- Email: [email protected]
The influence of extent and local management on the outcome of radiotherapy for brain metastases
Inr J Rudrarwn Oncology BIU/. Phy\ Vol. 19. pp. I I I-l Printed in the USA All rights reserved.
0360-3016/90 $3.00 + .Xl Copyright 0 1990 Pergamon Press plc
IS
0 Brief Communication THE INFLUENCE
P. J.
OF EXTENT AND LOCAL MANAGEMENT ON THE OUTCOME OF RADIOTHERAPY FOR BRAIN METASTASES
HOSKIN,
Department
M.R.C.P., F.R.C.R., J.
CROW,
MSc.
AND H. T. FORD,
F.R.C.R.
of Radiotherapy and Oncology, The Royal Marsden Hospital, Downs Road, Sutton. Surrey, U.K.
The results of cranial irradiation for brain metastases in 164 consecutive patients have been reviewed to evaluate a policy of localized high dose irradiation for solitary metastases. Fifty of the 164 patients receiving whole brain irradiation (35 Gy in 15 daily fractions) were selected for boosts delivering 15 Gy in 8 daily fractions to the site of solitary deposits. No difference in overall survival or the incidence of death from progressive brain metastases was seen between the patients receiving a boost and those who did not. Overall median survival was 112 days with 62% of patients dying from metastatic disease outside the brain. Factors associated with increased survival were early response to radiotherapy and, in breast cancer patients, a disease-free interval of 2 years. Palliation of presenting symptoms was achieved in 86% of patients at 3 weeks from starting radiotherapy. It is concluded that whereas whole brain irradiation results in useful symptom control, there is no advantage for high dose treatment in the majority of patients with solitary metastases even in the absence of metastatic disease elsewhere. Brain metastases,
Radiotherapy,
Boost, Survival, Palliation.
hospital with intracranial metastases has been carried out to assess both the value of cranial irradiation in reducing morbidity, and particularly to assess the role of high-dose irradiation using small volume boosts to the site of solitary metastases.
INTRODUCTION
are a common finding in progressive Brain metastases cancer and may result in considerable morbidity. Radiotherapy is well established as an effective palliative treatment for cerebral metastases to prevent progressive neurological deficits, restore function, and control other symptoms such as headache and confusion. Several series have looked at the influence of dose on outcome in the irradiation of multiple cranial metastases and neither retrospective series (2, 3, 5, 12) nor prospective studies (1, 7) have demonstrated an advantage for high doses even in selected subgroups where the brain is the sole site of relapse (6, 10). Single metastases are often considered to reflect a better prognosis and warrant more radical treatment, particularly when the central nervous system is the sole site of relapse ( 11). The routine use of CT scanning in the diagnosis of intracranial metastases enables identification of those patients with solitary deposits who might be expected to benefit from high dose irradiation. In a surgical series from the Mayo Clinic of patients receiving post operative radiotherapy following excision of a solitary brain metastasis a significantly better local control rate was seen in those receiving high dose (~39 Gy) irradiation (13). A retrospective analysis of patients presenting to this
METHODS
AND
MATERIALS
All patients treated for intracranial metastases under the care of one radiotherapist (HTF) between January 1978 and December 1987 have been traced. During this period the treatment policy was for patients with multiple intracranial metastases to receive 35 Gy in 15 daily fractions over 3 weeks, and for those with single metastases who either had no evidence of metastases outside the brain, or had primary breast cancer with only bone metastases, to receive an additional boost of 15 Gy in 8 daily fractions. Patients were treated using lateral opposed megavoltage beams with either a 6oCo machine or 6-8 MeV linear accelerator. Whole brain treatments were set up on the treatment machine using a standard base-line and verification films were taken and approved after the first treatment. Boosts were performed using immobilization in a cast and simulated 2 or 3 field planned volumes.
Reprint requests to: Dr. P. J. Hoskin, Department of Radiotherapy and Oncology, The Royal Marsden Hospital, Fulham Road, J ->-rdon SW3, UK. Acknowlcdgem~xts-We are indebted to Sue Ashley for help in
the statistical analyses and Mrs. Joan Staples who prepared the manuscript. Accepted for publication 24 January 1990.
111
112
I. J. Radiation Oncology 0 Biology 0 Physics
The diagnosis of cerebral metastases was confirmed by CT scan in 144 patients and by isotope brain scan in the remaining 20. Routine diagnostic CT scans were performed using 8 mm coronal cuts with and without intravenous contrast. Double dose contrast techniques were not used and magnetic resonance imaging was not available during this study. Data collected from the patient notes include primary site, presenting symptoms, site of metastases, treatment and response details, and cause of death. Response to treatment was assessed for both overall response and specific symptoms, in particular headache, confusion, and motor weakness. A four point scale was used: complete response, partial response, no response or worse. Complete response was defined by complete resolution of the presenting symptoms with no evolution of new symptoms which could be attributed to intracranial pathology. Any reported improvement less than a complete response has been designated a partial response. This does not necessarily imply the conventional meaning of at least 50% change, since this could not be accurately quantified from the assessments available in this study. Survival analysis was performed using the method of Kaplan-Meier (9). RESULTS
A total of 209 patients were identified who presented with intracranial metastases during the lo-year study period. Forty-five of these patients deteriorated before or shortly after starting irradiation, failing to complete at least 50% of the prescribed treatment. These cases have not been analyzed further. Of the remaining 164, 5 patients had insufficient data available in the clinical notes to assess response. Forty patients did not receive the standard whole-brain regimen of 35 Gy in 15 fractions; 32 received 30 Gy in 10 fractions and 8 received 40 Gy in 20 fractions. All patients receiving the boost were administered the standard dose of 15 Gray in 8 fractions. The study population was composed of 80 female and 84 male patients with a median age of 59 years (range 27-85 years). Eighty-six patients had primary lung cancer, 48 had primary breast cancer and the remainder had a variety of primary sites including melanoma ( 13), kidney
Table 1. Presenting
July 1990, Volume 19, Number 1
60
50 40 30 20
IO 0 0
I TM
2
3
4
SIKE 1ST CEREBRAL YTASTASIS fEARSI
Fig. 1. The effect of localized
boost on overall survival.
(5), colon (3), bladder (3), unknown primary (3), prostate
(I), lymphoma (l), and synovial sarcoma (1). The presenting symptoms are shown in Table 1. Five patients had partial excision of a solitary metastasis. Overall, 79 patients had a solitary metastasis, the position of which was parietal (41), posterior fossa (20), frontal ( 1 l), and occipital (7). Fifty of these patients were selected to receive a boost to the site of solitary metastasis on the basis of absent or stable disease elsewhere and good general condition. In 93 patients overall, the central nervous system was the first site of distant relapse. Dexamethasone was given with irradiation in 154 patients, of whom 152 started steroids within 24 hr of diagnosis. One hundred and twelve patients received an initial dose of 16 milligrams daily, 28 received an initial dose of 12 milligrams daily and 14 received an initial dose of 8 milligrams or less daily, and 92 patients continued with steroids from the time of irradiation until death. The median duration of administration of Dexamethasone was 8 weeks (range 2-40 weeks). Patients who did not receive Dexamethasone and those receiving low doses (~8 mg Dexamethasone) were a selected group with minimal dis-
signs and symptoms
Symptom
No.
(B)
Headache Hemiparesis Seizures Confusion Papilledema Ataxia Dysphasia Vomiting Cranial nerve palsy
72 46 28 22 21 21 19 19 14
(44) (28) (17) (13) (13) (13) (ll) (11) (8)
0
1 TIE
SINCE IS7 CERErnl.
Fig. 2. The influence
of primary
P wE,1s11.51s
3
4
fYEutsI
histology
on overall survival.
113
Outcome of radiotherapy for brain metastases 0 P. J. HOSKINet al.
0 0
z
1 TIE
0
4
3
SINCE 157 CERFBAAL METASTASIS WEARS1
Fig. 3. The effect of response
to treatment
90
SIKE
2
*ST CEREBW
*ETAsTASIS
3 IYEAnu
at 3 weeks (a) and 6 weeks (b) on overall survival.
ease and consequently in those patients a slightly better survival was seen. No difference in survival was seen in the 40 patients who responded to Dexamethasone prior to starting irradiation compared to those who did not. Thirty-three patients had systemic treatment, either chemotherapy or hormone therapy, instituted or changed within 1 month of the diagnosis of cerebral metastases. No difference in survival was seen in this subgroup of patients. Median overall survival was 112 days and the l-year survival was only 15%. No 5-year survivors are present. The survival of patients selected to receive a boost compared to those who did not is shown in Figure 1; the median survival (range) of those receiving a boost was 140 (32-974) days, and of those not receiving a boost was 116 ( 13-1436) days. The influence of primary histology upon survival is shown in Figure 2. A univariate analysis has shown no influence of any of the following factors upon survival: primary site, boost, performance status, presenting symptom or sign, status of primary site. The only factor associated with improved survival for the whole
100
I ,I*
population was a response to treatment at 3 weeks and 6 weeks as shown in Figure 3. Response at 1 week was not a survival indicator. In patients with primary breast cancer a disease-free interval of 2 years was associated with a better survival advantage 6 months after the diagnosis of cerebral metastases (p < 0.005) as shown in Figure 4. Although only 7 1 patients (43%) had distant metastases diagnosed at the time of intracranial metastases, 10 1 (62%) died with progressive metastatic disease outside the brain and in only 43 (26%) were progressive intracerebral metastases implicated in the cause of death. In patients receiving a boost 28% died with progressive intracranial metastases compared to 26% of those who did not receive a boost. Details of overall symptomatic response and the response of headache, confusion, and motor weakness to irradiation are shown in Table 2. Thirty-six patients had recurrence of symptoms following an initial response. The probability of symptom relapse with time is shown in Figure 5. At relapse the majority of patients were treated with increasing doses of Dexamethasone and analgesics.
,
II .
._____ 0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..-~............
0
1
TM
SINCE 151
3
2
CER&BRIl KTASTASIS
Fig. 4. The effect of disease-free
in primary breast cancer.
1
IIEIRSJ
interval (DF’I) on overall survival Fig. 5. Symptom
relapse-free
survival.
114
I. J. Radiation Oncology 0 Biology 0 Physics
July 1990, Volume 19, Number 1
Table 2. Symptom palliation Individual symptoms at 4 weeks Overall response at 3 weeks
Complete response Partial response Worse Dead
Headache (n = 72)
Confusion (n = 22)
Motor weakness (n = 57)
No.
(%)
No.
(u/o)
No.
(%)
No.
I%)
58 80 5 2
(36) (50) (3) (1)
60 6
(83) (8)
6 10 2 3
(27) (45) (9) (14)
13 27 0
(23) (47)
11
(19)
1
2
Six received further irradiation giving doses of 20-25 Gy over 2-3 weeks. DISCUSSION The results of this study highlight the extremely poor prognosis associated with cerebral metastases also demonstrated in other series (14). Although a small number of patients survive for several years, the majority deteriorate within a few months. The development of symptomatic intracranial metastases is frequently a herald of generalized metastatic disease. Although the majority of patients in this series had primary lung tumors, those with potentially treatable metastatic disease outside the brain, in particular those patients with breast cancer, fared no better when cerebral metastases were present. No effect of systemic therapy in these patients was seen on survival, and no difference in the cause of death was observed. Despite the poor prognosis for survival, radiotherapy appears an effective means of palliation. The overall response rate of 85% seen at 3 weeks from initiating radiotherapy is in agreement with other reported series (I, 7, 8, 10, 14). Headache in particular appears extremely well controlled with an 83% complete response rate at 4 weeks, although it is likely that concomitant Dexamethasone as given in these patients had a significant effect also. More complex problems such as confusion and motor weakness have, not surprisingly, a lower complete response rate at 4 weeks, but overall response rates of around 70% were still achieved. Selection of patients for more radical treatment on the basis of a solitary deposit and limited disease elsewhere outside the brain has failed to significantly improve either overall survival or disease-free survival, although those receiving a boost had a longer median survival ( 140 days, range 32 to 974) compared with those who did not receive
a boost who had a median survival of 116 days (range 13 to 1436). Selection for a boost did not reduce the likelihood of death with progressive intracranial disease. This may reflect the fact that true solitary metastases are extremely rare and that despite more accurate diagnosis using CT scanning, microscopic disease is invariably present elsewhere in the brain. It also may reflect the fact that cerebral metastases are very commonly associated with more widespread disease. Previous retrospective studies have defined various factors predicting long-term survival. These factors include the brain being the first site of relapse, surgical resection and a long disease-free interval (4), site of primary, and functional level at time of irradiation (I, 7). Meningeal disease and multilobe involvement ( 13) have been shown to be poor prognostic features. This study indicates that response at 3 and 6 weeks are predictors of a better overall survival, and confirms the importance of a long diseasefree interval in those with breast cancer. The use of steroids in the management of brain metastases is generally recommended, but previous studies have not documented their use in detail. Further prospective study is required to investigate both their relative importance in the management of brain metastases and the optimum dose, duration, and steroid preparation. In conclusion, cranial irradiation may be indicated for the palliation of brain metastases, but survival following treatment is poor. For most patients a short palliative dose of irradiation to the whole brain will be most appropriate. A small subgroup of patients with a better prognosis may exist in those with a long disease-free interval after primary breast cancer, good performance status at presentation, and in those patients who show a good initial response to irradiation for whom more radical treatment may be considered.
REFERENCES Borgelt, B.; Gelber, R.; Kramer, S.; Brady, L. W.; Chang, C. H.; Davis, L. W.; Perez, C. A.; Hendrickson, E. R. The palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int. J. Radiat. Oncol. Biol. Phys. 6:1-9; 1980. Chatani, M.; Teshima, T.; Hata, K.; Inoue, T.; Suzuki, T.
Whole brain irradiation for metastases from lung cancer. Acta Radio]. Oncol. 24:311-314; 1985. 3. D’Elia, E.; Bonucci, I.; Biti, G. P.; Pirtoli, L. Different fractionation schedules in radiation treatment of cerebral metastases. Acta Radiol. Oncol. 25: 18 l- 184; 1986. 4. Di Stefano, A.; Yong Yap, Y.; Hortobaggi, G. N.; Blumen-
Outcome of radiotherapy for brain metastases 0 P. J. HOSKINet al.
5.
6.
7.
8.
9.
10.
schein, G. R. The natural history of breast cancer patients with brain metastases. Cancer 44: 19 13- 19 18; 1979. Egawa, S.; Tukigama, I.; Akine, Y.; Kajiura, Y.; Yanagawa, S.; Watai, K.; Nomura, K. Radiotherapy of brain metastases. Int. J. Radiat. Oncol. Biol. Phys. 12:1621-1625; 1986. Gelber, R. D.; Lorson, M.; Borgelt, B.; Kramer, S. Equivalence of radiation schedules for the palliative treatment of brain metastases in patients with favourable prognosis. Cancer 48:1749-1753; 1981. Harwood, A. R.; Simpson, W. J. Radiation therapy of cerebral metastases: a randomised prospective clinical trial. Int. J. Radiat. Oncol. Biol. Phys. 2: 1091-1094; 1977. Hendrickson, E. R. The optimum schedule for palliative radiotherapy for metastatic brain cancer. Int. J. Radiat. Oncol. Biol. Phys. 2:165-168; 1977. Kaplan, E. L.; Meier, P. Non-parametric estimation for incomplete observations. J. Am. Stat. Assoc. 53:457-481; 1958. Kurtz, J. M.: Gelber, R.; Brady, L. W.; Corella, G. J.; Cooper,
115
J. S. The palliation of brain metastases in a favourable patient population: a randomized clinical trial by the Radiation Therapy Oncology Group. Int. J. Radiat. Oncol. Biol. Phys. 7:891-896; 1981. Sarma, D. P.; Weilbacher, T. G. Long term survival after brain metastasis from lung cancer. Cancer 58: 1366- 1370; 1986. Sham, J. S. T.; Lau, W. H.; Tung, Y. Radiotherapy ofbrain metastases from carcinoma of the bronchus. Clin. Radiol. 40:193-194; 1989. Smalley, S. R.; Schray, M. E.; Laws, E. R.; O’Fallon, J. R. Adjuvant radiation therapy after surgical resection of solitary brain metastasis: association with pattern of failure and survival. Int. J. Radiat. Oncol. Biol. Phys. 13:1611-1616; 1987. 14. West, J.; Maor, M. Intracranial metastases: behavioural patterns related to primary site and results of treatment by whole brain irradiation. Int. J. Radiat. Oncol. Biol. Phys. 6:11-15; 1980.
0360-3016/90 $3.00 + .Xl Copyright 0 1990 Pergamon Press plc
IS
0 Brief Communication THE INFLUENCE
P. J.
OF EXTENT AND LOCAL MANAGEMENT ON THE OUTCOME OF RADIOTHERAPY FOR BRAIN METASTASES
HOSKIN,
Department
M.R.C.P., F.R.C.R., J.
CROW,
MSc.
AND H. T. FORD,
F.R.C.R.
of Radiotherapy and Oncology, The Royal Marsden Hospital, Downs Road, Sutton. Surrey, U.K.
The results of cranial irradiation for brain metastases in 164 consecutive patients have been reviewed to evaluate a policy of localized high dose irradiation for solitary metastases. Fifty of the 164 patients receiving whole brain irradiation (35 Gy in 15 daily fractions) were selected for boosts delivering 15 Gy in 8 daily fractions to the site of solitary deposits. No difference in overall survival or the incidence of death from progressive brain metastases was seen between the patients receiving a boost and those who did not. Overall median survival was 112 days with 62% of patients dying from metastatic disease outside the brain. Factors associated with increased survival were early response to radiotherapy and, in breast cancer patients, a disease-free interval of 2 years. Palliation of presenting symptoms was achieved in 86% of patients at 3 weeks from starting radiotherapy. It is concluded that whereas whole brain irradiation results in useful symptom control, there is no advantage for high dose treatment in the majority of patients with solitary metastases even in the absence of metastatic disease elsewhere. Brain metastases,
Radiotherapy,
Boost, Survival, Palliation.
hospital with intracranial metastases has been carried out to assess both the value of cranial irradiation in reducing morbidity, and particularly to assess the role of high-dose irradiation using small volume boosts to the site of solitary metastases.
INTRODUCTION
are a common finding in progressive Brain metastases cancer and may result in considerable morbidity. Radiotherapy is well established as an effective palliative treatment for cerebral metastases to prevent progressive neurological deficits, restore function, and control other symptoms such as headache and confusion. Several series have looked at the influence of dose on outcome in the irradiation of multiple cranial metastases and neither retrospective series (2, 3, 5, 12) nor prospective studies (1, 7) have demonstrated an advantage for high doses even in selected subgroups where the brain is the sole site of relapse (6, 10). Single metastases are often considered to reflect a better prognosis and warrant more radical treatment, particularly when the central nervous system is the sole site of relapse ( 11). The routine use of CT scanning in the diagnosis of intracranial metastases enables identification of those patients with solitary deposits who might be expected to benefit from high dose irradiation. In a surgical series from the Mayo Clinic of patients receiving post operative radiotherapy following excision of a solitary brain metastasis a significantly better local control rate was seen in those receiving high dose (~39 Gy) irradiation (13). A retrospective analysis of patients presenting to this
METHODS
AND
MATERIALS
All patients treated for intracranial metastases under the care of one radiotherapist (HTF) between January 1978 and December 1987 have been traced. During this period the treatment policy was for patients with multiple intracranial metastases to receive 35 Gy in 15 daily fractions over 3 weeks, and for those with single metastases who either had no evidence of metastases outside the brain, or had primary breast cancer with only bone metastases, to receive an additional boost of 15 Gy in 8 daily fractions. Patients were treated using lateral opposed megavoltage beams with either a 6oCo machine or 6-8 MeV linear accelerator. Whole brain treatments were set up on the treatment machine using a standard base-line and verification films were taken and approved after the first treatment. Boosts were performed using immobilization in a cast and simulated 2 or 3 field planned volumes.
Reprint requests to: Dr. P. J. Hoskin, Department of Radiotherapy and Oncology, The Royal Marsden Hospital, Fulham Road, J ->-rdon SW3, UK. Acknowlcdgem~xts-We are indebted to Sue Ashley for help in
the statistical analyses and Mrs. Joan Staples who prepared the manuscript. Accepted for publication 24 January 1990.
111
112
I. J. Radiation Oncology 0 Biology 0 Physics
The diagnosis of cerebral metastases was confirmed by CT scan in 144 patients and by isotope brain scan in the remaining 20. Routine diagnostic CT scans were performed using 8 mm coronal cuts with and without intravenous contrast. Double dose contrast techniques were not used and magnetic resonance imaging was not available during this study. Data collected from the patient notes include primary site, presenting symptoms, site of metastases, treatment and response details, and cause of death. Response to treatment was assessed for both overall response and specific symptoms, in particular headache, confusion, and motor weakness. A four point scale was used: complete response, partial response, no response or worse. Complete response was defined by complete resolution of the presenting symptoms with no evolution of new symptoms which could be attributed to intracranial pathology. Any reported improvement less than a complete response has been designated a partial response. This does not necessarily imply the conventional meaning of at least 50% change, since this could not be accurately quantified from the assessments available in this study. Survival analysis was performed using the method of Kaplan-Meier (9). RESULTS
A total of 209 patients were identified who presented with intracranial metastases during the lo-year study period. Forty-five of these patients deteriorated before or shortly after starting irradiation, failing to complete at least 50% of the prescribed treatment. These cases have not been analyzed further. Of the remaining 164, 5 patients had insufficient data available in the clinical notes to assess response. Forty patients did not receive the standard whole-brain regimen of 35 Gy in 15 fractions; 32 received 30 Gy in 10 fractions and 8 received 40 Gy in 20 fractions. All patients receiving the boost were administered the standard dose of 15 Gray in 8 fractions. The study population was composed of 80 female and 84 male patients with a median age of 59 years (range 27-85 years). Eighty-six patients had primary lung cancer, 48 had primary breast cancer and the remainder had a variety of primary sites including melanoma ( 13), kidney
Table 1. Presenting
July 1990, Volume 19, Number 1
60
50 40 30 20
IO 0 0
I TM
2
3
4
SIKE 1ST CEREBRAL YTASTASIS fEARSI
Fig. 1. The effect of localized
boost on overall survival.
(5), colon (3), bladder (3), unknown primary (3), prostate
(I), lymphoma (l), and synovial sarcoma (1). The presenting symptoms are shown in Table 1. Five patients had partial excision of a solitary metastasis. Overall, 79 patients had a solitary metastasis, the position of which was parietal (41), posterior fossa (20), frontal ( 1 l), and occipital (7). Fifty of these patients were selected to receive a boost to the site of solitary metastasis on the basis of absent or stable disease elsewhere and good general condition. In 93 patients overall, the central nervous system was the first site of distant relapse. Dexamethasone was given with irradiation in 154 patients, of whom 152 started steroids within 24 hr of diagnosis. One hundred and twelve patients received an initial dose of 16 milligrams daily, 28 received an initial dose of 12 milligrams daily and 14 received an initial dose of 8 milligrams or less daily, and 92 patients continued with steroids from the time of irradiation until death. The median duration of administration of Dexamethasone was 8 weeks (range 2-40 weeks). Patients who did not receive Dexamethasone and those receiving low doses (~8 mg Dexamethasone) were a selected group with minimal dis-
signs and symptoms
Symptom
No.
(B)
Headache Hemiparesis Seizures Confusion Papilledema Ataxia Dysphasia Vomiting Cranial nerve palsy
72 46 28 22 21 21 19 19 14
(44) (28) (17) (13) (13) (13) (ll) (11) (8)
0
1 TIE
SINCE IS7 CERErnl.
Fig. 2. The influence
of primary
P wE,1s11.51s
3
4
fYEutsI
histology
on overall survival.
113
Outcome of radiotherapy for brain metastases 0 P. J. HOSKINet al.
0 0
z
1 TIE
0
4
3
SINCE 157 CERFBAAL METASTASIS WEARS1
Fig. 3. The effect of response
to treatment
90
SIKE
2
*ST CEREBW
*ETAsTASIS
3 IYEAnu
at 3 weeks (a) and 6 weeks (b) on overall survival.
ease and consequently in those patients a slightly better survival was seen. No difference in survival was seen in the 40 patients who responded to Dexamethasone prior to starting irradiation compared to those who did not. Thirty-three patients had systemic treatment, either chemotherapy or hormone therapy, instituted or changed within 1 month of the diagnosis of cerebral metastases. No difference in survival was seen in this subgroup of patients. Median overall survival was 112 days and the l-year survival was only 15%. No 5-year survivors are present. The survival of patients selected to receive a boost compared to those who did not is shown in Figure 1; the median survival (range) of those receiving a boost was 140 (32-974) days, and of those not receiving a boost was 116 ( 13-1436) days. The influence of primary histology upon survival is shown in Figure 2. A univariate analysis has shown no influence of any of the following factors upon survival: primary site, boost, performance status, presenting symptom or sign, status of primary site. The only factor associated with improved survival for the whole
100
I ,I*
population was a response to treatment at 3 weeks and 6 weeks as shown in Figure 3. Response at 1 week was not a survival indicator. In patients with primary breast cancer a disease-free interval of 2 years was associated with a better survival advantage 6 months after the diagnosis of cerebral metastases (p < 0.005) as shown in Figure 4. Although only 7 1 patients (43%) had distant metastases diagnosed at the time of intracranial metastases, 10 1 (62%) died with progressive metastatic disease outside the brain and in only 43 (26%) were progressive intracerebral metastases implicated in the cause of death. In patients receiving a boost 28% died with progressive intracranial metastases compared to 26% of those who did not receive a boost. Details of overall symptomatic response and the response of headache, confusion, and motor weakness to irradiation are shown in Table 2. Thirty-six patients had recurrence of symptoms following an initial response. The probability of symptom relapse with time is shown in Figure 5. At relapse the majority of patients were treated with increasing doses of Dexamethasone and analgesics.
,
II .
._____ 0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..-~............
0
1
TM
SINCE 151
3
2
CER&BRIl KTASTASIS
Fig. 4. The effect of disease-free
in primary breast cancer.
1
IIEIRSJ
interval (DF’I) on overall survival Fig. 5. Symptom
relapse-free
survival.
114
I. J. Radiation Oncology 0 Biology 0 Physics
July 1990, Volume 19, Number 1
Table 2. Symptom palliation Individual symptoms at 4 weeks Overall response at 3 weeks
Complete response Partial response Worse Dead
Headache (n = 72)
Confusion (n = 22)
Motor weakness (n = 57)
No.
(%)
No.
(u/o)
No.
(%)
No.
I%)
58 80 5 2
(36) (50) (3) (1)
60 6
(83) (8)
6 10 2 3
(27) (45) (9) (14)
13 27 0
(23) (47)
11
(19)
1
2
Six received further irradiation giving doses of 20-25 Gy over 2-3 weeks. DISCUSSION The results of this study highlight the extremely poor prognosis associated with cerebral metastases also demonstrated in other series (14). Although a small number of patients survive for several years, the majority deteriorate within a few months. The development of symptomatic intracranial metastases is frequently a herald of generalized metastatic disease. Although the majority of patients in this series had primary lung tumors, those with potentially treatable metastatic disease outside the brain, in particular those patients with breast cancer, fared no better when cerebral metastases were present. No effect of systemic therapy in these patients was seen on survival, and no difference in the cause of death was observed. Despite the poor prognosis for survival, radiotherapy appears an effective means of palliation. The overall response rate of 85% seen at 3 weeks from initiating radiotherapy is in agreement with other reported series (I, 7, 8, 10, 14). Headache in particular appears extremely well controlled with an 83% complete response rate at 4 weeks, although it is likely that concomitant Dexamethasone as given in these patients had a significant effect also. More complex problems such as confusion and motor weakness have, not surprisingly, a lower complete response rate at 4 weeks, but overall response rates of around 70% were still achieved. Selection of patients for more radical treatment on the basis of a solitary deposit and limited disease elsewhere outside the brain has failed to significantly improve either overall survival or disease-free survival, although those receiving a boost had a longer median survival ( 140 days, range 32 to 974) compared with those who did not receive
a boost who had a median survival of 116 days (range 13 to 1436). Selection for a boost did not reduce the likelihood of death with progressive intracranial disease. This may reflect the fact that true solitary metastases are extremely rare and that despite more accurate diagnosis using CT scanning, microscopic disease is invariably present elsewhere in the brain. It also may reflect the fact that cerebral metastases are very commonly associated with more widespread disease. Previous retrospective studies have defined various factors predicting long-term survival. These factors include the brain being the first site of relapse, surgical resection and a long disease-free interval (4), site of primary, and functional level at time of irradiation (I, 7). Meningeal disease and multilobe involvement ( 13) have been shown to be poor prognostic features. This study indicates that response at 3 and 6 weeks are predictors of a better overall survival, and confirms the importance of a long diseasefree interval in those with breast cancer. The use of steroids in the management of brain metastases is generally recommended, but previous studies have not documented their use in detail. Further prospective study is required to investigate both their relative importance in the management of brain metastases and the optimum dose, duration, and steroid preparation. In conclusion, cranial irradiation may be indicated for the palliation of brain metastases, but survival following treatment is poor. For most patients a short palliative dose of irradiation to the whole brain will be most appropriate. A small subgroup of patients with a better prognosis may exist in those with a long disease-free interval after primary breast cancer, good performance status at presentation, and in those patients who show a good initial response to irradiation for whom more radical treatment may be considered.
REFERENCES Borgelt, B.; Gelber, R.; Kramer, S.; Brady, L. W.; Chang, C. H.; Davis, L. W.; Perez, C. A.; Hendrickson, E. R. The palliation of brain metastases: final results of the first two studies by the Radiation Therapy Oncology Group. Int. J. Radiat. Oncol. Biol. Phys. 6:1-9; 1980. Chatani, M.; Teshima, T.; Hata, K.; Inoue, T.; Suzuki, T.
Whole brain irradiation for metastases from lung cancer. Acta Radio]. Oncol. 24:311-314; 1985. 3. D’Elia, E.; Bonucci, I.; Biti, G. P.; Pirtoli, L. Different fractionation schedules in radiation treatment of cerebral metastases. Acta Radiol. Oncol. 25: 18 l- 184; 1986. 4. Di Stefano, A.; Yong Yap, Y.; Hortobaggi, G. N.; Blumen-
Outcome of radiotherapy for brain metastases 0 P. J. HOSKINet al.
5.
6.
7.
8.
9.
10.
schein, G. R. The natural history of breast cancer patients with brain metastases. Cancer 44: 19 13- 19 18; 1979. Egawa, S.; Tukigama, I.; Akine, Y.; Kajiura, Y.; Yanagawa, S.; Watai, K.; Nomura, K. Radiotherapy of brain metastases. Int. J. Radiat. Oncol. Biol. Phys. 12:1621-1625; 1986. Gelber, R. D.; Lorson, M.; Borgelt, B.; Kramer, S. Equivalence of radiation schedules for the palliative treatment of brain metastases in patients with favourable prognosis. Cancer 48:1749-1753; 1981. Harwood, A. R.; Simpson, W. J. Radiation therapy of cerebral metastases: a randomised prospective clinical trial. Int. J. Radiat. Oncol. Biol. Phys. 2: 1091-1094; 1977. Hendrickson, E. R. The optimum schedule for palliative radiotherapy for metastatic brain cancer. Int. J. Radiat. Oncol. Biol. Phys. 2:165-168; 1977. Kaplan, E. L.; Meier, P. Non-parametric estimation for incomplete observations. J. Am. Stat. Assoc. 53:457-481; 1958. Kurtz, J. M.: Gelber, R.; Brady, L. W.; Corella, G. J.; Cooper,
115
J. S. The palliation of brain metastases in a favourable patient population: a randomized clinical trial by the Radiation Therapy Oncology Group. Int. J. Radiat. Oncol. Biol. Phys. 7:891-896; 1981. Sarma, D. P.; Weilbacher, T. G. Long term survival after brain metastasis from lung cancer. Cancer 58: 1366- 1370; 1986. Sham, J. S. T.; Lau, W. H.; Tung, Y. Radiotherapy ofbrain metastases from carcinoma of the bronchus. Clin. Radiol. 40:193-194; 1989. Smalley, S. R.; Schray, M. E.; Laws, E. R.; O’Fallon, J. R. Adjuvant radiation therapy after surgical resection of solitary brain metastasis: association with pattern of failure and survival. Int. J. Radiat. Oncol. Biol. Phys. 13:1611-1616; 1987. 14. West, J.; Maor, M. Intracranial metastases: behavioural patterns related to primary site and results of treatment by whole brain irradiation. Int. J. Radiat. Oncol. Biol. Phys. 6:11-15; 1980.