Faculty Bibliography

Introduction The lung is a common target of HIV-associated morbidity, but there are few studies of long-term pulmonary consequences of HIV infection after successful introduction of antiretroviral drugs (ARVs), particularly in developing countries with high rates of tuberculosis. Methods: A cross-sectional analysis of a prospective cohort of 152 participants with HIV infection, stable on antiretroviral therapy for at least 3 months, and without features of acute respiratory disease, was performed. Clinical and demographic information was recorded in a respiratory questionnaire, and pulmonary function tests including flow-volume loops before and after bronchodilator, body plethysmography and measurement of diffusing capacity were performed. Results: The participants were predominantly black African (81.3%) and female (53.4%); mean age was 38.4 (SD+/-8.4) years, and 32.9% had a history of tobacco smoking. The median duration of time on ART was 2.1 years (range 0.25-7.56), and the median CD₄ was 380 (range 25-1227). Forty-one percent reported a history of previous tuberculosis, and only 3 subjects previous Pneumocystis pneumonia. None were users of intravenous drugs. Respiratory symptoms were reported in 31% of participants, and any respiratory symptom (cough, phlegm, wheeze or dyspnea) was strongly associated with current smoking (OR 2.556, 95% CI 1.05-6.211, P=0.04). Irreversible airflow obstruction was present in 7%, and associated with ever smoking (OR 6.352, 95% CI 1.56-25.90, P=0.01). Impairment of diffusing capacity for carbon monoxide (DL_CO <70% predicted) was present in 57%, and associated with a history of previous tuberculosis (OR 2.01, 95% CI 1.00-4.00, P=0.47), but not with current or ever smoking. Conclusions: In this relatively young cohort of treated AIDS patients, a high proportion of whom were female, evidence of smoking-induced symptoms and airflow limitation was observed. The most common lung function abnormality, however, was an impairment in diffusing capacity, which was associated with previous tuberculosis. Further research to define structure/function relationships and the role of opportunistic infections and other risk factors is underway

Introduction The NYU Lung Cancer Biomarker Center recruited 1054 smokers (<=20 pack-years) between 03/2000 and 06/2010 and screened them with CT-scans and respiratory questionnaires. Questionnaires contained demographic characteristics, occupational exposures, lifestyle information including smoking history and alcohol use, personal and family medical history, and pulmonary symptoms. CT-scans showed the disease status and emphysema. We hypothesized that information collected in the questionnaires would be associated with the subject's disease status. Methods The 1054 Subjects were classified into 4 groups based on the results of enrollment CT-scans. This categorization showed 331 subjects with solid nodule(s) only, 95 with ground glass opacity (GGO) only, 126 with both solid nodule(s) and GGO(s), and the remaining 502 subjects with no nodules. Chi-squared tests were used to assess the relationship between disease category and each single predictor. Polytomous logistic regression was used to assess the association between disease category and multiple predictors. Forward/backward stepwise regression was used in model selection. Results Univariate analysis showed that gender (CMH =8.7092, p-value=0.0032), age (CMH=29.2219, p-value<0.0001), marijuana smoking (CMH=10.9024, p-value=0.0010), emphysema (CMH=9.0929, p-value=0.0026) and years of smoking (CMH=15.4712, p-value<0.0001) were significantly associated with disease categories. Further tests showed that years of smoking was highly associated with age, and did not contribute to prediction of disease category when age was included in the model, as shown by our multivariate analysis. In multivariate analysis, increasing age, gender (male) and history of emphysema were significant predictors of worsening disease class. Moreover, when we stratified by gender, the analysis indicated that in males, older men were more likely to develop solid nodules (OR=1.275, 95% CI [1.0063, 1.0490]) and more likely to have mixed nodules (OR= 1.0664, 95% CI= [1.0331, 1.1008]) than younger men. Males with emphysema were far more likely to have solid nodules compared with those without emphysema (OR= 1.9678, 95% CI= [1.3128, 2.9495]). Our analysis also revealed that in females, age was the only significant risk factor: older women were more likely to have mixed nodules than younger women (OR= 1.0593, 95% CI= [1.0274, 1.0922]). It is notable that all smoking related variables were not significantly associated with nodule class in our multivariable analysis. Conclusion Our analysis suggests that age, gender and emphysema play a critical role in the development of disease (solid nodule or GGO). Also, our analysis has shown that smoking is not likely to play a major role in the development of the disease in this high risk cohort

Introduction: Interferon gamma release assays (IGRA) have been widely accepted to replace the tuberculin skin test (TST). Data from the Botswana isoniazid preventative therapy (IPT) study showed that in HIV infected individuals only those who were TST positive benefitted from IPT (92% reduction in TB incidence.) We set out to compare the utility of IGRA's (QuantiFERON Gold-in-tube (QFT-GIT) and TSPOT-TB (TSPOT)) and the TST in a high TB-HIV burden setting. Methods: TST (2TU PPD), QFT-GIT and TSPOT-TB where performed in a prospective cohort of 200 HIV positive individuals on ARV's in Cape Town South Africa. Multivariate analysis was performed to assess associations with clinical, demographic and HIV associated (ARV usage, CD4 count, viral load) factors. Results: The participants had a mean (SD) age of 38.4(8.4) years, were predominantly black African (81.3%), female (53.4%) and 32.9% were ever smokers. The median (range) length of time on ARV's 2.1(0.25-7.56) years with median (range) CD4 count of 380(25-1227). Twenty eight (18.6%) participants were TST positive (>5mm), 45(30%) TSPOT-TB positive and 62(41.3%) QFT positive. The agreement between tests was generally fair/poor: QFT-TSPOT: 72%; kappa 0.39(95%CI 0.23-0.55), QFT-TST: 68%; kappa 0.27(95%CI 0.10- 0.44), TSPOT-TST 67%; kappa 0.13(95%CI -0.08 - 0.33). Using the TST as the gold standard, QFT had a sensitivity of 81.2%, specificity of 65.6% and TSPOT 47.6% & 70.8% respectively. LTBI test status was not different in those with/without a history of previous TB or history of smoking. Conclusions: IGRA's only correctly identified 2/3 of TST positive persons. If used alone, potentially a third of HIV infected individuals will receive unnecessary IPT. Longitudinal follow up studies are required to define the utility of IGRAs in the treatment of LTBI in HIV infected persons

Background: In the 7 years after the WTC disaster, the annual number of FDNY retirements attributable to pulmonary disability increased nearly 4-fold. FDNY rescue workers evaluated for pulmonary disability at our subspecialty center were previously found to have an obstructed phenotype based on FEV₁/FVC, Methacholine Challenge Test (MCT) reactivity, or bronchodilator (BD) reactivity (59% of N=1720). In this pilot study, we examined the long-term efficacy of treatment and its effect on disease progression in a subgroup (N=450) of individuals with subspecialty pulmonary testing who were referred for disability evaluation before and after a course of treatment. Methods: Our study cohort includes individuals that were symptomatic, had subspecialty pulmonary testing (pre-treatment), and were referred for disability evaluation after a full course of treatment (post-treatment). Disability evaluations were performed 30 days after weaning from inhaled medication. Individuals were excluded if their post-treatment PFT was less than three months after starting treatment. Change in lung function over time and post treatment were analyzed (N=415). Values are presented as medians with IQR. We defined statistical significance as P<0.05 using the paired t-test. Analyses were performed using GraphPad and SPSS software. Results: Individuals evaluated by MCT presented for disability evaluation earlier, with a shorter interval of treatment (7.5 months (5-10)) compared to those that received spirometry with BD, (14.5 months (8-25) p<0.0001). FEV₁ %Pred and FEV₁/FVC declined significantly; p<0.0001. FEV₁ observed declined at a slower rate in the treatment group than in the entire WTC-exposed FDNY cohort, including those not referred for subspecialty evaluation: 15cc/year vs. 26cc/year. After treatment course, BD response was unchanged, but methacholine reactivity improved significantly; Fig. 1. (Figure presented) Airway Hyperreactivity Post-Treatment. (A): BD response did not improve from pre- to post-treatment, (N=135). Shading represents a positive BD response. (B): Methacholine reactivity significantly improved, from failing at a cumulative dose of 1.4 mg/mL to failing at a cumulative dose of 64 mg/mL (N=35). There was no difference between individuals receiving bronchodilator response and methacholine challenge in FEV1 or FEV1/FVC both pre- and post treatment, the decline FEV1 from pre- to post-Treatment, time to entry into treatment cohort, or demographics (years of service, exposure, age at 9/11, or BMI). Conclusion: Treatment with removal from fire fighting, inhaled steroids and beta agonists did not improve spirometric measures, or BD response, but slowed the rate of decline and decreased hyper-reactivity to the MCT. Improvement in reactivity defined by MCT but not defined by BD response suggests that the pathways leading to BD response and MCT reactivity may be different and may respond differently to treatment

RATIONALE: In the first year post 9/11/2001, there was a decline of 439 mL in FEV1 in FDNY rescue workers, stabilizing to a 25 ml/year decline in the subsequent seven years. Airflow obstruction predominated in 1720 individuals who sought a subspecialty pulmonary evaluation for treatment. We noted that the most severely obstructed patients had significant weight gain. This well phenotyped cohort had serum drawn within the five months post 9/11/2001, allowing us to investigate relationships of metabolic syndrome biomarkers and decline in lung function. METHODS: The treatment cohort (N=1720) was stratified by FEV1 into either obstructed, FEV1<76% predicted (Lower limit of normal for this cohort), or normal airflow, FEV1>76% (Figure 1A). A pilot analysis assayed 41 patients' serum drawn 2.88+/-0.99 months following 9/11 for 15 biomarkers of metabolic syndrome by Luminex; 10 cases were obstructed and 31 were normal. All patients had normal pre-9/11 lung function. Serum cholesterol and triglycerides were available on 157 patients, 20/157 were obstructed. The subspecialty PFT was used to stratify patients, months post-9/11. Data represented as means +/-SD and significance was assessed based on p<=0.05 by t-test. RESULTS: BMIs at the time of serum sampling were no different between individuals with obstruction and normal FEV1. However, at subspecialty PFT, the obstructed patients had higher BMIs, (Figure 1B). The cohort with the greatest lung function decline had significantly greater cholesterol and Cholesterol/HDL ratios, with a trend towards elevated triglyceride levels, (Figure 1C-E). The subgroup with available Luminex data showed higher levels of sE-Selectin, tPAI-1, and s-ICAM in the obstructed patients, (Figure 2A-C). The same cohort had a trend towards elevated levels of C-peptide (Figure 2D). Individuals with obstruction at treatment entry had accelerated decline in lung function post 9/11, increased airway reactivity, and evidence of air trapping based on elevated RV when compared to those who maintained a normal FEV1%. (Figure presented) CONCLUSIONS: Blood drawn within 5 months of WTC exposure identified a subgroup of patients with markers of metabolic syndrome. This subgroup had more weight gain and greater decline in lung function during the next 7 years. The finding of metabolic syndrome biomarkers prior to decline in lung function raises the possibility that the combination of irritant exposure and mediators of metabolic syndrome interact and promote lung injury

BACKGROUND: Pulmonary function declined among 13,234 WTC exposed FDNY personnel in the first year after 9/11/2001. In subsequent years, the rate of decline in FEV1 returned to a group's mean about 25ml/yr (Aldrich et. al 2010). The subgroup who presented for treatment frequently had airway obstruction (Weiden et al. 2010) with some patients having accelerated decline of lung function compared to the total cohort. IgA and IgG4 mediate mucosal immunity while IgG3 drives lung inflammation. Serum from 2001-2002 was stored and tested for immunoglobulin expression to identify biomarkers that predict accelerated decline in lung function in WTC exposed individuals. METHODS: We conducted a pilot investigation of serum antibody expression in 40/1720 firefighters who received treatment for pulmonary complaints. All patients have at least 3 pulmonary function tests available for the study: (1) pre 9/11 (2) 3 months post 9/11 and (3) upon entry into treatment (32 months post 9/11/2001). All of these patients had normal pre-9/11 lung function (Fig. 1). At treatment entry, 9/40 patients had abnormal FEV1 (Mean 58 +/-13% predicted) and 31/40 had normal FEV1 (92 +/- 11% predicted). The serum samples, drawn 3 months post 9/11 were analyzed for IgM, IgA, IgE and IgG1 through IgG4 profiles by using Luminex multiplex technology. Immunoglobulins expression was then correlated with changes in FEV1 from the PFT done soon after 9/11/2001 to the PFT done at the time of presentation for treatment. RESULTS: In the serum collected prior to development lung function abnormalities, patients who maintained normal lung function have different Ig levels than those who developed airflow obstruction. Patients who went on to develop abnormal lung function post 9/11 had reduced IgA (p=0.06), IgG4 (p<0.0005) levels and increased IgG3 (p<0.01) level compared to those who maintained normal lung function (Fig. 2). There were no differences in IgM, IgG1, IgG2, or IgE levels. There were no significant differences between two groups in FEV1 prior to 9/11, age, BMI, exposure intensity status. 9 patients with abnormal lung function had accelerated decline in lung function post 9/11, increased airway reactivity and evidence of air trapping when compared to those who maintained normal lung function. (Figure presented) CONCLUSIONS: Reduced levels of IgG4 and IgA are biomarkers for future decline in lung function after an irritant exposure. Since IgG4 and IgA mediate mucosal immunity deficiency in mucosal immune response may predispose to non-resolving inflammation and progressive lung injury after an irritant exposure

Introduction: Macrolides are associated with a significant reduction in exacerbations compared with placebo in both chronic obstructive lung disease and cystic fibrosis. However, this may be due anti-bacterial and/or anti-inflammatory effects. The mechanism of the anti-inflammatory effects of macrolides in the lung are poorly understood. Alveolar macrophages (AM) have a central role on innate immune homeostasis of the lung; AM have developed many mechanisms to prevent autonomous induction and excess production of inflammatory cytokines through an increased expression of immune inhibitors such as C/EBPbeta and MafB. Interferon induces inhibitory C/EBPbeta, is regulated by MafB and has anti-inflammatory effect in the lung. Here, we test the hypothesis that azithromycin suppresses innate immune response by induction of inhibitory transcription factors. Methods: Alveolar macrophages were obtained from five subjects and incubated with lipopolysaccharide (LPS), interferon (IFN)-alpha, azithromycin (AZM) and AZM+LPS for 24hrs. Thirty nine supernatant cytokines were measured using Luminex. Protein extracts were obtained from AM and used for immunoblots for C/EBPbeta and MafB. Results: LPS strongly induced TNF-alpha production by AM, while the combination AZM and LPS inhibited LPS induced TNF-alpha production (Figure 1, p=0.02). AZM by itself had minimal effect. AZM also inhibited LPS induced CCL22, MDC, IL-1b, IL-6, IL10, G-GSF and GM-CSF. The inhibitory C/EBPbeta transcription factor and MafB expression was induced when AZM was added with LPS (Figure 2). Conclusions: AZM inhibits inflammatory cytokine production after LPS stimulation and induces inhibitory C/EBPbeta and MafB in AM. We will investigate if induction of innate immune inhibitors that blocks TLR signaling pathways underlies the broad anti-inflammatory effect achieved by macrolides. (Figure presented)

INTRODUCTION: Interferon-gamma (IFN-) is a Th-1 cytokine central to the immune response in pulmonary tuberculosis (TB). We have reported that TB patients treated with nebulized IFN- plus Isoniazid, Rifampin, Pyrazinamide and Ethambutol under Directly Observed Therapy (DOT) had significantly shorter time to sputum smear conversion, improved symptoms, and reduced inflammatory cytokines in Bronchoalveolar Lavage (BAL) obtained at 16 weeks, when compared to patients receiving DOT alone. We have also observed two genomic presentations in TB BAL, expressing transcription signals that correlate with either Th-1 or Th-2 responses. Furthermore, we observed that a Th-2 pattern could convert into Th-1, following anti-tuberculosis treatment. RATIONALE: We tested whether there are two presentations of TB based on cytokine production in BAL cells. We hypothesized that a Th-1 cytokine profile will be associated with less severe disease at presentation than patients with a Th-2 profile. METHODS: Luminex multiplex assays for 13 cytokines of BAL cell supernatants from 73 TB patients at the time of presentation. Patients segregated into Th-1 and Th-2 cytokine clusters. This was compared with clinical parameters. RESULTS: Cluster analysis of pre-treatment 24-hour BAL supernatants segregated 2 distinct cytokine patterns. Cluster 1 had higher mean levels of Th-1 cytokines; such as, IFN-, IL-12p40 and IP-10. More patients in Cluster 1 converted to a negative sputum smear by week 1 than those in Cluster 2 (OR= 2.89, p=0.032). Tuberculin Skin Test (TST) size and cell differential also distinguished the two groups. Cluster 1 had larger TST measurements (p=0.041), and possessed lymphocyte predominant BAL cell differentials (p=0.013). Conversely, the Cluster 2 cytokine pattern showed relatively elevated levels of IL-4 and IL-10, which denote an anti-inflammatory Th-2 response. When IL-4 levels alone were analyzed, we observed that patients tended to segregate into high and low levels. This separation became more distinct over time, and was independent of treatment arm. Upon further examination of IL-4 levels at baseline according to Cluster, we showed that Cluster 2 was composed of the group having higher IL-4 levels ( p=0.001). Members of this group also tended to have later sputum conversion, male predominance (p=0.016) and higher levels of neutrophils (p=0.0002) on BAL cell differentials. There were no differences between the 2 clusters with regard to age, smoking, and BMI. (Table Presented) CONCLUSION: A Th-1 cytokine patterns in BAL of TB infected patients predicts more cellular immunity and more rapid TB control than patients with a Th-2 pattern

Introduction:Regulatory T cells (Treg) play important roles in immunological self-tolerance, and are functionally immunosuppressive subsets of T cells. Th17 cells are critical in the defense against microbes, particularly at mucosal surfaces. It has been shown that the balance between Treg and Th17 cells is a key factor that regulates helper T-cell (CD4⁺ or Th) function. However, there is limited information on the balance between Treg and Th17 cells in bronchoalveolar lavage (BAL). We investigated the distribution of Th17 cells in relation to Treg in PBMC and BAL of healthy volunteers, emphysema patients and HIV-1 infected patients. Methods:BAL lymphocytes were obtained by plating BAL cells for >=1 hour to remove adherent alveolar macrophages. PBMC were isolated by Ficoll gradient fractionation. BAL and PB lymphocytes were stimulated with PMA and ionomycin, treated with monensin, permeabilized, then labeled with PerCP-anti-CD4, PE-anti-IL-17 and Alexa Fluor 647 anti-Foxp3, and analyzed on a FACSCalibur to determine the percentage of CD4⁺ lymphocytes expressing IL-17 and Foxp3 (mean+/-SD). Results:Ten subjects were available for analysis (2 healthy volunteers, 5 emphysema and 3 HIV-1 infected subjects). Both PB and BAL Th cells presented a wide range of FoxP3⁺ cells (19.2+/-18.6 and 6.3+/-5.9 respectively) and IL-17⁺ cells (9.5+/-8.1 and 8.8+/-4.6 respectively). There was a greater percentage of FoxP3⁺ in the HIV-infected group compared to the healthy volunteer and emphysema groups (11.7+/-6.4 vs. 1.7+/-1.1 and 4.9+/-4.7 respectively). Similarly, there was a trend towards higher percentage of IL-17⁺ cells in the BAL of HIV-infected subjects compared to the healthy volunteer and emphysema subjects (10.8+/-3.6 vs. 5.6+/-0.9 and 8.9+/-5.8 respectively). When the expression of FoxP3⁺ and IL-17⁺ cells in BAL was compared for each subject, a direct correlation with an r2 of 0.36 and a p-value of 0.06 was found Conclusions:We observed a relative preservation of FoxP3⁺ and IL-17⁺ cells in BAL CD4⁺ lymphocytes, despite prior evidence suggesting a preferential loss of both Treg and Th17 subsets in PBMC of HIV-1 infected subjects. The co-ordinated expression of these subsets in BAL Th cells warrants further investigation to evaluate its immune significance in the alveolar compartment

The spectrum of lung diseases associated with HIV is broad, and many infectious and noninfectious complications of HIV infection have been recognized. The nature and prevalence of lung complications have not been fully characterized since the Pulmonary Complications of HIV Infection Study more than 15 years ago, before antiretroviral therapy (ART) increased life expectancy. Our understanding of the global epidemiology of these diseases in the current ART era is limited, and the mechanisms for the increases in the noninfectious conditions, in particular, are not well understood. The Longitudinal Studies of HIV-Associated Lung Infections and Complications (Lung HIV) Study (ClinicalTrials.gov number NCT00933595) is a collaborative multi-R01 consortium of research projects established by the National Heart, Lung, and Blood Institute to examine a diverse range of infectious and noninfectious pulmonary diseases in HIV-infected persons. This article reviews our current state of knowledge of the impact of HIV on lung health and the development of pulmonary diseases, and highlights ongoing research within the Lung HIV Study