Impact of Initial Prognostic Factors and Intensity of Salvage Therapy on the Outcome of Progressive / Refractory High-Risk Neuroblastoma

Background: High-risk Neuroblastoma (N. B) patients have a poor outcome with 5-year survival rates of 50%. Patients with stage 4 disease or MYC-N amplification showed post-progression 5y O. S. of 7% to 8%. Other studies proved the same dismal outcome in high-risk relapsed patients. This study aimed to detect the O. S. and EFS of N. B patients post-progression. Secondary to explore, if initial prognostic factors, high-intensity salvage therapy and other treatment modalities could improve the outcome of progressive /refractory disease. Methods: Seventy patients of high-risk Neuroblastoma needed salvage therapy, either due to refractory/progressive disease or irresectability of the primary tumor. Initial prognostic factors and different treatment strategies were collected and correlated with the outcome. Results: Fifty-seven (57 /70) patients died from progressive disease with a median survival of 20.6 months with three y EFS and O. S. of 9.5% and 35.7%, respectively. Objective response (CR/VGPR/PR) post-induction, consolidation by HSCT, radiotherapy, and maintenance therapy; affected survival significantly post salvage therapy. Multivariate analysis revealed that the only independent factor that significantly affected O. S was maintenance therapy. The independent factors that affected the EFS negatively were the presence of liver metastases, poor response post-induction, and not administering radiotherapy. Conclusion: Response to induction had a significant impact on the outcome post salvage. Salvage therapy did not improve the outcome for those with inadequate induction response. Initial front-line targeted therapy like antiGD2 is needed to improve the outcome, especially for chemo-resistant ones.


Introduction
Neuroblastoma (N. B.) is an embryonal malignancy arising from neuroblasts of the sympathetic nervous system. Patients with metastatic disease or high-risk biologic features have a poor outcome, with 50% survival at five years [1].
Age, stage, histopathology, and biological features (MYC-N gene status, DNA ploidy) are the prognostic factors affecting the outcome [2]. Many studies were carried out to improve the outcome of this aggressive disease by using myeloablative conditioning, followed by stem cell rescue. Mega therapy showed improvement in event-free survival (EFS) compared to conventional treatment [3]. Treatment of high-risk Neuroblastoma consists of induction chemotherapy, local control by surgery, consolidation by autologous transplant, radiotherapy followed by maintenance with both immunotherapy and a differentiating agent (Isotretinoin) [3]. Cis retinoic acid and immunotherapy are used to eradicate minimal residual disease [4]. N. B is an aggressive disease with a significant portion showed refractory/progressive disease or inadequate response. The most common salvage regimen used is the ICE regimen (Ifosamamide, Carboplatin, and Etoposide) [5]. Patients with stage 4 disease or MYCN amplification showed post-progression 5y O. S. of 7% to 8%. Other studies proved the same dismal outcome in high-risk relapsed patients [6,7]. This study aimed to detect O. S. and EFS of N. B. patients post-progression. The Secondary aim was to explore if the initial prognostic factors, high-intensity salvage therapy, and other treatment modalities could improve the outcome of progressive /refractory disease.

Methods
The present study included seventy (70) patients diagnosed with N. B, with a progressive or refractory disease in the period between January 2007 and December 2015 and were followed till December 2017. All patients were younger than 18 years of age and diagnosed with high-risk N. B. classified according to COG ANBLOOB1 risk stratification. SFOP NB 97 protocol was given as induction for fear of toxicities from the more aggressive COG A3973 protocol. Progression was defined as the appearance of new lesion or 25% increase in any existing lesion after achieving CR/ VGPR [8]. Refractory patients included those who had not achieved at least a P. R. to induction therapy (minimum of four cycles) and did not show progression [9]. Initial prognostic factors like age, initial tumor site, pathology, MYCN status, stage, different metastatic sites, extent of resection, response to induction, tumor viability, consolidation, and maintenance therapy post salvage were analyzed and correlated with outcome post salvage.

Treatment
All patients received induction, 4-8 cycles of Vp16/CARBO (Etoposide, Carboplatin) alternating with CADO (Cyclophosphamide, Adriamycin, Vincristine) according to SFOP NB90 protocol [10]. Patients were evaluated after four cycles and at the end of treatment. International Neuroblastoma Response Criteria (INRC) was used to stratify response. If feasible, surgery was done after four cycles.

Statistical Analysis
Statistical analysis was done using SPSS© Statistics version 22. Survival analysis was done using the Kaplan-Meier method. The log-rank test was used to compare two survival curves. A p-value < 0.05 was considered significant. Event-free survival post-progression (EFS): was defined as the time between diagnosis of first progression/relapse to the date of any event (progression or relapse or death). O. S. post-progression was defined as the time between diagnosis of first progression/relapse and death or date of the last follow-up. Multivariate analysis was done using the Coxproportional Hazard regression model for the significant factors affecting the survival on univariate analysis. Hazard ratio (H. R.), with its 95% confidence interval (CI), was used for risk estimation.

Results
The study included 70 patients with progressive/refractory high-risk N. B. Patients' age ranged from 4 to 120 months (median: 36 months ± 22.8 SD) with slight male predominance with M/F ratio 1.29:1, All clinic pathological data were summarized in table 1. Salvage therapy (ICE) was given due to local inoperability in 11 (15.7%) patients, refractory disease at the end of induction in 26 (37.1%) patients, and due to progressive disease in 33 (47.1%) patients. Figure 1.

Outcome of Salvage Therapy
Eight patients died from septic shock. Sixty-two patients were evaluated; thirteen patients (21%) achieved C. R. Twenty-nine patients (46.7%) showed an objective response. Thirty-three (53.3%) patients showed progressive disease ( Figure 1). The included patients were followed for a median duration of 22 months (ranged 2.3 -115.9 months). The median O. S. and EFS of the whole group were 20.6 and 9.6 months, respectively. The cumulative O. S. and EFS 35.7% and 9.5% at 36 months, as shown in (Figures 2 and 3).

Discussion
In high-risk N. B., the outcome is more correlated to the biological characteristics than the type of chemotherapy or extent of resection [12]. The poorer prognosis was associated with the high serum ferritin, bone, and B. M. metastasis, as well as to persisting skeletal and B. M. disease post-induction therapy [13]. There is no much data in the literature about post-progression O. S. and PFS. In the present study, patients with amplified MYC-N gene showed lower 3y EFS and O. S. compared to non amplified ones, although it showed clinical significance, it lacked statistical significance due to small numbers. Jennifer concluded that the only disease characteristics associated with worse 5-year O. S. postprogression were MYC-N amplification and multifocal soft tissue disease [14]. In the present study, initial metastatic sites did not affect the survival except for metastatic liver disease. Basta reported that MYC-N amplification (P <0.0001) and liver metastases (P= 0.02) were the only factors associated with worse post relapse O. S. in the univariate analysis [15]. In contrast, MYC-N status and presence of bone marrow disease were the only independent poor prognostic factors that affected O. S. (P value=0.007 for both).
Response to induction therapy was proven to influence the outcome of high-risk Neuroblastoma [16]. There is no much data in the literature about the impact of response to induction on relapsing/progressive disease outcomes. In the present study, it was evident that patients with good induction response (C. R., VGPR, PR) showed better survival even when they progressed and were salvaged compared to those with inadequate induction response. Patients who achieved objective response post-induction (≥P. R.) had significantly better three y post-progression EFS and O. S. compared to those with less than P. R. response (P value<0.001 for EFS,<0.018 for O. S.). Response to salvage therapy is dependent -in our point of view-on the tumor biology, which affects tumor response initially and at relapse/progression. Intensification of induction/salvage may not improve the outcome of high-risk N. B. with poor biological features as they have poor outcome whatever kind of therapy given.
ICE chemotherapy was given for all patients with a progressive or refractory disease with 38.4% of the refractory patients showed an objective response (C. R., VGPR) compared to 35.2% of the patients in the progressive group. Kushner et al. who used ICE as salvage, reported significant responses (C. R., VGPR, and P. R.) in 53% of relapsing patients versus 15% in patients with refractory N. B. and just in 3% of patients with P. D [5].
An older study by Abdel Rahman reported better cumulative O. S. of 65% for progressive N. B. patients and 68% for refractory ones who received ICE regimen and less treatmentrelated mortality (3%) with objective response 36.5% compared to our results (OS 35.7% at three years) [17].
Consolidation by hematopoietic stem cell transplant improved outcome significantly post salvage therapy compared to non-transplanted group (P-value 0.002 for EFS and 0.003 for O. S.). A study carried by Simon et al.; demonstrated better median survival of 2.08 and 3-year O. S. of 43.5 ± 10.9% for the transplanted group compared to those who did not undergo ASCT (median survival of 0.89 years, 3year survival rate of 9.6 ± 2.8 P <0.001) [18].
Additional studies suggested that haploidentical stem cell transplantation may improve the outcome of children with relapsed and refractory N. B [19]. In the current study, objective response post-induction, consolidation by HSCT, radiotherapy, and maintenance post salvage showed a significant positive impact on the outcome. The presence of metastatic liver disease affected EFS negatively. Simultaneously, multivariate analysis showed that maintenance therapy was the only independent factor affecting the O. S. The independent factors that affected the EFS were the presence of liver metastases, the inadequate response at the end of induction (N. R., P. D.) and not administrating radiotherapy. A similar Italian study by Alberto Garverneta et al. proved that age less than 18 months, extra-abdominal primary, lower stage, and absence of MYC-N amplification had a significant positive impact on O. S. and the multivariate analysis revealed that age, stage, and MYC-N status are the independent prognostic factors [8].
In our view, improving the outcome in high-risk N. B. is not by intensifying induction/salvage as the problem is the tumor biology, which predicts tumor response. Major Responders (C. R., VGPR, PR) to low-intensity induction have better response to salvage therapy than poor or non-responders even if they received high-intensity salvage therapy.

Conclusions
Our results emphasize that N. B. is a salvageable disease. ICE chemotherapy is a good salvage regimen. More concerns should be paid for supportive measures. Response to induction had a significant impact on the outcome post salvage. Salvage therapy did not improve the outcome for those with inadequate induction response. Consolidation by HSCT, radiotherapy, and maintenance therapy improved the outcome for responders to salvage therapy. Initial front-line targeted therapy like antiGD2 is needed to improve the outcome, especially for chemo-resistant ones who do not benefit much from therapy intensification.

Declarations
This study was not funded by any organization and was approved by our hospital' scientific and ethical committee; the authors declared that they did not have any conflict of interest.