Choroid plexus tumours

Abstract

Choroid plexus tumours are rare epithelial brain tumours and limited information is available regarding their biology and the best treatment. A meta-analysis was done to determine prognostic factors and the influence of various treatment modalities. A thorough review of the medical literature (1966-1998) revealed 566 well-documented choroid plexus tumours. These were entered into a database, which was analysed to determine prognostic factors and treatment modalities. Most patients with a supratentorial tumour were children, while the most common sites in adults were the fourth ventricle and the cerebellar pontine angle. Cerebellar pontine angle tumours were more frequently benign. Histology was the most important prognostic factor, as one, five, and 10-year projected survival rates were 90, 81, and 77% in choroid plexus-papilloma (n=353) compared to only 71, 41, and 35% in choroid plexus-carcinoma respectively (P<0.0005). Surgery was prognostically relevant for both choroid plexus-papilloma (P=0.0005) and choroid plexus-carcinoma (P=0.0001). Radiotherapy was associated with significantly better survival in choroid plexus-carcinomas. Eight of 22 documented choroid plexus-carcinomas responded to chemotherapy. Relapse after primary treatment was a poor prognostic factor in choroid plexus-carcinoma patients but not in choroid plexus-papilloma patients. Treatment of choroid plexus tumours should start with radical surgical resection. This should be followed by adjuvant treatment in case of choroid plexus-carcinoma, and a "wait and see" approach in choroid plexus-papilloma.

Figure 1

Age distribution of choroid plexus tumours: Supratentorial tumours occur most frequently in infants (A) n=188, while the age distribution of infratentorial tumours is spread over all ages (B), n=71; P<0.005, Mann–Whitney U-test). This analysis included only patients with documented primary intracerebral locations.

Figure 2

Survival of patients published as CPP with CPC patients. Only patients with observation times >0 and clear histological allocation were included. Symbols indicate end of observation times of surviving patient (=censored data). Statistically significant: P<0,0005, log rank test.

Figure 3

Young children and patients over 59 years of age have a worse prognosis as compared to the age classes in between (P<0.0005 log rank test). This included patients, in whom the observation time was >0, and in whom age at diagnosis was documented. Survival rates were calculated as Kaplan–Meier estimates.

Figure 4

In both, CPC, (A) and CPP (B), the type of surgery was a statistically significant prognostic factor (P=0.0001, and P<0.00005, respectively, log rank tests).

Figure 5

Influence of irradiation on survival with incompletely resected CPC. Inclusion factors for this analysis were: Observation time >0, histology CPC, surgery well documented, less than gross total resection. The difference in favour of irradiation is marginally statistically significant (P=0.029, log rank test).