Smoking history is associated with reduced efficacy of neoadjuvant therapy in pancreatic adenocarcinoma.

BACKGROUND: Differential responses to neoadjuvant therapy (NAT) exist in pancreatic ductal adenocarcinoma (PDAC); however, contributing factors are poorly understood. Tobacco smoke is a common risk factor for PDAC, with nicotine-induced chemoresistance observed in other cancers. This study aimed to explore the potential association between tobacco use and NAT efficacy in PDAC. METHODS: A single-center, retrospective analysis was conducted that included all consecutive patients with PDAC who underwent surgical resection after NAT with a documented smoking history (N = 208). NAT response was measured as percentage fibrosis in the surgical specimen. Multivariable models controlled for covariates and survival were modeled using the Kaplan-Meier method. RESULTS: Postoperatively, major responses to NAT (>95% fibrosis) were less frequently observed in smokers than in nonsmokers (13.7% vs 30.4%, respectively; P = .021). Pathologic complete responses were similarly less frequent in smokers than in nonsmokers (2.1% vs 9.9%, respectively; P = .023). On multivariate analysis controlling for covariates, smoking history remained independently associated with lower odds of major fibrosis (odds ratio [OR], 0.25; 95% CI, 0.10-0.59; P = .002) and pathologic complete response (OR, 0.21; 95% CI, 0.03-0.84; P = .05). The median overall survival was significantly longer in nonsmokers than in smokers (39.1 vs 26.6 months, respectively; P = .05). CONCLUSION: Tobacco use was associated with diminished pathologic responses to NAT. Future research to understand the biology underlying this observation is warranted and may inform differential NAT approaches or counseling among these populations.

Type of neoadjuvant treatment strategy is associated with differential pathologic responses in pancreatic ductal adenocarcinoma.

BACKGROUND: Tumor fibrosis after neoadjuvant treatment (NAT) for pancreatic ductal adenocarcinoma (PDAC) correlates with treatment response. Herein we assessed how different NAT strategies influence pathologic responses and survival. METHODS: Patients with surgically resected PDAC who received NAT (1991-2020) were included. Descriptive statistics compared outcomes amongst fibrosis groups (none, minor <50 â€‹%, partial 51%-94 â€‹%, major ≥95 â€‹%) and NAT (chemotherapy alone, chemoradiation, or chemotherapy â€‹+ â€‹chemoradiation (total neoadjuvant therapy, TNT)). RESULTS: Patients with major fibrosis most often received TNT (65.8 â€‹%, p â€‹< â€‹0.001). Major fibrosis was associated with the greatest rate of downstaging (77.8 â€‹%, p â€‹< â€‹0.001), highest R0 margin rate (100 â€‹%, p â€‹< â€‹0.01), and lowest mean positive lymph node ratio (0.80, p â€‹< â€‹0.01). Amongst complete responders, 11/14 (78.6 â€‹%) received TNT. Median overall (66.3 months, p â€‹= â€‹0.003) and disease-free (54.7months, p â€‹= â€‹0.05) survival were highest with major fibrosis. CONCLUSIONS: Major fibrosis and complete pathologic responses after NAT are most frequent with a TNT strategy and are associated with improved outcomes.

A phase 1 study of nivolumab in combination with interferon-gamma for patients with advanced solid tumors.

This phase I, dose-escalation trial evaluates the safety of combining interferon-gamma (IFN-γ) and nivolumab in patients with metastatic solid tumors. Twenty-six patients are treated in four cohorts assessing increasing doses of IFN-γ with nivolumab to evaluate the primary endpoint of safety and determine the recommended phase two dose (RP2D). Most common adverse events are low grade and associated with IFN-γ. Three dose limiting toxicities are reported at the highest dose cohorts. We report only one patient with any immune related adverse event (irAE). No irAEs ≥ grade 3 are observed and no patients require corticosteroids. The maximum tolerated dose of IFN-γ is 75 mcg/m2, however based on a composite of safety, clinical, and correlative factors the RP2D is 50 mcg/m2. Exploratory analyses of efficacy in the phase I cohorts demonstrate one patient with a complete response, and five have achieved stable disease. Pre-planned correlative assessments of circulating immune cells demonstrate intermediate monocytes with increased PD-L1 expression correlating with IFN-γ dose and treatment duration. Interestingly, post-hoc analysis shows that IFN-γ induction increases circulating chemokines and is associated with an observed paucity of irAEs, warranting further evaluation. ClinicalTrials.gov Trial Registration: NCT02614456.

Folic Acid Supplementation Promotes Hypomethylation in Both the Inflamed Colonic Mucosa and Colitis-Associated Dysplasia.

PURPOSE: The purpose of this study was to assess the effect of folic acid (FA) supplementation on colitis-associated colorectal cancer (CRC) using the azoxymethane/dextran sulfate sodium (AOM/DSS) model. METHODS: Mice were fed a chow containing 2 mg/kg FA at baseline and randomized after the first DSS treatment to receive 0, 2, or 8 mg/kg FA chow for 16 weeks. Colon tissue was collected for histopathological evaluation, genome-wide methylation analyses (Digital Restriction Enzyme Assay of Methylation), and gene expression profiling (RNA-Seq). RESULTS: A dose-dependent increase in the multiplicity of colonic dysplasias was observed, with the multiplicity of total and polypoid dysplasias higher (64% and 225%, respectively) in the 8 mg FA vs. the 0 mg FA group (p < 0.001). Polypoid dysplasias were hypomethylated, as compared to the non-neoplastic colonic mucosa (p < 0.05), irrespective of FA treatment. The colonic mucosa of the 8 mg FA group was markedly hypomethylated as compared to the 0 mg FA group. Differential methylation of genes involved in Wnt/ß-catenin and MAPK signaling resulted in corresponding alterations in gene expression within the colonic mucosa. CONCLUSIONS: High-dose FA created an altered epigenetic field effect within the non-neoplastic colonic mucosa. The observed decrease in site-specific DNA methylation altered oncogenic pathways and promoted colitis-associated CRC.

TET1 and TDG Suppress Inflammatory Response in Intestinal Tumorigenesis: Implications for Colorectal Tumors With the CpG Island Methylator Phenotype.

BACKGROUND & AIMS: Aberrant DNA methylation is frequent in colorectal cancer (CRC), but underlying mechanisms and pathologic consequences are poorly understood. METHODS: We disrupted active DNA demethylation genes Tet1 and/or Tdg from ApcMin mice and characterized the methylome and transcriptome of colonic adenomas. Data were compared to human colonic adenocarcinomas (COAD) in The Cancer Genome Atlas. RESULTS: There were increased numbers of small intestinal adenomas in ApcMin mice expressing the TdgN151A allele, whereas Tet1-deficient and Tet1/TdgN151A-double heterozygous ApcMin colonic adenomas were larger with features of erosion and invasion. We detected reduction in global DNA hypomethylation in colonic adenomas from Tet1- and Tdg-mutant ApcMin mice and hypermethylation of CpG islands in Tet1-mutant ApcMin adenomas. Up-regulation of inflammatory, immune, and interferon response genes was present in Tet1- and Tdg-mutant colonic adenomas compared to control ApcMin adenomas. This up-regulation was also seen in murine colonic organoids and human CRC lines infected with lentiviruses expressing TET1 or TDG short hairpin RNA. A 127-gene inflammatory signature separated colonic adenocarcinomas into 4 groups, closely aligned with their microsatellite or chromosomal instability and characterized by different levels of DNA methylation and DNMT1 expression that anticorrelated with TET1 expression. Tumors with the CpG island methylator phenotype (CIMP) had concerted high DNMT1/low TET1 expression. TET1 or TDG knockdown in CRC lines enhanced killing by natural killer cells. CONCLUSIONS: Our findings reveal a novel epigenetic regulation, linked to the type of genomic instability, by which TET1/TDG-mediated DNA demethylation decreases methylation levels and inflammatory/interferon/immune responses. CIMP in CRC is triggered by an imbalance of methylating activities over demethylating activities. These mice represent a model of CIMP CRC.

Social Media Depictions of Cochlear Implants: An Instagram and TikTok Analysis.

OBJECTIVE: To identify populations of authors who post about cochlear implants (CIs) on Instagram and TikTok, to illustrate the content of these posts, and to elucidate factors that might help surgeons better educate CI patients. STUDY DESIGN: Qualitative study. SETTING: Instagram and TikTok social media platforms. PATIENTS: All public social media posts identified with the search terms below. Posts were excluded if unrelated to CIs or if written in a non-English language. INTERVENTION: Instagram and TikTok were searched for posts from March 2021 through September 2021 with the search terms #cochlearimplant, #cochlearimplants, #cochlearimplantkids, #cochlearkids, and #cochlearfamily. MAIN OUTCOME MEASURES: Posts were subclassified and analyzed for content including topics of posts, authorship, timeframe of posts, depiction of CIs, and popularity. RESULTS: Of 1,942 posts included in the final analysis, 1,400 were found on Instagram and 542 on TikTok. Authors were mostly patients (n = 771, 39.7%), companies (n = 568, 29.2%), and patients' family members (n = 482, 24.8%). Only 21 posts were made by physicians (1.1%). Out of 379 total educational posts examined, patients themselves were the most common authors (n = 219, 57.8%) followed by patients' family members (n = 139, 36.7%). Physicians authored only a small fraction of all educational posts (n = 19, 5.0%). CONCLUSIONS: This study showed minimal physician involvement in the CI social media spheres of Instagram and TikTok. In addition, there were few educational posts on either platform, revealing ample opportunity for physicians to become more involved with CI social media.

Harnessing choice architecture in urologic practice: Implementation of an opioid-sparing protocol grounded in cognitive behavioral theory.

PURPOSE: Opioids are prescribed excessively following surgery. As many urologic oncology procedures are performed minimally invasively, an opportunity exists to push forward initiatives to minimize postoperative opioid use. MATERIALS AND METHODS: A quality improvement initiative to reduce inpatient opioid prescribing was launched at a tertiary cancer center. In Phase I (December 2019-July 2020), providers were instructed to start standing acetaminophen. In Phase II (beginning August 2020), education was provided to the entire care team and ordersets were modified to an opioid sparing protocol (OSP). We analyzed the proportion of minimally invasive surgery (MIS) prostatectomy and nephrectomy patients that adhered to an OSP during each phase and compared them to controls from the preceding 2 years. RESULTS: A total of 303, 153, and 839 patients underwent MIS during the Phase I, Phase II, and control periods respectively. The proportion of patients adhering to an OSP increased from 16% at the beginning of Phase I to 76% at the end of Phase II (p-trend < 0.001). The median total oral morphine equivalents for oral opioids declined from 20 mg and 40 mg at baseline for prostatectomy and nephrectomy patients respectively to 0 mg for both groups (p-trends < 0.001). Multivariable analysis found that patients received 22% and 81% less oral morphine equivalents during Phase I and II respectively compared to the control period (P < 0.001). CONCLUSIONS: Adherence to an OSP is most effective when initiatives incorporate the entire team and are supported by nudge theory-based structural changes. Using these strategies, most patients following urologic MIS can dramatically reduce opioid use postoperatively.

Elaboration Models with Symmetric Information Divergence.

Various statistical methodologies embed a probability distribution in a more flexible family of distributions. The latter is called elaboration model, which is constructed by choice or a formal procedure and evaluated by asymmetric measures such as the likelihood ratio and Kullback-Leibler information. The use of asymmetric measures can be problematic for this purpose. This paper introduces two formal procedures, referred to as link functions, that embed any baseline distribution with a continuous density on the real line into model elaborations. Conditions are given for the link functions to render symmetric Kullback-Leibler divergence, Rényi divergence, and phi-divergence family. The first link function elaborates quantiles of the baseline probability distribution. This approach produces continuous counterparts of the binary probability models. Examples include the Cauchy, probit, logit, Laplace, and Student-t links. The second link function elaborates the baseline survival function. Examples include the proportional odds and change point links. The logistic distribution is characterized as the one that satisfies the conditions for both links. An application demonstrates advantages of symmetric divergence measures for assessing the efficacy of covariates.

High Variability in Cellular Proliferation, Gene Expression, and Cytokine Production in the Nonneoplastic Colonic Epithelium of Young Apc+/Min-FCCC Mice.

An urgent need exists to identify efficacious therapeutic preventive interventions for individuals who are at high risk of developing colorectal cancer. To maximize the benefits of preventive intervention, it is vital to identify the time interval during which the initiation of a preventive intervention will lead to an optimal outcome. The goal of the present study was to determine if oncogenic events can be detected in the nonneoplastic colonic mucosa of Apc+/Min-FCCC mice prior to formation of the first adenoma, thus defining an earlier point of intervention along the cancer continuum. Tissues taken at three potential points of intervention were characterized: prior to Apc mutation (wild type Apc+/+-FCCC mice); after initiation but prior to colon adenoma formation (tumor-free Apc+/Min-FCCC mice); and after formation of the first colon adenoma (tumor-bearing Apc+/Min-FCCC mice). Experimentation focused on molecular processes that are dysregulated in early colon lesions: 1) cellular proliferation (proliferative index and size of the proliferative zone); 2) cellular stemness (expression of Ascl2, Grem1, Lgr5 and Muc2); 3) EGFR signaling (expression of Ereg); and 4) inflammation (expression of Mmp9, Ptsg2, and Reg4, as well as secretion of 18 cytokines involved in immune activation and response). Interestingly, the nonneoplastic colonic mucosa of wild type, tumor-free Apc+/Min-FCCC , and tumor-bearing Apc+/Min-FCCC mice did not display significant differences in average epithelial cell proliferation (fold change 0.8-1.3, p≥0.11), mucosal gene expression (fold change 0.8-1.4, p≥0.22), or secretion of specific cytokines from colonic mucosa (fold change 0.2-1.5, p≥0.06). However, the level of cytokine secretion was highly variable, with many (22% of wild type, 31% of tumor-free Apc+/Min-FCCC , and 31% of tumor-bearing Apc+/Min-FCCC ) mice categorized as outliers (> 1.5 x interquartile ranges below the first quartile or above the third quartile) due to elevated expression of at least one cytokine. In summary, no differences were observed in proliferation, stemness, and EGFR signaling in the colonic mucosa of wild type vs Apc+/Min-FCCC mice, with low baseline cytokine expression, prior to the formation of the first colon adenoma. The results of this study provide valuable baseline data to inform the design of future cancer prevention studies.

Combined Inhibition of AKT and KIT Restores Expression of Programmed Cell Death 4 (PDCD4) in Gastrointestinal Stromal Tumor.

The majority of gastrointestinal stromal tumor (GIST) patients develop resistance to the first-line KIT inhibitor, imatinib mesylate (IM), through acquisition of secondary mutations in KIT or bypass signaling pathway activation. In addition to KIT, AKT is a relevant target for inhibition, since the PI3K/AKT pathway is crucial for IM-resistant GIST survival. We evaluated the activity of a novel pan-AKT inhibitor, MK-4440 (formerly ARQ 751), as monotherapy and in combination with IM in GIST cell lines and preclinical models with varying IM sensitivities. Dual inhibition of KIT and AKT demonstrated synergistic effects in IM-sensitive and -resistant GIST cell lines. Proteomic analyses revealed upregulation of the tumor suppressor, PDCD4, in combination treated cells. Enhanced PDCD4 expression correlated to increased cell death. In vivo studies revealed superior efficacy of MK-4440/IM combination in an IM-sensitive preclinical model of GIST compared with either single agent. The combination demonstrated limited efficacy in two IM-resistant models, including a GIST patient-derived xenograft model possessing an exon 9 KIT mutation. These studies provide strong rationale for further use of AKT inhibition in combination with IM in primary GIST; however, alternative agents will need to be tested in combination with AKT inhibition in the resistant setting.

A statistical framework for non-negative matrix factorization based on generalized dual divergence.

A statistical framework for non-negative matrix factorization based on generalized dual Kullback-Leibler divergence, which includes members of the exponential family of models, is proposed. A family of algorithms is developed using this framework, including under sparsity constraints, and its convergence proven using the Expectation-Maximization algorithm. The framework generalizes some existing methods for different noise structures and contrasts with the recently developed quasi-likelihood approach, thus providing a useful alternative for non-negative matrix factorization. A measure to evaluate the goodness-of-fit of the resulting factorization is described. The performance of the proposed methods is evaluated extensively using real life and simulated data and their utility in unsupervised and semi-supervised learning is illustrated using an application in cancer genomics. This framework can be viewed from the perspective of reinforcement learning, and can be adapted to incorporate discriminant functions and multi-layered neural networks within a deep learning paradigm.

Predicting positive peritoneal cytology in pancreatic cancer.

INTRODUCTION: Positive cytology from peritoneal washings obtained prior to potential resection of pancreatic cancer is associated with grim prognosis, equivalent to M1 disease. We examine our experience with pancreatic cancer patients who underwent pre-resection lavage in an attempt to predict who would have malignant cells on peritoneal cytology. METHODS: We conducted a retrospective review of patients undergoing pancreatectomy for pancreatic adenocarcinoma at a tertiary care institution from 1995 to 2019 and had pre-resection lavage performed. Demographic and clinicopathologic data were collected. Logistic regression models were used to identify predictors of positive cytology. RESULTS: Three hundred ninety-nine patients underwent pancreatic resection and had lavage performed. Forty-three (10.8%) had positive peritoneal cytology. Those with positive cytology had higher median Ca19-9 value than those with negative cytology at diagnosis (368.5 vs 200 U/mL, p = 0.007) and after neoadjuvant therapy (100.3 vs 43 U/mL, p = 0.013). After controlling for preoperative therapy received, an initial Ca19-9 greater than 1220 U/mL (OR 2.72, 95% CI 1.07-6.89, p = 0.035), locally advanced disease (OR 4.86, 95% CI 1.31-18.09, p = 0.018), and BMI ≥ 25 kg/m2 (OR 2.67, 95% CI 1.04-6.97, p = 0.042) were associated with positive cytology in multivariate logistic regression model. The associated ROC curve had an AUC of 0.7507, suggesting adequate discrimination of those with positive peritoneal cytology. CONCLUSION: Diagnostic laparoscopy remains an important adjunct to the workup, diagnosis, and staging of pancreatic adenocarcinoma. Patients with locally advanced disease, significantly elevated serum Ca19-9 at diagnosis, and BMI ≥ 25 kg/m2 may be at higher risk for positive peritoneal cytology, regardless of whether neoadjuvant therapy is administered.

Assessing Methods for Evaluating the Number of Components in Non-Negative Matrix Factorization.

Non-negative matrix factorization is a relatively new method of matrix decomposition which factors an m×n data matrix X into an m×k matrix W and a k×n matrix H, so that X≈W×H. Importantly, all values in X, W, and H are constrained to be non-negative. NMF can be used for dimensionality reduction, since the k columns of W can be considered components into which X has been decomposed. The question arises: how does one choose k? In this paper, we first assess methods for estimating k in the context of NMF in synthetic data. Second, we examine the effect of normalization on this estimate's accuracy in empirical data. In synthetic data with orthogonal underlying components, methods based on PCA and Brunet's Cophenetic Correlation Coefficient achieved the highest accuracy. When evaluated on a well-known real dataset, normalization had an unpredictable effect on the estimate. For any given normalization method, the methods for estimating k gave widely varying results. We conclude that when estimating k, it is best not to apply normalization. If underlying components are known to be orthogonal, then Velicer's MAP or Minka's Laplace-PCA method might be best. However, when orthogonality of the underlying components is unknown, none of the methods seemed preferable.

The Cancer Prevention Project of Philadelphia: preliminary findings examining diversity among the African diaspora.

Objective: Cancer mortality inequity among persons of African Ancestry is remarkable. Yet, Black inclusion in cancer biology research is sorely lacking and warrants urgent attention. Epidemiologic research linking African Ancestry and the African Diaspora to disease susceptibility and outcomes is critical for understanding the significant and troubling health disparities among Blacks. Therefore, in a cohort of diverse Blacks, this study examined differences in genetic ancestry informative markers (AIMs) in the DNA repair pathway and the cancer related biomarker 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL).Methods: Participants completed a questionnaire and provided bio-specimens. AIMs in or around DNA repair pathway genes were analyzed to assess differences in minor allele frequency (MAF) across the 3 ethnic subgroups. NNAL concentration in urine was measured among current smokers.Results: To date the cohort includes 852 participants, 88.3% being Black. Of the 752 Blacks, 51.3% were US-born, 27.8% were Caribbean-born, and 19.6% were Africa-born. Current and former smokers represented 14.9% and 10.0%, respectively. US-born Blacks were more likely to be smokers and poor metabolizers of NNAL. Two-way hierarchical clustering revealed MAF of AIMs differed across the 3 ethnic subgroups.Conclusion: Our findings are consistent with the emerging literature demonstrating Black heterogeneity underscoring African Ancestry genetic subgroup differences - specifically relevant to cancer. Further investigations, with data harmonization and sharing, are urgently needed to begin to map African Ancestry cancer biomarkers as well as race, and race by place\region comparative biomarkers to inform cancer prevention and treatment in the era of precision medicine.

Supervised Dimension Reduction for Large-Scale "Omics" Data With Censored Survival Outcomes Under Possible Non-Proportional Hazards.

The past two decades have witnessed significant advances in high-throughput "omics" technologies such as genomics, proteomics, metabolomics, transcriptomics and radiomics. These technologies have enabled simultaneous measurement of the expression levels of tens of thousands of features from individual patient samples and have generated enormous amounts of data that require analysis and interpretation. One specific area of interest has been in studying the relationship between these features and patient outcomes, such as overall and recurrence-free survival, with the goal of developing a predictive "omics" profile. Large-scale studies often suffer from the presence of a large fraction of censored observations and potential time-varying effects of features, and methods for handling them have been lacking. In this paper, we propose supervised methods for feature selection and survival prediction that simultaneously deal with both issues. Our approach utilizes continuum power regression (CPR) - a framework that includes a variety of regression methods - in conjunction with the parametric or semi-parametric accelerated failure time (AFT) model. Both CPR and AFT fall within the linear models framework and, unlike black-box models, the proposed prognostic index has a simple yet useful interpretation. We demonstrate the utility of our methods using simulated and publicly available cancer genomics data.

Identification of Wee1 as a target in combination with avapritinib for gastrointestinal stromal tumor treatment.

Management of gastrointestinal stromal tumors (GISTs) has been revolutionized by the identification of activating mutations in KIT and PDGFRA and clinical application of RTK inhibitors in advanced disease. Stratification of GISTs into molecularly defined subsets provides insight into clinical behavior and response to approved targeted therapies. Although these RTK inhibitors are effective in most GISTs, resistance remains a significant clinical problem. Development of effective treatment strategies for refractory GISTs requires identification of novel targets to provide additional therapeutic options. Global kinome profiling has the potential to identify critical signaling networks and reveal protein kinases essential in GISTs. Using multiplexed inhibitor beads and mass spectrometry, we explored the majority of the kinome in GIST specimens from the 3 most common molecular subtypes (KIT mutant, PDGFRA mutant, and succinate dehydrogenase deficient) to identify kinase targets. Kinome profiling with loss-of-function assays identified an important role for G2/M tyrosine kinase, Wee1, in GIST cell survival. In vitro and in vivo studies revealed significant efficacy of MK-1775 (Wee1 inhibitor) in combination with avapritinib in KIT mutant and PDGFRA mutant GIST cell lines as well as notable efficacy of MK-1775 as a monotherapy in the engineered PDGFRA mutant line. These studies provide strong preclinical justification for the use of MK-1775 in GIST.

Netrin G1 Promotes Pancreatic Tumorigenesis through Cancer-Associated Fibroblast-Driven Nutritional Support and Immunosuppression.

Pancreatic ductal adenocarcinoma (PDAC) has a poor 5-year survival rate and lacks effective therapeutics. Therefore, it is of paramount importance to identify new targets. Using multiplex data from patient tissue, three-dimensional coculturing in vitro assays, and orthotopic murine models, we identified Netrin G1 (NetG1) as a promoter of PDAC tumorigenesis. We found that NetG1+ cancer-associated fibroblasts (CAF) support PDAC survival, through a NetG1-mediated effect on glutamate/glutamine metabolism. Also, NetG1+ CAFs are intrinsically immunosuppressive and inhibit natural killer cell-mediated killing of tumor cells. These protumor functions are controlled by a signaling circuit downstream of NetG1, which is comprised of AKT/4E-BP1, p38/FRA1, vesicular glutamate transporter 1, and glutamine synthetase. Finally, blocking NetG1 with a neutralizing antibody stunts in vivo tumorigenesis, suggesting NetG1 as potential target in PDAC. SIGNIFICANCE: This study demonstrates the feasibility of targeting a fibroblastic protein, NetG1, which can limit PDAC tumorigenesis in vivo by reverting the protumorigenic properties of CAFs. Moreover, inhibition of metabolic proteins in CAFs altered their immunosuppressive capacity, linking metabolism with immunomodulatory function.See related commentary by Sherman, p. 230.This article is highlighted in the In This Issue feature, p. 211.

Interactive effect of TLR SNPs and exposure to sexually transmitted infections on prostate cancer risk in Jamaican men.

BACKGROUND: Prostate cancer (PC) risk increases with African ancestry and a history of sexually transmitted infections (STIs). Also, single-nucleotide polymorphisms (SNPs) in toll-like receptor (TLR) genes influence PC risk. This pilot study explores interactions between STIs and TLR-related SNPs in relation to PC risk among Jamaican men. METHODS: This case-control study evaluates two TLR related SNPs in 356 Jamaican men (194 controls and 162 cases) with or without history of STIs using stepwise penalized logistic regression in multivariable analyses. RESULTS: Age (odds ratio [OR] = 1.08; 95% confidence interval [CI]: 1.04-1>.12; p < .001) and IRF3_rs2304206 GG genotype (OR = 0.47; 95% CI: 0.29-0<.78; p = .003) modulated PC risk in people with history of STIs. In the population with no history of STIs, resulting interactions between risk factors did not survive correction for multiple hypothesis testing. CONCLUSION: Overall, an interaction between the IFR3_rs2304206 variant and a history of exposure to STIs leads to greater decrease of PC risk than the presence of polymorphic genotype alone. These findings are suggestive and require further validation. Identification of gene variants along with detection of lifestyle behaviors may contribute to identification of men at a greater risk of PC development in the population.

Unified methods for feature selection in large-scale genomic studies with censored survival outcomes.

MOTIVATION: One of the major goals in large-scale genomic studies is to identify genes with a prognostic impact on time-to-event outcomes which provide insight into the disease process. With rapid developments in high-throughput genomic technologies in the past two decades, the scientific community is able to monitor the expression levels of tens of thousands of genes and proteins resulting in enormous datasets where the number of genomic features is far greater than the number of subjects. Methods based on univariate Cox regression are often used to select genomic features related to survival outcome; however, the Cox model assumes proportional hazards (PH), which is unlikely to hold for each feature. When applied to genomic features exhibiting some form of non-proportional hazards (NPH), these methods could lead to an under- or over-estimation of the effects. We propose a broad array of marginal screening techniques that aid in feature ranking and selection by accommodating various forms of NPH. First, we develop an approach based on Kullback-Leibler information divergence and the Yang-Prentice model that includes methods for the PH and proportional odds (PO) models as special cases. Next, we propose R2 measures for the PH and PO models that can be interpreted in terms of explained randomness. Lastly, we propose a generalized pseudo-R2 index that includes PH, PO, crossing hazards and crossing odds models as special cases and can be interpreted as the percentage of separability between subjects experiencing the event and not experiencing the event according to feature measurements. RESULTS: We evaluate the performance of our measures using extensive simulation studies and publicly available datasets in cancer genomics. We demonstrate that the proposed methods successfully address the issue of NPH in genomic feature selection and outperform existing methods. AVAILABILITY AND IMPLEMENTATION: R code for the proposed methods is available at github.com/lburns27/Feature-Selection. CONTACT: karthik.devarajan@fccc.edu. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.