The two-stage mutation model in retinal hemangioblastoma

Abstract

Background: The two-stage mutation model involving successive inactivation of both alleles of a tumor suppressor gene was originally proposed by Knudson, who analyzed the age incidence curves for unilateral and bilateral retinoblastoma, and suggested that hereditary tumors arise by a single somatic event superimposed on a defective genetic background and sporadic tumors by a two-stage somatic process. In this study, the age-incidence curve of patients with retinal hemangioblastoma with and without associated von Hippel-Lindau disease were analyzed.

Methods: We reviewed the literature between 1964 and 1998 to find all reported cases of retinal hemangioblastoma and classified patients in a type A group (n = 223) when associated with von Hippel-Lindau disease and a type B group (n = 30) when not associated with von Hippel-Lindau disease. We analyzed and compared the age incidence of these two groups.

Results: There was a statistically significant difference between the mean age at diagnosis of retinal hemangioblastoma in the two groups, i.e., 48.4 +/- 16.6 years for type B patients and 24.9 +/- 12.0 years for type A patients (p < 0.0001). The age incidence curve for type A retinal hemangioblastoma fit a first-order equation (log S = 0.411-0.034t) with r = 0.97, indicating a single somatic mutation, whereas the age incidence curve for type B retinal hemangioblastoma fit a second-order equation (log S = 0.184-2.25 x 10(-4)t2) with r = 0.97, indicating two somatic mutations.

Conclusions: Type B (sporadic) retinal hemangioblastoma may arise from two separate somatic mutations inactivating both alleles at the von Hippel-Lindau locus, whereas patients with von Hippel-Lindau disease (type A) inherit a defective allele and require only one additional somatic mutation.