PATTERNS OF FAILURE FOLLOWING TOTAL-BODY IRRADIATION AND BONE-MARROW TRANSPLANTATION WITH OR WITHOUT A RADIOTHERAPY BOOST FOR ADVANCED NEUROBLASTOMA

Purpose: To evaluate the patterns of failure and outcome of patients u ndergoing high-dose chemotherapy, total body irradiation (TBI), and bo ne marrow transplantation (BMT) for advanced/relapsed pediatric neurob lastoma, with emphasis on the impact of a radiotherapy boost to primar y and metastatic sites. Methods and Materials: Between May 1986 and Ju ne 1993, 26 patients with advanced neuroblastoma underwent high-dose c hemotherapy and TBI followed by BMT at our institution. The majority o f patients were over the age of 2 years (73%) and were Stage IV at dia gnosis (81%). Multiple metastatic sites were involved including bone ( 17), bone marrow (15), distant nodes (11), liver (5), lung (4) and bra in (1). Twenty patients (77%) received cyclophhosphamide (50 mg/kg x 4 days) and TBI as consolidation therapy. TBI was delivered to a total dose of 12 Gy given in 2 Gy twice daily (b.i.d.) fractions over the 3 days preceding bone marrow infusion. A local radiotherapy boost of 8-2 4 Gy was given to 13 out of 26 patients (50%) to the primary and/or me tastatic sites immediately prior to or following induction chemotherap y according to physician judgement, Sites not amenable to a radiothera py boost included the bone marrow, diffuse/bilateral lung involvement, and multiple bone metastases (> four sites). Results: The actuarial o verall survival of the 26 patients was 40.4% at 3 and 5 years, with a progression-free survival at 5 years of 38.5%. Six patients died of tr ansplant-related toxicity (23%). The use of cyclophosphamide as high-d ose consolidation chemotherapy was significantly better than other mul tidrug regimens used in terms of overall survival (p < 0.0001) and pro gression-free survival (p = 0.0004). The presence of liver involvement prior to BMT was a significant adverse prognostic factor by multivari ate analysis. Of the 20 patients surviving the transplant, 10 (50%) un derwent a local radiotherapy boost, The patterns of failure were as fo llows: 3 out of 10 ''boost'' patients failed overall, none in previous (old) sites of disease only, 1 in new sites only, and 2 in old and ne w sites; 6 out of 10 ''no boost'' patients failed overall, 4 in old si tes only, none in new sites only, and 2 in old and new sites, There wa s a trend toward improved 5-year progression-free survival in patients surviving the transplant that received a boost (68% vs. 33%,p = 0.24) . A failure analysis was also performed for each of the 59 initially i nvolved sites, of which the majority (64%) were amenable to a radiothe rapy boost. Overall, there is a trend toward less failure in sites ame nable to a radiotherapy boost that were irradiated (1 out of 10) vs. t hose not irradiated (6 out of 28). Failure in the liver ocurred in thr ee out of four of the patients with liver involvement that did not rec eive boost radiotherapy, whereas all seven patients with distant nodal involvement were controlled without a boost. Long-term sequelae inclu de learning difficulties (2), cataract formation (1), and hearing loss (2). Sequelae attributable to a radiotherapy boost occurred in only o ne patient who received whole brain radiotherapy and developed a catar act and learning difficulties. Conclusion: We have found an actuarial 5-year survival rate of 40.4% for patients with advanced neuroblastoma treated with BMT, which compares favorably with results of other publ ished series. Disease recurrence following BMT was most common in prev ious sites of disease. The majority (64%) of these sites were amenable to a radiotherapy boost. An analysis of failure suggests that a low-d ose radiotherapy boost improves control of these sites.