Radiation therapy

Radiation therapy is usually reserved for patients with disease that does not durably respond to chemotherapy.[21,22,42,43]

For children with low-grade gliomas for whom radiation therapy is indicated, approaches that contour the radiation distribution to the tumor and avoid normal brain tissue (3-D conformal radiation therapy, intensity-modulated radiation therapy (IMRT), stereotactic radiation therapy, and proton radiation therapy [charged-particle radiation therapy]) can reduce the acute and long-term toxicities associated with these modalities.[44,45]; [46][Level of evidence C2] Radiation doses of 54 Gy in 1.8 Gy fractions are typically used.[47,48] In a prospective study of 174 patients treated with proton therapy, the 5-year actuarial rate of local control was 85% (95% confidence interval [CI], 78%–90%), the PFS rate was 84% (95% CI, 77%–89%), and the OS rate was 92% (95% CI, 85%–95%). Brain stem and spinal cord tumor locations and a dose of 54 Gy relative biological effectiveness (RBE) or less were associated with inferior local control (P < .01 for both).[49] In a separate study that included 100 patients with WHO grade 2 diffuse gliomas, a subset of patients (n = 16) were treated with radiation therapy. These patients had a 5-year PFS rate of 74% when patients with histone H3 variants were excluded.[14][Level of evidence B4]

Subsequent to radiation therapy administration, care must be taken to distinguish radiation-induced imaging changes, termed pseudoprogression or spurious progression,[50] from disease progression. The peak time to radiation therapy–induced imaging changes, often presenting as an apparent enlargement of the irradiated mass, is 4 to 6 months, but they can manifest even later.[51-54]; [55,56][Level of evidence B4]; [8,57,58][Level of evidence C2] In a report of 83 patients with low-grade astrocytomas, pseudoprogression was more common after radiation doses of higher than 50.4 Gy (RBE) (hazard ratio [HR], 2.61; P = .16). Pseudoprogression was also more common after proton radiation therapy than after photon IMRT (HR, 2.15; P = .048), presumably because of increased effects on the vasculature. Patients with pilocytic histology had lower rates of pseudoprogression than those with nonpilocytic low-grade gliomas (HR, 0.47; P = .037). There was no association with overall disease control.[50]

A report from the SIOP-LGG 2004 (NCT00276640) study and LGG-registry cohorts evaluated the following radiological criteria for pseudoprogression:[59]

The following results were observed:

Radiation therapy results in long-term radiographic disease control for most children with chiasmatic and posterior pathway chiasmatic gliomas. However, despite radiological control, visual outcomes are variable.

The management of unresectable circumscribed astrocytic gliomas, pediatric-type diffuse low-grade gliomas, glioneuronal tumors, and neuronal tumors is controversial. To identify negative prognostic features in patients treated with radiation therapy, the St. Jude Children’s Research Hospital assessed 150 children (median age, 8 years; range, 1.2–20 years) who received radiation therapy and were monitored for a median of 11.4 years (range, 0.24–29.4 years). Recursive positioning analysis yielded low-risk and high-risk prognostic groups. The 10-year OS rate was 95.6% for patients in the low-risk group, versus 76.4% for patients in the high-risk group. Low-risk tumors included pilocytic astrocytoma/ganglioglioma located outside of the midbrain/thalamus, while high-risk tumors included diffuse astrocytoma or those located in the midbrain/thalamus. Within the high-risk group of patients, delayed radiation therapy (defined as after at least one line of chemotherapy) was associated with a decrement in OS.[65]

Children with NF1 may be at higher risk of radiation-associated secondary tumors and morbidity resulting from vascular changes. Radiation therapy is used as a last resort in these patients, given the heightened risk of inducing neurological toxic effects and second malignancy.[66]