Recurrent or Refractory Neuroblastoma in Patients Initially Classified as High Risk

Chemotherapy combined with immunotherapy.

Chemotherapy combined with immunotherapy.

Temozolomide, irinotecan, and dinutuximab.[21]

131I-MIBG. 131I-MIBG alone, in combination with other therapy, followed by stem cell rescue.

Novel therapies.

Novel therapies.

ALK inhibitors for patients with ALK variants. In a trial of 20 patients with ALK aberrations treated with crizotinib, the response rate was 15%. Two patients had partial responses, and one patient had a complete response. All three patients had a somatic ALK Arg1275Gln variant.[22][Level of evidence C3] Lorlatinib has shown activity in patients with ALK-aberrant, relapsed neuroblastoma. The response rates were 13% in patients younger than 18 years and 47% in patients aged 18 years or older.[23] Additional patients in each cohort had minor responses, resulting in modified response rates of 30% and 67%, respectively. A single-institution series reported that 9 of 13 adult patients with relapsed, ALK-aberrant neuroblastoma responded to lorlatinib.[24]

Novel therapies.

WEE1 inhibitors. A phase II trial of adavosertib plus irinotecan reported that 3 of 20 patients with relapsed neuroblastoma had objective responses, which met the primary efficacy end point.[25]

Novel therapies.

Bevacizumab. The BEACON trial was a multiarm randomized trial with a factorial design for patients with relapsed or refractory high-risk neuroblastoma. Eighty patients were assigned to receive chemotherapy with the addition of bevacizumab and 80 patients were assigned to receive chemotherapy alone.[26] Patients randomly assigned to receive bevacizumab had a higher response rate (26% vs. 18%) that met a prespecified threshold for success. The 1-year PFS rate was nominally higher for patients who received bevacizumab (46% with bevacizumab vs. 38% for those who did not receive bevacizumab; HR, 0.89; 95% confidence interval [CI], 0.63–1.27). There was evidence for an interaction with the chemotherapy randomization (see below), such that patients randomly assigned to bevacizumab, irinotecan, and temozolomide had the most favorable 1-year PFS rates (67%).

Novel therapies.

RIST regimen. A randomized phase II trial included 129 patients with relapsed or refractory neuroblastoma. The study compared the combination of dasatinib and rapamycin added to irinotecan and temozolomide (RIST regimen) with irinotecan and temozolomide alone. In the full cohort, the median PFS was significantly longer in the RIST group (11 months) than in the chemotherapy-alone group (5 months; HR, 0.62; P = .019). On subgroup analysis, the benefit was largely seen in patients with MYCN-amplified disease (median PFS, 6 months with RIST vs. 2 months with control; HR, 0.45; P = .012) and not in patients with MYCN wild-type disease (median PFS, 14 months with RIST vs. 8 months with control; HR, 0.84; P = .49).[27]

Chemotherapy (phase I/II studies).

Chemotherapy (phase I/II studies).

Topotecan in combination with cyclophosphamide or etoposide.[28]

Chemotherapy (phase I/II studies).

Temozolomide with irinotecan.

Immunotherapy. Novel anti-GD2 drugs have been evaluated in patients with recurrent or refractory neuroblastoma. Hu14.18 anti-GD2 has been chemically linked with IL-2 and combined with granulocyte-macrophage colony-stimulating factor (GM-CSF), and a phase II trial of this regimen reported a few durable responses.[29]

The ANBL1221 (NCT01767194) trial was the first multicenter trial to evaluate anti-GD2 therapy combined with chemotherapy in a cohort of patients with relapsed or refractory neuroblastoma. Patients in first relapse or progression were randomly assigned to receive either irinotecan/temozolomide/dinutuximab/GM-CSF (I/T/DIN/GM-CSF) or temozolomide/irinotecan/temsirolimus.[21]; [30][Level of evidence C3]

The ANBL1221 (NCT01767194) trial was the first multicenter trial to evaluate anti-GD2 therapy combined with chemotherapy in a cohort of patients with relapsed or refractory neuroblastoma. Patients in first relapse or progression were randomly assigned to receive either irinotecan/temozolomide/dinutuximab/GM-CSF (I/T/DIN/GM-CSF) or temozolomide/irinotecan/temsirolimus.[21]; [30][Level of evidence C3]

Of the 17 patients treated with the combination that included dinutuximab, 9 patients (53%) had objective responses, compared with 1 of 18 patients treated with the regimen that contained temsirolimus.

The ANBL1221 (NCT01767194) trial was the first multicenter trial to evaluate anti-GD2 therapy combined with chemotherapy in a cohort of patients with relapsed or refractory neuroblastoma. Patients in first relapse or progression were randomly assigned to receive either irinotecan/temozolomide/dinutuximab/GM-CSF (I/T/DIN/GM-CSF) or temozolomide/irinotecan/temsirolimus.[21]; [30][Level of evidence C3]

In an expansion cohort consisting of 36 additional patients nonrandomly assigned to receive I/T/DIN/GM-CSF, objective responses were seen in 13 patients (36.1%). For all 53 patients enrolled on the study and treated with I/T/DIN/GM-CSF, there were 22 objective responses (41.5%).[30][Level of evidence C3] This outcome is superior to any other published outcome for patients with refractory or relapsed high-risk neuroblastoma.

In a retrospective cohort study of 146 patients with high-risk neuroblastoma who received chemoimmunotherapy with I/T/DIN/GM-CSF in first relapse, the following results were reported:[31][Level of evidence C2]

In a retrospective cohort study of 146 patients with high-risk neuroblastoma who received chemoimmunotherapy with I/T/DIN/GM-CSF in first relapse, the following results were reported:[31][Level of evidence C2]

A total of 49% of patients had an objective response, similar to the response rate seen in the ANBL1221 trial.

In a retrospective cohort study of 146 patients with high-risk neuroblastoma who received chemoimmunotherapy with I/T/DIN/GM-CSF in first relapse, the following results were reported:[31][Level of evidence C2]

Of the patients with stable disease or better at first disease evaluation after chemoimmunotherapy, 22% had an improved response (per International Neuroblastoma Response Criteria) on subsequent evaluation. Only 13% of patients with stable disease at first disease evaluation eventually had an objective response, whereas approximately 40% of patients with initial minimal response or partial response status achieved complete response after subsequent cycles. Patients who received more than six cycles of therapy and had continued stable disease were unlikely to achieve an objective response.

In a retrospective cohort study of 146 patients with high-risk neuroblastoma who received chemoimmunotherapy with I/T/DIN/GM-CSF in first relapse, the following results were reported:[31][Level of evidence C2]

The median PFS was 13.1 months from initiation of therapy. The 1-year PFS rate was 50%, and the 2-year PFS rate was 28%.

In a retrospective cohort study of 146 patients with high-risk neuroblastoma who received chemoimmunotherapy with I/T/DIN/GM-CSF in first relapse, the following results were reported:[31][Level of evidence C2]

The median duration of response was 15.9 months.

In a retrospective cohort study of 146 patients with high-risk neuroblastoma who received chemoimmunotherapy with I/T/DIN/GM-CSF in first relapse, the following results were reported:[31][Level of evidence C2]

The median PFS after discontinuation of all anticancer therapy, including I/T/DIN/GM-CSF, was 10.4 months.

Limited data are available about the use of chemotherapy backbones other than irinotecan/temozolomide as part of a chemoimmunotherapy strategy. A registry study included 24 patients with relapsed or progressive high-risk neuroblastoma who were treated with topotecan/cyclophosphamide and dinutuximab.[32]

Limited data are available about the use of chemotherapy backbones other than irinotecan/temozolomide as part of a chemoimmunotherapy strategy. A registry study included 24 patients with relapsed or progressive high-risk neuroblastoma who were treated with topotecan/cyclophosphamide and dinutuximab.[32]

The objective response rate was 42%.

For children with recurrent or refractory neuroblastoma, 131I-MIBG is an effective palliative agent and may be considered alone or in combination with chemotherapy and stem cell support in a clinical research trial.[33-38]; [39,40][Level of evidence C1]

A North American retrospective study of more than 200 patients treated with 131I-MIBG therapy compared children who had recurrence or progression of disease with children who had stable or persistent disease since diagnosis.[41]

A North American retrospective study of more than 200 patients treated with 131I-MIBG therapy compared children who had recurrence or progression of disease with children who had stable or persistent disease since diagnosis.[41]

The rate of immediate progression after 131I-MIBG therapy was lower, and the OS rate at 2 years was better (65% vs. 39%) in patients with stable, persistent disease.

Tandem consolidation using 131I-MIBG, vincristine, and irinotecan with autologous hematopoietic stem cell transplant (HSCT) followed by busulfan/melphalan with autologous HSCT was retrospectively reported in eight patients.[40]

Tandem consolidation using 131I-MIBG, vincristine, and irinotecan with autologous hematopoietic stem cell transplant (HSCT) followed by busulfan/melphalan with autologous HSCT was retrospectively reported in eight patients.[40]

This treatment resulted in three complete responses, two partial responses, and one minor response.

Single autologous HSCT with 131I-MIBG and carboplatin/etoposide/melphalan was studied in additional patients. Patients were treated after completing induction chemotherapy.[42]

Single autologous HSCT with 131I-MIBG and carboplatin/etoposide/melphalan was studied in additional patients. Patients were treated after completing induction chemotherapy.[42]

Response to induction therapy included refractory disease in 27 patients and progressive disease in 15 patients. Four of 41 evaluable patients had complete or partial responses after transplant. Eight patients with partial responses to induction therapy were treated, resulting in three responses.

Single autologous HSCT with 131I-MIBG and carboplatin/etoposide/melphalan was studied in additional patients. Patients were treated after completing induction chemotherapy.[42]

There was a 12% incidence of sinusoidal obstructive syndrome.

A randomized phase II trial included 105 evaluable patients who were treated with either 131I-MIBG alone, 131I-MIBG with irinotecan and vincristine, or 131I-MIBG with vorinostat.[43]

A randomized phase II trial included 105 evaluable patients who were treated with either 131I-MIBG alone, 131I-MIBG with irinotecan and vincristine, or 131I-MIBG with vorinostat.[43]

Patients enrolled in the vorinostat arm had the highest response rate (32%).

A randomized phase II trial included 105 evaluable patients who were treated with either 131I-MIBG alone, 131I-MIBG with irinotecan and vincristine, or 131I-MIBG with vorinostat.[43]

Patients treated with MIBG alone or with irinotecan/vincristine had response rates of 14%.

A single-arm, phase II trial included 30 patients with relapsed or refractory neuroblastoma who were treated with 131I-MIBG and topotecan.[44]

A single-arm, phase II trial included 30 patients with relapsed or refractory neuroblastoma who were treated with 131I-MIBG and topotecan.[44]

A 13% response rate was reported.

131I-MIBG is associated with risk of second malignancy. In one report of 563 5-year survivors of neuroblastoma, 15.5% had received prior 131I-MIBG therapy.[45]

131I-MIBG is associated with risk of second malignancy. In one report of 563 5-year survivors of neuroblastoma, 15.5% had received prior 131I-MIBG therapy.[45]

The risk of a second malignant neoplasm was higher for patients treated with prior 131I-MIBG, compared with survivors who were not treated with 131I-MIBG (sub-hazard ratio, 5.7; 95% CI, 1.8–17.8).

Irinotecan and temozolomide.

Irinotecan and temozolomide.

The combination of irinotecan and temozolomide had a 15% response rate in one study.[46][Level of evidence B4]

Irinotecan and temozolomide.

In the BEACON trial, patients were randomly assigned to receive either temozolomide or irinotecan and temozolomide, with or without bevacizumab, in a factorial trial design.[26]

Irinotecan and temozolomide. › In the BEACON trial, patients were randomly assigned to receive either temozolomide or irinotecan and temozolomide, with or without bevacizumab, in a factorial trial design.[26]

Response rates were similar in the two treatment groups (21% with temozolomide vs. 20% with irinotecan and temozolomide).

Irinotecan and temozolomide. › In the BEACON trial, patients were randomly assigned to receive either temozolomide or irinotecan and temozolomide, with or without bevacizumab, in a factorial trial design.[26]

The 1-year PFS rate was superior for patients who were treated with irinotecan and temozolomide (53%), compared with temozolomide (30%; HR, 0.59; 95% CI, 0.39–0.90).

A retrospective study reported on 74 patients who received 92 cycles of ifosfamide, carboplatin, and etoposide. The study included 37 patients who received peripheral blood stem cell rescue after responding to this drug combination.[47]

A retrospective study reported on 74 patients who received 92 cycles of ifosfamide, carboplatin, and etoposide. The study included 37 patients who received peripheral blood stem cell rescue after responding to this drug combination.[47]

Disease regressions (major and minor responses) were achieved in 14 of 17 patients (82%) with a new relapse, 13 of 26 patients (50%) with refractory neuroblastoma, and 12 of 34 patients (35%) who were treated for progressive disease during chemotherapy (P = .005).

A retrospective study reported on 74 patients who received 92 cycles of ifosfamide, carboplatin, and etoposide. The study included 37 patients who received peripheral blood stem cell rescue after responding to this drug combination.[47]

Grade 3 toxicities were rare.

Topotecan in combination with cyclophosphamide alone or with etoposide has been used in patients with recurrent disease who did not receive topotecan initially.[28][Level of evidence A2]

Topotecan in combination with cyclophosphamide alone or with etoposide has been used in patients with recurrent disease who did not receive topotecan initially.[28][Level of evidence A2]

The response rates were 32% (18 of 57) for patients who received topotecan and cyclophosphamide and 19% (11 of 59) for patients who received topotecan alone.

In the BEACON trial, patients were randomly assigned to receive either temozolomide or topotecan and temozolomide, with or without bevacizumab.[26]

In the BEACON trial, patients were randomly assigned to receive either temozolomide or topotecan and temozolomide, with or without bevacizumab.[26]

Response rates were similar in the two treatment groups (23% with temozolomide vs. 27% with topotecan and temozolomide).

In the BEACON trial, patients were randomly assigned to receive either temozolomide or topotecan and temozolomide, with or without bevacizumab.[26]

The 1-year PFS rate was nominally higher in patients who received topotecan and temozolomide (47%) than in patients who received temozolomide (23%; HR, 0.59; 95% CI, 0.33–1.08).

High-dose carboplatin, irinotecan, and/or temozolomide has been used to treat patients with refractory disease or new relapses (after treatment that included topotecan) occurring off therapy (68% objective response rate). However, this regimen is not used to treat patients whose disease progresses while on therapy.[48]