Systemic inflammation, the peripheral blood transcriptome and primary melanoma.

Peripheral blood transcriptomes from 383 newly-diagnosed melanoma patients were subjected to differential gene expression analysis. The hypotheses were that impaired systemic immunity is associated with poorer prognosis (thicker tumors and fewer tumor infiltrating lymphocytes (TILs)) and evidence of systemic inflammation (high-sensitivity C-reactive protein (hsCRP) and fibrinogen levels). Higher fibrinogen levels were associated with thicker primaries. In single gene analysis hsCRP levels were significantly associated with higher blood CD274 expression, (coding for PD-L1), but each was independently prognostic, with hsCRP associated with increased mortality, and higher CD274 protective, independent of age. Pathway analysis identified downregulation of immune cell signalling pathways in the blood of people with thicker tumors and notable upregulation of STAT1 in people with brisk TILs. Transcriptomic data provided evidence for increased NFB signalling with higher inflammatory markers but with reduction in expression of HLA class II molecules and higher CD274 suggesting that aberrant systemic inflammation is a significant mediator of reduced immune function in melanoma. In summary, transcriptomic data revealed evidence of reduced immune function in patients with thicker tumors and fewer TILs, at diagnosis. Inflammatory markers were associated with thicker primaries and independently with death from melanoma suggesting that systemic inflammation contributes to that reduced immune function.

Immune subtyping of melanoma whole slide images using multiple instance learning.

Determining early-stage prognostic markers and stratifying patients for effective treatment are two key challenges for improving outcomes for melanoma patients. Previous studies have used tumour transcriptome data to stratify patients into immune subgroups, which were associated with differential melanoma specific survival and potential predictive biomarkers. However, acquiring transcriptome data is a time-consuming and costly process. Moreover, it is not routinely used in the current clinical workflow. Here, we attempt to overcome this by developing deep learning models to classify gigapixel haematoxylin and eosin (H&E) stained pathology slides, which are well established in clinical workflows, into these immune subgroups. We systematically assess six different multiple instance learning (MIL) frameworks, using five different image resolutions and three different feature extraction methods. We show that pathology-specific self-supervised models using 10x resolution patches generate superior representations for the classification of immune subtypes. In addition, in a primary melanoma dataset, we achieve a mean area under the receiver operating characteristic curve (AUC) of 0.80 for classifying histopathology images into 'high' or 'low immune' subgroups and a mean AUC of 0.82 in an independent TCGA melanoma dataset. Furthermore, we show that these models are able to stratify patients into 'high' and 'low immune' subgroups with significantly different melanoma specific survival outcomes (log rank test, P< 0.005). We anticipate that MIL methods will allow us to find new biomarkers of high importance, act as a tool for clinicians to infer the immune landscape of tumours and stratify patients, without needing to carry out additional expensive genetic tests.

Pro- and anti-tumour activities of CD146/MCAM in breast cancer result from its heterogeneous expression and association with epithelial to mesenchymal transition.

CD146, also known as melanoma cell adhesion molecule (MCAM), is expressed in numerous cancers and has been implicated in the regulation of metastasis. We show that CD146 negatively regulates transendothelial migration (TEM) in breast cancer. This inhibitory activity is reflected by a reduction in MCAM gene expression and increased promoter methylation in tumour tissue compared to normal breast tissue. However, increased CD146/MCAM expression is associated with poor prognosis in breast cancer, a characteristic that is difficult to reconcile with inhibition of TEM by CD146 and its epigenetic silencing. Single cell transcriptome data revealed MCAM expression in multiple cell types, including the malignant cells, tumour vasculature and normal epithelium. MCAM expressing malignant cells were in the minority and expression was associated with epithelial to mesenchymal transition (EMT). Furthermore, gene expression signatures defining invasiveness and a stem cell-like phenotype were most strongly associated with mesenchymal-like tumour cells with low levels of MCAM mRNA, likely to represent a hybrid epithelial/mesenchymal (E/M) state. Our results show that high levels of MCAM gene expression are associated with poor prognosis in breast cancer because they reflect tumour vascularisation and high levels of EMT. We suggest that high levels of mesenchymal-like malignant cells reflect large populations of hybrid E/M cells and that low CD146 expression on these hybrid cells is permissive for TEM, aiding metastasis.

Molecular dynamics simulations reveal membrane lipid interactions of the full-length lymphocyte specific kinase (Lck).

The membrane-bound lymphocyte-specific protein-tyrosine kinase (Lck) triggers T cell antigen receptor signalling to initiate adaptive immune responses. Despite many structure-function studies, the mode of action of Lck and the potential role of plasma membrane lipids in regulating Lck's activity remains elusive. Advances in molecular dynamics simulations of membrane proteins in complex lipid bilayers have opened a new perspective in gathering such information. Here, we have modelled the full-length Lck open and closed conformations  using data available from different crystalographic studies and simulated its interaction with the inner leaflet of the T cell plasma membrane. In both conformations, we found that the unstructured unique domain and the structured domains including the kinase interacted with the membrane with a preference for PIP lipids. Interestingly, our simulations suggest that the Lck-SH2 domain interacts with lipids differently in the open and closed Lck conformations, demonstrating that lipid interaction can potentially regulate Lck's conformation and in turn modulate T cell signalling. Additionally, the Lck-SH2 and kinase domain residues that significantly contacted PIP lipids are found to be conserved among the Src family of kinases, thereby potentially representing similar PIP interactions within the family.

Chemistry Manufacturing and Controls Development, Industry Reflections on Manufacture, and Supply of Pandemic Therapies and Vaccines.

This publication provides some industry reflections on experiences from the Chemistry, Manufacturing, and Controls (CMC) development and manufacture and supply of vaccines and therapies in response to the COVID-19 pandemic. It integrates these experiences with the outcomes from the collaborative work between industry and regulators in recent years on innovative science- and risk-based CMC strategies to the development of new, high-quality products for unmet medical needs. The challenges for rapid development are discussed and various approaches to facilitate accelerated development and global supply are collated for consideration. Relevant regulatory aspects are reviewed, including the role of Emergency Use/Conditional Marketing Authorizations, the dialogue between sponsors and agencies to facilitate early decision-making and alignment, and the value of improving reliance/collaborative assessment and increased collaboration between regulatory authorities to reduce differences in global regulatory requirements. Five areas are highlighted for particular consideration in the implementation of strategies for the quality-related aspects of accelerated development and supply: (1) the substantial need to advance reliance or collaborative assessment; (2) the need for early decision making and streamlined engagement between industry and regulatory authorities on CMC matters; (3) the need to further facilitate 'post-approval' changes; (4) fully exploiting prior and platform knowledge; and (5) review and potential revision of legal frameworks. The recommendations in this publication are intended to contribute to the discussion on approaches that can result in earlier and greater access to high-quality pandemic vaccines and therapies for patients worldwide but could also be useful in general for innovative medicines addressing unmet medical needs.

TGFß limits Myc-dependent TCR-induced metabolic reprogramming in CD8+ T cells.

T cell activation is dependent upon the integration of antigenic, co-stimulatory and cytokine-derived signals and the availability and acquisition of nutrients from the environment. Furthermore, T cell activation is accompanied by reprogramming of cellular metabolism to provide the energy and building blocks for proliferation, differentiation and effector function. Transforming growth factor ß (TGFß) has pleiotropic effects on T cell populations, having both an essential role in the maintenance of immune tolerance but also context-dependent pro-inflammatory functions. We set out to define the mechanisms underpinning the suppressive effects of TGFß on mouse CD8+ T cell activation. RNA-sequencing analysis of TCR-stimulated T cells determined that Myc-regulated genes were highly enriched within gene sets downregulated by TGFß. Functional analysis demonstrated that TGFß impeded TCR-induced upregulation of amino acid transporter expression, amino acid uptake and protein synthesis. Furthermore, TCR-induced upregulation of Myc-dependent glycolytic metabolism was substantially inhibited by TGFß treatment with minimal effects on mitochondrial respiration. Thus, our data suggest that inhibition of Myc-dependent metabolic reprogramming represents a major mechanism underpinning the suppressive effects of TGFß on CD8+ T cell activation.

Oncolytic virus treatment differentially affects the CD56dim and CD56bright NK cell subsets in vivo and regulates a spectrum of human NK cell activity.

Natural killer (NK) cells protect against intracellular infection and cancer. These properties are exploited in oncolytic virus (OV) therapy, where antiviral responses enhance anti-tumour immunity. We have analysed the mechanism by which reovirus, an oncolytic dsRNA virus, modulates human NK cell activity. Reovirus activates NK cells in a type I interferon (IFN-I) dependent manner, inducing STAT1 and STAT4 signalling in both CD56dim and CD56bright NK cell subsets. Gene expression profiling revealed the dominance of IFN-I responses and identified induction of genes associated with NK cell cytotoxicity and cell cycle progression, with distinct responses in the CD56dim and CD56bright subsets. However, reovirus treatment inhibited IL-15 induced NK cell proliferation in an IFN-I dependent manner and was associated with reduced AKT signalling. In vivo, human CD56dim and CD56bright NK cells responded with similar kinetics to reovirus treatment, but CD56bright NK cells were transiently lost from the peripheral circulation at the peak of the IFN-I response, suggestive of their redistribution to secondary lymphoid tissue. Coupled with the direct, OV-mediated killing of tumour cells, the activation of both CD56dim and CD56bright NK cells by antiviral pathways induces a spectrum of activity that includes the NK cell-mediated killing of tumour cells and modulation of adaptive responses via the trafficking of IFN-γ expressing CD56bright NK cells to lymph nodes.

Ulcerated melanoma: Systems biology evidence of inflammatory imbalance towards pro-tumourigenicity.

Microscopic ulceration is an independent predictor of melanoma death. Here, we used systems biology to query the role of host and tumour-specific processes in defining the phenotype. Albumin level as a measure of systemic inflammation was predictive of fewer tumour-infiltrating lymphocytes and poorer survival in the Leeds Melanoma Cohort. Ulcerated melanomas were thicker and more mitotically active (with corresponding transcriptomic upregulated cell cycle pathways). Sequencing identified tumoural p53 and APC mutations, and TUBB2B amplification as associated with the phenotype. Ulcerated tumours had perturbed expression of cytokine genes, consistent with protumourigenic inflammation and histological and transcriptomic evidence for reduced adaptive immune cell infiltration. Pathway/network analysis of multiomic data using neural networks highlighted a role for the ß-catenin pathway in the ulceration, linking genomic changes in the tumour to immunosuppression and cell proliferation. In summary, the data suggest that ulceration is in part associated with genomic changes but that host factors also predict melanoma death with evidence of reduced immune responses to the tumour.

Tumour cell CD99 regulates transendothelial migration via CDC42 and actin remodelling.

Metastasis requires tumour cells to cross endothelial cell (EC) barriers using pathways similar to those used by leucocytes during inflammation. Cell surface CD99 is expressed by healthy leucocytes and ECs, and participates in inflammatory transendothelial migration (TEM). Tumour cells also express CD99, and we have analysed its role in tumour progression and cancer cell TEM. Tumour cell CD99 was required for adhesion to ECs but inhibited invasion of the endothelial barrier and migratory activity. Furthermore, CD99 depletion in tumour cells caused redistribution of the actin cytoskeleton and increased activity of the Rho GTPase CDC42, known for its role in actin remodelling and cell migration. In a xenograft model of breast cancer, tumour cell CD99 expression inhibited metastatic progression, and patient samples showed reduced expression of the CD99 gene in brain metastases compared to matched primary breast tumours. We conclude that CD99 negatively regulates CDC42 and cell migration. However, CD99 has both pro- and anti-tumour activity, and our data suggest that this results in part from its functional linkage to CDC42 and the diverse signalling pathways downstream of this Rho GTPase. This article has an associated First Person interview with the first author of the paper.

Multi-scale simulations of the T cell receptor reveal its lipid interactions, dynamics and the arrangement of its cytoplasmic region.

The T cell receptor (TCR-CD3) initiates T cell activation by binding to peptides of Major Histocompatibility Complexes (pMHC). The TCR-CD3 topology is well understood but the arrangement and dynamics of its cytoplasmic tails remains unknown, limiting our grasp of the signalling mechanism. Here, we use molecular dynamics simulations and modelling to investigate the entire TCR-CD3 embedded in a model membrane. Our study demonstrates conformational changes in the extracellular and transmembrane domains, and the arrangement of the TCR-CD3 cytoplasmic tails. The cytoplasmic tails formed highly interlaced structures while some tyrosines within the immunoreceptor tyrosine-based activation motifs (ITAMs) penetrated the hydrophobic core of the membrane. Interactions between the cytoplasmic tails and phosphatidylinositol phosphate lipids in the inner membrane leaflet led to the formation of a distinct anionic lipid fingerprint around the TCR-CD3. These results increase our understanding of the TCR-CD3 dynamics and the importance of membrane lipids in regulating T cell activation.

Non-redundant activity of GSK-3α and GSK-3ß in T cell-mediated tumor rejection.

Glycogen synthase kinase-3 (GSK-3) is a positive regulator of PD-1 expression in CD8+ T cells and GSK-3 inhibition enhances T cell function and is effective in the control of tumor growth. GSK-3 has two co-expressed isoforms, GSK-3α and GSK-3ß. Using conditional gene targeting, we demonstrate that both isoforms contribute to T cell function to different degrees. Gsk3b-/- mice suppressed tumor growth to the same degree as Gsk3a/b-/- mice, whereas Gsk3a-/- mice behaved similarly to wild-type, revealing an important role for GSK-3ß in regulating T cell-mediated anti-tumor immunity. The individual GSK-3α and ß isoforms have differential effects on PD-1, IFNγ, and granzyme B expression and operate in synergy to control PD-1 expression and the infiltration of tumors with CD4 and CD8 T cells. Our data reveal a complex interplay of the GSK-3 isoforms in the control of tumor immunity and highlight non-redundant activity of GSK-3 isoforms in T cells, with implications for immunotherapy.

Site-directed M2 proton channel inhibitors enable synergistic combination therapy for rimantadine-resistant pandemic influenza.

Pandemic influenza A virus (IAV) remains a significant threat to global health. Preparedness relies primarily upon a single class of neuraminidase (NA) targeted antivirals, against which resistance is steadily growing. The M2 proton channel is an alternative clinically proven antiviral target, yet a near-ubiquitous S31N polymorphism in M2 evokes resistance to licensed adamantane drugs. Hence, inhibitors capable of targeting N31 containing M2 (M2-N31) are highly desirable. Rational in silico design and in vitro screens delineated compounds favouring either lumenal or peripheral M2 binding, yielding effective M2-N31 inhibitors in both cases. Hits included adamantanes as well as novel compounds, with some showing low micromolar potency versus pandemic "swine" H1N1 influenza (Eng195) in culture. Interestingly, a published adamantane-based M2-N31 inhibitor rapidly selected a resistant V27A polymorphism (M2-A27/N31), whereas this was not the case for non-adamantane compounds. Nevertheless, combinations of adamantanes and novel compounds achieved synergistic antiviral effects, and the latter synergised with the neuraminidase inhibitor (NAi), Zanamivir. Thus, site-directed drug combinations show potential to rejuvenate M2 as an antiviral target whilst reducing the risk of drug resistance.

Anemia is a poor prognostic factor for stage I non-small cell lung cancer (NSCLC) patients treated with Stereotactic Body Radiation Therapy (SBRT).

•This study reports on the prognostic ability of haematological parameters for the largest known biopsy-proven stage-I medically inoperable cohort treated with SBRT.•After SBRT, the median values of Hb, ALC, ANC and TPC declined whereas the NLR and the PLR increased as compared to pre-SBRT.•Anemia along with other parameters was found to be a poor prognostic factor for local control despite treatment with SBRT to doses of >100 Gy BED10.•Simple and minimally invasive methods like a peripheral blood sample can provide prognostic information even for stage-I NSCLC patients.•Patient, tumor and treatment factors along with molecular markers should be used to create risk stratification models that can guide therapy.

Genetic and Environmental Determinants of Immune Response to Cutaneous Melanoma.

The immune response to melanoma improves the survival in untreated patients and predicts the response to immune checkpoint blockade. Here, we report genetic and environmental predictors of the immune response in a large primary cutaneous melanoma cohort. Bioinformatic analysis of 703 tumor transcriptomes was used to infer immune cell infiltration and to categorize tumors into immune subgroups, which were then investigated for association with biological pathways, clinicopathologic factors, and copy number alterations. Three subgroups, with "low", "intermediate", and "high" immune signals, were identified in primary tumors and replicated in metastatic tumors. Genes in the low subgroup were enriched for cell-cycle and metabolic pathways, whereas genes in the high subgroup were enriched for IFN and NF-κB signaling. We identified high MYC expression partially driven by amplification, HLA-B downregulation, and deletion of IFNγ and NF-κB pathway genes as the regulators of immune suppression. Furthermore, we showed that cigarette smoking, a globally detrimental environmental factor, modulates immunity, reducing the survival primarily in patients with a strong immune response. Together, these analyses identify a set of factors that can be easily assessed that may serve as predictors of response to immunotherapy in patients with melanoma. SIGNIFICANCE: These findings identify novel genetic and environmental modulators of the immune response against primary cutaneous melanoma and predict their impact on patient survival.See related commentary by Anichini, p. 2457.

Prophylactic Cranial Irradiation in Extensive Stage Small Cell Lung Cancer: Outcomes at a Comprehensive Cancer Centre.

PURPOSE: The role of prophylactic cranial irradiation (PCI) remains controversial in extensive stage small cell lung cancer (ES-SCLC) with the publication of 2 randomized control trials demonstrating differing outcomes in overall survival. The aim of this study is to determine the impact of PCI on survival and the development of brain metastasis while addressing the disparate use of postchemotherapy brain imaging in the aforementioned trials. METHODS AND MATERIALS: The medical records of 397 consecutive patients with ES-SCLC between Jan. 1, 2005 and Dec. 31, 2011 were retrospectively reviewed. In those eligible patients (n = 155) without baseline brain metastases and who had at least a partial response to chemotherapy, overall survival and time to brain metastasis were estimated using the Kaplan-Meier method comparing patients receiving PCI or not, using both univariate and multivariate analyses. Patients were stratified by their receipt of initial postchemotherapy brain imaging. Follow-up did not include serial brain imaging, which was performed when clinically indicated. Differences between the groups with covariates were analyzed using χ2 statistics and Student's t-tests. RESULTS: By multivariate analysis, statistically significant predictors of overall survival were the presence of extrathoracic metastases, performance status and use of PCI. There was a statistically significant difference in overall survival (HR 0.55; 95% CI: 0.39-0.77; P = .0005) and time to brain metastasis (HR 0.40; 95% CI: 0.23-0.66; P = .0004) with the use of PCI. Median survival for the PCI and non-PCI groups was 13.5 and 8.5 months respectively. A survival difference with PCI was observed in both patients that received postchemotherapy brain imaging (HR 0.55; 95% CI: 0.35-0.88; P = .012) and those who did not (HR 0.48; 95% CI: 0.29-0.77; P = .0025). CONCLUSIONS: PCI in the setting of at least a partial response to chemotherapy was found to have a survival benefit and prolongation of the time to development of brain metastases, when factoring in the use of initial postchemotherapy but not routine surveillance brain imaging.

ß-Catenin-mediated immune evasion pathway frequently operates in primary cutaneous melanomas.

Immunotherapy prolongs survival in only a subset of melanoma patients, highlighting the need to better understand the driver tumor microenvironment. We conducted bioinformatic analyses of 703 transcriptomes to probe the immune landscape of primary cutaneous melanomas in a population-ascertained cohort. We identified and validated 6 immunologically distinct subgroups, with the largest having the lowest immune scores and the poorest survival. This poor-prognosis subgroup exhibited expression profiles consistent with ß-catenin-mediated failure to recruit CD141+ DCs. A second subgroup displayed an equally bad prognosis when histopathological factors were adjusted for, while 4 others maintained comparable survival profiles. The 6 subgroups were replicated in The Cancer Genome Atlas (TCGA) melanomas, where ß-catenin signaling was also associated with low immune scores predominantly related to hypomethylation. The survival benefit of high immune scores was strongest in patients with double-WT tumors for BRAF and NRAS, less strong in BRAF-V600 mutants, and absent in NRAS (codons 12, 13, 61) mutants. In summary, we report evidence for a ß-catenin-mediated immune evasion in 42% of melanoma primaries overall and in 73% of those with the worst outcome. We further report evidence for an interaction between oncogenic mutations and host response to melanoma, suggesting that patient stratification will improve immunotherapeutic outcomes.

Blood Coagulation Factor X Exerts Differential Effects on Adenovirus Entry into Human Lymphocytes.

It has been proposed that blood coagulation factors, principally factor X (FX), enhance the uptake of human adenovirus type 5 (Ad5) into cultured epithelial cells by bridging the viral hexon capsid protein and cell-surface heparan sulphate proteoglycans (HSPGs). We studied the effects of FX on Ad transduction of lymphoid cell lines (NK92MI, a natural killer cell line; Daudi, a B-cell line and Jurkat, a T-cell line) as well as primary peripheral blood lymphocytes (PBL) and HeLa epithelial cells using either replication-deficient Ad5, or a derivative in which the Ad5 fiber was replaced with that of another Ad type, Ad35, termed Ad5F35. PBL and NK92MI were resistant to Ad5 transduction. Transduction of Jurkat and Daudi cells by Ad5 was reduced by FX but without discernible effects on cell-surface Ad5 binding. FX reduced virus binding and transduction of all lymphoid cell lines by Ad5F35, as well as transduction of the T- and Natural Killer (NK)-cell populations of PBL. Flow cytometry analysis showed that all lymphoid cell lines were negative for HSPG components, in contrast to HeLa cells. FX reduced transduction of an HSPG-negative mutant Chinese hamster ovary cell line (CHOpgsA745) by Ad5 and Ad5F35, with Ad5F35 binding also being reduced by FX. These results point to fiber-dependent differences (Ad5 versus Ad35 fiber) in Ad binding to and transduction of human lymphoid and epithelial cells in the presence of FX.

Age-not Charlson Co-morbidity Index-predicts for mortality after stereotactic ablative radiotherapy for medically inoperable stage I non-small cell lung cancer.

PURPOSE: In this single institution retrospective study of patients with stage I medically inoperable non-small cell lung cancer (NSCLC) treated with stereotactic ablative radiotherapy (SABR) we attempt to model overall survival (OS) using initial prognostic variables with specific attention on the Charlson co-morbidity index (CCI). METHODS: Between 2008 and 2013, 335 patients with medically inoperable stage I NSCLC were treated with SABR or hypofractionated radiotherapy (50-60 Gy in at least 5 Gy or 4 Gy fractions respectively) at our institution. Medical comorbidities and Charlson scores were determined by individual chart review. Patients were stratified into 3 groups based on the CCI score (0-1, 2-3, 4-9) and again based on the age-adjusted Charlson Comorbidity score (aCCI). Cumulative survival for each stratum was determined using the Kaplan-Meier method. Non-significant and confounding variables were identified and discounted from survival modeling. 3 sex stratified Cox regression models were tested: (1) aCCI with age and comorbidity combined; (2) age and CCI; (3) age alone, comorbidity removed. RESULTS: The median survival was 4.4 years and the median follow up 4.7 years. The median CCI and aCCI scores were 2 and 5 respectively. Patients with aCCI 7-12 had an increased hazard of death on univariate analysis HR 2.45 (1.15-5.22 95%CI, p = 0.02) and -excluding age as a competing variable- on multivariate analysis HR 2.25 (1.04-4.84 95%CI, p = 0.04). Patients with CCI 4-9 had an increased hazard of death on univariate analysis HR 1.57(1.30-2.90) but not on multivariate analysis. On formalized testing - with either continuous or categorical variables- all three survival models yielded similar coefficients of effect. CONCLUSION: We identify male gender, weight loss greater than 10% and age as independent prognostic factors for patients treated with medically inoperable NSCLC treated with SABR or hypofractionated radiotherapy. Based on our survival models, age alone can be used interchangeably with aCCI or CCI plus age with the same prognostic value. Age is more reliably recorded, less prone to error and therefore a more useful metric than Charlson score in this group of patients.

Quantitative texture analysis on pre-treatment computed tomography predicts local recurrence in stage I non-small cell lung cancer following stereotactic radiation therapy.

BACKGROUND: The prediction of local recurrence (LR) of stage I non-small cell lung cancer (NSCLC) after definitive stereotactic body radiotherapy (SBRT) remains elusive. The purpose of this study was to assess whether quantitative imaging features on pre-treatment computed tomography (CT) can predict LR beyond 18 (18F) fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/CT maximum standard uptake value (SUVmax). METHODS: This retrospective study evaluated 36 patients with 37 stage I NSCLC who had local tumor control (LC; n=19) and (LR; n=18). Textural features were extracted on pre-treatment CT. Mann-Whitney U tests were used to compare LC and LR groups. Receiver-operating characteristic (ROC) curves were constructed and the area under the curve (AUC) calculated with LR as outcome. RESULTS: Gray-level correlation and sum variance were greater in the LR group, compared with the LC group (P=0.02 and P=0.04, respectively). Gray-level difference variance was lower in the LR group (P=0.004). The logistic regression model generated using gray-level correlation and difference variance features resulted in AUC (SE) 0.77 (0.08) (P=0.0007). The addition of 18F-FDG PET/CT SUVmax did not improve the AUC (P=0.75). CONCLUSIONS: CT textural features were found to be predictors of LR of early stage NSCLC on baseline CT prior to SBRT.