Survival of Patients With Neuroblastoma After Assignment to Reduced Therapy Because of the 12- to 18-Month Change in Age Cutoff in Children's Oncology Group Risk Stratification

PURPOSE

In 2006, Children's Oncology Group (COG) reclassified subgroups of toddlers diagnosed with neuroblastoma from high-risk to intermediate-risk, when the age cutoff for high-risk assignment was raised from 365 days (12 months) to 547 days (18 months). The primary aim of this retrospective study was to determine if excellent outcome was maintained after assigned reduction of therapy.

PATIENTS AND METHODS

Children <3 years old at diagnosis, enrolled on a COG biology study from 1990 to 2018, were eligible (n = 9,189). Assigned therapy was reduced for two cohorts of interest on the basis of the age cutoff change: 365-546 days old with International Neuroblastoma Staging System (INSS) stage 4, MYCN not amplified (MYCN-NA), favorable International Neuroblastoma Pathology Classification (INPC), hyperdiploid tumors (12-18mo/Stage4/FavBiology), and 365-546 days old with INSS stage 3, MYCN-NA, and unfavorable INPC tumors (12-18mo/Stage3/MYCN-NA/Unfav). Log-rank tests compared event-free survival (EFS) and overall survival (OS) curves.

RESULTS

For 12-18mo/Stage4/FavBiology, 5-year EFS/OS (± SE) before (≤2006; n = 40) versus after (>2006; n = 55) assigned reduction in therapy was similar: 89% ± 5.1%/89% ± 5.1% versus 87% ± 4.6%/94% ± 3.2% (P = .7; P = .4, respectively). For 12-18mo/Stage3/MYCN-NA/Unfav, the 5-year EFS and OS were both 100%, before (n = 6) and after (n = 4) 2006. The 12-18mo/Stage4/FavBiology plus 12-18mo/Stage3/MYCN-NA/Unfav classified as high-risk ≤2006 had an EFS/OS of 91% ± 4.4%/91% ± 4.5% versus 38% ± 1.3%/43% ± 1.3% for all other high-risk patients <3 years old (P < .0001; P < .0001, respectively). The 12-18mo/Stage4/FavBiology plus 12-18mo/Stage3/MYCN-NA/Unfav classified as intermediate-risk >2006 had an EFS/OS of 88% ± 4.3%/95% ± 2.9% versus 88% ± 0.9%/95% ± 0.6% for all other intermediate-risk patients <3 years old (P = .87; P = .85, respectively).

CONCLUSION

Excellent outcome was maintained among subsets of toddlers with neuroblastoma assigned to reduced treatment after reclassification of risk group from high to intermediate on the basis of new age cutoffs. Importantly, as documented in prior trials, intermediate-risk therapy is not associated with the degree of acute toxicity and late effects commonly observed with high-risk regimens.

INTRODUCTION

Treatment of neuroblastoma, a clinically heterogeneous pediatric malignancy, is tailored on the basis of prognostic clinical and biologic factors. Children's Oncology Group (COG) biology trials enroll approximately 80% of all newly diagnosed patients in North America and include centrally assessed biomarkers and pathology review to stratify patients into low-, intermediate-, and high-risk groups. Most patients with low-risk neuroblastoma can be treated with surgery alone,¹ and a subset of infants may undergo spontaneous tumor regression without any treatment.² In COG, children assigned to intermediate‐risk group are prescribed two to eight initial cycles of chemotherapy, with surgical resection of the residual primary tumor when possible.³ In a single-institution experience, patients with MYCN-NA localized neuroblastoma were treated with surgery alone or surgery plus limited chemotherapy.⁴ In COG, treatment for high-risk neuroblastoma includes five to six cycles of induction chemotherapy and surgery, consolidation therapy with high-dose myeloablative chemotherapy with single or tandem autologous hematopoietic stem-cell transplantation (HSCT) and radiation, and postconsolidation immunotherapy including anti-GD2 antibody.⁵ The intensive, multimodality high-risk treatment is associated with significantly more acute and late toxicities compared with intermediate-risk therapy. Therefore, it is important to optimize risk stratification so that patients will be assigned to appropriate therapeutic groups.⁶

CONTEXT

• Key Objective

• Was excellent outcome maintained for biomarker-defined subgroups of toddlers with neuroblastoma who were assigned to a reduction of therapy, from high-risk to intermediate-risk?

• Knowledge Generated

• Changes in risk stratification that modify the standard of care must be validated. In the Children's Oncology Group (COG), prognostic biomarkers are used to assign frontline therapeutic intensity for neuroblastoma. In 2006, COG changed the standard of care for some subgroups of toddlers with neuroblastoma because of a change in risk stratifying age cutoff, from 12 months to 18 months. Only recently was the sample size large enough for validation analysis. Retrospective analyses demonstrated that these toddler subgroups were able to have the same excellent outcome with assignment to intermediate-risk therapy as they did with high-risk therapy. This is important because compared with intensive multimodality high-risk therapy, the lower intensity intermediate-risk therapy was likely associated with reduced risk of acute toxicity and late effects.

• Relevance

• The current study shows that excellent treatment outcomes were maintained for certain subgroups of neuroblastoma after a revised risk stratification and reduction of therapy. Toddlers within these neuroblastoma subgroups can be spared the toxicities associated with high-risk therapy.

Almost 50 years ago, Breslow and McCann observed that older age was associated with worse outcome for children with neuroblastoma.⁷ Thereafter, a clinically practical cutoff of 365 days was adopted for treatment stratification. From 2000 to 2006, the COG biology protocol ANBL00B1 formalized the 365-day age cutoff for risk stratification. In 2005, several studies described favorable outcome for those 12-18 months with high-risk disease, including George et al⁸ who demonstrated that children age 12-18 months old with metastatic neuroblastoma had favorable outcomes with high-dose therapy if their tumors were hyperdiploid and lacked MYCN amplification. In patients with MYCN-NA, stage 4 disease treated with intensive therapy, Schmidt et al observed that the 12-18–month old cohort had significantly better survival than children ≥18 months.⁹ In 2005, an analysis by London et al¹⁰ of 3,666 COG patients identified a more optimal age cutoff, between 15 and 19 months of age, for risk stratification. The COG ANBL00B1 risk stratification was revised in 2006 (Data Supplement [Tables 1-3], online only), including the application of a 547-day (18-month) age cutoff, which resulted in reclassification from high-risk to intermediate-risk and assignment to reduction of therapy for two patient cohorts: (1) 365-546 days of age at diagnosis with International Neuroblastoma Staging System (INSS) stage 4 disease, and tumors that were MYCN-NA, favorable International Neuroblastoma Pathology Classification (INPC), and hyperdiploid (favorable biology) (12-18mo/Stage4/FavBiology) and (2) 365-546 days of age at diagnosis with INSS stage 3 disease and tumors that were MYCN-NA and unfavorable INPC (12-18mo/Stage3/MYCN-NA/Unfav).

Age continues to be a strong prognostic factor in the context of modern risk–based treatments.¹¹ However, the 2006 increase in age cutoff, and its effect on the outcome of patients assigned to a reduction in therapy, has yet to be validated. The primary objective of our study was to determine if excellent outcome was maintained for the subgroups of children 12-18 months of age at diagnosis after assignment to reduction of therapy as a result of increasing the age cutoff from 365 days to 547 days. Secondary objectives included describing the event-free survival (EFS) and overall survival (OS) <2006 that motivated consideration for revising the risk stratification and describing the EFS and OS of patients age <1,096 days (3 years) at diagnosis for 6-monthly groups, before versus after the 2006 change in age cutoff.

PATIENTS AND METHODS

A total of 9,189 patients met eligibility criteria: (1) enrollment on a COG biology study (B003, B903, B954, 9047, or ANBL00B1) between 1990 and 2018, (2) age at diagnosis <1,096 days (3 years), and (3) known survival data (Fig 1). The cohorts of interest for the primary objective were 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav.

FIG 1.

Flow diagram of patient cohorts. ^aIncludes four patients diagnosed in 2007, before the age cutoff amendment at their site. COG, Children's Oncology Group.

The ANBL00B1 protocol amendment that changed the age cutoff was released on October 13, 2006 (Data Supplement [Tables 1-3]); adoption of the new risk stratification occurred over a period of time because each COG site had to obtain IRB approval of the amendment. For data analysis, an artificial approximate date, December 31, 2006, was applied to create two eras: ≤2006 and >2006. The 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav cohorts were assigned to high-risk before the age cutoff protocol amendment and to intermediate-risk after the age cutoff amendment. However, four patients in the 12-18mo/Stage4/FavBiology cohort were assigned to high-risk therapy during 2007, before activating the age cutoff amendment at their site. For the purposes of this study, these four high-risk patients were reassigned to the ≤2006 era (ie, analyzed as high-risk, the risk group to which they were assigned), although they were diagnosed in 2007. Patients were analyzed on the basis of risk group, that is, as per assigned treatment rather than actual treatment received, as treatment data were not available.

For patients who enrolled on frontline therapeutic trials the trials were CCG-3881 (1989-1995) or COG-ANBL0531 (2007-2011) for intermediate-risk neuroblastoma and CCG-3891 (1991-1996), POG-9341/9342 (1993-1995), POG-9640 (1998-2000), COG-P9963 (2000-2003), COG-A3973 (2001-2006), COG-ANBL00P1 (2001-2004), COG-ANBL0532 (2007-2012), and/or COG-ANBL0032 (2001-2009) for high-risk neuroblastoma (Data Supplement [Table 4]).

Statistical Considerations

In this retrospective cohort study, the primary end point was EFS time, defined as time from diagnosis until first occurrence of relapse, progression, secondary malignancy, or death from any cause. EFS was selected because, unlike OS, it avoids confounding from the effects of salvage treatment after first relapse or progression. OS was a secondary end point, defined as the time from diagnosis until death from any cause.

Kaplan-Meier curves of EFS and OS were generated, with 5-year point estimates and SEs per Greenwood. Curves were compared with a log-rank test. After confirming the assumption of proportional hazards, hazard ratios and associated P values were calculated using a Cox proportional hazards regression model, using Firth's correction for the hazard ratio if monotone likelihood (one group with no events). Patient characteristics were compared by subgroup using Fisher's exact test. P values <0.05 were considered statistically significant. Analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC).

RESULTS

Patient Characteristics

Of the 9,189 eligible patients, 105 comprised the toddler cohorts of interest: 95 were 12-18mo/Stage4/FavBiology (n = 40, ≤2006; n = 55, >2006) and 10 were 12-18mo/Stage3/MYCN-NA/Unfav (n = 6, ≤2006; n = 4, >2006; Fig 1). The proportion of patients by sex, diagnostic category, grade of differentiation, serum lactate dehydrogenase, and serum ferritin were similar between age cutoff eras (Table 1). Only 20 of the 105 patients were enrolled on a therapeutic trial; 85 (81%) patients enrolled only on a biology study (Fig 1; Data Supplement [Table 4]).

TABLE 1.

Characteristics of Patients With Neuroblastoma 365-546 Days of Age at Diagnosis (1990-2018): INSS Stage 4, MYCN-NA, Favorable INPC, and Hyperdiploid (n = 95); INSS Stage 3, MYCN-NA, Unfavorable INPC (n = 10): Before Versus After Assignment to a Reduction of Therapy Because of the Change in Age Cutoff

Of 8,523 patients with known risk group, 2,426 were classified as high-risk (n = 1,581, ≤2006; n = 845, >2006); 2,604 were intermediate-risk (n = 1,160, ≤2006; n = 1,444, >2006); and 3,493 were low-risk (n = 2,124, ≤2006; n = 1,369, >2006; Fig 1).

Outcome

The 5-year EFS and OS for the overall cohort of patients <3 years of age at diagnosis were 77% ± 0.5% and 83% ± 0.4%, respectively (Table 2; Data Supplement [Figure 1]). For the 12-18mo/Stage4/FavBiology cohort, EFS was similar before (≤2006; n = 40) versus after (>2006; n = 55) the assigned reduction in therapy: 89% ± 5.1% versus 87% ± 4.6% (P = .7; Table 2; Fig 2A). Furthermore, there was no significant difference in OS before versus after the reduction in therapy: 89% ± 5.1% versus 94% ± 3.2% (P = .4; Table 2; Fig 2B). For the 12-18mo/Stage3/MYCN-NA/Unfav cohort, no events occurred, neither before (n = 6) nor after (n = 4) the assigned reduction in therapy, resulting in 5-year EFS and OS of 100% (Table 2; Figs 2C and 2D).

TABLE 2.

Comparisons of EFS and OS Within Cohorts of Interest

FIG 2.

(A) EFS curves by age cutoff era in the 12-18mo/Stage4/FavBiology cohort: ≤2006 (n = 40) versus >2006 (n = 55). (B) OS curves by age cutoff era in the 12-18mo/Stage4/FavBiology cohort: ≤2006 (n = 40) versus >2006 (n = 55). (C) EFS curves by age cutoff era in the 12-18mo/Stage3/MYCN-NA/Unfav cohort: ≤2006 (n = 6) versus >2006 (n = 4). (D) OS curves by age cutoff era in the 12-18mo/Stage3/MYCN-NA/Unfav cohort: ≤2006 (n = 6) versus >2006 (n = 4). (E) EFS curves by age cutoff era in the combined 12-18mo/Stage4/FavBiology cohort and the 12-18mo/Stage3/MYCN-NA/Unfav cohort: ≤2006 (n = 46) versus >2006 (n = 59). (F) OS curves by age cutoff era in the combined 12-18mo/Stage4/FavBiology cohort and the 12-18mo/Stage3/MYCN-NA/Unfav cohort: ≤2006 (n = 46) versus >2006 (n = 59). EFS, event-free survival; OS, overall survival.

Outcome was compared for the combined cohorts of interest (12-18mo/Stage4/FavBiology plus 12-18mo/Stage3/MYCN-NA/Unfav) that were assigned to reduction in therapy. There was no significant difference in EFS before (≤2006; n = 46) versus after (>2006; n = 59) the reduction in therapy: 91% ± 4.4% versus 88% ± 4.3% (P = .6; Fig 2E). Similarly, there was no significant difference in OS for the combined cohorts of interest before versus after the assigned reduction in therapy: 91% ± 4.5% versus 95% ± 2.9% (P = .5; Fig 2F).

Within high-risk patients <3 years old diagnosed ≤2006, the combined cohorts of interest had an EFS of 91% ± 4.4% versus 38% ± 1.3% for other high-risk patients (P < .0001; Table 2; Fig 3A); similarly, in the combined cohorts-of-interest, the OS was 91% ± 4.5% versus 43% ± 1.3% for other high-risk patients (P < .0001; Table 2; Fig 3B). Within intermediate-risk patients <3 years old diagnosed >2006, there was neither a significant difference in EFS for the combined cohorts of interest versus other intermediate-risk patients: 88% ± 4.3% versus 88% ± 0.9%, respectively (P = .9; Table 2; Fig 3C) nor a significant difference in OS: 95% ± 2.9% versus 95% ± 0.6%, respectively (P = .9; Table 2; Fig 3D).

FIG 3.

(A) EFS of high-risk patients diagnosed ≤2006 who are 365-547 days of age at diagnosis: Combined 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav cohort (group 2 in red; n = 46) versus all other high-risk patients <3 years old (group 1 in blue; n = 1,535). (B) OS of high-risk patients diagnosed ≤2006 who are 365-547 days of age at diagnosis: Combined 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav cohort (group 2 in red; n = 46) versus all other high-risk patients <3 years old (group 1 in blue; n = 1,535). (C) EFS of intermediate-risk diagnosed >2006 patients who are 365-547 days of age at diagnosis: Combined 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav cohort (group 2 in red; n = 59) versus all other intermediate-risk patients <3 years old (group 1 in blue; n = 1,385). (D) OS of intermediate-risk patients diagnosed >2006 who are 365-547 days of age at diagnosis: Combined 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav cohort (group 2 in red; n = 59) versus all other intermediate-risk patients <3 years old (group 1 in blue; n = 1,385). EFS, event-free survival; OS, overall survival.

To provide context for the comparison of ≤2006 versus >2006 within the cohorts of interest, comparisons were performed for ≤2006 versus >2006 within each of the 6-monthly age groups for the overall cohort and for intermediate-risk (Data Supplement [Table 5]) and high-risk patients (Data Supplement [Table 6]). For patients in the overall cohort diagnosed ≤2006, both EFS and OS gradually decreased with increasing age group (Data Supplement [Figs 2A and 2B]). Among patients diagnosed >2006, two distinct clusters of curves are apparent on visual inspection, for both EFS and OS: one with patients 0-6, 6-12, and 12-18 months of age at diagnosis (ie, <18 months) and another with patients 18-24, 24-30, and 30-36 months (ie, ≥18 months; Data Supplement [Figs 2C and 2D]). For intermediate-risk patients, within each 6-monthly age group, there was no significant difference in EFS nor OS for ≤2006 versus >2006 ( Data Supplement [Table 5]). For high-risk patients, within each 6-monthly age group <18 months, there was no significant difference in EFS nor OS for ≤2006 versus >2006. Conversely, for all high-risk 6-monthly age groups ≥18 months, there was significant improvement >2006 compared with ≤2006 in both EFS and OS (Data Supplement [Table 6]).

DISCUSSION

In 2006, the COG risk stratification was revised based on the findings that an age cutoff of 547 days of age at diagnosis was more optimal for risk stratification than 365 days.¹⁰ This resulted in a reduction of assigned therapy for subgroups of patients 12-18 months of age, that is, a change from high-risk to intermediate-risk therapy. To our knowledge, this is the largest and only known study to focus on a comparison of toddlers with favorable biology tumors before versus after an assigned reduction of therapy for neuroblastoma. This study exemplifies an important validation step in risk stratification. We test experimental treatments in rigorous trials, so too, must we validate that changes to standard of care because of changes in risk stratification do not adversely affect outcomes, even if only small subgroups of patients are affected. Our research follows the guidance of Evans and D'Angio,⁶ who, in 2005, advised prospective trials to test the newly expanded low-risk group on the basis of age in patients with nonamplified MYCN tumors, using uniform treatment regimens. Herein, we have demonstrated that the change in age cutoff was appropriate and that COG's decision for assignment to reduction of therapy in these subgroups was successful.

Proportionately, the number of patients who received assignment to reduced therapy is small; however, in absolute numbers, 105 is appreciable. The benefits to these 105 patients and their families were substantial in terms of reduced risk of severe toxicity and late effects.^12,13

To our knowledge, this is the first study to summarize the changes in COG risk stratification that occurred in 2006 (Data Supplement [Table 3]). In addition to reduction of therapy for 12-18mo/Stage4/FavBiology and 12-18mo/Stage3/MYCN-NA/Unfav (from high-risk to intermediate-risk), eight subgroups had intensification of assigned initial therapy (five subgroups from low-risk to intermediate-risk; three subgroups from low-risk to high-risk). Our future research will determine whether the outcome of these subgroups was improved as a result of these changes in classification and assignment of intensified therapy.

Our analysis demonstrates that there was no significant difference in EFS and OS levels before versus after the change in age cutoff. Interestingly, OS levels appear slightly higher after the assigned reduction in therapy, which may indicate that toddlers with favorable biology who are initially treated with intermediate-risk therapy are more likely to be salvageable and responsive to treatment after relapse. Only a small proportion of the patients who were classified as intermediate-risk had a relapse and required salvage therapy; most patients were able to be cured with intermediate-risk therapy.

In Figures 3A and 3B, comparing the toddler cohorts of interest with the balance of high-risk patients <3 years old at diagnosis illustrates the dramatic difference in outcome that existed before 2006, contributing to the motivation and justification for moving the toddler cohorts of interest from high- to intermediate-risk. After 2006, the toddler cohorts of interest had outcome that was extremely similar to the outcome of the balance of intermediate-risk patients. This provides compelling visual evidence that the 2006 change in risk group assignment was appropriate.

Among all intermediate-risk patients included in the study, we found no significant differences in outcomes by 6-monthly age for ≤2006 versus >2006. This is consistent with the results of recent clinical trials, which have reported maintenance of excellent outcomes for intermediate-risk patients after reduction in therapy.^14,15 Maintenance of EFS and OS over time was considered a clinically appropriate benchmark for success in our cohorts of interest.

Our analysis of the overall high-risk cohort revealed improved outcomes, comparing ≤2006 versus >2006, within each of the 6-monthly age groups from 18 months to 3 years of age. Other studies have demonstrated gradual improvement in outcome over time for patients with high-risk neuroblastoma, although unfortunately overall outcome remains poor.^5,16 Before 2006, EFS and OS gradually decreased with increasing 6-monthly age group. After 2006, there were two distinct clusters of 6-monthly curves, one with similar EFS and OS for patients <18 months and another for patients >18 months, which likely reflects the improved therapy available after 2006 for high-risk patients.

Our analysis was limited by small sample size, especially within the 12-18mo/Stage3/MYCN-NA/Unfav cohort. The 12-18mo/Stage3/MYCN-NA/Unfav cohort was not originally considered a separate group but became a subgroup only after the change in age cutoff was applied. Testing for noninferiority by applying a noninferiority margin would have provided the strongest possible statistical evidence that outcome had been maintained after the assigned reduction of therapy; however, we were unable to take this approach because of the small sample size. An imbalance in the proportion of patients with low MKI was observed within the 12-18mo/Stage4/FavBiology cohort (74%, ≤2006; 53%, >2006); thus, patients in the ≤2006 cohort appear to have had slightly more favorable biology than those in the >2006 cohort. Otherwise, known prognostic risk factors were similar in the groups before versus after 2006.

Toxicity data were unavailable for most of the patients in our cohort. It is reasonable to assume that patients reclassified as intermediate-risk were spared the substantially increased toxicity associated with high-risk neuroblastoma treatment. For example, incidence of grade ≥ 3 anemia and febrile neutropenia toxicities were 74.2% and 71%, respectively, among patients receiving high-risk therapy,¹⁷ compared with 41.8% and 14.6%, respectively, among patients receiving intermediate-risk therapy¹⁴ on frontline clinical trials. Survivors of neuroblastoma treated from 1970 to 1985 had substantially higher risk of late effects, with increased rates of neurological, musculoskeletal, endocrine, and sensory conditions compared with sibling controls; chronic health conditions were more common in those treated with multimodal chemotherapy compared with surgery alone.¹⁸ Several studies of patients treated for high-risk neuroblastoma found an increased incidence of ototoxicity (prevalence ranging from 13% to 95%), second malignancy (0.7%-17.2% of survivors), and deficiencies of the endocrine system, including growth failure, hypothyroidism, ovarian failure, and prediabetes.^12,19-23 Intermediate-risk assignment spares patients treatment with HSCT, including exposure to cisplatin and myeloablative doses of carboplatin (Table 3). The Late Effects After High-Risk Neuroblastoma (LEAHRN; ClinicalTrials.gov identifier: NCT03057626) study, to characterize late effects of high-risk treatment, is ongoing.¹³

TABLE 3.

Cumulative Chemotherapy and Postconsolidation Therapy Exposure in Intermediate- Versus High-Risk Treatment and Potential Late Effects

Current and future changes to risk stratification will necessitate ongoing monitoring of patients that undergo a change in risk group. The International Neuroblastoma Risk Group (INRG) has led efforts to develop uniform approaches for pretreatment risk classification (the INRG classification), including the pretreatment International Neuroblastoma Group Staging System (INRGSS).^24,25 Segmental chromosome aberrations (SCAs; loss or gain of a portion of a chromosome arm) portend unfavorable outcome^16,26,27 and have already been used for treatment assignment in frontline clinical trials.¹⁴ In 2021, a revised COG classifier (v2) was developed which incorporates INRGSS and SCA to classify risk and stratify treatment.¹⁶ In the revised COG classifier, if SCA is unfavorable, patients 365-546 days of age at diagnosis with metastatic disease, MYCN-NA, favorable INPC, and hyperdiploid tumors will be classified as high-risk, rather than intermediate-risk.¹⁶

The neuroblastoma risk classifiers of SIOP-Europe (SIOPEN) and COG are well aligned in important ways, although subtle differences exist.²⁸ Applying the SIOPEN risk stratification to the cohorts of interest herein: Patients in the 12-18mo/Stage4/FavBiology cohort would have been eligible for the High-risk Neuroblastoma Study 1.8 of SIOP-Europe (SIOPEN) (ClinicalTrials.gov identifier: NCT01704716), in which all patients >12 months of age with metastatic disease are eligible; however, in this SIOPEN trial, treatment stopped after induction therapy and surgery for patients age 12-18 months whose tumors lacked MYCN amplification and SCAs. Patients in the 12-18mo/Stage3/MYCN-NA/Unfav cohort would not have been eligible for the SIOPEN high-risk trial.

Studies of response- and biology-based risk factor–guided therapy for non–high-risk neuroblastoma are ongoing (COG ANBL1232; ClinicalTrials.gov identifier: NCT02176967). Telomere maintenance mechanism (TMM)^29-31 or RAS/mitogen-activated protein kinase pathway alterations³² show promise to identify rare intermediate-risk patients with otherwise favorable features who ultimately relapse or to identify high-risk patients with TMM-negative tumors who may be amenable to a reduction of therapy.³⁰ Efforts are also underway to further stratify high-risk patients, including identification of ultra-high-risk groups.^33,34 In addition to identification of new prognostic biomarkers, the incorporation of nomograms³⁵ and machine learning strategies³⁶ could enable more precise algorithms to match patients with appropriate treatment intensity.

Our 29-year study demonstrates that excellent outcome is maintained among toddlers with Stage4/FavBiology and Stage3/MYCN-NA/Unfav neuroblastoma assigned to a significant reduction of therapy, from high- to intermediate-risk treatment. Importantly, patients were likely spared acute toxicity and late effects known to be associated with high-risk therapy. With recent changes in the COG classifier, identification of new prognostic biomarkers and the promise of more precise risk stratification algorithms, it will be important to continue to identify patients who are likely to benefit from reduction in therapy and then closely monitor if the treatment changes negatively affect survival. Efforts to identify additional patient cohorts who may not require high-risk therapy to achieve long-term survival are critical to improve the long-term health of children with neuroblastoma.

DATA SHARING STATEMENT

A data sharing statement provided by the authors is available with this article at DOI https://doi.org/10.1200/JCO.22.01946.

AUTHOR CONTRIBUTIONS

Conception and design: Meredith S. Irwin, Michael D. Hogarty, Robert Castleberry, John M. Maris, Susan L. Cohn, Wendy B. London

Administrative support: Robert Castleberry, Wendy B. London

Provision of study materials or patients: Meredith S. Irwin, Robert Castleberry, Susan L. Cohn

Collection and assembly of data: Meredith S. Irwin, Robert Castleberry, Fan F. Zhang, Arlene Naranjo, Wendy B. London

Data analysis and interpretation: Hannah G. Bender, Meredith S. Irwin, Michael D. Hogarty, Robert Castleberry, John M. Maris, Pei-Chi Kao, Arlene Naranjo, Susan L. Cohn, Wendy B. London

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Survival of Patients With Neuroblastoma After Assignment to Reduced Therapy Because of the 12- to 18-Month Change in Age Cutoff in Children's Oncology Group Risk Stratification

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).