paediatrics Brussels 17

I. J. Radiation Oncology ● Biology ● Physics

954

Volume 42, Number 5, 1998

Table 1. Tumor grade and site of origin

Table 2. Tumor type and site of origin

Tumor grade

Brain

Spine

Total

Tumor type Brain Upper spine Lumbar spine Total

1–2 3–4

36

34

70 10 80

Myxopapillary

0

0

12

12 68 80

9

1

Other types

45 45

16 16

7

Total

45

35

Total

19

performed in 17 (21%) patients, subtotal resections in 57 (71%) patients, and biopsy alone in the remaining 6 (8%) patients (Table 3). Of the 45 patients with tumors arising in the brain, gross total resections were performed in 7 (16%), subtotal resections in 36 (80%), and biopsies in 2 (4%). Of the 35 patients with tumors arising in the brain, gross total resections were performed in 10 (29%), subtotal resections in 21 (67%), and biopsies in 4 (11%). Tumors were graded on a 4 grade scale: grade 1 lesions lacked mitotic activity; grade 2 lesions exhibited occasional mitoses; grade 3 lesions showed brisk mitotic activity, and often some endothelial proliferation; grade 4 tumors showed high mitotic indices and prominent endothelial prolifera- tion. As a rule, high-grade (3 and 4) were markedly cellular and often showed nucleolar prominence. Palisading necro- sis was most often seen in grade 4 lesions. Cellular pleo- morphism and simple non-palisading necrosis were not considered useful grading parameters. Post-operative therapy All 80 patients underwent radiation therapy as a portion of their initial treatment sequence (Table 3). The most common indication for radiotherapy was the presence of residual tumor in 63 patients. The remaining 17 patients received adjuvant therapy following gross total resection, 78 of whom received treatment at Mayo Clinic and 2 at other institutions. Treatment was delivered with 4, 6, or 10 MV photons. Radiotherapy was delivered to the craniospi- nal axis in 20 patients, to the whole brain in 3 patients, to the total spine in 6 patients, and to the primary tumor bed (brain or spine) alone in the remaining 51 cases. Doses adminis- tered to the primary tumor bed and to areas of gross disease ranged from 24 to 59.4 Gy (median 49.7 Gy) in 1.8–2.0 Gy fractions. Twenty-five (31%) of the 80 patients received doses less than 45 Gy to the primary tumor bed. Prophy- lactic doses administered to uninvolved regions ranged from 28.5 to 40.0 Gy. The doses delivered and the field arrange- ments used were based on the treating physician’s prefer- ences.

METHODS AND MATERIALS

Study group Between 1963 and 1994, 80 patients with ependymomas were identified from the Radiation Oncology Tumor Reg- istry. The outcome of therapy was retrospectively analyzed. Included in the study were 33 females and 47 males ranging in age from 1 to 69 years, with a median age of 33.5 years. Statistical methods Survival and local control rates were the principle end- points of the analysis. Local failure was defined as progres- sion of local symptoms or as tumor growth documented on neuroimaging studies. Survival and local control rates were determined with the Kaplan-Meier product limit method. The statistical significance of differences between the curves was determined by the log-rank test. Multivariate analysis was performed with the Cox proportional hazards model. Follow-up ranged from 5 to 30 years, with a median follow-up of 10.4 years. Surgery and pathologic features Primary sites included the supratentorial brain in 13 pa- tients, infratentorial brain in 32 patients, and spinal cord in 35 patients. Seventy patients were found to have low-grade (1 and 2) tumors and 10 had high-grade (3 and 4) tumors. The distribution of tumor grade, sub-type, and site of origin are found in Tables 1 and 2. Myxopapillary tumors were uniformally grade 1 tumors found in the lumbar spine region. The following studies were performed to assess for seeding of the central nervous system (CNS): myelography in 40 patients, MRI in 21 patients, CT scans in 25 patients, and CSF cytology in 24 patients. Based on the findings of clinical studies (myelography, MRI, CT, or CSF cytology), 4 patients (6%) were found to have seeding of the CNS at the time of diagnosis, including 2 of 70 (3%) with low-grade tumors and 2 of 10 (20%) with high-grade tumors. The extent of tumor resection was determined by a care- ful review of operative reports. Gross total resections were

Table 3. Summary of therapy

Surgery

Radiotherapy

Tumor grade

Biopsy

STR GTR Local field

WBRT/WSRT

CSRT

Adjuvant chemotherapy

1–2 3–4

6 0

49

15

49

8 1

13

1 0

8

2

2

7

STR 5 subtotal resection; GTR 5 gross total resection; WBRT 5 whole brain radiotherapy; WSRT 5 whole spine radiotherapy; CSRT 5 craniospinal radiotherapy.