Competing risks regression for clustered data

A population average regression model is proposed to assess the marginal effects of covariates on the cumulative incidence function when there is dependence across individuals within a cluster in the competing risks setting. This method extends the Fine-Gray proportional hazards model for the subdistribution to situations, where individuals within a cluster may be correlated due to unobserved shared factors. Estimators of the regression parameters in the marginal model are developed under an independence working assumption where the correlation across individuals within a cluster is completely unspecified. The estimators are consistent and asymptotically normal, and variance estimation may be achieved without specifying the form of the dependence across individuals. A simulation study evidences that the inferential procedures perform well with realistic sample sizes. The practical utility of the methods is illustrated with data from the European Bone Marrow Transplant Registry.     

B-RAF mutant alleles associated with Langerhans cell histiocytosis, a granulomatous pediatric disease

Background

Langerhans cell histiocytosis (LCH) features inflammatory granuloma characterised by the presence of CD1a+ dendritic cells or ‘LCH cells’. Badalian-Very et al. recently reported the presence of a canonical ^V600EB-RAF mutation in 57% of paraffin-embedded biopsies from LCH granuloma. Here we confirm their findings and report the identification of two novel B-RAF mutations detected in LCH patients.

Methods and Results

Mutations of B-RAF were observed in granuloma samples from 11 out of 16 patients using ‘next generation’ pyrosequencing. In 9 cases the mutation identified was ^V600EB-RAF. In 2 cases novel polymorphisms were identified. A somatic ^600DLATB-RAF insertion mimicked the structural and functional consequences of the ^V600EB-RAF mutant. It destabilized the inactive conformation of the B-RAF kinase and resulted in increased ERK activation in 293 T cells. The ^600DLATB-RAF and ^V600EB-RAF mutations were found enriched in DNA and mRNA from the CD1a+ fraction of granuloma. They were absent from the blood and monocytes of 58 LCH patients, with a lower threshold of sequencing sensitivity of 1%–2% relative mutation abundance. A novel germ line ^T599AB-RAF mutant allele was detected in one patient, at a relative mutation abundance close to 50% in the LCH granuloma, blood monocytes and lymphocytes. However, ^T599AB-RAF did not destabilize the inactive conformation of the B-RAF kinase, and did not induce increased ERK phosphorylation or C-RAF transactivation.

Conclusions

Our data confirmed presence of the ^V600EB-RAF mutation in LCH granuloma of some patients, and identify two novel B-RAF mutations. They indicate that ^V600EB-RAF and ^600DLATB-RAF mutations are somatic mutants enriched in LCH CD1a⁺ cells and absent from the patient blood. Further studies are needed to assess the functional consequences of the germ-line ^T599AB-RAF allele.     

Inferring loss-of-heterozygosity from unpaired tumors using high-density oligonucleotide SNP arrays

Loss of heterozygosity (LOH) of chromosomal regions bearing tumor suppressors is a key event in the evolution of epithelial and mesenchymal tumors. Identification of these regions usually relies on genotyping tumor and counterpart normal DNA and noting regions where heterozygous alleles in the normal DNA become homozygous in the tumor. However, paired normal samples for tumors and cell lines are often not available. With the advent of oligonucleotide arrays that simultaneously assay thousands of single-nucleotide polymorphism (SNP) markers, genotyping can now be done at high enough resolution to allow identification of LOH events by the absence of heterozygous loci, without comparison to normal controls. Here we describe a hidden Markov model-based method to identify LOH from unpaired tumor samples, taking into account SNP intermarker distances, SNP-specific heterozygosity rates, and the haplotype structure of the human genome. When we applied the method to data genotyped on 100 K arrays, we correctly identified 99% of SNP markers as either retention or loss. We also correctly identified 81% of the regions of LOH, including 98% of regions greater than 3 megabases. By integrating copy number analysis into the method, we were able to distinguish LOH from allelic imbalance. Application of this method to data from a set of prostate samples without paired normals identified known regions of prevalent LOH. We have developed a method for analyzing high-density oligonucleotide SNP array data to accurately identify of regions of LOH and retention in tumors without the need for paired normal samples.     

Structural and functional interaction sites between Na,K-ATPase and FXYD proteins

Several members of the FXYD protein family are tissue-specific regulators of Na,K-ATPase that produce distinct effects on its apparent K(+) and Na(+) affinity. Little is known about the interaction sites between the Na,K-ATPase alpha subunit and FXYD proteins that mediate the efficient association and/or the functional effects of FXYD proteins. In this study, we have analyzed the role of the transmembrane segment TM9 of the Na,K-ATPase alpha subunit in the structural and functional interaction with FXYD2, FXYD4, and FXYD7. Mutational analysis combined with expression in Xenopus oocytes reveals that Phe(956), Glu(960), Leu(964), and Phe(967) in TM9 of the Na,K-ATPase alpha subunit represent one face interacting with the three FXYD proteins. Leu(964) and Phe(967) contribute to the efficient association of FXYD proteins with the Na,K-ATPase alpha subunit, whereas Phe(956) and Glu(960) are essential for the transmission of the functional effect of FXYD proteins on the apparent K(+) affinity of Na,K-ATPase. The relative contribution of Phe(956) and Glu(960) to the K(+) effect differs for different FXYD proteins, probably reflecting the intrinsic differences of FXYD proteins on the apparent K(+) affinity of Na,K-ATPase. In contrast to the effect on the apparent K(+) affinity, Phe(956) and Glu(960) are not involved in the effect of FXYD2 and FXYD4 on the apparent Na(+) affinity of Na,K-ATPase. The mutational analysis is in good agreement with a docking model of the Na,K-ATPase/FXYD7 complex, which also predicts the importance of Phe(956), Glu(960), Leu(964), and Phe(967) in subunit interaction. In conclusion, by using mutational analysis and modeling, we show that TM9 of the Na,K-ATPase alpha subunit exposes one face of the helix that interacts with FXYD proteins and contributes to the stable interaction with FXYD proteins, as well as mediating the effect of FXYD proteins on the apparent K(+) affinity of Na,K-ATPase.