Chemoimmunotherapy Combo Elicits Encouraging Responses in Relapsed High-Risk Neuroblastoma

The 4-drug combination of irinotecan, temozolomide, dinutuximab (Unituxin), and granulocyte-macrophage colony-stimulating factor (GM-CSF) elicited objective responses in almost half of patients with high-risk neuroblastoma, according to retrospective findings published in the Journal of Clinical Oncology.¹

In total, 48.6% (n = 71) of patients achieved an objective response, including a complete response (CR), partial response (PR), and minor response in 29%, 14%, and 5% of patients, respectively. This includes 47% of the 133 patients with metastatic soft tissue and bone disease. Additionally, investigators detected a marrow response in 73% of the 58 patients with available posttreatment bone marrow data.

“Data regarding response timing and duration will aid clinicians in balancing the risks and benefits of continuing treatment of [irinotecan/temozolomide/dinutuximab/GM-CSF] in patients with relapsed [high-risk neuroblastoma],” lead study author Benjamin J. Lerman, MD, MS, of the Children’s Hospital of Philadelphia and the University of Pennsylvania, and colleagues, wrote in the journal article. “The survival outcomes described in this study provide a new point of reference for evaluating novel therapeutics in relapsed [high-risk neuroblastoma].”

This retrospective study sought to describe response and progression-free survival (PFS) in patients with relapsed high-risk neuroblastoma who received the regimen of irinotecan, temozolomide, dinutuximab, and GM-CSF.

To be eligible, patients needed to be under 30 years old at diagnosis of high-risk neuroblastoma and received at least 1 cycle of irinotecan/temozolomide/dinutuximab/GM-CSF for progressive disease (PD) or relapsed disease. Patients also needed to have an objective response or stable disease after standard initial frontline therapy.

Patients were excluded if they had primary refractory disease and progressed through induction therapy. Those who received the combination for suboptimal response to induction treatment were also excluded.

In total, 146 patients were analyzed in this study, which included those who received at least 1 cycle of irinotecan/temozolomide/dinutuximab/GM-CSF outside the context of a clinical trial from January 1, 2015, to June 1, 2020. The study regimen consisted of intravenous (IV) irinotecan at 50 mg/m2 once daily, temozolomide at 100 mg/m2 once daily on days 1 though 5, IV dinutuximab at 17.5 mg/m2 once daily on days 2 through 5, and subcutaneous GM-CSF at 250 µg/m2 once daily on days 6 through 12 of a 21-day cycle.

Regarding patients who received the combination for multiple distinct relapse events, the study only used data related to the first relapse event treated with the combination.

Patients received a median of 4.5 cycles of study treatment (interquartile range, 2-7; range, 1-31) over a median of 87 days. Of the 23% (n = 34) patients who received concurrent treatment at some point during their time on the combination, 5 received surgery, 23 received radiation, 2 received meta-iodobenzylguanidine therapy, and 5 received another anticancer therapy.

The median age of patients at diagnosis was 51 months (interquartile range, 31-72), and 58.2% (n = 85) were male. A total of 67.8% (n = 99) of patients received the study treatment after their first relapse or progression. In addition, 63.0% (n = 92) of patients had measurable disease at relapse or progression, and 55.5% (n = 81) had received anti-GD2 therapy prior to study treatment.

In total, 50 (37%) of the 134 patients had known tumor MYCN status, and 43 of those patients (32%) had tumors with MYCN amplification. Additionally, 94 patients had known tumor ALK status, and 21 of those patients (22%) had tumors with ALK amplification or activating mutation.

A total of 58.2% (n = 85) of patients discontinued study treatment because of stable disease (SD) or PD, 13.0% (n = 19) discontinued because of unacceptable acute or chronic toxicity, 23.3% (n = 34) discontinued because of a transition to another anticancer therapy, and 19.9% (n = 29) discontinued because of planned treatment completion. Patients who stopped treatment with the combination because of toxicity received a median of 3 cycles (range, 1-15) before discontinuing. At the time of data collection, 4 patients were still on the study therapy.

The primary end point was the proportion of patients with objective response after receiving the combination, which was defined as CR, PR, or minor response on a disease evaluation after initiation of study treatment; PFS; and duration of response (DOR) after objective response. Investigators sought to evaluate timing and durability of response and to identify molecular and clinical factors associated with PFS in these patients.

Patients who achieved an objective response received a median of 2 cycles (range, 1-9) from the beginning of study treatment to their first disease evaluation showing objective response. One patient required over 6 cycles to reach a first objective response; this patient received 9 cycles of treatment.

The median time between diagnosis and first relapse was 26 months (range, 17-38) in patients who achieved an objective response vs 18 months (range, 12-35) in those who did not respond (P = .04). Additionally, White patients were most likely to respond, as 71.2% (n = 40) of White patients responded and 53.3% (n = 51) did not respond (P = .03). Responders were also more likely to have received prior anti-GD2 therapy, at 64.8% (n = 46) vs 46.7% (n = 35; P = .03) in those who did not respond, and tumor ALK aberrations were less frequent in responders, at 11.3% (n = 8) vs 17.3% (n = 13; P = .02) in those who did not respond.

Furthermore, the median age of responders (54 months; range, 33.0-80.5) was older than that of non-responders (44 months; range, 30.5-64.5; P = .15), and more responders had stage M disease, at 97.2% (n = 69) vs 89.3% (n = 67; P = .13). Responders were also less likely to have measurable disease at the time of receiving study therapy, at 56.3% (n = 40) vs 69.3% (n = 52; P = .12).

Although all the above distinctions met the criteria for inclusion in the multivariate model, after multivariate adjustment, only the comparison between White and non-Black and non-White patients retained statistical significance. No covariates were independently associated with PFS.

Of the total population, 21% (n = 31) had SD, and 30% (n = 44) progressed on the study treatment. Of the 79 patients who initially had SD, minor response, or PR, 72% (n = 57) achieved their best response by the first disease evaluation. Of the 22% of patients with SD or better at first post–study treatment disease evaluation who experienced an improvement in response designation in a subsequent evaluation, 5 had initial SD, 3 had initial minor response, and 14 had initial PR.

In the overall study population, the median PFS was 13.1 months (95% CI, 9.8-16.5), the 6-month PFS rate was 62% (95% CI, 55%-71%), the 1-year PFS rate was 51% (95% CI, 43%-60%), and the 2-year PFS rate was 28% (95% CI, 20%-40%).

In all patients who had an objective response, the median DOR was 15.9 months (95% CI, 14-not available [NA]), the 6-month PFS rate was 86% (95% CI, 78%-95%), the 1-year PFS rate was 68% (95% CI, 57%-81%), and the 2-year PFS rate was 44% (95% CI, 32%-61%).

Additionally, 21 patients who achieved an objective response discontinued study treatment and received no further anticancer therapy, with a median additional PFS of 10.4 months (95% CI, 7.4-NA). Of those patients, 14 had CRs and 7 had PRs. Of these patients, the 6-month PFS rate was 73% (95% CI, 54%-99%), and the 1-year PFS rate was 26% (95% CI, 10%-67%).

In total, 108 patients relapsed or progressed 0 to 30 months after initiating study treatment. Of these, 56% (n = 60) experienced relapse while actively receiving study treatment, 31% (n = 33) relapsed after transitioning from study treatment to another anticancer therapy, and 14% (n = 15) relapsed after discontinuation of study treatment without additional therapy.

Bone was the most common site of relapse (76.9%; n = 83), followed by metastatic soft tissue 54.6% (n = 59) and lymph nodes in 54.6% (n = 59), bone marrow in 30.6% (n = 33), lung in 16.7% (n = 18), and brain in 5.6% (n = 6).

“These data establish a new baseline of response and survival for comparison during future trials in patients with relapsed/progressive neuroblastoma in the chemoimmunotherapy era,” the study authors concluded.

Reference

1. Lerman BJ, Li Y, Carlowicz C, et al. Progression-free survival and patterns of response in patients with relapsed high-risk neuroblastoma treated with irinotecan/temozolomide/dinutuximab/granulocyte-macrophage colony-stimulating factor. J Clin Oncol. Published online October 7, 2022. doi:10.1200/JCO.22.01273