Therapy for ependymoma includes aggressive surgical intervention and radiotherapy administered by use of methods that keep the risk of side-effects to a minimum. We extended this treatment approach to include children under the age of 3 years with the aim of improving tumour control.

Between July 11, 1997, and Nov 18, 2007, 153 paediatric patients (median age 2·9 years [range 0·9–22·9 months]) with localised ependymoma were treated. 85 patients had anaplastic ependymoma; the tumours of 122 were located in the infratentorial region, and 35 had received previous chemotherapy. Patients received conformal radiotherapy after definitive surgery (125 patients had undergone gross total, 17 near total, and 11 subtotal resection). Doses of 59·4 Gy (n=131) or 54·0 Gy (n=22) were prescribed to a 10 mm margin around the target volume. Disease control, patterns of failure, and complications were recorded for patients followed over 10 years. Overall survival, event-free survival (EFS), cumulative incidence of local recurrences, and cumulative incidence of distant recurrences were assessed. Variables considered included tumour grade, tumour location, ethnic origin, sex, age when undergoing conformal radiotherapy, total radiotherapy dose, number of surgical procedures, surgery extent, and preradiotherapy chemotherapy.

After a median follow-up of 5·3 years (range 0·4–10·4), 23 patients had died and tumour progression noted in 36, including local (n=14), distant (n=15), and combined failure (n=7). 7-year local control, EFS, and overall survival were 87·3% (95% CI 77·5–97·1), 69·1% (56·9–81·3), and 81·0% (71·0–91·0), respectively. The cumulative incidences of local and distant failure were 16·3% (9·6–23·0) and 11·5% (5·9–17·1), respectively. In the 107 patients treated with immediate postoperative conformal radiotherapy (without delay or chemotherapy), 7-year local control, EFS, and overall survival were 88·7% (77·9–99·5), 76·9% (63·4–90·4), and 85·0% (74·2–95·8), respectively; the cumulative incidence of local and distant failure were 12·6% (5·1–20·1), and 8·6% (2·8–14·3), respectively. The incidence of secondary malignant brain tumour at 7 years was 2·3% (0–5·6) and brainstem necrosis 1·6% (0–4·0). Overall survival was affected by tumour grade (anaplastic vs differentiated: HR 3·98 [95% CI 1·51–10·48]; p=0·0052), extent of resection (gross total vs near total or subtotal: 0·16 [0·07–0·37]; p<0·0001), and ethnic origin (non-white vs white: 3·0 [1·21–7·44]; p=0·018). EFS was affected by tumour grade (anaplastic vs differentiated: 2·52 [1·27–5·01]; p=0·008), extent of resection (gross total vs near total or subtotal: 0·20 [0·11–0·39]; p<0·0001]), and sex (male vs female: 2·19 [1·03–4·66]; p=0·042). Local failure was affected by extent of resection (gross total vs near total or subtotal: 0·16 [0·067–0·38]; p<0·0001), sex (male vs female: 3·85 [1·10–13·52]; p=0·035), and age (<3 years vs ≥3 years: 3·25 [1·30–8·16]; p=0·012). Distant recurrence was only affected by tumour grade (anaplastic vs differentiated: 4·1 [1·2–14·0]; p=0·017).

Treatment of ependymoma should include surgery with the aim of gross-total resection and conformal, high-dose, postoperative irradiation. Future trials might consider treatment stratification based on sex and age.

American Cancer Society and American Lebanese Syrian Associated Charities (ALSAC).