Age-linked prognostic categorization based on a new histologic grading system of neuroblastomas. A clinicopathologic study of 211 cases from the Pediatric Oncology Group.

Histologic sections (minimum of four sections per patient) from 211 patients with neuroblastoma were reviewed. The tumors were resected before therapy, which was standardized according to age and stage. Low mitotic rate (MR) (less than or equal to ten per ten high-power fields) and calcification emerged as the most significant prognostic features after statistical analysis by stepwise log-rank tests (P less than 0.0001 and P = 0.0065, respectively). Histologic Grades 1, 2, and 3 were defined on the basis of the presence of both, any one, or none of these two prognostic features, respectively (Grade 3 had absence of low MR, i.e., these tumors had high MR [greater than ten per ten high-power fields]). Statistically significant differences in survival were observed in the grades after adjusting for age and stage (P less than 0.001). The degree of differentiation, although significant by itself, was no longer significant after adjusting for the grades. Age groups (less than or equal to 1 versus greater than 1 year of age), which also emerged as an independent prognostic feature (P less than 0.001), were linked with the grades to define two risk groups as follows: (1) a low-risk (LR) group consisting of patients in both age groups with Grade 1 tumors and patients 1 year of age or younger with Grade 2 tumors and (2) a high-risk (HR) group consisting of patients older than 1 year of age with Grade 2 tumors and patients in both age groups with Grade 3 tumors. The difference in survival between LR (160 cases) and HR groups (51 cases) was statistically significant (P less than 0.001). Concordance between these LR and HR groups and the Shimada classification was observed in 84% of cases. The new histologic grading system has the following advantages: (1) use of familiar terminology and histologic features in the grading system and (2) relative ease of assessment because the degree of differentiation does not need to be determined. The grading system should be tested on a new data set with an appropriate histologic sample of similar size to confirm these results.     

Sample size calculations for the log rank test: a Gompertz model approach.

Sample size calculations for clinical trials dealing with survivorship are often based on an exponential model. This model is inappropriate when a non-zero proportion of the population is expected to have indefinite survival. In such cases the Gompertz model offers a reasonable alternative. A method for calculating the required accrual time for a clinical trial in which the treatment arms have Gompertz survival distributions satisfying the proportion hazards assumption is developed. A computer program to perform this method is given, as well as an iterative method that can be used when a computer is not available.