Children’s Oncology Group risk groups sort neuroblastoma by age, biology, and stage
The Children’s Oncology Group framework for neuroblastoma risk assignment is built around a simple clinical idea with major treatment consequences: the same diagnosis can behave very differently depending on age, stage, tumor biology, and resectability. The current grouping divides patients into high-risk, low-risk, and intermediate-risk disease, using features that are already central to bedside decision-making rather than relying on a single anatomic label. In a pediatric cancer where overtreatment can leave lasting toxicity and undertreatment can invite relapse, the risk table functions as more than a classification exercise. It determines how aggressively clinicians escalate therapy, how much weight they place on tumor genetics, and where surgery fits into the sequence of care.
Age alone can move a child into the highest category. Stage M disease in a patient aged 18 months or older is classified as high-risk regardless of other features, a stark reminder that metastatic spread in an older infant or child carries a different biological meaning than it does in very young infants. The same stage M designation becomes high-risk in children younger than 18 months when MYCN amplification is present. Stage MS or stage L2 disease also enters the high-risk group when MYCN amplification is detected, and stage L2 disease with unfavorable histology is considered high-risk at age 18 months or older. In practice, this means that tumor genetics and microscopic appearance can override what might otherwise look like a more limited anatomic situation.
One adverse molecular event can move a child across the treatment threshold.
The high-risk category also captures stage M or MS disease in children aged 12 to 18 months when at least one unfavorable feature is present. Those features include unfavorable histology, segmental chromosomal aberrations, or a diploid tumor. Stage L1 incompletely resected disease with MYCN amplification is also placed in the high-risk group. Each of these entries reflects a different route to aggressive biology, but the clinical consequence is the same: the disease is not being judged by size or location alone. Instead, the tumor’s capacity for dissemination, genomic instability, and resistance potential is treated as the central danger signal.
Low-risk disease is defined with equal precision. Stage L1 disease with nonamplified MYCN status remains low-risk regardless of other features, and completely resected stage L1 disease with MYCN amplification is also categorized as low-risk. Stage MS disease in children younger than 12 months is low-risk only when every favorable feature is present: the child is asymptomatic, histology is favorable, no segmental chromosomal aberrations are present, and the tumor is hyperdiploid. This cluster of requirements makes clear that favorable outcome in neuroblastoma is rarely the result of a single reassuring marker. It is the convergence of age, clinical stability, tumor differentiation, and genomic profile that permits a conservative strategy.
For the youngest patients, biology can be mercifully different.
The intermediate-risk group functions as the residual category for all cases that do not meet high-risk or low-risk definitions. In clinical terms, this is where many real-world gray zones live. A patient may not have the clearly adverse profile that mandates the most intensive treatment, yet still lack the combination of favorable features needed to justify observation or limited therapy. Intermediate-risk disease therefore serves as a practical bridge between two extremes. It captures the children whose tumors require active treatment, but not necessarily the full intensity reserved for the most ominous biology.
This structure matters because neuroblastoma is not a uniform disease, even when it presents under the same diagnosis code. Stage, age, MYCN amplification, histology, chromosomal pattern, and ploidy each supply a different piece of prognostic information. The table used by the Children’s Oncology Group operationalizes those variables into a bedside algorithm that can be applied before treatment begins. That approach gives pediatric oncologists a shared language for planning chemotherapy intensity, surgical timing, and the extent of surveillance. It also limits the risk of treating low-risk tumors as though they were biologically aggressive, a mistake that can add long-term harm without improving cure rates.
Conversely, the table prevents biologically dangerous tumors from being undercalled because they appear limited on first glance. A small primary lesion with MYCN amplification does not behave like an indolent mass, and an older child with metastatic disease does not carry the same outlook as an infant with a favorable stage MS tumor. Neuroblastoma risk stratification is therefore an exercise in integrating what is seen on imaging with what the tumor is doing at the molecular level. The classification reflects a disease in which anatomy, development, and cancer genetics are tightly intertwined.
The table also emphasizes how resection status enters the picture without dominating it. Completely resected stage L1 disease with MYCN amplification is low-risk, while incompletely resected stage L1 disease with the same amplification is high-risk. Surgical completeness matters, but it does not act in isolation. Once again, tumor biology determines whether surgery is the final step or merely one part of a broader treatment plan. That distinction affects not only prognosis but also how clinicians think about postoperative therapy and follow-up intensity.
The published classification directs readers to a more complete system for full categorization, but the table itself already captures the essential clinical message: neuroblastoma risk is assigned by a layered assessment of age, stage, histology, genomic features, and resection status. In a pediatric malignancy where treatment toxicity can be substantial, that layered assessment is not a technicality. It is the foundation for matching therapy to disease behavior, and it remains the key to separating children who may be observed or treated conservatively from those who need immediate escalation.