Do different clinical evidence bases lead to discordant health-technology assessment decisions? An in-depth case series across three jurisdictions.

BACKGROUND: Health-technology assessment (HTA) plays an important role in informing drug-reimbursement decision-making in many countries. HTA processes for the Pharmaceutical Benefits Advisory Committee (PBAC) in Australia, the Common Drug Review (CDR) in Canada, and the National Institute for Health and Clinical Excellence (NICE) in England and Wales are among the most established in the world. In this study, we performed nine in-depth case studies to assess whether different clinical evidence bases may have influenced listing recommendations made by PBAC, CDR, and NICE. METHODS: Nine drugs were selected for which the three agencies had provided listing recommendations for the same indication between 2007 and 2010. We reviewed the evidence considered for each listing recommendation, identified the similarities and differences among the clinical evidence bases considered, and evaluated the extent to which different clinical evidence bases could have contributed to different decisions based on HTA body comments and public assessment of the evidence. RESULTS: HTA agencies reached the same recommendation for reimbursement (recommended for listing) for four drugs and different recommendations for five drugs. In all cases, each agency used different evidence bases in their recommendations. The agencies considered overlapping sets of clinical comparators and trials when evaluating the same drug. While PBAC and NICE considered indirect and/or mixed-treatment comparisons, CDR did not. In some cases, CDR and/or NICE excluded trials from review if the drug and/or the comparator were not administered according to the relevant marketing authorization. CONCLUSIONS: In the listing recommendations reviewed, considerable variability exists in the clinical evidence considered by PBAC, CDR, and NICE for drug-listing recommendations. Differences in evidence resulted from differences in the consideration of indirect and mixed-treatment comparison data and differences in medical practice in each jurisdiction.

Genomic convergence to identify candidate genes for Parkinson disease: SAGE analysis of the substantia nigra.

Genomic convergence is a multistep approach that combines gene expression with genomic linkage to identify and prioritize susceptibility genes for complex disease. As a first step, we previously performed linkage analysis on 174 multiplex Parkinson's disease (PD) families, identifying five peaks for PD risk and two for genes affecting age at onset (AAO) in PD [Hauser et al., Hum Mol Genet 2003;12:671-677]. We report here the next step: serial analysis of gene expression [SAGE; Scott et al., JAMA 2001;286:2239-2242] to analyze substantia nigra tissue from three PD patients and two age-matched controls. We find 933 differentially expressed genes (P<0.05) between PD and controls, but of these, only 50 genes represented by unique SAGE tags map within our previously described PD linkage regions. Furthermore, genes encoded by mitochondrial DNA are expressed 1.5-fold higher in PD patients versus controls, without an increase in the corresponding nuclear-encoded mitochondrial components, suggesting an increase in mtDNA genomes in PD or a disjunction with nuclear expression. The next step in the genomic convergence process will be to screen these 50 high-quality candidate genes for association with PD risk susceptibility and genetic effects on AAO.

Somatic mutation of vascular endothelial growth factor receptors in juvenile hemangioma.

Juvenile hemangiomas are the most common tumors of infancy, occurring in as many as 10% of all births. These benign vascular lesions enlarge rapidly during the first year of life by hyperplasia of endothelial cells and attendant pericytes and then spontaneously involute over a period of years, leaving loose fibrofatty tissue. Several hypotheses have been put forth concerning hemangiogenesis, including the possibility that the tumor is the result of somatic mutation in one or more components of critical vascular growth-regulatory pathways. To test this hypothesis, we obtained 15 proliferative-phase hemangiomas after surgical resection and dissected them to enrich for the lesional (endothelial and pericytic) components of each specimen. To determine whether hemangiomas represent a clonal expansion from a single progenitor cell, we assayed X-inactivation patterns for each lesion by using the polymorphic X-linked human androgen receptor gene. Twelve of 14 informative hemangiomas showed a significant degree of allelic loss after methylation-based and transcription-based polymerase chain reaction clonality assays, suggesting a nonrandom X-inactivation pattern and, thus, a monoclonal origin. We then sequenced genes encoding the receptors of the vascular endothelial growth factors (VEGFs) as candidates for potential somatic mutation. Mutations were found in two of the 15 hemangioma specimens: a missense mutation (P1147S) in the kinase domain of the VEGFR2 (FLK1/KDR) gene in one specimen and a missense mutation (P954S) in the kinase insert of the VEGFR3 (FLT4) gene in another specimen. In each case, the mutation was detected in tumor tissue but not in adjacent normal tissue. These results suggest that one potential mechanism involved in hemangioma formation is the alteration of the VEGF signaling pathway in endothelial and/or pericytic cells.