Yan Lin

Background: We previously reported early on-treatment significant modulation in circulating regulatory T cell (Treg), myeloid derived suppressor cells (MDSC) and antigen-specific type I CD4+ and CD8+ T cells that correlated with clinical outcome in regionally advanced melanoma patients treated with neoadjuvant ipilimumab. Here, we investigated the long term immunologic impact of CTLA4 blockade.Methods: Patients were treated with ipilimumab given at 10 mg/kg IV every 3 weeks for 2 doses bracketing surgery. Blood specimens were collected at baseline and during treatment for up to 9 months. We tested immune responses at 3, 6, and 9 months utilizing multicolor flow cytometry. We compared frequencies of circulating Treg and MDSC on-study to baseline levels, as well as frequencies of CD4+ and CD8+ T cells specific to shared tumor-associated antigens (Gp-100, MART-1, NY-ESO-1).Results: Levels of Treg significantly increased when measured at 6 weeks following ipilimumab but returned to baseline by 3 months, with no significant difference in Treg levels between relapsed and relapse-free groups at 3, 6 or 9 months. However, lower baseline levels of circulating Treg (CD4+CD25hi+CD39+) were significantly associated with better relapse free survival (RFS) (p = 0.04). Levels of circulating monocytic HLA-DR+/loCD14+ MDSC were lower at baseline in the relapse-free group and further decreased at 6 weeks, though the differences did not reach statistical significance including measurements at 3, 6 or 9 months. We detected evidence of type I (interferon-γ producing), activated (CD69+) CD4+ and CD8+ antigen-specific T cell immunity against cancer-testis (NY-ESO-1) as well as melanocytic lineage (MART-1, gp100) antigens in the absence of therapeutic vaccination. These responses were significantly boosted at 6 weeks and persisted at 3, 6 and 9 months following the initiation of ipilimumab.Conclusions: Lower Treg levels at baseline are significantly associated with RFS and increased Treg frequency after CTLA4 blockade was only transient. Lower MDSC was also associated with RFS and MDSC levels were further decreased after ipilimumab. Tumor specific effector immune responses are boosted with CTLA4 blockade and tend to be durable.Trial registrationClinicalTrials.gov Identifier: NCT00972933

Background Veliparib (ABT-888) is an oral PARP inhibitor expected to increase gemcitabine activity. This phase I determined the maximal tolerable dose (MTD), dose-limiting toxicities (DLT), antitumor activity, pharmacokinetics (PK), and pharmacodynamics (PD) of veliparib combined with gemcitabine. Methods Patients with advanced solid tumors received veliparib (10–40-mg PO BID) on chemotherapy weeks with gemcitabine 500–750-mg/m² IV on days 1, 8, and 15 (28-day cycle), or on days 1 and 8 (21-day cycle). The MTD, DLT, adverse events, PK, and PD were evaluated. Results Eleven patients were enrolled on the 28-day schedule. The 28-day schedule was considered intolerable and amended to a 21-day schedule, with 20 patients enrolled. Grade ≥ 3 adverse events were myelosuppression-related. The MTD was determined to be 750-mg/m² gemcitabine IV on days 1 and 8- and 20-mg PO veliparib BID days 1–14 on a 21-day schedule. Of 27 patients evaluable for response, 3 had PR and 15 had SD. There was no evidence of any major drug–drug interaction, and PK parameter values for veliparib, gemcitabine, and dFdU were as expected. Analysis of PBMCs showed evidence of PARP inhibition and DNA damage associated with therapy. Conclusions Gemcitabine at 750-mg/m² IV on days 1 and 8 combined with veliparib at a dose of 20-mg PO BID days 1–14 on a 21-day schedule is relatively well-tolerated, with manageable, expected toxicities. Clinical responses were observed in a pretreated population of patients, suggesting that this combination should be further evaluated in the phase II setting.

In a previously reported study, patients with regionally advanced melanoma were treated with neoadjuvant ipilimumab (ipi) (Tarhini in PLoS ONE 9(2): e87705, 3). Significant changes in circulating myeloid derived suppressor cells (MDSC), regulatory T cells (Treg) and peptide specific type I CD4+ and CD8+ T cells were noted at week 6 that correlated with clinical outcome. Characterization of antigen-specific effector T cell secreted cytokines may shed insights into ipi associated T cell activation and function. Patients were treated with neoadjuvant ipi (10 mg/kg every 3 weeks ×2) administered intravenously before and after surgery. Peripheral blood mononuclear cells (PBMC) that were collected at baseline and week 6 (after ipi) were tested here. Each sample was divided into 5 groups and stimulated with controls or shared melanoma antigen overlapping peptide pools (NY-ESO 1, gp-100, MART-1). Secreted cytokines, chemokines and growth factors were assessed using Luminex. Cytokine expression levels between the 3 antigen groups were compared using the Wilcox rank-sum test. Seventeen cytokines were differentially expressed with stimulation by each antigen at baseline (p value <0.05): IL1β, MIP1β, IL1RA, VEGF, IL13, IL17, MIP1α, GM-CSF, MCP1, IL5, IL2R, IL4, IL10, IFNγ, TNFα, IL8 and IL2. At week 6, 15 cytokines were differentially expressed (p < 0.05): IL1β, VEGF, G-CSF, HGF, IL13, IL17, GM-CSF, MCP1, IL5, IL7, IL4, IL10, IFNγ, IL8 and IL2. Patients were later clustered based on cytokine expression levels at baseline and at week 6, and recurrence free survival (RFS) was compared. Clear differences in RFS were noted based on cytokine level clustering both at baseline and at week 6: Patients whose PBMCs secreted more cytokines in response to NY-ESO-1 showed a trend towards better RFS. PBMCs of patients treated with ipi secreted significantly more cytokines, chemokines and growth factors in response to NY-ESO-1 than to gp-100 or MART-1. These cytokines belonged to different functional groups, including inflammatory, type 1, type 2 and regulatory, that warrant further study. Patients whose PBMCs secreted more cytokines (particularly in response to NY-ESO-1) tended to have better RFS, supporting further exploration in terms of therapeutic predictive value.

Background Within-household transmission of COVID-19 is responsible for a significant number of infections. The risk of within-household infection is greatly increased among those from Black Asian and minority ethnic (BAME) and low income communities. Efforts to protect these communities are urgently needed. The aim of this study is to explore the acceptability of the availability of accommodation to support isolation among at risk populations. Methods Our study used a mixed methods design structured in two phases. In phase 1, we conducted a survey study of a sample of volunteers from our existing database of 300 individuals who had provided consent to be contacted about ongoing research projects into infection control. In phase 2, we conducted semi-structured interviews with 19 participants from BAME communities and low income communities recruited through social media. Results Participants from the survey and interview phase of the study viewed the provision of accommodation as important and necessary. Factors influencing likely uptake of accommodation included perceived 1) vulnerability of household 2) exposure to the virus and 3) options for isolation at home. Barriers to accepting the offer of accommodation included 1) being able to isolate at home 2) wanting to be with family 3) caring responsibilities 4) concerns about mental wellbeing 5) upheaval of moving when ill and 6) concerns about infection control. Participants raised a series of issues that should be addressed before accommodation is offered. These included questions regarding who should use temporary accommodation and at what stage to effectively reduce transmission in the home, and how infection control in temporary accommodation would be managed. Conclusion This research provides evidence that the provision of accommodation to prevent within household transmission of the virus is viewed as acceptable, feasible and necessary by many people who are concerned about infection transmission in the home. We explore ways in which accommodation might be offered. In particular, vulnerable members of the household could be protected if accommodation is offered to individuals who are informed through test trace and isolate that they have been in contact with the virus.

In modern biomedical research of complex diseases, a large number of demographic and clinical variables, herein called phenomic data, are often collected and missing values (MVs) are inevitable in the data collection process. Since many downstream statistical and bioinformatics methods require complete data matrix, imputation is a common and practical solution. In high-throughput experiments such as microarray experiments, continuous intensities are measured and many mature missing value imputation methods have been developed and widely applied. Numerous methods for missing data imputation of microarray data have been developed. Large phenomic data, however, contain continuous, nominal, binary and ordinal data types, which void application of most methods. Though several methods have been developed in the past few years, not a single complete guideline is proposed with respect to phenomic missing data imputation. In this paper, we investigated existing imputation methods for phenomic data, proposed a self-training selection (STS) scheme to select the best imputation method and provide a practical guideline for general applications. We introduced a novel concept of "imputability measure" (IM) to identify missing values that are fundamentally inadequate to impute. In addition, we also developed four variations of K-nearest-neighbor (KNN) methods and compared with two existing methods, multivariate imputation by chained equations (MICE) and missForest. The four variations are imputation by variables (KNN-V), by subjects (KNN-S), their weighted hybrid (KNN-H) and an adaptively weighted hybrid (KNN-A). We performed simulations and applied different imputation methods and the STS scheme to three lung disease phenomic datasets to evaluate the methods. An R package "phenomeImpute" is made publicly available. Simulations and applications to real datasets showed that MICE often did not perform well; KNN-A, KNN-H and random forest were among the top performers although no method universally performed the best. Imputation of missing values with low imputability measures increased imputation errors greatly and could potentially deteriorate downstream analyses. The STS scheme was accurate in selecting the optimal method by evaluating methods in a second layer of missingness simulation. All source files for the simulation and the real data analyses are available on the author's publication website.

E1694 tested GM2-KLH-QS21 vaccine versus high-dose interferon-α2b (HDI) as adjuvant therapy for operable stage IIB-III melanoma. We tested banked serum specimens from patients in the vaccine arm of E1694 for prognostic biomarkers. Aushon Multiplex Platform was used to quantitate baseline serum levels of 115 analytes from 40 patients. Least absolute shrinkage and selection operator proportional hazard regression (Lasso PH) was used to select markers that are most informative for relapse-free survival (RFS) and overall survival (OS). Regular Cox PH models were then fit with the markers selected by the Lasso PH. Survival receiver operating characteristic (ROC) analysis was used to evaluate the ability of the models to predict 1-year RFS and 5-year OS. Four markers that include Tumor Necrosis Factor alpha Receptor II (TNF-RII), Transforming Growth Factor alpha (TGF-α), Tissue Inhibitor of Metalloproteinases 1 (TIMP-1), and C-reactive protein (CRP) were found to be most informative for the prediction of OS (high levels correlate with worse prognosis). The dichotomized risk score based on the four markers could significantly separate the OS curves (p = 0.0005). When using the four-marker PH model to predict 5-year OS, we achieved an area under the curve (AUC) of 89% (cross validated AUC = 72%). High baseline TNF-RII was also significantly associated with worse RFS. The RFS with high (above median) TNF-RII was significantly lower than low TNF-RII (p = 0.01). The biomarker signature consisting of TNFR-II, TGF-α, TIMP-1 and CRP is significantly prognostic of survival in patients with high-risk melanoma and warrants further investigation.

Serum lactate dehydrogenase (LDH) is a prognostic factor for patients with stage IV melanoma. To gain insights into the biology underlying this prognostic factor, we analyzed total serum LDH, serum LDH isoenzymes, and serum lactate in up to 49 patients with metastatic melanoma. Our data demonstrate that high serum LDH is associated with a significant increase in LDH isoenzymes 3 and 4, and a decrease in LDH isoenzymes 1 and 2. Since LDH isoenzymes play a role in both glycolysis and oxidative phosphorylation (OXPHOS), we subsequently determined using tissue microarray (TMA) analysis that the levels of proteins associated with mitochondrial function, lactate metabolism, and regulators of glycolysis were all elevated in advanced melanomas compared with nevic melanocytes. To investigate whether in advanced melanoma, the glycolysis and OXPHOS pathways might be linked, we determined expression of the monocarboxylate transporters (MCT) 1 and 4. Analysis of a nevus-to-melanoma progression TMA revealed that MCT4, and to a lesser extend MCT1, were elevated with progression to advanced melanoma. Further analysis of human melanoma specimens using the Seahorse XF24 extracellular flux analyzer indicated that metastatic melanoma tumors derived a large fraction of energy from OXPHOS. Taken together, these findings suggest that in stage IV melanomas with normal serum LDH, glycolysis and OXPHOS may provide metabolic symbiosis within the same tumor, whereas in stage IV melanomas with high serum LDH glycolysis is the principle source of energy.

Background: Detecting candidate markers in transcriptomic studies often encounters difficulties in complex diseases, particularly when overall signals are weak and sample size is small. Covariates including demographic, clinical and technical variables are often confounded with the underlying disease effects, which further hampers accurate biomarker detection. Our motivating example came from an analysis of five microarray studies in major depressive disorder (MDD), a heterogeneous psychiatric illness with mostly uncharacterized genetic mechanisms.Results: We applied a random intercept model to account for confounding variables and case-control paired design. A variable selection scheme was developed to determine the effective confounders in each gene. Meta-analysis methods were used to integrate information from five studies and post hoc analyses enhanced biological interpretations. Simulations and application results showed that the adjustment for confounding variables and meta-analysis improved detection of biomarkers and associated pathways.Conclusions: The proposed framework simultaneously considers correction for confounding variables, selection of effective confounders, random effects from paired design and integration by meta-analysis. The approach improved disease-related biomarker and pathway detection, which greatly enhanced understanding of MDD neurobiology. The statistical framework can be applied to similar experimental design encountered in other complex and heterogeneous diseases.

Gene expression microarrays have become powerful tools in many areas of biological and biomedical research. These technologies allow researchers to measure the expression levels of thousands of genes in a tissue or cell sample simultaneously. One of the most common types of microarray experiments is simply an exploratory study to compare two samples (e.g. tumor and normal tissue) and look for a list of genes that might be differentially expressed between the two. Differential expression is typically measured by computing a t-statistic or similar statistic for each gene. The genes are then ranked according to the absolute value of the t-statistic, and the twenty or fifty best candidates might be studied in follow-up experiments. When sample sizes are small, the t-statistic can be problematic, because variances are estimated poorly and the "top 20" list is often dominated by the genes with the lowest variance estimates. Lönnstedt and Speed (2001) proposed an empirical Bayes method for avoiding this problem, and showed that their approach has lower false positive and false negative rates than t-statistic-based methods. However, their method was designed mainly for the two-channel microarray, which produces paired data. It is not suitable for technologies such as one-channel arrays that produce unpaired data. We propose a simplification for Lönnstedt and Speed's method, and then extend it to unpaired data with two or more independent treatments. We demonstrate our method on both simulated and real data. When the number of replicates is small, gene rankings based on our statistic appear to be much more reliable than rankings based on the t-statistic.