Modeling nonstationary longitudinal data.

An important theme of longitudinal data analysis in the past two decades has been the development and use of explicit parametric models for the data's variance-covariance structure. A variety of these models have been proposed, of which most are second-order stationary. A few are flexible enough to accommodate nonstationarity, i.e., nonconstant variances and/or correlations that are not a function solely of elapsed time between measurements. We review five nonstationary models that we regard as most useful: (1) the unstructured covariance model, (2) unstructured antedependence models, (3) structured antedependence models, (4) autoregressive integrated moving average and similar models, and (5) random coefficients models. We evaluate the relative strengths and limitations of each model, emphasizing when it is inappropriate or unlikely to be useful. We present three examples to illustrate the fitting and comparison of the models and to demonstrate that nonstationary longitudinal data can be modeled effectively and, in some cases, quite parsimoniously. In these examples, the antedependence models generally prove to be superior and the random coefficients models prove to be inferior. We conclude that antedependence models should be given much greater consideration than they have historically received.

Sex inequality in kidney transplantation rates.

BACKGROUND: Men in the United States undergoing renal replacement therapy are more likely than women to receive a kidney transplant. However, the ability to pay may, in part, be responsible for this finding. OBJECTIVE: To compare adult male and female transplantation rates in a setting in which equal access to medical treatment is assumed. METHODS: Using data from the Canadian Organ Replacement Register, the rate of first transplantations was computed for the 20, 131 men and the 13,458 women aged 20 years or older who initiated renal replacement therapy between January 1, 1981, and December 31, 1996. Poisson regression analysis was used to estimate the male-female transplantation rate ratio, adjusting for age, race, province, calendar period, underlying disease leading to renal failure, and dialytic modality. Actuarial survival methods were used to compare transplantation probability for covariable-matched cohorts of men and women. RESULTS: Men experienced 20% greater covariable-adjusted kidney transplantation rates relative to women (rate ratio, 1.20; 95% confidence interval, 1.13-1.27). The sex disparity was stronger for cadaveric transplants (rate ratio, 1.23) compared with those from living donors (rate ratio, 1.10). The 5-year probability of receiving a transplant was 47% for men and 39% for women within covariable-matched cohorts (P<.001). The sex disparity in transplantation rates increased with increasing age. The sex effect was weaker among whites and Oriental persons (Chinese, Japanese, Vietnamese, Cambodian, Laotian, Filipino, Malaysian, Indonesian, and Korean) and stronger among blacks, Asian Indians (Indian, Pakistani, and Sri Lankan), and North American Indians (aboriginal). CONCLUSION: Since survival probability and quality of life are superior for patients who undergo transplantation relative to those who undergo dialysis, an increased effort should be made to distribute kidneys available for transplantation more equitably by sex among patients undergoing renal replacement therapy.

Geographically masking health data to preserve confidentiality.

The conventional approach to preserving the confidentiality of health records aggregates all records within a geographical area that has a population large enough to ensure prevention of disclosure. Though this approach normally protects the privacy of individuals, the use of such aggregated data limits the types of research one can conduct and makes it impossible to address many important health problems. In this paper we discuss the design and implementation of geographical masks that not only preserve the security of individual health records, but also support the investigation of questions that can be answered only with some knowledge about the location of health events. We describe several alternative methods of masking individual-level data, evaluate their performance, and discuss both the degree to which we can analyse masked data validly as well as the relative security of each approach, should anyone attempt to recover the identity of an individual from the masked data. We conclude that the geographical masks we describe, when appropriately used, protect the confidentiality of health records while permitting many important geographically-based analyses, but that further research is needed to determine how the power of tests for clustering or the strength of other associative relationships are adversely affected by the characteristics of different masks.

Evidence for peptide transport across microsomal membranes.

Antigenic peptides bound to class I molecules of the major histocompatibility complex (MHC) are recognized by T-cell receptors during development of an antiviral immune response. T cells respond to peptides derived from cytoplasmic viral proteins as well as viral membrane proteins, indicating that a pathway exists for the transport of proteins or peptides from the cytosol into the compartment(s) where the MHC class I molecules assemble. To investigate this pathway, we have developed an in vitro assay for the transport of peptides into microsomal vesicles. This assay provides evidence for the transport of chemically synthesized peptides (13-21 amino acids) containing N-linked glycosylation acceptor sequences, which serve as glycosylation substrates. Their transport results in depletion of the pool of available dolichol high-mannose oligosaccharides in the lumen of the microsomal vesicles. We have observed transport of peptides derived from antigenic human immunodeficiency virus gag and influenza B nucleoprotein sequences, but transport of a third randomly selected peptide was not detected, suggesting specificity of the transport process. We were not able to demonstrate ATP dependence of this peptide transport process by using apyrase and an ATPase inhibitor. This result was unexpected in light of the recent identification of MHC-linked genes with homology to ATP-binding cassette transporters, which have been proposed to mediate peptide transport.

An ATP-binding membrane protein is required for protein translocation across the endoplasmic reticulum membrane.

The role of nucleotides in providing energy for polypeptide transfer across the endoplasmic reticulum (ER) membrane is still unknown. To address this question, we treated ER-derived mammalian microsomal vesicles with a photoactivatable analogue of ATP, 8-N3ATP. This treatment resulted in a progressive inhibition of translocation activity. Approximately 20 microsomal membrane proteins were labeled by [alpha 32P]8-N3ATP. Two of these were identified as proteins with putative roles in translocation, alpha signal sequence receptor (SSR), the 35-kDa subunit of the signal sequence receptor complex, and ER-p180, a putative ribosome receptor. We found that there was a positive correlation between inactivation of translocation activity and photolabeling of alpha SSR. In contrast, our data demonstrate that the ATP-binding domain of ER-p180 is dispensable for translocation activity and does not contribute to the observed 8-N3ATP sensitivity of the microsomal vesicles.

Characterization of the rough endoplasmic reticulum ribosome-binding activity.

The rough endoplasmic reticulum membranes of mammalian cells contain specific ribosome-binding sites. A purification to apparent homogeneity of a negatively charged protein (ERp180) of relative molecular mass 180,000 (180 K) was reported which was proposed to function as a rough endoplasmic reticulum ribosome receptor. We report here that ribosome-binding site activity quantitatively solubilized from rough endoplasmic reticulum membranes does not cofractionate with ERp180. By contrast, ribosome-binding site activity fractionates as a much smaller, positively charged protein.

Antitumor activity of the monoclonal antibody-Vinca alkaloid immunoconjugate LY203725 (KS1/4-4-desacetylvinblastine-3-carboxhydrazide) in a nude mouse model of human ovarian cancer.

The activity of the conjugate of monoclonal antibody KS1/4 with 4-desacetylvinblastine-3-carboxhydrazide (KS1/4-DAVLB-HY) was explored in the OVCAR-3 human ovarian xenograft tumor model. Multiple schedules of KS1/4-DAVLB-HY administration were employed, including a comparison of i.p. and i.v. routes of treatment. When inoculates of 6 x 10(7) OVCAR-3 cells were injected i.p. into female athymic nude mice, untreated control animals had a mean survival of 18-34 days, with the development of massive ascites and large intraabdominal tumors. Significant increases in survival were observed in KS1/4-DAVLB-HY conjugate-treated animals with all schedules utilized. Parallel therapy with equivalent doses of free DAVLB-HY or a non-antigen-binding immunoconjugate did not significantly increase the survival of the animals. These data demonstrate that the immunoconjugate KS1/4-DAVLB-HY significantly increases the survival of OVCAR-3 tumor-bearing mice and indicates that this immunoconjugate may be useful in the treatment of human ovarian cancer.

Reconstitution of protein translocation activity from partially solubilized microsomal vesicles.

We have used a reconstitution assay to demonstrate that protein translocation activity can be recovered after microsomal vesicles derived from the rough endoplasmic reticulum have been partially solubilized with n-octyl-beta-glucopyranoside. Two independent approaches were used to establish conditions for partially solubilizing microsomal membranes. When the lipid bilayer was disrupted by detergents to the extent that the integrity of the lipid bilayer had been perturbed, membranes were inactive for translocation. However, detergent-treated membranes could be reconstituted in good yield into a translocation competent form once the detergent was removed.

Induction of surface exclusion (entry exclusion) by Streptococcus faecalis sex pheromones: use of monoclonal antibodies to identify an inducible surface antigen involved in the exclusion process.

The Streptococcus faecalis plasmid pCF-10 is representative of a class of plasmids that enables its host cells to respond to sex pheromones produced by other S. faecalis cells. The pheromone response has been previously shown to result in increased conjugal plasmid transfer, cell clumping, and multiple cell-surface antigenic changes. To test for other effects of pheromone induction, cells carrying pCF-10 were used as recipients in matings with an isogenic donor strain carrying a derivative of pCF-10, tagged with a transposon to provide an additional selective marker. Pheromone induction of the "male recipients" decreased their recipient ability by a factor of 10-300 in comparison to uninduced cells or plasmid-free recipients. These results indicate that an entry exclusion (surface exclusion) function, similar to that described in studies of plasmids in Gram-negative bacteria, is induced during the S. faecalis pheromone response process. The exclusion operates only against homologous plasmids. Immunological, biochemical, and genetic experiments using monoclonal antibodies reactive with C130, the predominant protein antigen associated with the pheromone response of cells carrying pCF-10, indicate that this antigen is involved in surface exclusion. The data also support the notion that synthesis of C130 involves a posttranslational modification of a precursor of C130 to a final product of higher molecular weight form.