High-dose spatially fractionated GRID radiation therapy (SFGRT): a comparison of treatment outcomes with Cerrobend vs. MLC SFGRT.

PURPOSE: Spatially fractionated GRID radiotherapy (SFGRT) using a customized Cerrobend block has been used to improve response rates in patients with bulky tumors. The clinical efficacy of our own multileaf collimator (MLC) technique is unknown. We undertook a retrospective analysis to compare clinical response rates attained using these two techniques. METHODS AND MATERIALS: Seventy-nine patients with bulky tumors (median diameter, 7.6 cm; range, 4-30 cm) treated with SFGRT were reviewed. Between 2003 and late 2005, the Cerrobend block technique (n = 39) was used. Between late 2005 and 2008, SFGRT was delivered using MLC-shaped fields (n = 40). Dose was prescribed to dmax (depth of maximum dose) and was typically 15 Gy. Eighty percent of patients in both groups received external beam radiotherapy in addition to SFGRT. The two-sided Fisher-Freeman-Halton test was used to compare pain and mass effect response rates between the two groups. RESULTS: Sixty-one patients (77%) were treated for palliative intent and 18 (23%) for curative intent. The majority of patients had either lung or head-and-neck primaries in both groups; the most frequent site of SFGRT application was the neck. The majority of patients complained of either pain (65%) or mass effect (58%) at intake. Overall response rates for pain and mass response were no different between the Cerrobend and MLC groups: pain, 75% and 74%, respectively (p = 0.50), and mass effect, 67% and 73%, respectively (p = 0.85). The majority of toxicities were Grade 1 or 2, and only 3 patients had late Grade 3-4 toxicities. CONCLUSIONS: MLC-based and Cerrobend-based SFGRT have comparable and encouraging response rates when used either in the palliative or curative setting. MLC-based SGFRT should allow clinics to more easily adopt this novel treatment approach for the treatment of bulky tumors.

High-dose chemotherapy and autologous hematopoietic progenitor cell rescue in children with recurrent medulloblastoma and supratentorial primitive neuroectodermal tumors: the impact of prior radiotherapy on outcome.

BACKGROUND: The role of myeloablative chemotherapy in children with recurrent medulloblastoma and supratentorial primitive neuroectodermal tumors (MB/ST-PNET) is controversial, in particular in patients who develop recurrent disease after craniospinal radiotherapy. METHODS: In this retrospective analysis, the authors investigated the outcome of children with recurrent MB/ST-PNET who were referred for myeloablative chemotherapy and autologous hematopoietic progenitor cell rescue at Childrens Hospital Los Angeles. RESULTS: Thirty-three children were referred for myeloablative chemotherapy: Fourteen of those children were never transplanted because of pre-transplant adverse events, and 19, including 6 without and 13 with previous irradiation, underwent transplant. Conditioning regimens included a backbone of thiotepa, which was given either in a single cycle or in multiple sequential cycles. The 3-year post-transplant event-free survival rate in unirradiated versus previously irradiated children was 83% +/- 15% versus 20% +/- 12%, respectively (P = .04). One child who had never been exposed to radiotherapy died of toxicity; the other children received post-transplant radiotherapy and remained disease free. Nine previously irradiated children experienced 4 toxic deaths and 6 tumor recurrences (1 patient had both): An interval of <1 year between initial radiotherapy and myeloablative chemotherapy predicted a greater risk of toxic death (P = .02), whereas a history of meningeal metastases at diagnosis and a poor response to the initial rescue therapy predicted a greater risk of post-transplant recurrence (P = .03 and P = .08, respectively). CONCLUSIONS: Myeloablative doses of thiotepa-based chemotherapy and radiotherapy were able to cure most children who had radiotherapy-naive, chemoresponsive recurrences. Children who developed recurrences after craniospinal radiotherapy had poorer outcomes; however, cure was possible in those who had good prognostic features at presentation, chemoresponsive recurrences, and a long interval between initial radiotherapy and myeloablative chemotherapy.