Preoperative breast MRI and mortality in older women with breast cancer.

PURPOSE: The survival benefit from detecting additional breast cancers by preoperative magnetic resonance imaging (MRI) continues to be controversial. METHODS: We followed a cohort of 4454 women diagnosed with non-metastatic breast cancer (stage I-III) from 2/2005-6/2010 in five registries of the breast cancer surveillance consortium (BCSC). BCSC clinical and registry data were linked to Medicare claims and enrollment data. We estimated the cumulative probability of breast cancer-specific and all-cause mortality. We tested the association of preoperative MRI with all-cause mortality using a Cox proportional hazards model. RESULTS: 917 (20.6%) women underwent preoperative MRI. No significant difference in the cumulative probability of breast cancer-specific mortality was found. We observed no significant difference in the hazard of all-cause mortality during the follow-up period after adjusting for sociodemographic and clinical factors among women with MRI (HR 0.90; 95% CI 0.72-1.12) compared to those without MRI. CONCLUSION: Our findings of no breast cancer-specific or all-cause mortality benefit supplement prior results that indicate a lack of improvement in surgical outcomes associated with use of preoperative MRI. In combination with other reports, the results of this analysis highlight the importance of exploring the benefit of preoperative MRI in patient-reported outcomes such as women's decision quality and confidence levels with decisions involving treatment choices.

Healthcare utilization after liver transplantation is highly variable among both centers and recipients.

The relationship between healthcare utilization before and after liver transplantation (LT), and its association with center characteristics, is incompletely understood. This was a retrospective cohort study of 34 402 adult LTs between 2002 and 2013 using Vizient inpatient claims data linked to the United Network for Organ Sharing (UNOS) database. Multivariable mixed-effects linear regression models evaluated the association between hospitalization 90 days pre-LT and the number of days alive and out of the hospital (DAOH) 1 year post-LT. Of those patients alive at LT discharge, 24.7% spent ≥30 days hospitalized during the first year. Hospitalization in the 90 days pre-LT was inversely associated with DAOH (ß = -3.4 DAOH/week hospitalized pre-LT; P = .002). Centers with >30% of their liver transplant recipients hospitalized ≥30 days in the first LT year were typically smaller volume and/or transplanting higher risk recipients (Model for End-Stage Liver Disease [MELD] score ≥35, inpatient or ventilated pre-LT). In conclusion, pre-LT hospitalization predicts 1-year post-LT hospitalization independent of MELD score at the patient-level, whereas center-specific post-LT healthcare utilization is associated with certain center behaviors and selection practices.

Using semi-Markov processes to study timeliness and tests used in the diagnostic evaluation of suspected breast cancer.

Diagnostic evaluation of suspected breast cancer due to abnormal screening mammography results is common, creates anxiety for women and is costly for the healthcare system. Timely evaluation with minimal use of additional diagnostic testing is key to minimizing anxiety and cost. In this paper, we propose a Bayesian semi-Markov model that allows for flexible, semi-parametric specification of the sojourn time distributions and apply our model to an investigation of the process of diagnostic evaluation with mammography, ultrasound and biopsy following an abnormal screening mammogram. We also investigate risk factors associated with the sojourn time between diagnostic tests. By utilizing semi-Markov processes, we expand on prior work that described the timing of the first test received by providing additional information such as the mean time to resolution and proportion of women with unresolved mammograms after 90 days for women requiring different sequences of tests in order to reach a definitive diagnosis. Overall, we found that older women were more likely to have unresolved positive mammograms after 90 days. Differences in the timing of imaging evaluation and biopsy were generally on the order of days and thus did not represent clinically important differences in diagnostic delay. Copyright © 2016 John Wiley & Sons, Ltd.

A comparison of non-homogeneous Markov regression models with application to Alzheimer's disease progression.

Markov regression models are useful tools for estimating the impact of risk factors on rates of transition between multiple disease states. Alzheimer's disease (AD) is an example of a multi-state disease process in which great interest lies in identifying risk factors for transition. In this context, non-homogeneous models are required because transition rates change as subjects age. In this report we propose a non-homogeneous Markov regression model that allows for reversible and recurrent disease states, transitions among multiple states between observations, and unequally spaced observation times. We conducted simulation studies to demonstrate performance of estimators for covariate effects from this model and compare performance with alternative models when the underlying non-homogeneous process was correctly specified and under model misspecification. In simulation studies, we found that covariate effects were biased if non-homogeneity of the disease process was not accounted for. However, estimates from non-homogeneous models were robust to misspecification of the form of the non-homogeneity. We used our model to estimate risk factors for transition to mild cognitive impairment (MCI) and AD in a longitudinal study of subjects included in the National Alzheimer's Coordinating Center's Uniform Data Set. Using our model, we found that subjects with MCI affecting multiple cognitive domains were significantly less likely to revert to normal cognition.

Modeling nonhomogeneous Markov processes via time transformation.

Longitudinal studies are a powerful tool for characterizing the course of chronic disease. These studies are usually carried out with subjects observed at periodic visits giving rise to panel data. Under this observation scheme the exact times of disease state transitions and sequence of disease states visited are unknown and Markov process models are often used to describe disease progression. Most applications of Markov process models rely on the assumption of time homogeneity, that is, that the transition rates are constant over time. This assumption is not satisfied when transition rates depend on time from the process origin. However, limited statistical tools are available for dealing with nonhomogeneity. We propose models in which the time scale of a nonhomogeneous Markov process is transformed to an operational time scale on which the process is homogeneous. We develop a method for jointly estimating the time transformation and the transition intensity matrix for the time transformed homogeneous process. We assess maximum likelihood estimation using the Fisher scoring algorithm via simulation studies and compare performance of our method to homogeneous and piecewise homogeneous models. We apply our methodology to a study of delirium progression in a cohort of stem cell transplantation recipients and show that our method identifies temporal trends in delirium incidence and recovery.

Detection of Chlamydia trachomatis and Neisseria gonorrhoeae in swab specimens by the Hybrid Capture II and PACE 2 nucleic acid probe tests.

BACKGROUND AND OBJECTIVES: The Digene Hybrid Capture II (HC II) CT/GC Test (Digene Corp., Beltsville, MD) is a new nucleic acid signal amplification-based test for the detection of Chlamydia trachomatis and Neisseria gonorrhoeae in specimens from the genital tract. For optimal results, the HC II CT/GC Test employs a special conical shaped brush for cervical specimen collection from nonpregnant women and swabs from pregnant women. GOALS: To validate a protocol for HC II C. trachomatis and N. gonorrhoeae testing of specimens collected for the GenProbe PACE 2 System. STUDY DESIGN: Specimens were collected from 1,746 patients with a swab and placed in GenProbe transport media according to the manufacturer's recommended procedure. The specimens were first tested at two clinical laboratories by the PACE 2 system, and then blindly tested by HC II CT/GC using an adjusted cutoff value. Discrepant specimens were adjudicated by polymerase chain reaction (PCR), and the result common to two of the three testing methods (HC II, PACE 2, and PCR) was defined as the consensus result. RESULTS: Combining the data from both sites, the relative sensitivity of the HC II Test compared with the consensus result for the detection of 1,761 specimens for C. trachomatis and 1,750 specimens for N. gonorrhoeae was 100% for both organisms. The relative specificities for C. trachomatis and N. gonorrhoeae detection were 99.8% and 99.7%, respectively. The relative sensitivities of the PACE 2 CT and GC Systems were 86.5% and 87.1%, respectively, with relative specificities of 99.9% and 100%. The difference in sensitivity between HC II and PACE 2 for C. trachomatis detection was significant (P < 0.016). CONCLUSION: The HC II CT/GC Test can be performed using specimens collected in GenProbe transport media and has a significantly greater sensitivity for C. trachomatis detection than the PACE 2 System.

T-cell receptors of man and mouse studied with antibodies against synthetic peptides.

We used polyclonal rabbit antibodies directed against synthetic peptides predicted from the gene sequence of the human T-cell receptor (TCR) beta-chain YT35 to study the antigen receptor on human helper T-cell leukemia lines and on normal mouse thymocytes. Antibodies were raised to peptides corresponding to joining segment (J beta) and to a conserved stretch of sequence around the first cysteine in the constant region (C beta). These peptides were selected on the basis of homology with corresponding segments of immunoglobulin light chains. The specificity of the antibodies was established using synthetic overlapping peptides that modelled the complete TCR beta-chain. Western blot analysis was performed against detergent lysates of T cells. Both of the antibodies reacted strongly with 2-3 polypeptides in the mass range 40-45 kDa in mouse and human cells. Clearance experiments using monoclonal antibodies against murine TCR alpha- and beta-chains and against human TCR beta-chain and immunoprecipitations with monoclonal antibody to the murine T3 complex established that these components represented the alpha/beta heterodimer. An additional component around 31 kDa was detected by anti-J beta antibodies in murine thymus extracts. The use of the affinity-purified antipeptide antibody in two-dimensional Western blot analyses allows the clear discrimination between the characteristic individual receptors of monoclonal neoplastic T cells and the polydisperse patterns representative of heterogeneous normal populations. Antigenic cross-reactions between T-cell receptor beta-chains of man and mouse observed with monoclonal antibodies and rabbit antisera to peptides are consistent with the homology in gene sequence between the two species.

A monoclonal antigen-binding T cell immunoprotein: antigenic relatedness to T cell receptor beta chain FR1 V and J peptide segments: physicochemical distinctiveness from classical immunoglobulins and T cell receptor heterodimers.

The monoclonal murine T cell hybridoma, 51H7D, was previously shown to bind the arsazobenzene hapten and to produce a soluble antigen-binding molecule. In this paper we characterize this antigen-binding immunoprotein for its relationship to known T cell receptors serologically, using antibodies specific for variable region framework, or joining region peptides predicted from gene sequence and by biochemical means. The 51H7D cell expresses a protein with subunit size of approximately 31,000, that reacts antigenically with affinity-purified antibodies directed against synthetic first framework and joining segment peptides, corresponding to the gene sequence of the T cell receptor beta chain, YT35. This molecule does not react with affinity-purified antibodies directed against murine immunoglobulin, framework 1 sequences of alpha and gamma T cell receptors, or with antibodies against synthetic heavy chain joining segments. The subunit of mol. wt. 31,000 can form higher aggregates, notably in the mol. wt range of 60,000-70,000, depending upon extraction conditions. The soluble form of the antigen-binding molecule bears the J beta cross-reactive determinant and occurs predominantly as a charge restricted molecular species of approximate mol. wt 60,000-70,000. The purified molecule has a blocked N-terminus, but quantitative statistical analysis of its amino acid composition indicates a closer relatedness to T cell receptor beta chains and other antigen-binding T cell products, than it has to alpha, gamma or delta TCR chains. No evidence for more than one type of polypeptide chain was found and the polymerization is not dependent upon the formation of disulfide bonds. These studies raise the possibility that antigen-binding soluble T cell molecules might belong to a new family of immunoproteins, that is related to, but distinct from, classical immunoglobulins and alpha beta or gamma delta heterodimers.

Antibodies to synthetic peptides corresponding to variable-region first-framework segments of T cell receptor. Detection of T cell products and cross-reactions with classical immunoglobulins.

Recent studies at the gene level have shown that T cells express rearranged genes for four types of T cell receptors that are strongly homologous to classical immunoglobulins in the joining region and in the framework 1 (Fr1) and 3 segments of the variable region. Based upon the homologies in gene sequence, it follows that the gene products would show similarities in amino acid sequence and in the folding of the proteins so that cross-reactivities in antigenic determinants would be expected between variable regions of the T cell receptors and classical immunoglobulins. We have synthesized peptides corresponding to predicted protein sequences of the Fr1 residues of T cell receptor alpha, beta- and gamma-chains and have produced antibodies in rabbits against these synthetic peptides. Use of antisera and affinity-purified antipeptide antibodies indicated that high-titer antibodies could be raised that were specific for individual Fr1 peptides. Cross-reactions among Fr1 peptides of T cell receptors and immunoglobulin light chains were observed. In addition, some rabbit antisera raised against classical polyclonal immunoglobulins or affinity-purified immunoglobulin-like T cell receptors were found to exhibit binding activity against Fr1 peptides of T cell receptor beta- and gamma-chains. The sequence homology, although real among the Fr1 of T cell receptors and immunoglobulin light chains, is moderate and the antigenic cross-reaction must reflect the configuration and types of amino acids present. The development of antipeptide antibodies holds promise for the characterization of T cell receptors of various T cell sources and also offers a new means for the identification of molecules related to rearranging immunoglobulins.

Immunoglobulin epitopes defined by synthetic peptides corresponding to joining region sequence: conservation of determinants and dependence upon the presence of an arginyl or lysyl residue for cross-reaction between light chains and T-cell receptor chains.

Joining or J region sequences of rearranging immunoglobulins and T-cell receptors show considerable sequence homology, particularly in their C-terminal portion corresponding to the fourth framework region of immunoglobulin variable regions. In order to test the question of whether serological cross-reactions between immunoglobulin variable regions and T-cell receptors were due to antigenic similarities in their J regions, we synthesized synthetic peptides corresponding to immunoglobulin J regions and to J regions predicted from gene sequence of the T-cell receptor beta chain. We found that antibodies produced against a synthetic 16-mer J beta sequence reacted with T-cell receptor chains and also with immunoglobulin light chains. The cross-reactivity was dependent upon the J signature sequence FG()GT(R or K)L where the presence of a positively charged lysyl or arginyl residue was essential for cross-reactivity. We were able to classify J region determinants into two distinct antigenic sets; one corresponding to JH and the other corresponding to J kappa, J lambda, J beta and J alpha. Although considerable homology occurs between JH and JL (or J beta) sequences, little cross-reactivity was observed between these two J subsets. Antibodies raised against polyclonal murine IgG immunoglobulins contained antibody subpopulations specifically reactive with either JH or J beta peptides. The serological data derived here using antipeptide antibodies are consistent with computer modeling studies that indicate that the conformations of T-cell receptor variable regions resemble those of classical immunoglobulins. Our data comparing cross-reactivities restricted to the J region indicate that the expression of the J region by intact T-cell receptor beta chains is probably more similar to that of light chains than it is to the corresponding region of heavy chains.

Conservation among vertebrate immunoglobulin chains detected by antibodies to a synthetic joining segment peptide.

We used affinity purified antibodies produced against a synthetic peptide sequence corresponding to the entire J beta of a human T cell receptor gene to screen sera of man, mouse and other vertebrates to determine the presence of cross-reactive molecules. Little evidence for free alpha/beta heterodimers was found, but the antibody reacted with light chains of many vertebrate species, including characterized myeloma proteins of man and mouse. Some vertebrate orders, notably Aves, lacked polypeptide chains cross-reactive with J beta, but detectable determinants occurred in primitive vertebrates such as the galapagos shark (Carcharhinus galapagensis). In addition to the strong cross-reaction with purified light chains, human heavy chains reacted weakly with the antibody. The cross-reaction correlated with the sequence of the denatured immunoglobulins and was inhibitable with free peptide. These results establish the similarity of T cell receptor beta chains to immunoglobulin chains and support the conclusion that J region sequences were conserved, not only within mammalian immunoglobulins and T cell receptors, but in vertebrate evolution.

Antibody-dependent cell-mediated cytotoxicity by human lymphocytes: a regulatory role for the sheep red blood cell receptor.

Antibody-dependent cell-mediated cytotoxicity (ADCC) is a cytolytic mechanism whereby unimmunized lymphoid cells lyse antibody-sensitized target cells. The cells mediating ADCC are referred to functionally as killer cells (K cells). K cells are a heterogeneous population of lymphocytes based on cell membrane marker criteria but may be morphologically homogeneous. It is well established that K cells must have an Fc receptor (FcR) appropriate for the class of target cell-bound antibody, and that physical interaction between the K cell's FcR and the Fc portion of target cell-bound antibody is required for the initiation of lysis. Relatively little is known about the regulatory mechanisms presumed to exist and modulate ADCC reactions. We have made 2 observations that led us to investigate the possibility that the sheep erythrocyte (SRBC) receptor (ER) on human cells might be involved as a regulator of ADCC. First, SRBC are more efficiently lysed by human lymphocytes than are other erythrocyte target cells, and second, although nonsensitized bystander cells are usually not lysed during ADCC reactions, SRBC are consistently lysed. Removal of E-rosetting capacity of lymphocytes by trypsin treatment selectively decreases lysis of IgG sensitized SRBC. Furthermore, lysis of nonsensitized bystander SRBC is also inhibited by trypsin treatment of effector cells. ADCC is also decreased by blocking the ER with simple sugars or monoclonal antibody so that E-rosettes are inhibited. In contrast to trypsin treatment, which selectively inhibits lysis of SRBC, ER-specific monoclonal antibody and simple sugars decrease cytolysis against all target cells tested.

Nonspecific human lymphocyte-mediated cytotoxicity induced by immobilized IgG aggregate: regulation by suppressor cells.

Normal human lymphocytes lyse a variety of erythrocytes when effector cells and target cells are incubated on immobilized, normal human IgG aggregates (10). Lymphocyte-mediated cytolysis induced by immobilized aggregate requires an overnight incubation with adherent mononuclear cells. This study was carried out to determine whether preincubation with adherent cells effects the activation of a cytolytic cell or inactivation of a suppressor cell. It was determined that a suppressor cell function is inactivated, and the suppressor cells were characterized as T gamma and T mu lymphocytes.

Antibody-dependent cell-mediated cytotoxicity by human lymphocytes. II. Comparison with polymorphonuclear leukocytes against erythrocyte and nucleated target cells.

Human peripheral blood lymphocytes were compared to neutrophils with respect to their cytotoxic activity against IgG-sensitized erythrocytes (ox; ORBC) and nucleated (Raji) target cells in vitro. Overall, neutrophils were more efficient than lymphocytes in lysing erythrocytes, but lymphocytes were more efficient in lysing Raji cells. The antibody concentrations producing maximal levels of lysis against erythrocyte target cells were similar for both kinds of effector cells and guinea pig complement; however, in order to obtain Raji cell lysis mediated by neutrophils at a level comparable to that produced by lymphocytes a higher concentration of antibody was required. Another difference found between the lytic activity of these effector cell populations was kinetics of lysis against erythrocyte target: a 15 min lag phase was consistently observed only with neutrophil effector cells. Both effector cells lyse Raji cells with similar kinetics and no lag phase. Lysis by lymphocytes and neutrophils was directly proportional to the effector:target cell (E:T) ratio against both target cells. The effector cells were similarly inhibited from lysis of erythrocytes by IgG aggregate, but neutrophil lysis of Raji cells appeared to be more sensitive to inhibition. The neutrophils did not lyse nonsensitized bystander cells, and lymphocytes lysed sheep erythrocyte bystander cells only.

Correlation if circulating immune complexes and disease status in patients with leukaemia.

The occurrence of soluble immune complexes (IC) was investigated in 177 serum samples from 92 patients with various leukaemias using the Raji cell immunoassay. In general, patients with myeloproliferative diseases had a higher incidence and higher quantities of IC than did patients with lymphoproliferative disorders. Elevated levels of IC were found in the sera of patients as follows: 17% with chronic lymphocytic leukaemia (mean value of 13.1 microgram/ml), 67% with acute lymphocytic leukaemia (54.1 microgram/ml), 65% with chronic myelocytic leukaemia (86.7 microgram/ml), 70% with acute myelocytic leukaemia (202.5 microgram/ml) and 56% with acute myelomonocytic leukaemia (41.9 microgram/ml). Patients in terminal blastic crisis of chronic myelocytic leukaemia had the highest levels, with a mean level of 1,364.1 microgram/ml. Serial samples were obtained, as available, from individual patients during the course of the disease in an attempt to relate severity with the incidence and quantity of IC. No significant correlation could be made between the occurrence or levels of IC and the presence of absence of systemic symptoms. Similarly, no correlations could be made between levels of IC and haematological parameters, infection, or therapy. However, the data does indicate a positive relationship between the levels of IC and the progressive state of the leukaemia, especially, the myelocytic leukaemias.

Antibody--antigen complex stimulated lysis of non-sensitized sheep red cells by human lymphocytes. I. Requirements for IgG complexes.

IgG antibody--antigen complexes stimulated lysis of non-sensitized sheep erythrocytes (SRBC) by normal human peripheral blood lymphocytes (PBL). Heat-aggregated human IgG, rabbit IgG-ovalbumin complexes and rabbit IgG-sensitized ox erythrocytes (ORBC) were effective in the induction of SRBC lysis by PBL. However, IgM-sensitized ORBC and IgM-complement-sensitized ORBC were ineffective. As only SRBC and not ORBC or chicken erythrocytes (CRBC) were lysed under identical experimental conditions, it is conceivable that the SRBC receptor present on the T cell is involved. Furthermore, 45% inhibition of lysis was obtained by pretreating the effector cells with anti-human thymocyte globulin (ATG) and complete inhibition was obtained by adding SRBC stroma to the reaction mixture. The requirement for the inclusion of IgG complexes and the absence of specific anti-target cell antibody distinguish this reaction from natural cell-mediated cytotoxicity and antibody-dependent cell-mediated cytotoxicity (ADCC). Immune killer T cells would not appear to be responsible as eight different donors were used and none of these were cytotoxic to SRBC in the absence of IgG complexes. The induction of this cytotoxic reaction appears to require the recognition and interaction by the effector cells of two separate molecular entities, i.e. the SRBC membrane by the T cell and the IgG Fc region by an IgG-Fc receptor-bearing cell.