DISCUSSION
In this study, the 5-year EFS rates of all high-risk neuroblastoma patients who underwent the double-conditioning regimen comprising thiotepa and melphalan and of CR/VGPR/PR patients before HDC were 41.5% and 50.0%, respectively. Considering the poor prognosis of patients who did not undergo anti-GD2 antibody immunotherapy, the outcome of our patients who showed CR/VGPR/PR before HDC was comparable to previous studies.3,22 It is partly becauseMYCN -amplified high-risk neuroblastoma patients show a superior survival rate.
Comprehensive studies have considered MYCN amplification as a poor prognostic factor in high-risk neuroblastoma.23,24 In contrast, in this study,MYCN amplification was a favorable prognostic factor, althoughMYCN -amplified high-risk neuroblastoma patients showed a higher response rate after induction chemotherapy compared toMYCN -non-amplified high-risk neuroblastoma patients. This result indicated that the double-conditioning regimen might be appropriate for the treatment of MYCN -amplified high-risk neuroblastoma.
Some studies have shown favorable outcomes in MYCN -amplified high-risk neuroblastoma patients after intensive treatment, including tandem HDC.5,6 This finding suggests thatMYCN -amplified high-risk neuroblastoma might be more chemosensitive and more likely to benefit from treatment intensification. The double-conditioning regimen is quite unique in that it consists of two cycles of a drug combination (thiotepa and melphalan) for 2 consecutive weeks to safely administer the maximum dose of these drugs.8 Therefore, this highly potent regimen could be especially effective in MYCN -amplified high-risk neuroblastoma, as shown in some studies using tandem HDC.
In this study, MYCN -amplified high-risk neuroblastoma patients who showed CR/VGPR before HDC especially had good prognosis. Kushner et al. (2014) reported that MYCN -amplified high-risk neuroblastoma patients show extreme dichotomy in the clinical course;MYCN -amplified and MYCN -non-amplified high-risk neuroblastoma patients who showed good response to induction chemotherapy also showed similar good long-term outcomes, whileMYCN -amplified high-risk neuroblastoma patients who did not show CR/VGPR developed earlier progression with a significantly poor outcome compared to MYCN -non-amplified high-risk neuroblastoma patients.25 The high frequency of chemosensitiveMYCN -amplified high-risk neuroblastoma in our cohort might have led to our positive results.
Other favorable prognostic factors shown in this study, such as good remission status (CR/VGPR) and bone metastasis negativity, were similar to previous reports.15,26 High MSI (>1) and older age (≥5 years), extracted as variables for risk stratification from the analysis of the HR-NBL-1/SIOPEN study, didn’t have any prognostic impact.25
The toxicity of the double-conditioning regimen is relatively severe. In this study, 2 patients died of regimen-related toxicity, and 1 patient who developed grade 4 TMA died from a pulmonary hemorrhage 1 year post-HDC. Acute mucositis was frequently observed. The dose-finding experience of the double-conditioning regimen for several solid tumors showed severe gastrointestinal toxicity, microangiopathy, renal tubular acidosis, and neurological toxcity8. Another study also reported excessive gastrointestinal toxicity and delayed platelet recovery.27 Okada et al. (2019) reported the successful prevention of renal toxicity by decreasing the doses of thiotepa and melphalan in patients less than 2 years old or in those showing low renal function while gastrointestinal toxicity was still severe.9 Therefore, the double-conditioning regimen warrants special care when applied to high-risk neuroblastoma patients who undergo intensive induction therapy.
Recently, thiotepa as a double-conditioning regimen therapeutic was approved in Japan along with melphalan with a reduced cumulative dose from 280 to 210 mg/m2. The feasibility of this modified double-conditioning regimen has been confirmed in early clinical trials.28 Further studies to confirm the efficacy of this modified double-conditioning regimen are required.
This retrospective study had a few limitations. First, there was selection bias of patients. Especially, the difference in the response rate between MYCN -amplified and MYCN -non-amplified high-risk neuroblastoma patients warrants careful interpretation. Second, treatments other than the double-conditioning regimen were heterogeneous. However, treatment components did not exhibit any prognostic effects, except for second SCT. Third, we could not cover the entire range of toxic profiles because data collection was based on only medical records. Finally, because of a lack of a common long-term follow-up method, this study did not report long-term toxicity. Therefore, long-term complications in patients who underwent this potent treatment merit considerable care.