In recent years some progress has been made in the therapy of EPs. However, the prognosis for patients with EPs remains rather poor, and sound evidence regarding the main prognostic factors and the optimal therapeutic management strategy for EPs is still lacking.
The main primary treatment for EP is surgery, however, a complete removal of the tumor is rarely possible. Based on historical data and most recent series, there is a trend to consider the extent of surgery as a major prognostic factor [4, 5]. Therefore, surgery is commonly the first procedure performed in patients with EP:
Survival rates for childhood EPs reported in the literature vary from 14% to 70% at 5 years [4, 11, 16, 25–31]. In series including infratentorial lesions in children only, survival rates tend to be lower, whereas in series including adult EPs survival rates are much better [7, 32]. In the present analysis survival rates at 5 and 10 years were 77% and 64%, respectively. In analogy to a variety of published data, age ≥ 18 seems to be associated with higher overall survival. Patients ≥ 18 yrs. showed 5 and 10 year overall survival rates of 88% and 74% as opposed to 68% and 54% in children < 18 years of age. Patients under 4 years of age demonstrated significantly worse overall survival times compared to patients ≥ 4 years of age (p = 0.04).
In a variety of studies, histologic grade of the tumor was also identified as a prognostic factor [16, 26, 33–38]. However, a number of studies including the present analysis did not reveal a significant influence of histologic grade on treatment outcome. Therefore, the possible correlation between tumor grade and survival is discussed controversially [4, 7, 11, 16, 29, 34, 39–41]. The different findings reported might be attributable to sample size, anatomic tumor location, variability in the definition of anaplasia, discrepancies in histological diagnosis, and the inclusion in some series of ependymoblastoma and subependymoma, which exhibit different biologic behaviour and should be analyzed separately [5, 42–44].
Furthermore, there is still some controversy about the definition of the optimal treatment volume for EP. Large studies have shown no survival benefit for patients treated with whole-brain radiotherapy (WBRT) and irradiation of the craniospinal axis (CSA) as opposed to RT administered to smaller treatment volumes [25, 33, 45, 46]. Kovalic reported a 10 year survival rate of 75% for patients treated with local RT as compared to 28% in patients treated with WBRT . Furthermore, several investigations observed a lack of statistically significant improvement of outcome for patients treated with craniospinal irradiation [25, 33, 46–48]. However, RT of the CSA was frequently used in the treatment of high-grade ependymomas due to their anticipated higher risk of metastatic cerebrospinal fluid (CSF) seeding [15, 25, 34, 49, 50].
However, in several series the most common area of recurrence was locally, within the former RT field [25, 33, 46–48]. To improve local control, high local doses to the defined target volume are essential. In the past, a number of studies have proven a significant correlation between adequate RT dose and local control. An RT dose between 50 Gy and 60 Gy has been shown to be effective and high local control rates can be achieved, with 54 Gy being considered the standard dose and higher doses being prescribed in investigational studies. Garrett et al. showed overall survival rates of 14% in patients treated with less than 45 Gy, and a survival rate of 50% in patients treated with more than 45 Gy . Goldwein et al reported similar results, with overall survival rates of 18% in patients treated with less than 45 Gy as compared to survival rates of 51% in patients treated with doses ≥ 45 Gy . Therefore, based on the known dose-response relationship of EP, an improvement in RT techniques might help improve local control, enabling escalation of local RT doses while minimizing irradiation of surrounding normal tissue.
This might be achieved by applying modern RT techniques such as stereotactic radiosurgery (SRS) and fractionated stereotactic radiotherapy (FSRT). High precision radiotherapy allows the application of high doses to a defined target volume while sparing normal tissue . In the past, high precision radiotherapy techniques were not applied routinely in the management of EP. With SRS, high local doses are applied in a single fraction. SRS has been applied in a small number of studies in patients with EP: Hodgson et al. published the largest series applying SRS in patients with EP, reporting a 3-year progression-free survival rate of 22% in 28 paediatric patients. Jawahar et al. report the outcome of 22 patients treated with SRS for anaplastic EP, with a 5-year actuarial progression-free survival rate of 32.4%. A recent study published by Mansur et al. showed a 3-year relapse free survival rate of 55.6%, and 3-year overall survival rate of 71.1% in nine patients with EP treated with SRS [51, 52]. However, SRS is limited to small targets, and therefore this technique is preferred especially in small recurrent lesions or as a boost modality for macroscopic residuals in combination with external beam RT to the tumor bed . With fractionated stereotactic radiotherapy (FSRT), larger target areas can be treated while exploiting the biological effect of fractionation. However, in the literature little data on FSRT in patients with EP can be found. The results of the present study support the idea, that FSRT can be applied safely in patients with primary or recurrent EP. Overall survival and local control rates are comparable to data found in the literature and show that FSRT can help improving local control without leading to an increased rate of marginal recurrences.
The role of chemotherapy (CHT) in the management of EP still remains unclear [27, 53]. In general, it would be reserved for patients after failure of previously performed surgery and RT [35, 54, 55]. The German Society of Paediatric Hematology and Oncology is conducting a large study for children with EP, investigating concomitant chemotherapy with vincristine for WHO grade II and III EP in children between age 4 and 21 with persisting tumor after initial RT. Adjuvant chemotherapy with cisplatin, vincristine, carboplatin and VP16 is recommended for Grade III EP. In children younger than age of 4, chemotherapy including cyclophosphamide, vincristine and methotrexate is recommended after surgery . Needle et al. report a benefit in overall survival for patients treated with carboplatin, vincristine, ifosfamide and etoposide . In spite of incomplete surgical resection, progression-free survival after 5 years was 80%. However, hyperfractionated RT was also included into this regime. Kühl et al. could demonstrate a rate of 55% of partial or complete remissions after postoperative CHT before the initiation of RT in patients with anaplastic EP . In a study performed by the Royal Marsden Hospital, London, 10-year overall survival was 54% in patients treated with CCNU as adjuvant CHT as opposed to 34% in patients treated with RT only . In children < 2–3 years of age, the up-front application of CHT can help delay the need to initiate RT [46, 58–62]. In the present analysis, three children were treated according to studies performed by the GPOH on multimodality treatment of EP (HIT) and received chemotherapy sequentially . A group of three patients received chemotherapy for tumor progression, including vincristine, carboplatin, cisplatine, ifosfamide and etoposide.
The results of large randomized trials evaluating the role of adjuvant CHT are anticipated.