The Journal of Bone and Joint Surgery, Vol 72, Issue 5 643-653, Copyright © 1990 by Journal of Bone and Joint Surgery, Inc
The effect of local extent of the tumor on prognosis in osteosarcoma
SS Spanier, JJ Shuster and RA Vander Griend
Orthopaedic Oncology Division, University of Florida, Gainesville 32611.
The effect of local extent of the tumor on disease-free survival was
studied in fifty-one patients who had an osteosarcoma, all of whom were
treated according to the same protocol. Disease-free survival was defined
as the interval between entry into the protocol and local recurrence,
metastasis, or death (whichever occurred first). All patients had a
Stage-IIB intramedullary osteosarcoma--that is, a high-grade malignant
tumor (Stage II) with cortical penetration (B). The tumors were subdivided
on the basis of the amount of local extension: E1--the tumor touches but
does not elevate or penetrate the periosteum; E2--the tumor elevates but
does not penetrate the periosteum; E3--the tumor penetrates into, but not
through, the periosteum; E4--there is minimum extraperiosteal extension
without invasion of another structure, such as a muscle, tendon, or
ligament; E5--the tumor invades one additional structure (a muscle,
ligament, or tendon); and E6--the tumor invades two or more structures
adjacent to the bone. When the tumor had invaded two or more adjacent
structures (E6), the instantaneous risk of failure was estimated to be 5.9
times greater than when the tumor had a lesser extent (E1 through E5). The
relative prognostic importance of the amount of local extension, the size
of the tumor, and the sex and age of the patient was compared with forward
step-wise Cox multivariate analysis. Only the amount of local extension had
a significant effect on disease-free survival. The estimated cumulative
probabilities of disease-free survival at five years, according to the
Kaplan-Meier method, were 49.1 +/- 8.5 per cent for patients who had a
Stage-IIB tumor, 79.8 +/- 9.3 per cent for those who had a Stage-IIB but
not E6 tumor, and 17.6 +/- 11.3 per cent for those who had a Stage-IIB E6
tumor. These data suggest that, with respect to prognosis, there are two
distinct subgroups of Stage-IIB osteosarcomas that can be objectively
defined by the amount of local extension.