Two-sample test with g-modeling and its applications.

Many real data analyses involve two-sample comparisons in location or in distribution. Most existing methods focus on problems where observations are independently and identically distributed in each group. However, in some applications the observed data are not identically distributed but associated with some unobserved parameters which are identically distributed. To address this challenge, we propose a novel two-sample testing procedure as a combination of the g $$ g $$ -modeling density estimation introduced by Efron and the two-sample Kolmogorov-Smirnov test. We also propose efficient bootstrap algorithms to estimate the statistical significance for such tests. We demonstrate the utility of the proposed approach with two biostatistical applications: the analysis of surgical nodes data with binomial model and differential expression analysis of single-cell RNA sequencing data with zero-inflated Poisson model.

Genetic analysis of sinonasal undifferentiated carcinoma discovers recurrent SWI/SNF alterations and a novel PGAP3-SRPK1 fusion gene.

BACKGROUND: Sinonasal Undifferentiated Carcinoma (SNUC) is a rare and aggressive skull base tumor with poor survival and limited treatment options. To date, targeted sequencing studies have identified IDH2 and SMARCB1 as potential driver alterations, but the molecular alterations found in SMARCB1 wild type tumors are unknown. METHODS: We evaluated survival outcomes in a cohort of 46 SNUC patients treated at an NCI designated cancer center and identify clinical and disease variables associated with survival on Kaplan-Meier and Cox multivariate survival analysis. We performed exome sequencing to characterize a series of SNUC tumors (n = 5) and cell line (MDA8788-6) to identify high confidence mutations, copy number alterations, microsatellite instability, and fusions. Knockdown studies using siRNA were utilized for validation of a novel PGAP3-SRPK1 gene fusion. RESULTS: Overall survival analysis revealed no significant difference in outcomes between patients treated with surgery +/- CRT and CRT alone. Tobacco use was the only significant predictor of survival. We also confirmed previously published findings on IDH and SMARC family mutations and identified novel recurrent aberrations in the JAK/STAT and PI3K pathways. We also validated a novel PGAP3-SRPK1 gene fusion in the SNUC cell line, and show that knockdown of the fusion is negatively associated with EGFR, E2F and MYC signaling. CONCLUSION: Collectively, these data demonstrate recurrent alterations in the SWI/SNF family as well as IDH, JAK/STAT, and PI3K pathways and discover a novel fusion gene (PGAP3-SRPK1). These data aim to improve understanding of possible driver mutations and guide future therapeutic strategies for this disease.

Rationale for Using Irreversible Epidermal Growth Factor Receptor Inhibitors in Combination with Phosphatidylinositol 3-Kinase Inhibitors for Advanced Head and Neck Squamous Cell Carcinoma.

Head and neck squamous cell carcinoma (HNSCC) is a common and debilitating form of cancer characterized by poor patient outcomes and low survival rates. In HNSCC, genetic aberrations in phosphatidylinositol 3-kinase (PI3K) and epidermal growth factor receptor (EGFR) pathway genes are common, and small molecules targeting these pathways have shown modest effects as monotherapies in patients. Whereas emerging preclinical data support the combined use of PI3K and EGFR inhibitors in HNSCC, in-human studies have displayed limited clinical success so far. Here, we examined the responses of a large panel of patient-derived HNSCC cell lines to various combinations of PI3K and EGFR inhibitors, including EGFR agents with varying specificity and mechanistic characteristics. We confirmed the efficacy of PI3K and EGFR combination therapies, observing synergy with α isoform-selective PI3K inhibitor HS-173 and irreversible EGFR/ERBB2 dual inhibitor afatinib in most models tested. Surprisingly, however, our results demonstrated only modest improvement in response to HS-173 with reversible EGFR inhibitor gefitinib. This difference in efficacy was not explained by differences in ERBB target selectivity between afatinib and gefitinib; despite effectively disrupting ERBB2 phosphorylation, the addition of ERBB2 inhibitor CP-724714 failed to enhance the effect of HS-173 gefitinib dual therapy. Accordingly, although irreversible ERBB inhibitors showed strong synergistic activity with HS-173 in our models, none of the reversible ERBB inhibitors were synergistic in our study. Therefore, our results suggest that the ERBB inhibitor mechanism of action may be critical for enhanced synergy with PI3K inhibitors in HNSCC patients and motivate further preclinical studies for ERBB and PI3K combination therapies.

Associations between mixtures of urinary phthalate metabolites with gestational age at delivery: a time to event analysis using summative phthalate risk scores.

BACKGROUND: Preterm birth is a significant public health concern and exposure to phthalates has been shown to be associated with an increased odds of preterm birth. Even modest reductions in gestational age at delivery could entail morbid consequences for the neonate and analyzing data with this additional information may be useful. In the present analysis, we consider gestational age at delivery as our outcome of interest and examine associations with multiple phthalates. METHODS: Women were recruited early in pregnancy as part of a prospective, longitudinal birth cohort at the Brigham and Women's Hospital in Boston, Massachusetts. Urine samples were collected at up to four time points during gestation for urinary phthalate metabolite measurement, and birth outcomes were recorded at delivery. From this population, we selected all 130 cases of preterm birth (< 37 weeks of gestation) as well as 352 random controls. We conducted analysis with both geometric average of the exposure concentrations across the first three visits as well as using repeated measures of the exposure. Two different time to event models were used to examine associations between nine urinary phthalate metabolite concentrations and time to delivery. Two different approaches to constructing a summative phthalate risk score were also considered. RESULTS: The single-pollutant analysis using a Cox proportional hazards model showed the strongest association with a hazard ratio (HR) of 1.21 (95% confidence interval (CI): 1.09, 1.33) per interquartile range (IQR) change in average log-transformed mono-2-ethyl-5-carboxypentyl phthalate (MECPP) concentration. Using the accelerated failure time model, we observed a 1.19% (95% CI: 0.26, 2.11%) decrease in gestational age in association with an IQR change in average log-transformed MECPP. We next examined associations with an environmental risk score (ERS). The fourth quartile of ERS was significantly associated with a HR of 1.44 (95% CI: 1.19, 1.75) and a reduction of 2.55% (95% CI: 0.76, 4.30%) in time to delivery (in days) compared to the first quartile. CONCLUSIONS: On average, pregnant women with higher urinary metabolite concentrations of individual phthalates have shorter time to delivery. The strength of the observed associations are amplified with the risk scores when compared to individual pollutants.

Testing the performance of a prototype lateral flow device using bronchoalveolar lavage fluid for the diagnosis of invasive pulmonary aspergillosis in high-risk patients.

The diagnosis of invasive pulmonary aspergillosis (IPA) increasingly relies on non-culture-based biomarkers in bronchoalveolar lavage (BAL) fluid. The Aspergillus lateral flow device (LFD) is a rapid immunoassay that uses a novel Aspergillus monoclonal antibody to gain specificity. The objective of the study is to compare specificity and sensitivity of the prototype LFD and the galactomannan (GM) enzyme immunoassay in BAL fluid in high-risk patients. A total of 114 BAL samples from 106 patients at high risk for IPA were studied: 8 patients had proven/probable IPA, 16 had possible IPA and 82 did not have IPA. In patients with proven/probable IPA, specificity of LFD was 94% and GM was 89%; sensitivity of LFD was 38% and GM was 75%. Negative predictive value (NPV) for LFD was 94% and for GM was 98%; positive predictive value (PPV) was 38% for both tests. The use of anti-mould prophylaxis did not affect specificity but resulted in decreased NPV of both LFD and GM. Union and intersection analysis showed no improvement in the performance by using both tests. Among patients at risk for IPA, the diagnostic performance of LFD and GM in BAL fluid appears comparable; specificity is high, but sensitivity of both LFD and GM is poor.

A Genome Wide CRISPR Profiling Approach Identifies Mechanisms of Cisplatin Resistance in Head and Neck Squamous Cell Carcinoma.

Background: Head and neck squamous cell carcinoma (HNSCC) is a lethal disease with poor survival rates, especially for cancers arising in the oral cavity or larynx. Cisplatin is a key chemotherapeutic for HNSCC; however poor survival rates may be partially due to cisplatin resistance observed in some HNSCCs. Here, we examined the utility of genome-wide CRISPR knockout profiling for nominating pivotal mechanisms of cisplatin resistance in HNSCC models. Methods: We characterized the cisplatin sensitivity of 18 HNSCC cell lines. Next, we used a genome-wide CRISPR/Cas9 library to identify genes involved in cisplatin resistance. We next performed validation assays in the UM-SCC-49 cell line model. Results: Our data prioritized 207 genes as pivotal for cisplatin resistance in HNSCC, including novel genes VGLL3, CIRHA1, NCOR1, SPANXA1, MAP2K7, ULK1, and CDK16. Gene set enrichment analysis identified several NOTCH family genes comprising the top pathway driving cisplatin resistance, which we then validated using a targeted NOTCH1 knockout model. Interestingly, we noted that HNSCC models with natural NOTCH pathway alterations including single allele mutations and/or frameshift alterations had diverse responses to cisplatin treatment suggesting that complex and multi-faceted mechanisms contribute to cisplatin resistance in HNSCC. Conclusions: Collectively, our study validates a genome-wide CRISPR/Cas9 approach for the discovery of resistance mechanisms in HNSCC, adds to the growing evidence that NOTCH1 status should be evaluated as a biomarker of cisplatin response and provides a framework for future work aimed at overcoming cisplatin resistance.

Reconstruction of Single-Cell Trajectories Using Stochastic Tree Search.

The recent advancement in single-cell RNA sequencing technologies enables the understanding of dynamic cellular processes at the single-cell level. Using trajectory inference methods, pseudotimes can be estimated based on reconstructed single-cell trajectories which can be further used to gain biological knowledge. Existing methods for modeling cell trajectories, such as minimal spanning tree or k-nearest neighbor graph, often lead to locally optimal solutions. In this paper, we propose a penalized likelihood-based framework and introduce a stochastic tree search (STS) algorithm aiming at the global solution in a large and non-convex tree space. Both simulated and real data experiments show that our approach is more accurate and robust than other existing methods in terms of cell ordering and pseudotime estimation.

Multi-kinase compensation rescues EGFR knockout in a cell line model of head and neck squamous cell carcinoma.

BACKGROUND: Head and neck squamous cell carcinoma (HNSCC) is a debilitating disease with poor survival rates. While the epidermal growth factor receptor (EGFR)-targeting antibody Cetuximab is approved for treatment, responses are limited and the molecular mechanisms driving resistance remain incompletely understood. METHODS: To better understand how cells survive without EGFR activity, we developed an EGFR knockout derivative of the UM-SCC-92 cell line using CRISPR/Cas9 technology. We then characterized changes to the transcriptome with RNAseq and changes in response to kinase inhibitors with resazurin cell viability assays. Finally, we tested if inhibitors with activity in the EGFR knockout model also had synergistic activity in combination with EGFR inhibitors in either wild type UM-SCC-92 cells or a known Cetuximab-resistant model. RESULTS: Functional and molecular analysis showed that knockout cells had decreased cell proliferation, upregulation of FGFR1 expression, and an enhanced mesenchymal phenotype. In fact, expression of common EMT genes including VIM, SNAIL1, ZEB1 and TWIST1 were all upregulated in the EGFR knockout. Surprisingly, EGFR knockout cells were resistant to FGFR inhibitor monotherapies, but sensitive to combinations of FGFR and either XIAP or IGF-1R inhibitors. Accordingly, both wild type UM-SCC-92 and Cetuximab-resistant UM-SCC-104 cells with were sensitive to combined inhibition of EGFR, FGFR and either XIAP or IGF-1R. CONCLUSIONS: These data offer insights into EGFR inhibitor resistance and show that resistance to EGFR knockout likely occurs through a complex network of kinases. Future studies of cetuximab-resistant HNSCC tumors are warranted to determine if this EMT phenotype and/or multi-kinase resistance is observed in patients.

Advancement of PI3 Kinase Inhibitor Combination Therapies for PI3K-Aberrant Chordoma.

Objectives Targeted inhibitors of the PI3 kinase (PI3K) pathway have shown promising but incomplete antitumor activity in preclinical chordoma models. The aim of this study is to advance methodology for a high-throughput drug screen using chordoma models to identify new combination therapies for chordoma. Study Design Present work is an in vitro study. Setting The study conducted at an academic research laboratory. Materials and Methods An in vitro study on automated high-throughput screening of chordoma cells was performed using a library of 1,406 drugs as both mono- and combination therapies with PI3K inhibitors. Combination indices were determined for dual therapies and synergistic outliers were identified as potential therapeutic agents. T (brachyury) siRNA knockdown in combination with PI3K pathway inhibition was also assessed. Results Fifty-nine combination therapies were identified as having potential therapeutic efficacy. Effective combinations included PI3K inhibitors with GSK1838705A (ALK/IGF-1R inhibitor), LY2874455 (VEGFR/FGFR inhibitor), El1 (selective Ezh2 inhibitor), and (-)-p-bromotetramisole oxalate (alkaline phosphatase inhibitor). The top ranking targets identified included ALK, PDGFR, VEGFR, aurora kinase, and BCL-2. T (brachyury) inhibition produced significant reduction in cell viability and growth; however PI3K inhibition in combination with T (brachyury) knockdown did not result in further reduction in growth and viability in vitro. Conclusion High throughput with in vitro combination screening is feasible with chordoma cells and allows for rapid identification of synergistic dual-therapies. Potential combination therapies and targetable pathways were identified. T (brachyury) knockdown produced significant reduction in cell viability, but did not show additional benefit with PI3K pathway inhibition in this model. Further in vitro and in vivo validation of these therapeutic combinations is warranted.

Small molecule profiling to define synergistic EGFR inhibitor combinations in head and neck squamous cell carcinoma.

BACKGROUND: Head and neck squamous cell carcinoma (HNSCC) is a debilitating disease with poor survival. Although epidermal growth factor receptor (EGFR)-targeting antibody cetuximab improves survival in some settings, responses are limited suggesting that alternative approaches are needed. METHODS: We performed a high throughput drug screen to identify EGFR inhibitor-based synergistic combinations of clinically advanced inhibitors in models resistant to EGFR inhibitor monotherapies, and then performed downstream validation experiments on prioritized synergistic combinations. RESULTS: From our screen, we re-discovered known synergistic EGFR inhibitor combinations with FGFR or IGF-1R inhibitors that were broadly effective and also discovered novel synergistic combinations with XIAP inhibitor and DNMT inhibitors that were effective in only a subset of models. CONCLUSIONS: Conceptually, our data identify novel synergistic combinations that warrant evaluation in future studies, and suggest that some combinations, although highly synergistic, will require parallel companion diagnostic development to be effectively advanced in patients.

Microbe-Mediated Activation of Toll-like Receptor 2 Drives PDL1 Expression in HNSCC.

As immunotherapies targeting the PDL1 checkpoint have become a mainstay of treatment for a subset of head and neck squamous cell carcinoma (HNSCC) patients, a detailed understanding of the mechanisms underlying PDL1-mediated immune evasion is needed. To elucidate factors regulating expression of PDL1 in HNSCC cells, a genome-wide CRISPR profiling approach was implemented to identify genes and pathways conferring altered PDL1 expression in an HNSCC cell line model. Our screen nominated several candidate PDL1 drivers, including Toll-like Receptor 2 (TLR2). Depletion of TLR2 blocks interferon-γ-induced PDL1 expression, and stimulation of TLR2 with either Staphylococcus aureus or a bacterial lipopeptide mimetic, Pam3CSK4, enhanced PDL1 expression in multiple models. The data herein demonstrate a role for TLR2 in modulating the expression of PDL1 in HNSCC models and suggest that microbiota may directly modulate immunosuppression in cancer cells. Our study represents a step toward disentangling the diverse pathways and stimuli regulating PDL1 expression in HNSCC and underscores a need for future work to characterize the complex microbiome in HNSCC patients treated with immunotherapy.

A conserved intratumoral regulatory T cell signature identifies 4-1BB as a pan-cancer target.

Despite advancements in targeting the immune checkpoints program cell death protein 1 (PD-1), programmed death ligand 1 (PD-L1), and cytotoxic T lymphocyte-associated protein 4 (CTLA-4) for cancer immunotherapy, a large number of patients and cancer types remain unresponsive. Current immunotherapies focus on modulating an antitumor immune response by directly or indirectly expanding antitumor CD8 T cells. A complementary strategy might involve inhibition of Tregs that otherwise suppress antitumor immune responses. Here, we sought to identify functional immune molecules preferentially expressed on tumor-infiltrating Tregs. Using genome-wide RNA-Seq analysis of purified Tregs sorted from multiple human cancer types, we identified a conserved Treg immune checkpoint signature. Using immunocompetent murine tumor models, we found that antibody-mediated depletion of 4-1BB-expressing cells (4-1BB is also known as TNFRSF9 or CD137) decreased tumor growth without negatively affecting CD8 T cell function. Furthermore, we found that the immune checkpoint 4-1BB had a high selectivity for human tumor Tregs and was associated with worse survival outcomes in patients with multiple tumor types. Thus, antibody-mediated depletion of 4-1BB-expressing Tregs represents a strategy with potential activity across cancer types.

Validation of a serum 4-microRNA signature for the detection of lung cancer.

BACKGROUND: Our previous studies have identified a serum-based 4-microRNA (4-miRNA) signature that may help distinguish patients with lung cancer (LC) from non-cancer controls (NCs). Here, we used an extended independent cohort of 398 subjects to further validate the diagnostic ability of this 4-miRNA signature. METHODS: Using quantitative reverse transcription polymerase chain reaction (qRT-PCR), expression of the 4-miRNAs was assessed in a total of 398 sera that included 213 LC patients and 185 NCs. A logistic regression model using training-test sets, receiver operating characteristic (ROC) curve analysis and t-test were used to test the impact of varying expression of these miRNAs on its diagnostic accuracy for LC. The cell proliferation and colony formation affected by these miRNAs, as well as gene ontology (GO) analysis of miRNA target genes were performed. RESULTS: The levels of the 4-miRNAs were significantly higher in the serum of patients with LCs as compared to NCs. Using a logistic regression prediction model based on training and test sets analysis, we obtained the area under the curve (AUC) of 0.921 [95% confidence interval (CI), 0.876-0.966] on the test set with specificity 90.6%, sensitivity 77.9%, accuracy 84.1%, positive predictive value (PPV) 89.8% and negative predictive value (NPV) 79.5%. CONCLUSIONS: We have verified that this serum 4-miRNA signature could provide a promising noninvasive biomarker for the prediction of LC, particularly in patients with indeterminate lung nodules on screening CT scans.