paediatrics Brussels 17

J Neurooncol (2006)

Table 3 Treatment characteristics and incidence of recurrence in selected spinal MPE series

E/MPE a

Reference

Number of cases

Time period

Mean age (years)

GTR/STR-Bx Adjuvant RT (%)

Mean follow-up (month)

Relapse rate (%)

OS rate at 10 years (%)

Sonneland et al. [9] 77

1924–1985 36.7 1963–1983 47 1950–1987 40 1958–1987 37 1960–1984 36 1963–1994 33.5 1971–1990 36 1968–2002 33

0/77

45/32

46

N/A 17

N/A

Shaw et al. [25]

22

12/10 34/24 43/16

8/14

100

156

31.8 22.4 18.7

95

Whitaker et al. [16] 58 Waldron et al. [18] 59

14/44 16/43

74

70

100

100

130

75 86

Wen et al. [27] Schild et al. [14] Clover et al. [26]

20 35 11 35

20/9

7/13

65

N/A 30

N/A 100 b

23/12

10/25

100

124.8

50 c

11/6 0/35

1/10

73 63

88

80 97

Present study

21/14

127

34

a E/MPE = Ependymoma/myxopapillary ependymoma; GTR = Gross total resection; STR = Subtotal resection; Bx = Biopsy; RT = Radiotherapy; OS = Overall survival; N/A = Not available

b 5-year OS rate for MPE subgroup c Relapse rate for MPE subgroup

survival rates and decreased recurrences in patients treated with postoperative RT after STR or GTR in various series of spinal ependymoma or MPE [8, 13, 15, 17, 26, 30, 32]. The optimal dose of radiation for spinal ependy- momas has been debated in the literature. This has been extrapolated to MPE, which is usually located in the cauda equina. Most authors recommend doses of 40–50 Gy [16, 26, 27, 29]. In a study by Garcia et al. [33], a dose–response effect was seen. Patients who received doses greater than 40 Gy had signifi- cantly improved PFS rates. Similarly, Marks and Adler [32] recommended a dose of 40-Gy for totally resected MPEs. In our study, which analyzed the effect of total radiation dose using 45 Gy as the cutoff dose; we did not find any significant differ- ences between the two dose groups, most likely because most patients received 50 Gy to the tumor site, and only four patients received doses equal to or less than 45 Gy. In conclusion, the long-term survival for spinal MPE managed with surgery and adjuvant radiotherapy is favorable. Failures occur exclusively in the neural axis, mainly in the primary site. Regardless of the extent of surgery, adjuvant RT to the primary disease site appears to significantly reduce the rate of tumor progression.

amenable to complete resection in up to 94% of cases [19, 30, 31]. In our study, GTR was possible in a somewhat smaller percentage of patients with MPE, in keeping with the more irregular shape of MPEs and their contact with multiple nerve roots. However, in MPE patients in our study undergoing GTR, the data suggest that GTR alone may not be sufficient, and that adjuvant RT should be given and is indicated even in the setting of GTR. This is in contrast to spinal epen- dymoma in which GTR without RT may be deemed sufficient treatment. In this study, 12 patients (34%) experienced disease recurrence, 8 at the local site alone (10-year LC rate was 72%). A similar pattern of failure has been noted in other retrospective series. Whitaker et al. [16] showed that in 43 patients treated with postoperative radiation, 6 of 8 with recurrent tumors had failures at the primary site. Similarly, of the 22 patients who underwent surgery and postoperative RT in the study by Shaw et al. [25], 6 of 7 failures were at the primary site. We did not observe any extraneural metastases in our study but did observe four recurrent tumors that were distant from the site of the primary tumor. Two patients had recurrences in the brain. Although rare, this pattern of failure has been noted in other series. Whitaker et al. [16] found a 5.8% incidence of cranial relapse in 259 patients with spinal ependymomas; the pattern of failure in our series support the use of local field irradiation for localized spinal MPE, regardless of the extent of the resection. In our current study, the addition of postoperative RT to surgery was associated with significantly better 10-year PFS rates (75% for surgery + RT vs. 37% for surgery alone, P = 0.04) and 10-year LC rates (86% for surgery + RT versus 46% for surgery alone, P = 0.03). Several authors have similarly reported improved

References

1. Damjanov I, Linder J (eds) (1996) Anderson’s pathology, 10th edn. S Mosby, St. Louis, pp 2753–2755 2. Peschel RE, Kapp DS, Cardinale F, Manuelidis EE (1983) Ependymomas of the spinal cord. Int J Radiat Oncol Biol Phys 9:1093–1096 3. Shuman RM, Alvord EC, Leech RW (1975) The biology of childhood ependymomas. Arch Neurol 32:731–739

123