Dean Follmann

VARIANCE IMPUTATION FOR OVERVIEWS OF CLINICAL TRIALS WITH CONTINUOUS RESPONSE

Overviews of clinical trials are an efficient and important means of summarizing information about a particular scientific area. When the outcome is a continuous variable, both treatment effect and variance estimates are required to construct a confidence interval for the overall treatment effect. Often, only partial information about the variance is provided in the publication of the clinical trial. This paper provides heuristic suggestions for variance imputation based on partial variance information. Both pretest-posttest (parallel groups) and crossover designs are considered. A key idea is to use separate sources of incomplete information to help choose a better variance estimate. The imputation suggestions are illustrated with a data set.

Correlation between immunologic responses to a recombinant glycoprotein 120 vaccine and incidence of HIV-1 infection in a phase 3 HIV-1 preventive vaccine trial

BACKGROUND An objective of the first efficacy trial of a candidate vaccine containing recombinant human immunodeficiency virus (HIV) type 1 envelope glycoprotein 120 (rgp120) antigens was to assess correlations between antibody responses to rgp120 and the incidence of HIV-1 infection. METHODS Within the randomized trial (for vaccinees, n=3598; for placebo recipients, n=1805), binding and neutralizing antibody responses to rgp120 were quantitated. A case-cohort design was used to study correlations between antibody levels and HIV-1 incidence. RESULTS Peak antibody levels were significantly inversely correlated with HIV-1 incidence. The relative risk (RR) of infection was 0.63 (95% confidence interval, 0.45-0.89) per log(10) higher neutralization titer against HIV-1(MN), and the RRs of infection for second-, third-, and fourth-quartile responses of antibody blocking of gp120 binding to soluble CD4 versus first-quartile responses (the lowest responses) were 0.35, 0.28, and 0.22, respectively. CONCLUSIONS Despite inducing a complex, robust immune response, the vaccine was unable to reduce the incidence of HIV-1. Two interpretations of the correlative results are that the levels of antibodies (i) caused both an increased (low responders) and decreased (high responders) risk of HIV-1 acquisition or (ii) represented a correlate of susceptibility to HIV-1 but had no causal effect on susceptibility. Although the data cannot definitively discriminate between these 2 explanations, (ii) appears to be more likely.

An approximate generalized linear model with random effects for informative missing data.

This paper develops a class of models to deal with missing data from longitudinal studies. We assume that separate models for the primary response and missingness (e.g., number of missed visits) are linked by a common random parameter. Such models have been developed in the econometrics (Heckman, 1979, Econometrica 47, 153-161) and biostatistics (Wu and Carroll, 1988, Biometrics 44, 175-188) literature for a Gaussian primary response. We allow the primary response, conditional on the random parameter, to follow a generalized linear model and approximate the generalized linear model by conditioning on the data that describes missingness. The resultant approximation is a mixed generalized linear model with possibly heterogeneous random effects. An example is given to illustrate the approximate approach, and simulations are performed to critique the adequacy of the approximation for repeated binary data.

Characterization of the defective interaction between a subset of natural killer cells and dendritic cells in HIV-1 infection

In this study, we demonstrate that the in vitro interactions between a CD56neg/CD16pos (CD56neg) subset of natural killer (NK) cells and autologous dendritic cells (DCs) from HIV-1–infected viremic but not aviremic individuals are markedly impaired and likely interfere with the development of an effective immune response. Among the defective interactions are abnormalities in the process of reciprocal NK–DC activation and maturation as well as a defect in the NK cell–mediated editing or elimination of immature DCs (iDCs). Notably, the lysis of mature DCs (mDCs) by autologous NK cells was highly impaired even after the complete masking of major histocompatibility complex I molecules, suggesting that the defective elimination of autologous iDCs is at the level of activating NK cell receptors. In this regard, the markedly impaired expression/secretion and function of NKp30 and TNF-related apoptosis-inducing ligand, particularly among the CD56neg NK cell subset, largely accounts for the highly defective NK cell–mediated lysis of autologous iDCs. Moreover, mDCs generated from HIV-1 viremic but not aviremic patients are substantially impaired in their ability to secrete interleukin (IL)-10 and -12 and to prime the proliferation of neighboring autologous NK cells, which, in turn, fail to secrete adequate amounts of interferon-γ.

Genetic Variation in OAS1 Is a Risk Factor for Initial Infection with West Nile Virus in Man

West Nile virus (WNV) is a re-emerging pathogen that can cause fatal encephalitis. In mice, susceptibility to WNV has been reported to result from a single point mutation in oas1b, which encodes 2′–5′ oligoadenylate synthetase 1b, a member of the type I interferon-regulated OAS gene family involved in viral RNA degradation. In man, the human ortholog of oas1b appears to be OAS1. The ‘A’ allele at SNP rs10774671 of OAS1 has previously been shown to alter splicing of OAS1 and to be associated with reduced OAS activity in PBMCs. Here we show that the frequency of this hypofunctional allele is increased in both symptomatic and asymptomatic WNV seroconverters (Caucasians from five US centers; total n = 501; OR = 1.6 [95% CI 1.2–2.0], P = 0.0002 in a recessive genetic model). We then directly tested the effect of this SNP on viral replication in a novel ex vivo model of WNV infection in primary human lymphoid tissue. Virus accumulation varied markedly among donors, and was highest for individuals homozygous for the ‘A’ allele (P<0.0001). Together, these data identify OAS1 SNP rs10774671 as a host genetic risk factor for initial infection with WNV in humans.

Compromised B cell responses to influenza vaccination in HIV-infected individuals.

BACKGROUND Yearly influenza vaccination, although recommended for human immunodeficiency virus (HIV)-infected individuals, has not received thorough evaluation in the era of antiretroviral therapy. We assessed the impact of HIV disease on B cell responses to influenza vaccination. METHODS Sixty-four HIV-infected and 17 HIV-negative individuals received the 2003-2004 trivalent inactivated influenza vaccine. Frequencies of influenza-specific antibody-secreting cells (ASCs) were measured by enzyme-linked immunospot (ELISPOT) assay, and antibody responses were measured by hemagglutination-inhibition (HI) assay. Memory responses to influenza were measured by ELISPOT assay after polyclonal activation of B cells in vitro. RESULTS Prevaccination HI titers were significantly higher in HIV-negative than in HIV-infected individuals. Peak HI titers and influenza-specific ASC frequencies were directly correlated with CD4+ T cell counts in HIV-infected individuals. Influenza-specific memory B cell responses were significantly lower in HIV-infected than in HIV-negative individuals and were directly correlated with CD4+ T cell counts. CONCLUSIONS HIV infection is associated with a weak antibody response to influenza vaccination that is compounded by a poor memory B cell response. CD4+ T cell count is a critical determinant of responsiveness to influenza vaccination, and the contribution of plasma HIV RNA level is suggestive and warrants further investigation.

Lysis of endogenously infected CD4+ T cell blasts by rIL-2 activated autologous natural killer cells from HIV-infected viremic individuals.

Understanding the cellular mechanisms that ensure an appropriate innate immune response against viral pathogens is an important challenge of biomedical research. In vitro studies have shown that natural killer (NK) cells purified from healthy donors can kill heterologous cell lines or autologous CD4+ T cell blasts exogenously infected with several strains of HIV-1. However, it is not known whether the deleterious effects of high HIV-1 viremia interferes with the NK cell-mediated cytolysis of autologous, endogenously HIV-1-infected CD4+ T cells. Here, we stimulate primary CD4+ T cells, purified ex vivo from HIV-1-infected viremic patients, with PHA and rIL2 (with or without rIL-7). This experimental procedure allows for the significant expansion and isolation of endogenously infected CD4+ T cell blasts detected by intracellular staining of p24 HIV-1 core antigen. We show that, subsequent to the selective down-modulation of MHC class-I (MHC-I) molecules, HIV-1-infected p24pos blasts become partially susceptible to lysis by rIL-2-activated NK cells, while uninfected p24neg blasts are spared from killing. This NK cell-mediated killing occurs mainly through the NKG2D activation pathway. However, the degree of NK cell cytolytic activity against autologous, endogenously HIV-1-infected CD4+ T cell blasts that down-modulate HLA-A and –B alleles and against heterologous MHC-Ineg cell lines is particularly low. This phenomenon is associated with the defective surface expression and engagement of natural cytotoxicity receptors (NCRs) and with the high frequency of the anergic CD56neg/CD16pos subsets of highly dysfunctional NK cells from HIV-1-infected viremic patients. Collectively, our data demonstrate that the chronic viral replication of HIV-1 in infected individuals results in several phenotypic and functional aberrancies that interfere with the NK cell-mediated killing of autologous p24pos blasts derived from primary T cells.

High Dose Atorvastatin Decreases Cellular Markers of Immune Activation without Affecting HIV-1 RNA Levels: Results of a Double-Blind Randomized Placebo Controlled Clinical Trial

BACKGROUND 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitors (statins) exhibit antiviral activity against human immunodeficiency virus type 1 (HIV-1) in vitro and may modulate the immune response to HIV infection. Studies evaluating the antiviral activity of statins have yielded conflicting results. METHODS We conducted a randomized, double-blind, placebo-controlled crossover trial to investigate the effect of atorvastatin on HIV-1 RNA (primary objective) and cellular markers of immune activation (secondary objective). HIV-infected individuals not receiving antiretroviral therapy were randomized to receive either 8 weeks of atorvastatin (80 mg) or placebo daily. After a 4-6 week washout phase, participants switched treatment assignments. The study had 80% power to detect a 0.3 log(10) decrease in HIV-1 RNA level. Expression of CD38 and HLA-DR on CD4(+) and CD8(+) T cells was used to measure immune activation. RESULTS Of 24 randomized participants, 22 completed the study. Although HIV-1 RNA level was unaffected by the intervention (-0.13 log(10) copies/mL; P = .85), atorvastatin use resulted in reductions in circulating proportions of CD4(+) HLA-DR(+) (-2.5%; P = .02), CD8(+) HLA-DR(+) (-5%; P = .006), and CD8(+) HLA-DR(+) CD38(+) T cells (-3%; P = .03). Reductions in immune activation did not correlate with declines in serum levels of low-density lipoprotein cholesterol. CONCLUSIONS Short-term use of atorvastatin was associated with modest but statistically significant reductions in the proportion of activated T lymphocytes.

Generalizing Logistic Regression by Nonparametric Mixing

Abstract Logistic regression is a common technique for analyzing the effect of a covariate vector x on the number of successes y in m trials when y has a binomial distribution. But at times either the logistic curve does not describe the probability of success p(x) adequately, or m is larger than 1 and y is more variable than the binomial distribution allows. Overdispersion relative to the binomial distribution is possible if the m trials in a set or “litter” are positively correlated, an important covariate is omitted, or x is measured with error. A simple way to accommodate departures from the logit link and overdispersion is to introduce a random intercept α and thus permit a random propensity toward success. When α varies between individual binary trials according to a discrete or multimodal distribution, p(x) has smooth steps and y has a binomial(m, p(x)) distribution. When the random α is constant for a set of m binary trials and varies between sets of m trials according to a discrete or multimodal ...

HIV infection-associated immune activation occurs by two distinct pathways that differentially affect CD4 and CD8 T cells

HIV infection is characterized by a brisk immune activation that plays an important role in the CD4 depletion and immune dysfunction of patients with AIDS. The mechanism underlying this activation is poorly understood. In the current study, we tested the hypothesis that this activation is the net product of two distinct pathways: the inflammatory response to HIV infection and the homeostatic response to CD4 T cell depletion. Using ex vivo BrdU incorporation of PBMCs from 284 patients with different stages of HIV infection, we found that CD4 proliferation was better predicted by the combination of CD4 depletion and HIV viral load (R2 = 0.375, P < 0.001) than by either parameter alone (CD4 T cell counts, R2 = 0.202, P < 0.001; HIV viremia, R2 = 0.302, P < 0.001). Interestingly, CD8 T cell proliferation could be predicted by HIV RNA levels alone (R2 = 0.334, P < 0.001) and this predictive value increased only slightly (R2 = 0.346, P < 0.001) when CD4 T cell depletion was taken into account. Consistent with the hypothesis that CD4 T cell proliferation is driven by IL-7 as a homeostatic response to CD4 T cell depletion, levels of phosphorylated STAT-5 were found to be elevated in naive subsets of CD4 and CD8 T cells from patients with HIV infection and in the central memory subset of CD4 T cells. Taken together these data demonstrate that at least two different pathways lead to immune activation of T cells in patients with HIV infection and these pathways differentially influence CD4 and CD8 T cell subsets.