Intensity Modulated Radiotherapy (IMRT) and Fractionated Stereotactic Radiotherapy (FSRT) for children with head-and-neck-rhabdomyosarcoma

The distribution of the tumor sites is shown in table 1. The median age of the children was 5 years at the time of RT (range 2–15 years). The male to female ratio was 1,4 : 1.

Before radiation therapy, all patients had undergone at least one surgical procedure. At primary diagnosis, 3 patients had a subtotal tumor resection, and in 16 patients only a biopsy was performed.

Patients were classified according to the surgical-pathologic grouping system used in the Intergroup Rhabdomyosarcoma Study Group (IRSG) trials [18, 19]. The patients' characteristics can be found in table 1.

All patients included into this study received systemic chemotherapy according to the German Soft Tissue Sarcoma Study protocols CWS 91, CWS 96 and CWS 2002 [20, 21]. The aggressive interdisciplinary treatment concepts are performed in close collaboration between the radiation oncologist and the pediatric oncologist. All patients were registered at the CWS study center.

The median time interval between primary diagnosis and radiation therapy was 5 months (range 3–9 months).

Patients were immobilized using an individual mask fixation system made of Scotch-Cast™; for treatment planning contrast-enhanced MRI- and CT-scans were acquired with the mask attached to a non-invasive stereotactic localization frame with an overall geometrical uncertainty of 1–2 mm [22].

For FSRT, we used the three-dimensional treatment planning system Voxelplan [23]. Three to five isocentric non-coplanar irregularly shaped fields formed by a midsize multileaf collimator were applied.

For IMRT, the inverse treatment-planning program Konrad (MRC Systems GmbH, Heidelberg, Germany) was used [24, 25]. An integrated multi-leaf collimator allowed for IMRT using the "step-and-shoot" technique described previously [26]. Fig. 1 shows an IMRT treatment plan for a child with head and neck RMS.

Median size of the planning target volume (PTV) for radiation was 93,4 ml (range 20,8 – 354,8 ml), consisting of the area of contrast uptake in T1-weighted sequences and the enhancing lesion and edema visible in T2-weighted MRI scans prior to chemotherapy including a 20 mm safety margin. It was our policy to define the PTV according to the common guidelines outlined in the treatment protocols of the GPOH. Gross tumor volume (GTV) and clinical target volume (CTV) were not defined separately for planning purposes.

We applied a median total dose of 45 Gy (range 32 Gy – 54 Gy) using a median fractionation of 5 × 1,8 Gy/week (range 1,6 Gy – 1,8 Gy).

In 4 patients, the total dose was applied in a hyperfractionated regimen using single fraction doses of 1,6 Gy.

Deep sedation or general anaesthesia was required in 5 out of 19 children.

Patients were seen for follow-up 6 weeks after radiation, thereafter in 3 to 12 months intervals, including contrast-enhanced MRI-scans, anamnesis and clinical examination. The clinical oncological follow-up was performed by the attending pediatric oncologist. Median follow-up time was 17 months (range 3 months 12 years).

Overall survival was calculated from primary diagnosis, survival after RT was calculated from initiation of RT. Progression-free survival after RT was calculated from the first day of RT until tumor progression (local or distant) or death (by any cause), whichever happened first, using the Kaplan Meier method [27]. All statistical analyses were performed using the Statistica 6.1 Software (StatSoft, Tulsa, OK, USA).

Overall survival rate after primary diagnosis of the RMS was 94% at 3 and 5 years, respectively (Fig. 2). One patient died of tumor progression.

The 3- and 5 year actuarial local control rate after RT was 89% (Fig. 3A). We observed 2 local recurrences.

With respect to tumor localization, the 3- and 5-year local control rates were 100% for patients with RMS of the orbit and 91% for patients with parameningeal tumors. One patient with parameningeal RMS developed local progression within the prior RT field of the tumor after 4 months. The patient had received a total RT dose of 54 Gy, RT was initiated 5 months after primary diagnosis after inclusion into the CWS-91 protocol. The tumor had shown no response to the CWS chemotherapy (ifosfamide, vincristine, actinomycin-D), and the chemotherapy regimen was switched to carboplatin/etoposide, which showed also no response. Thereafter, RT was initated.

One patient with RMS (ISG Group 4) of the mylohyoid muscle developed local failure 2 months after completion of RT simultaneously with distant bone metastases and progression of the lung metastases present at primary diagnosis. This patient was 1 year old at primary diagnosis, was included into the CWS 2002 protocol and showed no response to chemotherapy prior to RT, which was initiated 4 months after diagnosis; a total RT dose of 54 Gy was applied. At primary diagnosis, lung metastases were already present.

The rate of local control in children with parameningeal and head- and-neck tumors was 81% at 3- and 5-years as compared to 100% in RMS involving the orbit only (Fig. 3B).

The actuarial freedom of distant metastases rate at 3- and 5-years was 89% for all patients at 3 and 5 years (Fig. 3C). For patients with RMS of the orbit only, the actuarial distant progression-free rate was 100% at 5 year. One patient with parameningeal RMS developed distant metastases in both lungs 6 months after completion of RT and 1 month after local tumor progression. The actuarial 3- and 5-year distant disease-free survival rate was 91%, respectively in children with parameningeal RMS. As noted above, one girl diagnosed with ISRG 4 RMS of the mylohyoid muscle developed local and distant progression simultaneously 2 months after completion of RT. The rate of distant-metastases-free survival in children with parameningeal and head-and -neck RMS was 81% at 5-years as compared to 100% in children with involvement of the orbit only (Fig. 3D).

Radiotherapy was well tolerated in all children and could be completed without interruptions > 4 days.

Minor acute radiation-induced side effects occurring within 3 months from RT included skin erythema in 8 patients (42%), alopecia in 4 patients (21%), conjunctivitis in 3 patients (16%), mucositis in 8 patients (42%) and nausea/vomiting in 2 patients (11%); no toxicities >CTC grade 2 developed.

To date, the long-term toxicity, developing per definition <6 months after RT, developed in 9 out of 19 patients (47%). We observed three patients with reduced lacrimation and one patient with chronic swelling of the left eye lid with slight erythema CTC grade 1. One patient developed trismus. However, the poor cooperation of the patient in training mouth opening after RT could have contributed to this outcome. In one patient growth hormone deficiency developed with subsequent microsomia requiring growth hormone replacement. This patient had been treated with the initial CWS-96 protocol followed by treatment according to the CWS-protocol for tumor recurrence. Radiotherapy had been applied during the initial protocol with 32 Gy, followed by 8 Gy brachytherapy and 50,4 Gy FSRT for tumor recurrence. In one patient with orbital RMS visual deficiency present prior to RT declined leading to amaurosis. Optic pathway structures received a dose of 34 Gy in 1,8 Gy fractions. Discrete erythema and swelling of the eye lid was observed in the same patient with reduced lacrimation. One patient presented with reduced mouth-opening, and facial asymmetry developed in one patient only (5%). One patient with RMS of the pterygoid fossa developed asymmetric growth of the face with growth reduction on the left side. Until now, no secondary malignancies developed.