Can the prognosis of individual patients with glioblastoma be predicted using an online calculator

This study was designed to evaluate an online prognosis prediction calculator intended for use in patients with glioblastoma by comparison with actual outcomes in 2 UK-based neurosurgical units. Glioblastoma multiforme (GBM) is the most common and most aggressive primary brain tumor in adults. With an incidence rate of 2–3 cases per 100 000 per year, GBM accounts for 22.6% of all brain and central nervous system tumors.1 The overall median survival is reported as 7.6–9.4 months,2,3 but this can reach 14 months with a 2-year survival rate of 27% in clinical trial groups.4 It is therefore a conspicuous clinical entity and conveys a poor prognosis. In 2005, a phase III randomized trial organized by the European Organisation for Research and Treatment of Cancer (EORTC) and the National Cancer Institute of Canada (NCIC) Clinical Trials Group demonstrated a survival benefit conveyed by radiotherapy plus concomitant and adjuvant temozolomide in the treatment of patients with newly diagnosed glioblastoma.4 In an exploratory subanalysis of the EORTC/NCIC trial data, Gorlia et al. subsequently identified a variety of factors that were predictive of overall survival, including therapy administered, age, extent of surgery, mini-mental score examination, administration of corticosteroids, World Health Organization (WHO) performance status, and O-methylguanine-DNA methyltransferase (MGMT) promoter methylation status.5 This is in agreement with the most consistently reported prognostic factors for GBM (age, extent of surgical resection, and performance status),6–8 but promoter region methylation of the MGMT enzyme, which repairs DNA damaged by alkylating agents, and MMSE have also previously been associated with improved prognosis.9,10 On the basis of their findings, Gorlia et al. developed 3 nomograms, each intended to predict the survival times among patients with newly diagnosed GBM on the basis of individual-specific combinations of prognostic factors.5 These nomograms have been made available online as a “GBM Calculator” and are intended for use in patient counseling and treatment selection and in the design and interpretation of clinical trials.11 The nomograms were developed by correlating all of the identified predictive factors with outcome and combining them to give a predicted median survival and a 2-year survival probability. This was done in 3 distinct populations to separate the potential skewing effect of MGMT promoter methylation status and treatment given. Population 1 was the entire group (n = 573). Population 1 was stratified using the factors shown in Table 1. Table 1: Stratification criteria for GBM calculator model 1 In the authors' experience, many patients would like to know their prognosis at diagnosis to understand their condition and to plan practical and financial affairs, and it certainly helps them and their families to start the anticipatory grieving process necessary in terminally ill patients.12 The GBM calculator promised to be a very helpful clinical tool in this respect. Before offering this information to patients, the authors wanted to validate its use in a local patient population, because it was felt that giving potentially incorrect information would be unhelpful and insensitive and would undermine confidence in the treating team.