0360-3016190 $3.00 + .OO Copyright 0 1990 Pergamon Press plc
Inr J. Radiamn Oncology Btd. Phys., Vol. 18, pp. 927-932 Ptinted in the U.S.A. All rights mawd.
??Brief Communication
RADIATION THERAPY FOR GLIOMAS OF THE OPTIC NERVE AND CHIASM JEFFREY J. KOVALIC, M.D., PERRY W. GRIGSBY, M.D.,* MARK J. SHEPARD, B.S., BARBARA B. FINEBERG, B.A. AND PATRICK R. THOMAS, M.B., M.R.C.P., F.R.C.R. Mallinckrodt
Institute of Radiology, Radiation Oncology Center, Washington University School of Medicine. St. Louis, MO
Thirty-three patients with optic glioma seen over a 30-year period were reviewed. Five patients (15%) had tumor confined to the optic nerve, 8 patients (24%) had optic nerve and chiasmal involvement, and the remaining 20 patients (61%) had invasion of contiguous structures as well as chiasmal involvement. Eleven patients (33%) had a history of neurofibromatosis. Two-thirds of the patients had either a biopsy or a partial resection of the tumor, with the remaining one-third being clinically diagnosed. All patients received irradiation to local fields. The median dose was 5040 cGy in 160 cGy fractions. Of patients alive at last follow-up, the median time of follow-up was 12.3 years. The 5, lo-, and 15-year overall actuarial survivals were 94, 81, and 74%, respectively. Univariate and multivariate analysis were performed on the following clinical variables: extent of primary tumor, extent of surgery, dose of radiation, gender, race, age, and presence or absence of neurofibromatosis. Extension of the primary lesion to the optic chiasm and age less. than or equal to 15 years were the only two variables to have statistically significantly inferior 15-year progression free survivals by multivariate analysis. Eighteen (55%) patients had treatment related complications with most involving the pituitary gland. We conclude that postoperative radiotherapy is beneficial in patients with chiasmal involvement and those with incomplete resections. A minimum tumor dose of 4000 cGy is recommended. Radiation therapy, Optic nerve glioma, Pediatric tumors.
to follow-up, the current analysis is of the remaining 33 patients. The median age was 4-years old with a range from 1 to 34 years. Most patients (29) were under 16 years of age. The female-to-male ratio was 2 1:12 with 27 patients being white and six patients being black. Eleven patients (33%) had a history of neurofibromatosis. Presenting signs and symptoms are shown in Table 1. The median duration of symptoms prior to presentation was 1 year (mean = 23 months). Optic atrophy and visual field defects were the most common presenting signs. Visual acuity in the affected eye was recorded at diagnosis in 19 patients (Table 2). Endocrine abnormalities were noted in two patients at diagnosis. One patient had growth hormone deficiency and another patient had precocious puberty. Radiographic evaluation revealed that 23 out of 26 patients who underwent a pneumoencephalogram had an abnormality, whereas 1 l/ 14 patients with an angiogram showed vascular aberration. Twenty-six patients had plain films done with the Reese view that revealed optic foramen widening in 25/26 (96%) involvement of the pituitary fossa in 12/26 (46%), and chiasmal involvement in 18/ 26 (69%). Computerized tomography (CT) scan was per-
INTRODUCTION
Chiasmal and optic nerve gliomas are rare tumors of childhood. Their general incidence is 1 in 100,000 patients (2) with 90% presenting in the first two decades of life (28). Most of these neoplasms are low-grade astrocytomas with a long natural history. The untreated clinical course is that of progressive blindness or death from local tumor progression in a significant number of patients (1). Lesions that extend outside of the optic nerve to involve the chiasm and neighboring structures cannot be totally resected without substantial morbidity. It is in these patients that radiotherapy is most useful. This paper analyzes our 30year experience in treating predominantly locally advanced gliomas of the anterior visual system with radiation therapy.
METHODS
AND
MATERIALS
Patient characteristics Thirty-five patients completed cranial irradiation for optic nerve glioma at the Mallinckrodt Institute of Radiology between 1956 and 1986. With two patients lost * Supported by the American Cancer Society Clinical Oncology Career Development Award. Reprint requests to: Perry W. Grigsby, M.D., Radiation Oncology Center, 4939 Audubon, Suite 5500, St. Louis, MO 63110.
Acknowledgements-The authors wish to thank Elaine Pirkey for her help in preparation of this manuscript and Leslie MacConnell-Clubbs for her illustration. Accepted for publication 12 October 1989. 927
928
I. J. Radiation Oncology 0 Biology 0 Physics
April 1990, Volume 18, Number 4
Table 3. Disease extent
Table 1. Signs and symptoms at presentation
No. (%)
No. (%)*
Optic atrophy Visual field defect Strabismus Headache Total blindness Nystagmus Proptosis Light perception only Papilledema
20 20 9 9 g g 7 6 5
(64) (61) (30) (30) (26) (26) (23) (20) (16)
* Per cent of total number of patients for which data were available.
formed in all patients presenting after 1975 and showed chiasmal involvement in 7/9 (78%) patients. Twenty-two patients (67%) had histologic confirmation of the diagnosis and 11 patients (33%) were clinically diagnosed. Of the 22 surgically treated patients, 14 (64%) had biopsy only and 8 (36%) underwent a partial excision. None had total excision. Nineteen of 22 patients (86%) with a tissue diagnosis had low-grade astrocytomas, 1 had a high grade astrocytoma, and 2 had mixed oligodendroglioma/astrocytoma. The clinical diagnosis of optic nerve glioma was given to those patients who presented with visual field deficits, optic atrophy, and radiographic evidence of this disease. In these cases it was felt unsafe to biopsy the lesion. The largest tumor diameter was documented from the operative notes or radiographic studies. The median was 4 cm, mean 3.7 cm, and range from 1 to 7 cm. Only 5 ( 15%) patients had disease limited to the optic nerve (Table 3), with most patients exhibiting extrachiasmal extension. Radiation parameters Twenty-two patients were treated with postoperative irradiation and 11 underwent irradiation alone. All patients had local fields only to encompass known disease with a l-2 cm margin. The median field size was 6 X 7 cm. The median dose was 5040 cGy in 160 cGy fractions over 4 1 treatment days. The range of minimal tumor doses was 3786 cGy to 6100 cGy (excluding one patient who received 1522 cGy and failed to complete therapy). Fraction size ranged from 76-200 cGy. One patient was treated with orthovoltage (in 1959) and 32 patients with megavoltage therapy. Eight of these patients had cobalt tele-
Limited to ON ON and chiasm ON, chiasm, and CS
5 (15) g (24) 20 (61)
ON = Optic nerve; CS = Contiguous structures.
therapy with the remainder receiving 4 MV or higher energy. None received adjuvant chemotherapy. Follow-up Of patients alive at last follow-up visit, the median time of follow-up was 12.3 years with a range from 2 to 3 1 years. Statistical methods Survival analysis and comparisons were made with statistical software (9). Actuarial overall and progression-free survival were performed using the product limit method (16). The one patient who died from intercurrent disease was considered dead of disease and not censored. Tests of significance utilized the generalized Wilcoxon (Breslow) statistic (4). Multivariate analysis was based on the Cox Proportional Hazards Regression Model (7). RESULTS
Survival The overall actuarial survivals at 5, 10, and 15 years were 94, 81, and 74%, respectively. The overall progression-free survivals at 5, 10, and 15 years were 85, 75, and 75%, respectively (Fig, 1). The patient status at last followup was 22 (67%) alive and 11 (33%) dead. Ten of the 11
ii
5 60 a 2 _ !5 g 40 -
Actuonol
Surww/ (n =33/
2
Table 2. Visual acuity at diagnosis in the affected eye Visual acuity 20/20-20140 20/50-20/200 >20/200 Total
20
No. 9 7 3 19
01
1 0
I
I
1
I
5
IO
15
20
TIME (years)
Fig. I. Overall actuarial and progression-free survival.
929
Gliomas of the optic nerve and chiasm 0 J. J. KOVALIC et al. Table 4. Prognostic variables and their survival significance
15 yr
Univariate analysis p value*
Multivariate analysis p value+
100 86 51
100 86 57
0.11
0.052
74 92 75
46 92 15
46 92 75
0.12
0.518
3 30
0 90
0 79
0 79
12 21
75 91
15 75
75 75
0.92
0.276
27 6
89 67
77 67
17 67
0.25
0.065
25 8
84 88
71 88
71 88
0.48
0.029+
11 22
90 76
70 76
70 76
0.78
0.590
Progression-free survival (%) Variable Extent of primary Optic nerve only ON and chiasm ON, chiasm, and CS Extent of surgery None Biopsy only Partial resection Dose of radiation 15 years Neurofibromatosis Present Absent
No. of patients
5
10
5 8 20
100 100 68
11 14 8
Inr J. Radiamn Oncology Btd. Phys., Vol. 18, pp. 927-932 Ptinted in the U.S.A. All rights mawd.
??Brief Communication
RADIATION THERAPY FOR GLIOMAS OF THE OPTIC NERVE AND CHIASM JEFFREY J. KOVALIC, M.D., PERRY W. GRIGSBY, M.D.,* MARK J. SHEPARD, B.S., BARBARA B. FINEBERG, B.A. AND PATRICK R. THOMAS, M.B., M.R.C.P., F.R.C.R. Mallinckrodt
Institute of Radiology, Radiation Oncology Center, Washington University School of Medicine. St. Louis, MO
Thirty-three patients with optic glioma seen over a 30-year period were reviewed. Five patients (15%) had tumor confined to the optic nerve, 8 patients (24%) had optic nerve and chiasmal involvement, and the remaining 20 patients (61%) had invasion of contiguous structures as well as chiasmal involvement. Eleven patients (33%) had a history of neurofibromatosis. Two-thirds of the patients had either a biopsy or a partial resection of the tumor, with the remaining one-third being clinically diagnosed. All patients received irradiation to local fields. The median dose was 5040 cGy in 160 cGy fractions. Of patients alive at last follow-up, the median time of follow-up was 12.3 years. The 5, lo-, and 15-year overall actuarial survivals were 94, 81, and 74%, respectively. Univariate and multivariate analysis were performed on the following clinical variables: extent of primary tumor, extent of surgery, dose of radiation, gender, race, age, and presence or absence of neurofibromatosis. Extension of the primary lesion to the optic chiasm and age less. than or equal to 15 years were the only two variables to have statistically significantly inferior 15-year progression free survivals by multivariate analysis. Eighteen (55%) patients had treatment related complications with most involving the pituitary gland. We conclude that postoperative radiotherapy is beneficial in patients with chiasmal involvement and those with incomplete resections. A minimum tumor dose of 4000 cGy is recommended. Radiation therapy, Optic nerve glioma, Pediatric tumors.
to follow-up, the current analysis is of the remaining 33 patients. The median age was 4-years old with a range from 1 to 34 years. Most patients (29) were under 16 years of age. The female-to-male ratio was 2 1:12 with 27 patients being white and six patients being black. Eleven patients (33%) had a history of neurofibromatosis. Presenting signs and symptoms are shown in Table 1. The median duration of symptoms prior to presentation was 1 year (mean = 23 months). Optic atrophy and visual field defects were the most common presenting signs. Visual acuity in the affected eye was recorded at diagnosis in 19 patients (Table 2). Endocrine abnormalities were noted in two patients at diagnosis. One patient had growth hormone deficiency and another patient had precocious puberty. Radiographic evaluation revealed that 23 out of 26 patients who underwent a pneumoencephalogram had an abnormality, whereas 1 l/ 14 patients with an angiogram showed vascular aberration. Twenty-six patients had plain films done with the Reese view that revealed optic foramen widening in 25/26 (96%) involvement of the pituitary fossa in 12/26 (46%), and chiasmal involvement in 18/ 26 (69%). Computerized tomography (CT) scan was per-
INTRODUCTION
Chiasmal and optic nerve gliomas are rare tumors of childhood. Their general incidence is 1 in 100,000 patients (2) with 90% presenting in the first two decades of life (28). Most of these neoplasms are low-grade astrocytomas with a long natural history. The untreated clinical course is that of progressive blindness or death from local tumor progression in a significant number of patients (1). Lesions that extend outside of the optic nerve to involve the chiasm and neighboring structures cannot be totally resected without substantial morbidity. It is in these patients that radiotherapy is most useful. This paper analyzes our 30year experience in treating predominantly locally advanced gliomas of the anterior visual system with radiation therapy.
METHODS
AND
MATERIALS
Patient characteristics Thirty-five patients completed cranial irradiation for optic nerve glioma at the Mallinckrodt Institute of Radiology between 1956 and 1986. With two patients lost * Supported by the American Cancer Society Clinical Oncology Career Development Award. Reprint requests to: Perry W. Grigsby, M.D., Radiation Oncology Center, 4939 Audubon, Suite 5500, St. Louis, MO 63110.
Acknowledgements-The authors wish to thank Elaine Pirkey for her help in preparation of this manuscript and Leslie MacConnell-Clubbs for her illustration. Accepted for publication 12 October 1989. 927
928
I. J. Radiation Oncology 0 Biology 0 Physics
April 1990, Volume 18, Number 4
Table 3. Disease extent
Table 1. Signs and symptoms at presentation
No. (%)
No. (%)*
Optic atrophy Visual field defect Strabismus Headache Total blindness Nystagmus Proptosis Light perception only Papilledema
20 20 9 9 g g 7 6 5
(64) (61) (30) (30) (26) (26) (23) (20) (16)
* Per cent of total number of patients for which data were available.
formed in all patients presenting after 1975 and showed chiasmal involvement in 7/9 (78%) patients. Twenty-two patients (67%) had histologic confirmation of the diagnosis and 11 patients (33%) were clinically diagnosed. Of the 22 surgically treated patients, 14 (64%) had biopsy only and 8 (36%) underwent a partial excision. None had total excision. Nineteen of 22 patients (86%) with a tissue diagnosis had low-grade astrocytomas, 1 had a high grade astrocytoma, and 2 had mixed oligodendroglioma/astrocytoma. The clinical diagnosis of optic nerve glioma was given to those patients who presented with visual field deficits, optic atrophy, and radiographic evidence of this disease. In these cases it was felt unsafe to biopsy the lesion. The largest tumor diameter was documented from the operative notes or radiographic studies. The median was 4 cm, mean 3.7 cm, and range from 1 to 7 cm. Only 5 ( 15%) patients had disease limited to the optic nerve (Table 3), with most patients exhibiting extrachiasmal extension. Radiation parameters Twenty-two patients were treated with postoperative irradiation and 11 underwent irradiation alone. All patients had local fields only to encompass known disease with a l-2 cm margin. The median field size was 6 X 7 cm. The median dose was 5040 cGy in 160 cGy fractions over 4 1 treatment days. The range of minimal tumor doses was 3786 cGy to 6100 cGy (excluding one patient who received 1522 cGy and failed to complete therapy). Fraction size ranged from 76-200 cGy. One patient was treated with orthovoltage (in 1959) and 32 patients with megavoltage therapy. Eight of these patients had cobalt tele-
Limited to ON ON and chiasm ON, chiasm, and CS
5 (15) g (24) 20 (61)
ON = Optic nerve; CS = Contiguous structures.
therapy with the remainder receiving 4 MV or higher energy. None received adjuvant chemotherapy. Follow-up Of patients alive at last follow-up visit, the median time of follow-up was 12.3 years with a range from 2 to 3 1 years. Statistical methods Survival analysis and comparisons were made with statistical software (9). Actuarial overall and progression-free survival were performed using the product limit method (16). The one patient who died from intercurrent disease was considered dead of disease and not censored. Tests of significance utilized the generalized Wilcoxon (Breslow) statistic (4). Multivariate analysis was based on the Cox Proportional Hazards Regression Model (7). RESULTS
Survival The overall actuarial survivals at 5, 10, and 15 years were 94, 81, and 74%, respectively. The overall progression-free survivals at 5, 10, and 15 years were 85, 75, and 75%, respectively (Fig, 1). The patient status at last followup was 22 (67%) alive and 11 (33%) dead. Ten of the 11
ii
5 60 a 2 _ !5 g 40 -
Actuonol
Surww/ (n =33/
2
Table 2. Visual acuity at diagnosis in the affected eye Visual acuity 20/20-20140 20/50-20/200 >20/200 Total
20
No. 9 7 3 19
01
1 0
I
I
1
I
5
IO
15
20
TIME (years)
Fig. I. Overall actuarial and progression-free survival.
929
Gliomas of the optic nerve and chiasm 0 J. J. KOVALIC et al. Table 4. Prognostic variables and their survival significance
15 yr
Univariate analysis p value*
Multivariate analysis p value+
100 86 51
100 86 57
0.11
0.052
74 92 75
46 92 15
46 92 75
0.12
0.518
3 30
0 90
0 79
0 79
12 21
75 91
15 75
75 75
0.92
0.276
27 6
89 67
77 67
17 67
0.25
0.065
25 8
84 88
71 88
71 88
0.48
0.029+
11 22
90 76
70 76
70 76
0.78
0.590
Progression-free survival (%) Variable Extent of primary Optic nerve only ON and chiasm ON, chiasm, and CS Extent of surgery None Biopsy only Partial resection Dose of radiation 15 years Neurofibromatosis Present Absent
No. of patients
5
10
5 8 20
100 100 68
11 14 8
