[en] Purpose: To evaluate efficacy and tolerance of HFRT (BID) in children treated for Nb with curative intent. Material and method: Retrospective analysis of a series of 29 children treated from July 1989 through March 1995 at the Institut Gustave Roussy. Results: Mean age was 38.4 months (range 3-103) with only 1 patient under 12 months, and M/F sex ratio 1. Initial primary was abdominal in 28 and pelvic in 1. Ten children had limited disease at presentation (stage II = 1, stage III = 9) while 19 had disseminated disease (stage IVs = 1, stage IV = 18). Nmyc expression was assessed in 17 of the latest patients and was found amplified in 13. Initial therapy consisted in induction chemotherapy (CAdO = CPM, ADM, VCR ; VP16 + CDDP or carboplatine) 4-8 cycles, followed by resection of the primary and regional lymphatic drainage. The patient with stage II had primary total tumor resection and no chemotherapy. One patient was inoperable for medical reason. Intensive chemotherapy with autologous or allogenous BMT was conducted in children with metastatic disease. HFRT was administered for gross residual disease in most patients (= 22) or microscopic disease with Nmyc amplification (= 7). Three patients were in local progression before initiation of radiotherapy. Target volume varied throughout the time (postoperative or preoperative tumor volume). Total dose was adapted to the children's age and extension of the disease and 30-35 Gy were generally delivered (range 20-40). Two daily fractions of 1 Gy (range 0.8, 1) with at least a 6 hours interval were delivered 5 days a week using 4.5 MV X-rays (28), 18 MV X-rays (= 1). Gastrointestinal toxicity was very limited and few children experienced mild thombopenia, all of whom had receive intensive chemotherapy before. With a median follow-up of 37 months (range 23-74), 15 children are alive of whom 13 in CR, and 2 with distant metastases. Eight patients (27,5%) failed loco-regionally : 4 (13,7%) in field, 3 marginally and 1 outside the fields. When Nmyc was amplified, local failure rate was (4(13)) (in field = 3). These results can be compared to those of 37 children treated earlier at our institution with conventional fractionation RT and of whom 7 failed locally (in field = 5). This population presented more limited disease with 62% stage II-III. At last follow-up, no major growth impairment or other late sequelae was noted in HFRT group. Conclusion: We conclude that HFRT is probably as effective as conventional RT if the treated volume encompasses the preoperative residual disease. A longer follow-up will tell if HFRT reduces effectively the late morbidity in these children