Julio M. Singer

Restaging of colorectal cancer based on the identification of lymph node micrometastases through immunoperoxidase staining of CEA and cytokeratins.

The present study was performed to identify tumor cells in lymph nodes from colorectal adenocarcinomas considered free of disease by the classic hematoxylin-eosin stain, based on the detection of the carcinoembryonic antigen (CEA) and cytokeratins in neoplastic epithelial cells. For this purpose, 603 lymph nodes from 46 lesions were stained by the peroxidase-antiperoxidase technique. Tumor cells were detected in 22 nodes from 12 patients, mainly in the subcapsular sinuses, permitting a restaging of these patients into two groups: those now considered to have metastatic disease and those free of metastases. However, the 5-year follow-up showed no statistical differences in survival between the two groups.

M-Methods in multivariate linear models

For the multivariate linear model, coordinatewise M-estimators as well as an extension of the Maronna-type M-estimators are considered. Based on the Jureckova (asymptotic) linearity of M-statistics, the asymptotic distribution theory of the proposed estimators is studied under appropriate regularity conditions, and incorporated in the formulation of some (asymptotic) M-tests of linear hypotheses. Finally, robustness properties of both types of estimators are discussed.

Modelos MLG e MAG para análise da associação entre poluição atmosférica e marcadores de morbi-mortalidade: uma introdução baseada em dados da cidade de São Paulo

Este estudo, descreve e compara duas classes de modelos - os Modelos Lineares Generalizados (MLG) e os Modelos Aditivos Generalizados (MAG) - que podem ser utilizadas para avaliar a associacao entre poluicao atmosferica e marcadores de morbi-mortalidade. Enfoca os MAG como uma alternativa para a modelagem de relacoes nao lineares nao especificadas, e mostra que essa classe de modelos constitui uma boa opcao para representar tanto a sazonalidade quanto a relacao entre o desfecho e as variaveis meteorologicas. Como exemplo de aplicacao e avaliada a associacao entre mortalidade em idosos e poluicao atmosferica na cidade de Sao Paulo no periodo de 1994 a 1997. Os dados de mortalidade foram obtidos do Programa de Aprimoramento das Informacoes de Mortalidade (PRO-AIM) e as concentracoes diarias de poluentes (PM10, SO2, CO, e ozonio) foram obtidas da Companhia de Tecnologia de Saneamento Ambiental (CETESB). Informacoes acerca da temperatura e umidade relativa do ar foram obtidas do Instituto Astronomico e Geofisico da Universidade de Sao Paulo (IAG-USP). As duas classes de modelos produziram resultados coerentes, mas os modelos estatisticamente mais sofisticados tiveram mais poder para detectar efeitos significantes. Foram observadas associacoes entre mortalidade e os niveis de CO, SO2 e, em menor escala, PM10. As associacoes observadas foram dose-dependente e evidentes apos um curto periodo de exposicao.

Effects of flunitrazepam on memory and their reversal by two antagonists

The amnestic effects of flunitrazepam (2 mg intravenously) were studied in normal volunteers with emphasis on their relationship to sleep and their reversal by two specific benzodiazepine receptor antagonists (Ro 15–1788 and Ro 15–3505). The test battery was based on available clinical tests to assess various aspects of encoding and recall. The results suggest that flunitrazepam impairs acquisition of new information by interfering with encoding, and that these effects are clearly independent of sleep. Flunitrazepam effects on memory were fully reversed by both antagonists, as were the subjective and objective signs of sedation. This speaks against the hypothesis of different receptors for sedative and amnestic effects. Ro 15–3505 had shorter lasting effects than Ro 15–1788 and interfered with some tests; this is discussed in relation to its inverse agonistic effects.

Regression Models for the Analysis of Pretest/Posttest Data

SUMMARY The standard repeated measures ANOVA and ANCOVA models for data from pretest/posttest experiments may not be completely adequate either when a null pretest measurement implies that the posttest measurement is also null or when the data are heteroscedastic. We illustrate such a situation with an example in the field of dentistry involving the evaluation of children of both sexes with respect to a dental plaque index observed before and after toothbrushing with two types of toothbrushes. We propose a third alternative based on regression models having the posttoothbrushing index as response and the pre-toothbrushing index as explanatory variable, which may incorporate both the above requirements as well as the repeated measures nature of the data. Using the toothbrush data, we compare the results of the three analyses indicating how they can be implemented computationally.

GENERALIZED LEAST SQUARES METHODS FOR BIVARIATE POISSON REGRESSION

We consider bivariate Poisson regression models to analyse bivariate counts obtained under a stratified sampling scheme. A hybrid maximum likelihood (ML)/generalized least squares (GLS) method is used to obtain estimates of the relevant parameters. The proposed two stage procedure is asymptotically equivalent to and computationally simpler than that based exclusively on maximum likelihood. We compare the results obtained under both methods via a numerical illustration with real data as well as via a simulation study.

Null intercept measurement error regression models

We consider measurement error regression models with null intercepts for the analysis of pretest/posttest repcated measurements data. The within sample units correlation structure is induced by a mixed model adopted for the true values of the covariate. We indicate how maximum likelihood estimators of the regression parameters may be calculated and derive their asymptotic distribution. The proposed procedures are numerically illustrated with data previously analyzed in the literature via classical regression techniques.

Pericytes Extend Survival of ALS SOD1 Mice and Induce the Expression of Antioxidant Enzymes in the Murine Model and in IPSCs Derived Neuronal Cells from an ALS Patient

AbstractAmyotrophic Lateral Sclerosis (ALS) is one of the most common adult-onset motor neuron disease causing a progressive, rapid and irreversible degeneration of motor neurons in the cortex, brain stem and spinal cord. No effective treatment is available and cell therapy clinical trials are currently being tested in ALS affected patients. It is well known that in ALS patients, approximately 50% of pericytes from the spinal cord barrier are lost. In the central nervous system, pericytes act in the formation and maintenance of the blood-brain barrier, a natural defense that slows the progression of symptoms in neurodegenerative diseases. Here we evaluated, for the first time, the therapeutic effect of human pericytes in vivo in SOD1 mice and in vitro in motor neurons and other neuronal cells derived from one ALS patient. Pericytes and mesenchymal stromal cells (MSCs) were derived from the same adipose tissue sample and were administered to SOD1 mice intraperitoneally. The effect of the two treatments was compared. Treatment with pericytes extended significantly animals survival in SOD1 males, but not in females that usually have a milder phenotype with higher survival rates. No significant differences were observed in the survival of mice treated with MSCs. Gene expression analysis in brain and spinal cord of end-stage animals showed that treatment with pericytes can stimulate the host antioxidant system. Additionally, pericytes induced the expression of SOD1 and CAT in motor neurons and other neuronal cells derived from one ALS patient carrying a mutation in FUS. Overall, treatment with pericytes was more effective than treatment with MSCs. Our results encourage further investigations and suggest that pericytes may be a good option for ALS treatment in the future. Graphical Abstractᅟ

Predicting Random Effects From Finite Population Clustered Samples With Response Error

In many situations there is interest in parameters (e. g., mean) associated with the response distribution of individual clusters in a finite clustered population. We develop predictors of such parameters using a two-stage sampling probability model with response error. The probability model stems directly from finite population sampling without additional assumptions and thus is design-based. The predictors are closely related to best linear unbiased predictors (BLUP) that arise from common mixed-model methods, as well as to model-based predictors obtained via super population approaches for survey sampling. The context assumes clusters of equal size and equal size sampling of units within clusters. Target parameters may correspond to clusters realized in the sample, as well as nonrealized clusters. In either case, the predictors are linear and unbiased, and minimize the expected mean squared error. They correspond to the sum of predictors of responses for realized and nonrealized units in the cluster, accounting directly for the second-stage sampling fraction. In contrast, the BLUP commonly used in mixed models can be interpreted as predicting only the responses of second-stage units not observed for a cluster, not the cluster mean. The development reveals that two-stage sampling does not give rise to a more general variance structure often assumed in superpopulation models, even when variances within clusters are heterogeneous. With response error present, we predict target random variables defined as an expected (or average) response over units in a cluster.

Leverage analysis for linear mixed models

We consider a generalized leverage matrix useful for the identification of influential units and observations in linear mixed models and show how a decomposition of this matrix may be employed to identify high leverage points for both the marginal fitted values and the random effect component of the conditional fitted values. We illustrate the different uses of the two components of the decomposition with a simulated example as well as with a real data set.