Search

Background Prognostic modeling in health care has been predominantly statistical, despite a rapid growth of literature on machine-learning approaches in biological data analysis. We aim to assess the relative importance of variables in predicting overall survival among patients with non-small cell lung cancer using a Variable Importance (VIMP) approach in a machine-learning Random Survival Forest (RSF) model for posttreatment planning and follow-up. Methods A total of 935 non-small cell lung cancer patients were randomly and equally divided into 2 training and testing cohorts in an RFS model. The prognostic variables included age, sex, race, the TNM Classification of Malignant Tumors (TNM) stage, smoking history, Eastern Cooperative Oncology Group performance status, histologic type, treatment category, maximum standard uptake value of whole-body tumor (SUVmaxWB), whole-body metabolic tumor volume (MTVwb), and Charlson Comorbidity Index. The VIMP was calculated using a permutation method in the RSF model. We further compared the VIMP of the RSF model to that of the standard Cox survival model. We examined the order of VIMP with the differential functional forms of the variables. Results In both the RSF and the standard Cox models, the most important variables are treatment category, TNM stage, and MTVwb. The order of VIMP is more robust in RSF model than in Cox model regarding the differential functional forms of the variables. Conclusions The RSF VIMP approach can be applied alongside with the Cox model to further advance the understanding of the roles of prognostic factors, and improve prognostic precision and care efficiency.

T-cell/histiocyte-rich large B-cell lymphoma (TCRLBCL) is an aggressive variant of diffuse large B-cell lymphoma (DLBCL) characterized by rare malignant B cells within a robust but ineffective immune cell infiltrate. The mechanistic basis of immune escape in TCRLBCL is poorly defined and not targeted therapeutically. We performed a genetic and quantitative spatial analysis of the PD-1/PD-L1 pathway in a multi-institutional cohort of TCRLBCLs and found that malignant B cells harbored PD-L1/PD-L2 copy gain or amplification in 64% of cases, which was associated with increased PD-L1 expression (P = .0111). By directed and unsupervised spatial analyses of multiparametric cell phenotypic data within the tumor microenvironment, we found that TCRLBCL is characterized by tumor-immune "neighborhoods" in which malignant B cells are surrounded by exceptionally high numbers of PD-L1-expressing TAMs and PD-1+ T cells. Furthermore, unbiased clustering of spatially resolved immune signatures distinguished TCRLBCL from related subtypes of B-cell lymphoma, including classic Hodgkin lymphoma (cHL) and DLBCL-NOS. Finally, we observed clinical responses to PD-1 blockade in 3 of 5 patients with relapsed/refractory TCRLBCL who were enrolled in clinical trials for refractory hematologic malignancies (NCT03316573; NCT01953692), including 2 complete responses and 1 partial response. Taken together, these data implicate PD-1 signaling as an immune escape pathway in TCRLBCL and also support the potential utility of spatially resolved immune signatures to aid the diagnostic classification and immunotherapeutic prioritization of diverse tumor types.

PURPOSE: We hypothesized that whole-body metabolic tumor volume (MTVwb) could be used to supplement non-small cell lung cancer (NSCLC) staging due to its independent prognostic value. The goal of this study was to develop and validate a novel MTVwb risk stratification system to supplement NSCLC staging. METHODS: We performed an IRB-approved retrospective review of 935 patients with NSCLC and FDG avid tumor divided into modeling and validation cohorts based on the PET/CT scanner type used for imaging. In addition, sensitivity analysis was conducted by dividing the patient population into two randomized cohorts. Cox regression and Kaplan-Meier survival analyses were performed to determine the prognostic value of the MTVwb risk stratification system. RESULTS: The cutoff values (10.0, 53.4 and 155.0 mL) between the MTVwb quartiles of the modeling cohort were applied to both modeling and validation cohorts to determine each patient’s MTVwb risk stratum. The survival analyses showed that a lower MTVwb risk stratum was associated with better overall survival (all p

Stage IIIA non-small cell lung cancer (NSCLC) is heterogeneous in tumor burden, and its treatment is variable. Whole-body metabolic tumor volume (MTVWB) has been shown to be an independent prognostic index for overall survival (OS). However, the potential of MTVWB to risk-stratify stage IIIA NSCLC has previously been unknown. If we can identify subgroups within the stage exhibiting significant OS differences using MTVWB, MTVWB may lead to adjustments in patients’ risk profile evaluations and may, therefore, influence clinical decision making regarding treatment. We estimated the risk-stratifying capacity of MTVWB in stage IIIA by comparing OS of stratified stage IIIA with stage IIB and IIIB NSCLC. We performed a retrospective review of 330 patients with clinical stage IIB, IIIA, and IIIB NSCLC diagnosed between 2004 and 2014. The patients’ clinical TNM stage, initial MTVWB, and long-term survival data were collected. Patients with TNM stage IIIA disease were stratified by MTVWB. The optimal MTVWB cutoff value for stage IIIA patients was calculated using sequential log-rank tests. Univariate and multivariate cox regression analyses and Kaplan-Meier OS analysis with log-rank tests were performed. The optimal MTVWB cut-point was 29.2 mL for the risk-stratification of stage IIIA. We identified statistically significant differences in OS between stage IIB and IIIA patients (p

The staging and management of patients with newly diagnosed nonsmall cell lung cancer (NSCLC) in the setting of recently diagnosed other (metachronous or synchronous) primary cancer are challenging. This retrospective cohort study was carried out to test our hypothesis that baseline 2-deoxy-2-[F]fluoro-D-glucose (F-FDG) PET/CT parameters, including whole-body metabolic tumor volume (MTVWB), total lesion glycolysis (TLGWB), and maximum standardized uptake value (SUVmaxWB), are associated with the overall survival (OS) of such patients.A total of 110 NSCLC patients (52 men and 58 women, aged 68.6±7.8 years) with other primary malignant cancers who had baseline F-FDG PET/CT scans were retrospectively reviewed. MTVWB, TLGWB, and SUVmaxWB were measured. Kaplan-Meier analysis with the log-rank test and Cox regression models were used to assess the association of OS with F-FDG PET/CT parameters and clinical risk factors.Kaplan-Meier analysis and univariate Cox regression models showed significant associations of OS with ln(MTVWB), ln(TLGWB), ln(SUVmaxWB), TNM stage, and treatment type (surgery vs. no treatment). Multivariable Cox regression models showed a significant relationship of OS with ln(MTVWB) [hazard ratio (HR)=1.368, P=0.001], ln(TLGWB) (HR=1.313, P0.001), and ln(SUVmaxWB) (HR=1.739, P=0.006), adjusted for age, treatment type, tumor histology, and TNM stage. The TNM stage was not associated significantly with OS when MTVWB, TLGWB, or SUVmaxWB were included in the multivariable models.MTVWB, TLGWB, and SUVmaxWB from baseline F-FDG PET/CT are associated individually with OS of patients with both NSCLC and other primary malignant tumors independent of age, treatment type, tumor histology, and TNM stage.

As real-time and immediate feedback becomes increasingly important in tasks related to mobile information, big data stream processing systems are increasingly applied to process massive amounts of mobile data. However, when processing a drastically fluctuating mobile data stream, the lack of an elastic resource-scheduling strategy limits the elasticity and scalability of data stream processing systems. To address this problem, this paper builds a flow-network model, a resource allocation model, and a data redistribution model as the foundation for proposing Flink with an elastic resource-scheduling strategy (Flink-ER), which consists of a capacity detection algorithm, an elastic resource reallocation algorithm, and a data redistribution algorithm. The strategy improves the performance of the platform by dynamically rescaling the cluster and increasing the parallelism of operators based on the processing load. The experimental results show that the throughput of a cluster was promoted under the premise of meeting latency constraints, which verifies the efficiency of the strategy.

e21085 Background: The aim of this study is to model and validate a post-treatment PET/CT volumetric prognostic (ptPVP) index for risk assessment after definitive therapy for NSCLC. Methods: This IRB-approved retrospective cohort study included 1028 NSCLC patients diagnosed between 2004 and 2017. Variables included: pre-treatment whole-body metabolic tumor volume (MTVwb) measured on baseline FDG PET/CT, patient age, clinical TNM stage, ECOG performance status (PS), gender, tumor histology, treatment type, and survival outcomes. Patients were assigned to a modeling cohort (531 patients) for modeling ptPVP index and a clinical index with a multivariate Cox regression model. The ptPVP index was validated with a separate validation cohort (497 patients) by comparing the prognostic value of ptPVP index against its individual independent variables and the clinical index. Results: The ptPVP index generated from the modeling cohort was comprised of 5 variables including treatment type (surgery, no surgery, and no therapy), age, clinical TNM stage (stage I or II, III, and IV), PS ( 0, 1, and ≥ 2) and MTVwb. The clinical index was comprised of above 4 clinical variables without MTVwb. Univariate Cox models showed significant association of the ptPVP index with OS with Hazard ratio (HR) of 2.72 (95% CI = 2.39-3.09, p < 0.001). The C-statistic of ptPVP index (0.71) was significantly greater than that of clinical index (0.68) (p = 0.001). In the validation cohort, univariate Cox models showed significant association of the ptPVP index with OS (HR = 3.09, 95% CI = 2.64-3.61, p < 0.001). The ptPVP index showed greater prognostic power (C-statistic = 0.72) than that of the clinical index (C-statistic = 0.70, p = 0.001), treatment type (C-statistic = 0.66, p < 0.001), MTVwb (C-statistic = 0.67, p < 0.001), age (C-statistic = 0.58, p < 0.001), clinical 9-group TNM staging (C-statistic = 0.66, p < 0.001), and PS (C-statistic = 0.57, p < 0.001). Multivariate Cox regression analysis demonstrated a significant association of ptPVP index with OS (HR = 3.06, 95% CI of 2.60-3.61, p < 0.001) after adjusting gender and NSCLC histology. A better separation of Kaplan-Meier curves of patients grouped by ptPVP index than those grouped by clinical index, MTVwb, or clinical TNM stage was evident. Conclusions: The ptPVP index developed for NSCLC patients demonstrated moderately prognostic power and better prognostic ability than its individual independent variables and clinical index without MTVwb. It provides a practical quantitative risk assessment for the patient follow-up and management after initial treatment.

Objectives Whole-body metabolic tumor volume (MTVWB) and TNM staging are independent prognostic factors for overall survival (OS) in non-small cell lung cancer (NSCLC). We aimed to update and validate the PET/CT volumetric prognostic index (PVP index) using the new 8th edition TNM staging system to evaluate its prognostic power versus TNM staging and MTVWB alone. Materials and Methods This study was a retrospective analysis of 949 non-small cell lung cancer (NSCLC) patients diagnosed between 2004 and 2014. Clinical TNM stage, MTVWB, age and gender, tumor histology type at the initial staging PET/CT exam, as well as treatment history and long-term survival data were obtained. Patients were randomly assigned to modeling or validation group. Univariate and multivariate Cox regression analyses were performed to compare PVP index, TNM stage, and MTVWB in the validation group. Results The updated PVP index included the 3 variables TNM stage, and MTVWB and age. Univariate Cox models showed significant association of PVP index with overall survival (OS) in patients with NSCLC (with Hazard ratio HR = 2.88 in the validation group, p Conclusions The updated PVP index provides a quantitative risk assessment for NSCLC patients using 8th edition TNM staging, MTVWB, and age. The index provides a simple and practical way for the care team to incorporate the independent prognostic value of both the TNM stage and MTVWB. This approach can further improve the accuracy of overall survival prognosis.

To test the hypothesis that whole-body metabolic tumor burden (MTBWB) on postsurgical fluorodeoxyglucose (FDG) positron emission tomographic (PET)/computed tomographic (CT) images in patients with non-small cell lung cancer (NSCLC) is associated with their overall survival (OS).The institutional review board approved this study and waived the requirement for obtaining informed consent. One hundred forty-two patients with NSCLC (69 men, 73 women; median age, 67.7 years) who underwent postsurgical FDG PET/CT were retrospectively reviewed. The whole-body metabolic tumor volume (MTVWB), whole-body total lesion glycolysis (TLGWB), and whole-body maximum standardized uptake value (SUVWBmax) were measured. OS served as the primary end point of the study. Kaplan-Meier curves and Cox regression were used to assess the association between PET/CT markers and OS.The interobserver variability was low, as demonstrated with intraclass correlation coefficients higher than 0.94 for SUVWBmax, MTVWB, and TLGWB. When compared with those with negative postsurgical FDG PET/CT findings, a significant decrease of OS was found in patients with the presence of FDG-avid tumor on the basis of both a log-rank test (P = .001) and a univariate Cox model (hazard ratio = 2.805, P = .001). In patients with FDG-avid tumor, there was a significant association between OS and ln MTVWB (P.001), ln TLGWB (P.001), and ln SUVWBmax (P.010) in either univariate or multivariate analysis, after adjusting for patient age, sex, TNM restage, and therapy after postsurgical PET/CT studies. The OS differences between the groups dichotomized by the median value of MTVWB (11.54 mL, P = .004), TLGWB (32.38 mL, P.001), or SUVWBmax (4.93, P = .023) were significant.MTBWB and tumor maximum standardized uptake at postsurgical FDG PET/CT are related to the patient's OS in NSCLC, independent of age, sex, TNM restaging, and therapy after postsurgical PET/CT studies.

Cerebral vasospasm (CVS) is a severe complication that occurs following aneurysmal subarachnoid hemorrhage (SAH). Magnetic resonance angiography (MRA) has been used to evaluate brain injury following SAH in humans. The present study was designed to assess a rabbit model of symptomatic CVS (SCVS) and the utility of MRA in evaluating SCVS in rabbits. Japanese white rabbits (n=24) were randomly divided into 2 equal groups: A sham group and a SAH group. Neurological scores were evaluated for 7 days following SAH. Basilar artery (BA) diameters were measured using MRA preoperatively and 7 days postoperatively. Rabbits were sacrificed 7 days following SAH and the BA diameter of each rabbit was determined using histological evaluation. Compared with the Sham group, neurological function was significantly reduced in the SAH group at all time points (P

Purpose TNM Stage 3B encompasses a wide range of primary tumor and nodal metastatic tumor burden. This study aimed to evaluate the prognostic value of quantitative FDG PET/CT parameters in patients with newly diagnosed Stage 3B Non-Small Cell Lung Cancer (NSCLC). Materials and Methods Institutional review board approved retrospective study identified patients diagnosed with Stage 3B NSCLC (8th edition TNM classification) on baseline FDG PET/CT at two medical centers (Medical centers A and B), between Feb 2004 and Dec 2014. Patients were excluded if they had prior NSCLC treatment or recent diagnosis of a second primary cancer. Quantitative FDG PET/CT parameters including whole body metabolic tumor volume (MTVwb), total lesion glycolysis (TLGwb), and maximum standardized uptake value (SUVmaxwb) were measured from baseline PET/CT using Edge method with Mimvista software. The primary endpoint was overall survival (OS). Cox proportional hazard regression and Kaplan-Meier overall survival analyses were used to test for an association between OS and quantitative FDG PET/CT parameters. The distributions of MTVwb, TLGwb, SUVmaxwb were skewed, so a natural logarithm transformation was applied and the transformed variables [(ln(MTVwb), ln(TLGwb), and ln(SUVmaxwb)] were used in the analysis. Results The training set included 110 patients from center A with Stage 3B NSCLC. 78.2% of patients expired during follow-up. Median OS was 14 months. 1-year, 2-year, and 5-year OS was 56.5%, 34.6% and 13.9%, respectively. Univariate Cox regression analysis showed no significant difference in OS on the basis of age, gender, histology, ln(TLGwb), or ln(SUVmaxwb). ln(MTVwb) was positively associated with OS [hazard ratio (HR) of 1.23, p = 0.037]. This association persisted on multivariate Cox regression analysis (HR 1.28, p = 0.043), with adjustments for age, gender, treatment and tumor histology. External validation with 44 patients from center B confirmed increasing MTVwb was associated significantly worse OS. An MTVwb cut-off point of 85.6 mL significantly stratified Stage 3B NSCLC patient prognosis. Conclusion MTVwb is a prognostic marker for OS in patients with Stage 3B NSCLC, independent of age, gender, treatment, and tumor histology.

Background The purpose of this study is to describe the preliminary findings of 99mTc-labeled ethylene dicysteine deoxyglucose (99mTc-EC-DG) performed four weeks after chemoradiotherapy in patients with locally advanced head and neck squamous cell carcinoma. Methods Review of nine patients with locally advanced head and neck squamous cell carcinomas imaged with 99mTc-EC-DG single photon emission computed tomography-computed tomography (SPECT-CT) at baseline before treatment and at four weeks after treatment completion was performed. Results At four weeks post-treatment, five patients had either decreased activity or no significant activity on 99mTc-EC-DG SPECT-CT and were considered to have responded to treatment, whereas four patients did not have significantly decreased uptake on 99mTc-EC-DG SPECT-CT and were considered to have not adequately responded to treatment. Among the five patients considered to have treatment response at four weeks, all were free of disease (true-negative). Among the four patients considered to have stable activity on 99mTc-EC-DG SPECT-CT at four weeks, two were designated as having no response or incomplete response (true-positive), and two were designated as having complete response (false-positive) on subsequent composite assessment. Conclusions The pilot data is promising but warrants further investigation of 99mTc-EC-DG SPECT-CT for the assessment of locoregional treatment response at four weeks in patients with locally advanced head and neck squamous cell carcinomas.

Cerebral vasospasm (CVS) is a severe complication that occurs following aneurysmal subarachnoid hemorrhage (SAH). Magnetic resonance angiography (MRA) has been used to evaluate brain injury following SAH in humans. The present study was designed to assess a rabbit model of symptomatic CVS (SCVS) and the utility of MRA in evaluating SCVS in rabbits. Japanese white rabbits (n=24) were randomly divided into 2 equal groups: A sham group and a SAH group. Neurological scores were evaluated for 7 days following SAH. Basilar artery (BA) diameters were measured using MRA preoperatively and 7 days postoperatively. Rabbits were sacrificed 7 days following SAH and the BA diameter of each rabbit was determined using histological evaluation. Compared with the Sham group, neurological function was significantly reduced in the SAH group at all time points (P

e21164Background: Comorbidities reflect patients’ underlying multiple-disease growth curve and influence outcomes of treatment. However, it is unclear if there is and how much independent prognosti...

e21118Background: It is unclear how comorbidities are associated with the stage of non-small cell lung cancer (NSCLC) at cancer diagnosis. We aimed to evaluate the association between Charlson Como...

OBJECTIVES The objective of this study was to test the hypothesis that the metabolic tumor volume (MTV) of primary non-small-cell lung cancer is not sensitive to differences in F-fluorodeoxyglucose (F-FDG) uptake time, and to compare this consistency of MTV measurements with that of standardized uptake value (SUV) and total lesion glycolysis (TLG). METHODS Under Institutional Review Board approval, 134 consecutive patients with histologically proven non-small-cell lung cancer underwent F-FDG PET/computed tomography scanning at about 1 h (early) and 2 h (delayed) after intravenous injection of F-FDG. MTV, SUV, and TLG of the primary tumor were all measured. Student's t-test and Wilcoxon's signed-rank test for paired data were used to compare MTV, SUV, and TLG between the two scans. The intraclass correlation coefficient (ICC) was used to assess agreement in PET parameters between the two scans and between the measurements made by two observers. RESULTS MTV was not significantly different (P=0.17) between the two scans. However, SUVmax, SUVmean, SUVpeak, and TLG increased significantly from the early to the delayed scans (P

8548Background: We developed a PET/CT volumetric prognostic staging system (PVP index) using risk-based models integrating hazard of death from both whole-body metabolic tumor volume (MTVWB ) and T...

Background Radiolabeled metaiodobenzylguanidine (MIBG) is sensitive and specific for detecting neuroblastoma. The extent of MIBG-avid disease is assessed using Curie scores. Although Curie scoring is prognostic in patients with high-risk neuroblastoma, there is no standardized method to assess the response of specific sites of disease over time. The goal of this study was to develop approaches for Curie scoring to facilitate the calculation of scores and comparison of specific sites on serial scans. Procedure We designed three semiautomated methods for determining Curie scores, each with increasing degrees of computer assistance. Method A was based on visual assessment and tallying of MIBG-avid lesions. For method B, scores were tabulated from a schematic that associated anatomic regions to MIBG-positive lesions. For method C, an anatomic mesh was used to mark MIBG-positive lesions with automatic assignment and tallying of scores. Five imaging physicians experienced in MIBG interpretation scored 38 scans using each method, and the feasibility and utility of the methods were assessed using surveys. Results There was good reliability between methods and observers. The user-interface methods required 57 to 110 seconds longer than the visual method. Imaging physicians indicated that it was useful that methods B and C enabled tracking of lesions. Imaging physicians preferred method B to method C because of its efficiency. Conclusions We demonstrate the feasibility of semiautomated approaches for Curie score calculation. Although more time was needed for strategies B and C, the ability to track and document individual MIBG-positive lesions over time is a strength of these methods.

8516Background: We previously developed a PET/CT volumetric prognostic staging system (PVP index) using Cox regression models to predict the hazard of death using both whole-body metabolic tumor vo...