The assessment of Ki-67 as a prognostic marker in neuroendocrine tumours: a systematic review and meta-analysis.

INTRODUCTION: Gastroenteropancreatic neuroendocrine tumours (GEP NETs) are classified according to tumour mitotic count or Ki-67 labelling index (LI). AIMS: To systematically review articles reporting the prognosis of patients by Ki-67 LI and thereby improve the ability of clinicians to prognosticate for their patients. METHOD: 265 abstracts were identified relating Ki-67 and survival. After exclusion criteria were applied, 22 articles remained. Articles were excluded if they described non-human specimens, were non-English language, published prior to 2000, reported non-GEP NETs, reported subgroups selected by treatment modality or included <20 cases. Random-effects meta-analysis was used to combine studies to estimate survival proportions. RESULTS: Authors used varied methods in which to present 5-year survival, with often limited survival information. This reduced the number of studies that could be included in the meta-analysis. 5-year survival for patients with grade 1 and 2 GEP NETs were estimated to be 89% (95% CI 85% to 92%, m=12 studies, n=977 participants) and 70% (95% CI 62% to 79%, m=9, n=726), respectively. Using an alternative grade 1/2 boundary of 5%, 5-year survival rates for Ki-67≤5% and 5-20% were estimated as 89% (95% CI 84% to 94%, m=7, n=654) and 51% (95% CI 44% to 59%, m=4, n=183), respectively. For Ki-67>20%, 5-year survival was estimated to be 25% (95% CI 12% to 38%, m=10, n=208). CONCLUSIONS: Standardisation of grade boundaries has allowed us to combine data from multiple studies and amass a body of evidence linking Ki-67 and survival.

A Bayesian network for modelling blood glucose concentration and exercise in type 1 diabetes.

This article presents a new statistical approach to analysing the effects of everyday physical activity on blood glucose concentration in people with type 1 diabetes. A physiologically based model of blood glucose dynamics is developed to cope with frequently sampled data on food, insulin and habitual physical activity; the model is then converted to a Bayesian network to account for measurement error and variability in the physiological processes. A simulation study is conducted to determine the feasibility of using Markov chain Monte Carlo methods for simultaneous estimation of all model parameters and prediction of blood glucose concentration. Although there are problems with parameter identification in a minority of cases, most parameters can be estimated without bias. Predictive performance is unaffected by parameter misspecification and is insensitive to misleading prior distributions. This article highlights important practical and theoretical issues not previously addressed in the quest for an artificial pancreas as treatment for type 1 diabetes. The proposed methods represent a new paradigm for analysis of deterministic mathematical models of blood glucose concentration.