 Gy. 3. Fellow-up After radiation therapy, patients were followed at Stanford University Hospita1 with periodic physical examination and laboratory, radiographic, and scintigraphic studies. Follow-up ranged from 14 months to 16 years after radiotherapy with a mean follow-up of 7 years. Altogether, 1,5 13 bone stans, plain x-ray examinations, and CT-stans were performed over the course of follow-up in the two groups of patients. Chest and KUB roentgenograms were usually obtained at each examination visit, whereas bone scintigrams were scheduled either yearly or
Reduction of spinal metastases after XRT in prostatic cancer 0 1. D.
KAPLAN
et al.
1021
Fig. 3. (a) The anterior bone scan obtained later in the patient illustrated in Figure 1, which demonstrates the absente of metastatic lesions in the previously irradiated bones. Extensive metastases are found in the ribs, shoulder
girdle, proximal humeri and femora, and peripheral innominate bones. A lack of uptake is demonstrated in the lumbar spine, sacrum, pubis, ischium and femoral necks, areas which had been included in previous radiation fields; (b) Posterior
scan confirms
the anterior.
at symptoms of clinical bone involvement. The incidence of in-field and distant bone metastases was analyzed in both groups of patients. Metastatic disease was scored in ten sites: (a) skull, (b) upper extremity, (c) scapula and clavicle, (d) lower extremity, (e) ribs, (f) pelvis, in-field (g) pelvis, out-of-field, (h) cervical spine, (i) thoracic spine, and (j) lumbar spine. The lumbar spine field included LI to L4; L5 was included in the pelvic irradiation and was scored as “pelvic in-field”. The diagnostic imaging studies were graded as demonstrating metastases as follows: (a) no evidente of disease, (b) new disease (fìrst time documented), (c) stable disease, (d) regression of disease, or, (e) progression of disease. Clinical status was also scored for (a) bone pain, (b) spinal cord compression, (c) impending fracture.
RESULTS
Incidence and progression of bone metastases The actuarial risk of development and rank order of site-specific bone metastases for the entire group of 136 patients is described in Table 3. Lower extremity, ribs, and thoracic spine were the most common sites of bone metastases, whereas upper extremity and the skull occurred least frequently. To test the hypothesis that the development of metastatic disease is delayed in irradiated bone, the incidence of bone metastases in the lumbar spine was compared between patients who received extended-field (including paraaortic) irradiation and those who received pelvic irradiation only.
1. J. Radiation Oncology 0 Biology 0 Physics
May 1990, Volume 18, Number 5
Fig. 3. Diagnostic x-ray examination of the pelvis (Fig. 3a) and the lumbar spine (Fig. 3b) demonstrate osteoblastic metastases in the lumbar spine, sacrum, pubis, descending ischium, and femoral necks.
Tables 1 and 2 demonstrate that the adverse prognostic factors of higher Gleason pattem score, advanced stage, and increased incidence of nodal involvement were more frequent in the patients who received extended-field irradiation. In other words, patients who received paraaortic irradiation were pre-selected because of more advanced disease. To adjust for this disparity between groups, the
paucity of
actuarial freedom from relapse was tested using the interval from time of fust recurrence to relapse at a particular site. Among the 10 skeletal sites examined, the lumbar spine was the only one which demonstrated a significant differente with respect to patients receiving pelvic irradiation or extended-field irradiation. Specifically, the rate of pelvis in-field, pelvis out-of-field, skull, upper extremity,
Reduction of spinal metastases after XRT in prostatic cancer 0 1. D. Table 1. Patient characteristics:
number
of patients
(%)
No.
(%)
0 11 19 35 0
(0) (17) (30) (52) (0)
1 4 9 14 20 4 10 3
(1) (6) (14) (22) (38) (6) (15) (5)
No.
(%)
stage*
T3
T4 Gleason 3 4 5 6 7 8 9 10
1023
et al.
Extended field
Pelvic field
Tumor Til TI Tz
KAPLAN
pattern
* Stanford
10 :: 3
(0) (14) (15) (66) (4)
score 0 0 2 10 13 9 4
(0) (0) (4) (17) (22) (28) (20) (9)
Staging System (20).
shoulder girdle, thoracic spine, cervical spine, rib, and lower extremity recurrences did not differ between patients treated by either technique (Fig. 5). Conversely, the relapse rate, after initial relapse, in the lumbar spine was significantly reduced in those patients receiving radiation to the paraaortic lymph nodes (Fig. 6). The untreated lumbar spine was the fust site of recurrence in 15% of the patients who received pelvic-field irradiation only, as shown in Figure 6. In contrast, the lumbar spine was the first site of metastases in only 3% of the patients who received extended field irradiation. Once metastatic disease developed, subsequent treatment was the same for patients who received extendedfield or pelvic-field irradiation. Some mode of androgen deprivation therapy was given for palliation in 55 out of 61 patients who failed after extended-field irradiation and in 27 out of 32 patients who developed metastatic disease after pelvic-field treatment. Five instances of spinal cord compression occurred in the patients treated with extended-field irradiation, but none of these were in the lumbar spine. Also, five cases of cord compression occurred among the patients treated with the pelvic field only: 4 in the thoracic spine and 1 in the lumbar spine.
Table 2. Lymph
node status Evidente
By lymphangiogram By biopsy
for lymph node metastasis
Pelvic field
Extended
field
12/53 (23%) 13/32 (41%)
53167 (79%) 37/37 (100%)
Fig. 4. Dosimetry of the paraaortic fields, at the leve1 of S2 (bottorn).
fields (top) and also the pelvic
DISCUSSION
The finding that the rate of metastasis is reduced in incidentally irradiated bone has been observed in breast cancer. Hercbergs et al. observed a reduced rate of metastasis to the thoracic spine in women who received irradiation to an intemal mammary field (9). These investigators analyzed sites of bone metastases when a patient presented with metastatic disease. Hazra and Giri reported a pilot study in which a single 8 Gy dose delivered to the pelvic girdle reduced the incidence of osseous metastases to the pelvis (8). This large single fraction, however, led to unacceptable morbidity and the technique was aban-
Table 3. Actuarial risk of development and rank order of site-specific bone metastases 3 years Lower extremity Ribs Thoracic spine Cervical spine Pelvis out-of-field Scapula/clavicle Lumbar spine Pelvis in-field Upper extremity Skull
.237 .226 .223 .212 .186 .177 .159 .138 .080 .073
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
5 years .385 (1) .365 (2) .356 (3) .321 (4) .269 (6) .253 (7) .242 (9) .244 (8) .190(10) .311 (5)
10 years .509 .499 .492 .321 .473 .364 .440 .304 .452 .398
(1) (2) (3) (9) (4) (8) (6) (10) (5) (7)
1024
1. J. Radiation Oncology 0 Biology 0 Physics
LOWER EXTREMITY
May 1990, Volume 18, Number 5
I~.I o.nm
t....~....~....~...,,.,.,,,,,,,,,,,,,,,,,
1
2
3
4
5
6
1
l-
m
2
3
4
THORACIC
2
80
g 0 5
60
5 SPINE
40
6
7 G*u” P-“.Ius 11% * 0.1228
,o
t....,....,..,....,.. 1
2
3
4
5
6
7
1234567
L”““““““l”“““““““““““1
LI
1
2
3
4
5
6
CERVICAL
SPINE
‘?%“”
7 YEARS
Fig. 5. The actuarial freedom from metastases after the first evidente of recurrent disease, using the actuarial method of Kaplan and Meier (12) in six specific sites. Curve 1 represents patients irradiated to the pelvis only. Curve 2 represents patients irradiated to the pelvic and paraaortic regions.
In our study, the bones of the pelvis or the lumbar spine were not the primary targets for radiation. The incidental dose received by bone ranged from 30 to 50 Gy, delivered in 1.2-2.0 Gy fractions, usually over 35-50 days. This fractionation scheme did not lead to significant neutropenia or thrombocytopenia. The mechanism by which this dose of radiation delays the subsequent development of metastatic disease is uncertain. There appear to be at least two possibilities: (a) Micrometastases can be controlled with lower dose irradiation. The lower malignant cel1 number in occult micrometastasis requires less irradiation for sterilization. Furthermore, microscopic aggregates of malignant cells doned.
in a normal cellular environment are more likely to be wel1 oxygenated, making them more radiosensitive (5). (b) Alternatively, the microenvironment of bone could be altered by irradiation. In experimental models, adhesion between tumor cells and endothelium has been shown to be organ-specific ( 16, 17). Solid tumor proliferation is dependent upon the ability of the vasculature to proliferate and supply nutrients and oxygen (2 1). Unlike normal endothelium which has a low labeling index, in experimental models, non-transformed endothelium in tumor vasculature has a high labeling index and a cel1 loss fraction much like the tumors themselves (10). Preemptive irradiation could alter the ability of metastatic prostatic cells
Reduction of spinal metastases after XRT in prostatic cancer 0 1. D.
KAPLAN
et ai.
1025
1”’ “““““““““““““““““‘1todevelopment LUMBAR
1
2
3
4
5
6
cells in bone or inhibit the of tumor stroma and delay the development of clinically detectable metastases (4). Metastases to bone are common in prostate cancer. This study demonstrates that when bones have been irradiated, the subsequent development of bone metastasis is delayed; therefore, preemptive irradiation of bone may have a role in forestalling the development of metastasis. Elective irradiation to areas that are frequently involved with symptomatic metastases could be performed, and might significantly reduce the incidence of vertebral collapse and spinal cord compression, a devastating event for a patient with prostatic cancer. Preemptive irradiation of the spine to a modest but as yet undetermined radiation dose could significantly delay metastases to these critical areas, and have a significant positive impact on the quality of life of patients with high Gleason pattern scores, advanced stage, or lymphadenopathy, who are at high risk to develop metastatic disease. adhere
O,hm P.“.Iu+I 1 “I. 2 00250
SPINE
7
YEARS
Fig. 6. The actuarial freedom from metastases after the first evidence of recurrent disease, using the actuarial method of Kaplan and Meier ( 12). Curve 1 represents patients irradiated to the pelvis only. Curve 2 represents patients irradiated to the pelvic and paraaortic regions. A statistically significant reduction in the incidence of metastatic involvement in the lumbar field was demonstrated among patients receiving extended field irradiation.
to the endothelial
REFERENCES 1. Bagshaw, M. A. Radiation therapy for cancer of the prostate. In: Skinner, D. G., deKernion, J. B., eds. Diagnosis and management of genitourinary cancer. Philadelphia, PA: W. B. Saunders; 1978:355-379. 2. Bagshaw, M. A.; Cox, R. S.; Ray, G. R. Status of radiation treatment of prostate cancer at Stanford University. NCI Monogr. 7:47-60; 1988. 3. Batson, 0. V. The role of the vertebral veins in metastatic processes. Ann. Intern. Med. 16:38-43; 1942. 4. Dvorak, H. F. Tumors: wounds that do not heal-similarities between tumor stroma generation and wound healing. N. Engl. J. Med. 315:1650-1659; 1986. 5. Fletcher, G. H. Subclinical disease. Cancer 53: 1274- 1284; 1984. 6. Gleason, D. F. Histologie grading and clinical staging of prostatic carcinoma. In: Tannenbaum, M., ed. Urologie pathology: the prostate. Philadelphia, PA: Lea and Febiger: 1977:171-197. 7. Hanks, G. E.; Krall, J. M.; Muntz, K. L.; Diamond, J. J.; Kramer, S. The outcome of treatment of 3 13 patients with T-1 (UICC) prostate cancer treated with extemal beam irradiation. Int. J. Radiat. Oncol. Biol. Phys. 14:245-248; 1988. 8. Hazra, T. A.; Giri, S. Prophylactic pelvic girdle irradiation in the treatment of prostatic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 7:8 17-8 19; 198 1. 9. Hercbergs, A.; Wemer, A.; Brenner, H. J. Reduced thoracic vertebrae metastases following post mastectomy parastemal irradiation. Int. J. Radiat. Oncol. Biol. Phys. 11:773-776; 1985. 10. Hirst, D. G.; Denekamp, J.; Hobson, B. Proliferation kinetics of endothelial and tumor cells in three mouse mammary carcinomas. Cel1 Tissue Kinetics 15:25 1-261; 1982. 11. Kaplan, H. S. Hodgkin’s disease. Cambridge: Cambridge University Press; 1972. 12. Kaplan. E. L.; Meier, P. Nonparametric estimation from
13.
14.
15.
16.
17.
18.
19.
20.
21
incomplete observation. J. Am. Stat. Assoc. 53:457-481; 1958. Kempson, R. L.; Levine, G. The relationship of grade to prognosis in carcinoma of the prostate. Front. Radiat. Ther. Oncol. 9:267-273; 1974. Kumar, S.; Arnold, F. Can metastasis be restrained? In: Stoll, B. A., ed. Breast cancer: treatment and prognosis. Palo Alto, CA: Blackwell Scientific Pub.; 1986:287-299. Leibel, S. A.; Hunter, G. E.; Kramer, S. Patterns of care outcome studies: results of the national practice in adenocarcinoma of the prostate. Int. J. Radiat. Oncol. Biol. Phys. 10:401-409: 1984. Netlund, P. A.: Zehen, B. R. Organ-specific adhesion of metastatic tumor cells in vitro. Science 224: 1113-1114; 1984. Nicolson, G. L. Metastatic tumor cel1 attachment and invasion assay utilizing vascular endothelial monolayers. J. Histochem. Cytochem. 30:214-220; 1982. Pilepich, M. V.: Bagshaw, M. A.; Asbell, S. 0.: Hanks, G. E.; Krall, J. M.; Emami, B. N.; Bard, R. H. Definitive radiation therapy in resectable (Stage A2 and B) carcinoma of the prostate-results of a nationwide overview. Int. J. Radiat. Oncol. Biol. Phys. 13:659-663; 1987. Pilepich, M. V.; Krall, J. M.; Sause, W. T.: Johnson, R. J.; Russ, H. H.: Hanks, G. E.; Perez, C. A.; Zinninger, M.; Martz, K. L. Prognostic factors in carcinoma of the prostate-analysis of RTOG study 75-06. Int. J. Radiat. Oncol. Biol. Phys. 13:339-349; 1987. Pistenma, D. A.; Bagshaw, M. A.; Freiha, F. S. Extendedfield radiation for prostatic adenocarcinoma: status report of a limited prospective trial. In: Johnson, D. E., Samuels, M. L., eds. Cancer of the genitourinary tract. New York: Raven Press: 1979:229-247. Thomlinson, R. H.; Gray, L. H. Histologie structures of some human lung cancers and possible implication for radiotherapy. Br. J. Cancer 9:539-549: 1955.
18, pp.
1019-1025
0360-3016/90 $3.00 + .W Copyright IC) 1990 Pergamon Pres5 plc
??Original Contribution
REDUCTION
OF SPINAL IRRADIATION
METASTASES AFTER PREEMPTIVE IN PROSTATIC CANCER
IRVING D. KAPLAN, M.D.,’ RICCARDO VALDAGNI, M.D.,’ RICHARD S. Cox, PH.D.’ AND MALCOLM A. BAGSHAW, M.D.’ ‘Departmentof Radiation Oncology, StanfordUniversitySchool of Medicine, Stanford,CA 94305; and *Departmentof Radiation Oncology, Centro S. Pio X, via F. Nava, 3 1, 20 159 Milano, Italy For this study, 136 patients treatedat Stanford University Hospita1 for prostatic cancer between 1971 and 1980 were selected for review. The patients had received no prior therapy, and had no evidente of bone metastases at time of radiation treatment based on radiographic studies and bone scan. Of this group, 71 patients received extended-field irradiation (paraaortic and pelvic fields), and 65 patients received pelvic irradiation. The pelvic field was treated to 50 Gy and the paraaortic field received 45 Gy to 60 Gy. Al1 patients subsequently underwent routine follow-up examinations and studies at Stanford University Hospital: 1,513 follow-up X rays, bone stans, and CTstans were analyzed for site-specific recurrence. The follow-up ranged from 14 months to 16 yrs from the time of initial treatment, with a mean follow-up of 7 yrs. Lower extremities and ribs were found to be the most common sites of bone metastases. Irradiation of the lumbar spine to a dose of 35 to 60 Gy, coincidental to irradiation of the paraaortic lymph nodes prevented or delayed the development of lumbar spine metastases. The potential mechanism and clinical implications are discussed. Prostatic cancer, Bone metastases.
INTRODUCI’ION
that metastatic deposits would not develop in irradiated bone.
Radiation therapy has been shown to control primary prostatic neoplasms and to palliate metastatic sites of diseases ( 1,2, 7, 15, 18, 19, 20). Batson postulated that prostatic cancer spreads via a venous plexus that drains the prostate and communicates directly with the paravertebml and intravertebral veins, thus explaining the high incidence of vertebral metastases (3). In the mid 197Os, one of the authors of the present study (Bagshaw) observed that when the lumbar spine received irradiation incidental to paraaortic lymph node treatment, subsequent bone stans were often devoid of increased uptake in the irr& diated field, while other bones showed multiple sites of metastatic disease (Figs. la, b; 2a, b; 3a, b). In other conditions, such as in breast cancer, incidental bone irradiation has been shown to decrease the subsequent development of in-field bone metastases (9). This phenomenon could be explained by the lethal irradiation of subclinical disease-early eradication of the “seed” (5, 11). Alternatively, clinically detectable metastases might be delayed by the effect of radiation on the tumor bed (4, 14)-in other words, modification of the “soil” by radiation so
METHODS
AND MATERIALS
1. Patient characteristics A group of one hundred thirty-six patients treated between 197 1 and 1980 were reviewed for this study. None had received prior treatment. At the initiation of radiation therapy, none of the patients had detectable bone metastases based on radiographic studies and bone stans. The Stanford TNM staging system was used (20). Al1 initial biopsies were graded using the Broders system as modified by Kempson and Levine (13); in one hundred eleven patients, the histopathology was also scored by the Gleason method (6) (Table 1). Al1 patients were followed at Stanford. Of this group, 7 1 patients received paraaortic irradiation which coincidentally included the lumbar spine, in addition to pelvic and prostatic irradiation (extendedfield irradiation). Sixty-five patients were irradiated to the pelvis and prostate only (pelvic irradiation). Many of these patients had laparotomy-proven paraaortic adenopathy and received extended-field irradiation. Because of an ex-
Reprint request to: Irving D. Kaplan, M.D. authorswish to thankJoyce Ramback, MargeKeskin, Pat Bird, and Jody Kaplan for their assistance, coopexation,encouragement,and patience.
Acknowledgements-The
1019
Accepted for publication 16 November 1989.
1.J. RadiationOncology0 Biology0 Physics
1020
May 1990,Volume18, Number5
Fig. 1. Radiation portals of a patient who subsequently developed extensive metastasis to bone disease.
isting protocol, patients with laparotomy-proven pelvic nodes were randomized to receive either extended-field or pelvic irradiation. Lymphangiography was performed on 120 of the 136 patients. Histologie confirmation of lymph node status by biopsy or lymphadenectomy was obtained in 69 patients (Table 2). 2. Radiation doses The patients were irradiated at 4 MV with a linear accelerator.* In the pelvic irradiation technique which has been previously described ( 1, 2, 18) 26 Gy was delivered by four fields to the pelvis, including the lymphatics, the seminal vesicles, and the prostate. Next, the prostatic region was irradiated for 20 Gy (prostatic boost) using a right and left lateral 120-degree moving beam. Then, an additiona124 Gy was delivered to the pelvis by the original four fields. Figure 4 shows a typical treatment plan for the pelvic fields. The in-field pelvic bone received 30-50 Gy. The paraaortic region was treated from L 1 through L4 by a four-field technique. Lateral obliques were used in
* Varian Clinac IV.
addition to anterior and posterior portals to minimize the dose to the kidneys and spinal cord. A skin gap was calculated between the pelvic and paraaortic fields to abut fields without producing an overlap. A typical treatment plan is shown in Figure 4. The prescribed dose to the paraaortic region ranged from 45 Gy-60 Gy. With the exception of two patients who received 20 and 44 Gy, 54 patients received 45-50 Gy, and 15 patients received 50.560 Gy. Typically, the incidental dose to the vertebral bodies ranged from 35-50 Gy. 3. Fellow-up After radiation therapy, patients were followed at Stanford University Hospita1 with periodic physical examination and laboratory, radiographic, and scintigraphic studies. Follow-up ranged from 14 months to 16 years after radiotherapy with a mean follow-up of 7 years. Altogether, 1,5 13 bone stans, plain x-ray examinations, and CT-stans were performed over the course of follow-up in the two groups of patients. Chest and KUB roentgenograms were usually obtained at each examination visit, whereas bone scintigrams were scheduled either yearly or
Reduction of spinal metastases after XRT in prostatic cancer 0 1. D.
KAPLAN
et al.
1021
Fig. 3. (a) The anterior bone scan obtained later in the patient illustrated in Figure 1, which demonstrates the absente of metastatic lesions in the previously irradiated bones. Extensive metastases are found in the ribs, shoulder
girdle, proximal humeri and femora, and peripheral innominate bones. A lack of uptake is demonstrated in the lumbar spine, sacrum, pubis, ischium and femoral necks, areas which had been included in previous radiation fields; (b) Posterior
scan confirms
the anterior.
at symptoms of clinical bone involvement. The incidence of in-field and distant bone metastases was analyzed in both groups of patients. Metastatic disease was scored in ten sites: (a) skull, (b) upper extremity, (c) scapula and clavicle, (d) lower extremity, (e) ribs, (f) pelvis, in-field (g) pelvis, out-of-field, (h) cervical spine, (i) thoracic spine, and (j) lumbar spine. The lumbar spine field included LI to L4; L5 was included in the pelvic irradiation and was scored as “pelvic in-field”. The diagnostic imaging studies were graded as demonstrating metastases as follows: (a) no evidente of disease, (b) new disease (fìrst time documented), (c) stable disease, (d) regression of disease, or, (e) progression of disease. Clinical status was also scored for (a) bone pain, (b) spinal cord compression, (c) impending fracture.
RESULTS
Incidence and progression of bone metastases The actuarial risk of development and rank order of site-specific bone metastases for the entire group of 136 patients is described in Table 3. Lower extremity, ribs, and thoracic spine were the most common sites of bone metastases, whereas upper extremity and the skull occurred least frequently. To test the hypothesis that the development of metastatic disease is delayed in irradiated bone, the incidence of bone metastases in the lumbar spine was compared between patients who received extended-field (including paraaortic) irradiation and those who received pelvic irradiation only.
1. J. Radiation Oncology 0 Biology 0 Physics
May 1990, Volume 18, Number 5
Fig. 3. Diagnostic x-ray examination of the pelvis (Fig. 3a) and the lumbar spine (Fig. 3b) demonstrate osteoblastic metastases in the lumbar spine, sacrum, pubis, descending ischium, and femoral necks.
Tables 1 and 2 demonstrate that the adverse prognostic factors of higher Gleason pattem score, advanced stage, and increased incidence of nodal involvement were more frequent in the patients who received extended-field irradiation. In other words, patients who received paraaortic irradiation were pre-selected because of more advanced disease. To adjust for this disparity between groups, the
paucity of
actuarial freedom from relapse was tested using the interval from time of fust recurrence to relapse at a particular site. Among the 10 skeletal sites examined, the lumbar spine was the only one which demonstrated a significant differente with respect to patients receiving pelvic irradiation or extended-field irradiation. Specifically, the rate of pelvis in-field, pelvis out-of-field, skull, upper extremity,
Reduction of spinal metastases after XRT in prostatic cancer 0 1. D. Table 1. Patient characteristics:
number
of patients
(%)
No.
(%)
0 11 19 35 0
(0) (17) (30) (52) (0)
1 4 9 14 20 4 10 3
(1) (6) (14) (22) (38) (6) (15) (5)
No.
(%)
stage*
T3
T4 Gleason 3 4 5 6 7 8 9 10
1023
et al.
Extended field
Pelvic field
Tumor Til TI Tz
KAPLAN
pattern
* Stanford
10 :: 3
(0) (14) (15) (66) (4)
score 0 0 2 10 13 9 4
(0) (0) (4) (17) (22) (28) (20) (9)
Staging System (20).
shoulder girdle, thoracic spine, cervical spine, rib, and lower extremity recurrences did not differ between patients treated by either technique (Fig. 5). Conversely, the relapse rate, after initial relapse, in the lumbar spine was significantly reduced in those patients receiving radiation to the paraaortic lymph nodes (Fig. 6). The untreated lumbar spine was the fust site of recurrence in 15% of the patients who received pelvic-field irradiation only, as shown in Figure 6. In contrast, the lumbar spine was the first site of metastases in only 3% of the patients who received extended field irradiation. Once metastatic disease developed, subsequent treatment was the same for patients who received extendedfield or pelvic-field irradiation. Some mode of androgen deprivation therapy was given for palliation in 55 out of 61 patients who failed after extended-field irradiation and in 27 out of 32 patients who developed metastatic disease after pelvic-field treatment. Five instances of spinal cord compression occurred in the patients treated with extended-field irradiation, but none of these were in the lumbar spine. Also, five cases of cord compression occurred among the patients treated with the pelvic field only: 4 in the thoracic spine and 1 in the lumbar spine.
Table 2. Lymph
node status Evidente
By lymphangiogram By biopsy
for lymph node metastasis
Pelvic field
Extended
field
12/53 (23%) 13/32 (41%)
53167 (79%) 37/37 (100%)
Fig. 4. Dosimetry of the paraaortic fields, at the leve1 of S2 (bottorn).
fields (top) and also the pelvic
DISCUSSION
The finding that the rate of metastasis is reduced in incidentally irradiated bone has been observed in breast cancer. Hercbergs et al. observed a reduced rate of metastasis to the thoracic spine in women who received irradiation to an intemal mammary field (9). These investigators analyzed sites of bone metastases when a patient presented with metastatic disease. Hazra and Giri reported a pilot study in which a single 8 Gy dose delivered to the pelvic girdle reduced the incidence of osseous metastases to the pelvis (8). This large single fraction, however, led to unacceptable morbidity and the technique was aban-
Table 3. Actuarial risk of development and rank order of site-specific bone metastases 3 years Lower extremity Ribs Thoracic spine Cervical spine Pelvis out-of-field Scapula/clavicle Lumbar spine Pelvis in-field Upper extremity Skull
.237 .226 .223 .212 .186 .177 .159 .138 .080 .073
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
5 years .385 (1) .365 (2) .356 (3) .321 (4) .269 (6) .253 (7) .242 (9) .244 (8) .190(10) .311 (5)
10 years .509 .499 .492 .321 .473 .364 .440 .304 .452 .398
(1) (2) (3) (9) (4) (8) (6) (10) (5) (7)
1024
1. J. Radiation Oncology 0 Biology 0 Physics
LOWER EXTREMITY
May 1990, Volume 18, Number 5
I~.I o.nm
t....~....~....~...,,.,.,,,,,,,,,,,,,,,,,
1
2
3
4
5
6
1
l-
m
2
3
4
THORACIC
2
80
g 0 5
60
5 SPINE
40
6
7 G*u” P-“.Ius 11% * 0.1228
,o
t....,....,..,....,.. 1
2
3
4
5
6
7
1234567
L”““““““l”“““““““““““1
LI
1
2
3
4
5
6
CERVICAL
SPINE
‘?%“”
7 YEARS
Fig. 5. The actuarial freedom from metastases after the first evidente of recurrent disease, using the actuarial method of Kaplan and Meier (12) in six specific sites. Curve 1 represents patients irradiated to the pelvis only. Curve 2 represents patients irradiated to the pelvic and paraaortic regions.
In our study, the bones of the pelvis or the lumbar spine were not the primary targets for radiation. The incidental dose received by bone ranged from 30 to 50 Gy, delivered in 1.2-2.0 Gy fractions, usually over 35-50 days. This fractionation scheme did not lead to significant neutropenia or thrombocytopenia. The mechanism by which this dose of radiation delays the subsequent development of metastatic disease is uncertain. There appear to be at least two possibilities: (a) Micrometastases can be controlled with lower dose irradiation. The lower malignant cel1 number in occult micrometastasis requires less irradiation for sterilization. Furthermore, microscopic aggregates of malignant cells doned.
in a normal cellular environment are more likely to be wel1 oxygenated, making them more radiosensitive (5). (b) Alternatively, the microenvironment of bone could be altered by irradiation. In experimental models, adhesion between tumor cells and endothelium has been shown to be organ-specific ( 16, 17). Solid tumor proliferation is dependent upon the ability of the vasculature to proliferate and supply nutrients and oxygen (2 1). Unlike normal endothelium which has a low labeling index, in experimental models, non-transformed endothelium in tumor vasculature has a high labeling index and a cel1 loss fraction much like the tumors themselves (10). Preemptive irradiation could alter the ability of metastatic prostatic cells
Reduction of spinal metastases after XRT in prostatic cancer 0 1. D.
KAPLAN
et ai.
1025
1”’ “““““““““““““““““‘1todevelopment LUMBAR
1
2
3
4
5
6
cells in bone or inhibit the of tumor stroma and delay the development of clinically detectable metastases (4). Metastases to bone are common in prostate cancer. This study demonstrates that when bones have been irradiated, the subsequent development of bone metastasis is delayed; therefore, preemptive irradiation of bone may have a role in forestalling the development of metastasis. Elective irradiation to areas that are frequently involved with symptomatic metastases could be performed, and might significantly reduce the incidence of vertebral collapse and spinal cord compression, a devastating event for a patient with prostatic cancer. Preemptive irradiation of the spine to a modest but as yet undetermined radiation dose could significantly delay metastases to these critical areas, and have a significant positive impact on the quality of life of patients with high Gleason pattern scores, advanced stage, or lymphadenopathy, who are at high risk to develop metastatic disease. adhere
O,hm P.“.Iu+I 1 “I. 2 00250
SPINE
7
YEARS
Fig. 6. The actuarial freedom from metastases after the first evidence of recurrent disease, using the actuarial method of Kaplan and Meier ( 12). Curve 1 represents patients irradiated to the pelvis only. Curve 2 represents patients irradiated to the pelvic and paraaortic regions. A statistically significant reduction in the incidence of metastatic involvement in the lumbar field was demonstrated among patients receiving extended field irradiation.
to the endothelial
REFERENCES 1. Bagshaw, M. A. Radiation therapy for cancer of the prostate. In: Skinner, D. G., deKernion, J. B., eds. Diagnosis and management of genitourinary cancer. Philadelphia, PA: W. B. Saunders; 1978:355-379. 2. Bagshaw, M. A.; Cox, R. S.; Ray, G. R. Status of radiation treatment of prostate cancer at Stanford University. NCI Monogr. 7:47-60; 1988. 3. Batson, 0. V. The role of the vertebral veins in metastatic processes. Ann. Intern. Med. 16:38-43; 1942. 4. Dvorak, H. F. Tumors: wounds that do not heal-similarities between tumor stroma generation and wound healing. N. Engl. J. Med. 315:1650-1659; 1986. 5. Fletcher, G. H. Subclinical disease. Cancer 53: 1274- 1284; 1984. 6. Gleason, D. F. Histologie grading and clinical staging of prostatic carcinoma. In: Tannenbaum, M., ed. Urologie pathology: the prostate. Philadelphia, PA: Lea and Febiger: 1977:171-197. 7. Hanks, G. E.; Krall, J. M.; Muntz, K. L.; Diamond, J. J.; Kramer, S. The outcome of treatment of 3 13 patients with T-1 (UICC) prostate cancer treated with extemal beam irradiation. Int. J. Radiat. Oncol. Biol. Phys. 14:245-248; 1988. 8. Hazra, T. A.; Giri, S. Prophylactic pelvic girdle irradiation in the treatment of prostatic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 7:8 17-8 19; 198 1. 9. Hercbergs, A.; Wemer, A.; Brenner, H. J. Reduced thoracic vertebrae metastases following post mastectomy parastemal irradiation. Int. J. Radiat. Oncol. Biol. Phys. 11:773-776; 1985. 10. Hirst, D. G.; Denekamp, J.; Hobson, B. Proliferation kinetics of endothelial and tumor cells in three mouse mammary carcinomas. Cel1 Tissue Kinetics 15:25 1-261; 1982. 11. Kaplan, H. S. Hodgkin’s disease. Cambridge: Cambridge University Press; 1972. 12. Kaplan. E. L.; Meier, P. Nonparametric estimation from
13.
14.
15.
16.
17.
18.
19.
20.
21
incomplete observation. J. Am. Stat. Assoc. 53:457-481; 1958. Kempson, R. L.; Levine, G. The relationship of grade to prognosis in carcinoma of the prostate. Front. Radiat. Ther. Oncol. 9:267-273; 1974. Kumar, S.; Arnold, F. Can metastasis be restrained? In: Stoll, B. A., ed. Breast cancer: treatment and prognosis. Palo Alto, CA: Blackwell Scientific Pub.; 1986:287-299. Leibel, S. A.; Hunter, G. E.; Kramer, S. Patterns of care outcome studies: results of the national practice in adenocarcinoma of the prostate. Int. J. Radiat. Oncol. Biol. Phys. 10:401-409: 1984. Netlund, P. A.: Zehen, B. R. Organ-specific adhesion of metastatic tumor cells in vitro. Science 224: 1113-1114; 1984. Nicolson, G. L. Metastatic tumor cel1 attachment and invasion assay utilizing vascular endothelial monolayers. J. Histochem. Cytochem. 30:214-220; 1982. Pilepich, M. V.: Bagshaw, M. A.; Asbell, S. 0.: Hanks, G. E.; Krall, J. M.; Emami, B. N.; Bard, R. H. Definitive radiation therapy in resectable (Stage A2 and B) carcinoma of the prostate-results of a nationwide overview. Int. J. Radiat. Oncol. Biol. Phys. 13:659-663; 1987. Pilepich, M. V.; Krall, J. M.; Sause, W. T.: Johnson, R. J.; Russ, H. H.: Hanks, G. E.; Perez, C. A.; Zinninger, M.; Martz, K. L. Prognostic factors in carcinoma of the prostate-analysis of RTOG study 75-06. Int. J. Radiat. Oncol. Biol. Phys. 13:339-349; 1987. Pistenma, D. A.; Bagshaw, M. A.; Freiha, F. S. Extendedfield radiation for prostatic adenocarcinoma: status report of a limited prospective trial. In: Johnson, D. E., Samuels, M. L., eds. Cancer of the genitourinary tract. New York: Raven Press: 1979:229-247. Thomlinson, R. H.; Gray, L. H. Histologie structures of some human lung cancers and possible implication for radiotherapy. Br. J. Cancer 9:539-549: 1955.
