To evaluate the impact of radiation treatment parameters on cancer control outcomes for children with parameningeal rhabdomyosarcoma (PM-RMS) treated on Intergroup Rhabdomyosarcoma Study Group protocols II through IV (including IRS-IV pilot).

Radiation therapy (RT) treatment quality was assessed by contemporary review of portal radiographs, simulation films, treatment plans, and, in most cases, cross-sectional diagnostic imaging data for patients treated on Intergroup Rhabdomyosarcoma Study Group protocols II through IV. Five hundred ninety-five patients with PM-RMS were registered on these 4 studies between 1978 and 1997. Most of these patients (95%) had Group III disease. Radiation doses varied over the span of these trials with protocol doses ranging from 40 Gy to 50.4 Gy on IRS-II and IRS-III and 50.4 Gy to 59.4 Gy (hyperfractionated) on IRS-IV pilot and IRS-IV. Patients with high-risk signs of meningeal impingement, including cranial nerve palsy (CNP) or cranial base bone erosion (CBBE) with or without intracranial extension (ICE), were required to start radiotherapy at the time of study entry (Day 0). Among 595 patients reviewed, 385 (65%) had diagnostic images submitted to the Quality Assurance Review Center for assessment of target volume coverage. Only 123 (21%) patients, 49 (40%) of whom were treated on IRS-II, received whole brain RT.

The estimated overall survival and failure-free survival rates were 73% and 69% at 5 years, respectively. The estimated 5-year local failure (LF) rate was 17%. The detection of ICE increased from 24% to 41% as more cross-sectional diagnostic images became available. For patients with any sign of meningeal impingement, starting RT <2 weeks after diagnosis (n = 315) had 18% LF compared to 33% LF if started >2 weeks after diagnosis (n = 43) (p = 0.03). For patients with ICE, starting RT <2 weeks after diagnosis (n = 177) resulted in LF in 16% compared to 37% among those who started >2 weeks after (n = 19) (p = 0.07). For patients with CNP and/or CBBE, starting RT <2 weeks after diagnosis (n = 138) resulted in 21% LF compared to 30% among those that started >2 weeks (n = 23) (p = 0.23). In none of these circumstances was the 5-year failure-free survival significantly impacted by this increase in LF. The estimated 3-year survival after local failure was 17% (95% CI, 10%-25%). For patients without signs of meningeal impingement, there was no difference in local control whether they started radiation therapy earlier or later than 10 weeks. Patients with large (> or =5 cm) Group III tumors had an LF rate of 35% if they received less than 47.5 Gy compared to an LF rate of 18% in patients who received less than 47.5 Gy with smaller tumors or a rate of 15% if they received more than 47.5 Gy, irrespective of tumor size (p = 0.14). There was no evidence that whole brain radiation therapy affected LF or reduced central nervous system (CNS) relapse. Multivariate analysis of RT parameters and clinical factors demonstrated that a radiation dose of >47.5 Gy was associated with lower LF. The presence of ICE, CNP, or CBBE and age >10 years at diagnosis were significantly associated with higher rates of local failure.

The availability of cross-sectional diagnostic images (CT or MRI) has improved detection of ICE. Starting radiation therapy within 2 weeks of diagnosis for patients with signs of meningeal impingement was associated with lower rates of local failure. When no signs of meningeal impingement were present, delay of radiation therapy for more than 10 weeks did not impact local failure rates. Whole brain radiation therapy is unnecessary in PM-RMS. A dose of at least 47.5 Gy seems to be associated with lower rates of local failure, especially when tumor diameter is > or =5 cm.