The national infarct angioplasty project: UK experience and subsequent developments.

The UK had previously established a comprehensive strategy for in-hospital nurse-led thrombolysis for patients with ST-elevation myocardial infarction, with a growing use of pre-hospital thrombolysis by paramedical staff in the ambulance services. The National Infarct Angioplasty Project was sponsored by the government and examined the introduction of primary percutaneous coronary angioplasty (PPCI) in a variety of urban, rural and mixed communities. The project found that PPCI could be delivered within acceptable timelines, would be cost-effective, and could be delivered to the majority of the population. A project was therefore undertaken in England to transform services. There has been a rapid change and by 2012/13 over 95% of eligible patients received PPCI. Survival of patients with STEMI has improved over time and length of stay in hospital halved. However, nearly a quarter of STEMI patients do not receive reperfusion therapy (often because of late presentation) and additional work is needed to minimise delays to treatment. There are unexplained differences between regions in numbers of PPCI procedures per million population, and there is also variance between centres in the proportion of patients who are in shock or on a ventilator. Additional research is needed to ensure a consistent approach for these sick patients, who might have the most to gain from early treatment. The national audit programmes have been instrumental in measuring the changes in strategies, monitoring performance and highlighting the associated improvements in outcomes. A new risk model is being developed to allow a more comprehensive comparison of outcomes in different hospitals.

Gene expression profiling and subgroup identification of oligodendrogliomas.

The histological diagnosis of low-grade astrocytomas and oligodendrogliomas (WHO grade II) is often challenging, particularly in cases that show both astrocytic and oligodendroglial differentiation. We carried out gene expression profiling on 17 oligodendrogliomas (93% with LOH 1p and/or 19q) and 15 low-grade astrocytomas (71% with a TP53 mutation), using a cDNA array containing 1176 cancer-related genes. In oligodendrogliomas, 40 genes showed on average higher expression (at least a two-fold increase) than in astrocytomas, including DES, TDGF1, TGF-beta, GABA-BR1A, Histone H4, CDKN1A, PCDH43, Rho7 and Jun-D, while 39 genes were expressed at lower levels (at least a two-fold decrease), including JNK2, ITGB4, JNK3A2, RhoC, IFI-56K, AAD14 and EGFR. Immunohistochemistry revealed nuclear staining of Jun-D in oligodendrogliomas, in contrast to the immunoreactivity of cytoplasm and cell processes in low-grade astrocytomas. Partial least-squares analysis of the 79 genes at least two-fold differentially expressed between oligodendrogliomas and low-grade astrocytomas demonstrated perfect separation of oligodendrogliomas from low-grade astrocytomas and normal cerebral white matter. Clustering analysis based on the entire gene set divided the 17 subjects with oligodendrogliomas into two subgroups with significantly different survival (log-rank test, P=0.0305; survival to 5-years, 80 vs 0%, P=0.048). These results demonstrate that oligodendrogliomas and low-grade astrocytomas differ in their gene expression profiles, and that there are subgroups of oligodendroglioma with distinct expression profiles related to clinical outcome.

A simple method to improve probe set estimates from oligonucleotide arrays.

A popular commercially available oligonucleotide microarray technology employs sets of 25 base pair oligonucleotide probes for measurement of gene expression levels. A mathematical algorithm is required to compute an estimate of gene expression from the multiple probes. Previously proposed methods for summarizing gene expression data have either been substantially ad hoc or have relied on model assumptions that may be easily violated. Here we present a new algorithm for calculating gene expression from probe sets. Our approach is functionally related to leave-one-out cross-validation, a non-parametric statistical technique that is often applied in limited data situations. We illustrate this approach using data from our study seeking a molecular fingerprint of STAT3 regulated genes for early detection of human cancer.