Faculty Bibliography

The aim of this study is to evaluate whether NIS radiomics can distinguish lung adenocarcinomas from granulomas on non-contrast CT scans, and also to improve the performance of Lung-RADS by reclassifying benign nodules that were initially assessed as suspicious. The screening or standard diagnostic non-contrast CT scans of 362 patients was divided into training (S_t, N = 145), validation (S_v, N = 145), and independent validation (S_iv, N = 62) sets from different institutions. Nodules were identified and manually segmented on CT images by a radiologist. A series of 264 features relating to the edge sharpness transition from the inside to the outside of the nodule were extracted. The top 10 features were used to train a linear discriminant analysis (LDA) machine learning classifier on St. In conjunction with the LDA classifier, NIS radiomics classified nodules with an AUC of 0.82 ± 0.04, 0.77, and 0.71 respectively on S_t, S_v, and S_iv. We evaluated the ability of the NIS classifier to determine the proportion of the patients in S_v that were identified initially as suspicious by Lung-RADS but were reclassified as benign by applying the NIS scores. The NIS classifier was able to correctly reclassify 46% of those lesions that were actually benign but deemed suspicious by Lung-RADS alone on S_v.

Evidence from real-world clinical settings is lacking with regard to first-line immunotherapy plus chemotherapy for the treatment of non-small cell lung cancer (NSCLC). Our aim was to describe outcomes for patients treated with first-line pembrolizumab-combination therapy for metastatic nonsquamous NSCLC in US oncology practices. Using an anonymized, nationwide electronic health record-derived database, we identified patients who initiated pembrolizumab plus pemetrexed-carboplatin in the first-line setting (May 2017 to August 2018) after diagnosis of metastatic nonsquamous NSCLC that tested negative for EGFR and ALK genomic aberrations. Eligible patients had ECOG performance status of 0-1. An enhanced manual chart review was used to collect outcome information. Time-to-event analyses were performed using the Kaplan-Meier method. Of 283 eligible patients, 168 (59%) were male; median age was 66 years (range 33-84); and the proportions of patients with PD-L1 tumor proportion score (TPS) of ≥ 50%, 1-49%, < 1%, and unknown were 28%, 27%, 28%, and 17%, respectively. At data cutoff on August 31, 2019, median patient follow-up was 20.3 months (range 12-28 months), and median real-world times on treatment (rwToT) with pembrolizumab and pemetrexed were 5.6 (95% CI 4.5-6.4) and 2.8 months (95% CI 2.2-3.5), respectively. Median overall survival (OS) was 16.5 months (95% CI 13.2-20.6); estimated 12-month survival was 59.5% (95% CI 53.3-65.0); rwProgression-free survival was 6.4 months (95% CI 5.4-7.8); and rwTumor response rate (complete or partial response) was 56.5% (95% CI 50.5-62.4). Median OS was 20.6, 16.3, 13.2, and 13.7 months for patient cohorts with PD-L1 TPS ≥ 50%, 1-49%, < 1%, and unknown, respectively. These findings demonstrate the effectiveness of pembrolizumab plus pemetrexed-carboplatin by describing clinical outcomes among patients with metastatic nonsquamous NSCLC who were treated at US oncology practices.

BACKGROUND:Previous studies have indicated Coronavirus disease 2019 (COVID-19) patients with cancer have a high fatality rate. METHODS:We conducted a systematic review of studies that reported fatalities in COVID-19 patients with cancer. A comprehensive meta-analysis that assessed the overall case fatality rate and associated risk factors was performed. Using individual patient data, univariate and multivariate logistic regression analyses were used to estimate odds ratios (OR) for each variable with outcomes. RESULTS:We included 15 studies with 3019 patients, of which 1628 were men; 41.0% were from the UK and Europe, followed by the USA and Canada (35.7%) and Asia (China, 23.3%). The overall case fatality rate of COVID-19 patients with cancer measured 22.4% (95% confidence interval [CI] = 17.3% to 28.0%). Univariate analysis revealed age (odds ratio [OR] = 3.57; 95% CI = 1.80 to 7.06), male (OR = 2.10; 95% CI = 1.07 to 4.13), and comorbidity (OR = 2.00; 95% CI = 1.04 to 3.85) were associated with increased risk of severe events (defined as the individuals being admitted to the intensive care unit, or requiring invasive ventilation, or death). In multivariate analysis, only age greater than 65 years (OR = 3.16; 95% CI = 1.45 to 6.88) and being male (OR = 2.29; 95% CI = 1.07 to 4.87) were associated with increased risk of severe events. CONCLUSION/CONCLUSIONS:Our analysis demonstrated that COVID-19 patients with cancer have a higher fatality rate when compared with that of COVID-19 patients without cancer. Age and gender appear to be risk factors associated with a poorer prognosis.

Background: Time on treatment, also called time to treatmentdiscontinuation, is a readily available real-world effectiveness endpointhighly correlated at the patient-level with progression-free survival andmoderately to highly correlated with overall survival in clinical trials andreal-world data. In October 2016, pembrolizumab received FDA approvalbased on results from KEYNOTE-024, as a first-line monotherapy forpatients with metastatic NSCLCwith PD-L1 tumor proportion score (TPS)>=50%andnoEGFR/ALKgenomicaberrations, administereduntil diseaseprogression, unacceptable toxicity, or up to 24 months. In KEYNOTE-024,25%(39/154) of patients received 35 cycles (2 years) of pembrolizumabas initially assigned therapy. Our objective was to describe rwToT withfirst-line pembrolizumab in real-world oncology practice.
Method(s): Using the US nationwide Flatiron Health electronic healthrecord-derived, de-identified database, we included adult patients withpathologically confirmed advanced, PD-L1 TPS >=50% NSCLC whoinitiated first-line pembrolizumab monotherapy from November 2016-September 2019, with follow-up through September 2020. Eligibilitycriteria included ECOG performance status 0-2, PD-L1 TPS >=50%, noEGFR/ALK genomic aberration, and no known ROS1 aberration. Patientsenrolled in a clinical trial were excluded. Median rwToT and landmarkon-treatment rates were estimated using Kaplan-Meier method.
Result(s):(Table Presented)
Conclusion(s): Patients with key trial-eligible characteristics (ECOG 0-1,PD-L1 TPS >=50%, EGFR/ALK negative) experienced rwToT with firstlinepembrolizumab similar to the phase III pivotal clinical trial.Approximately 23% received at least 2 years of treatment, suggestinglong-term benefit of pembrolizumab monotherapy for PD-L1 TPS >=50%advanced NSCLC in a real-world setting.
Copyright International Association for the Study of Lung Cancer. Published by Elsevier Inc

OBJECTIVE:To identify stable and discriminating radiomic features on non-contrast CT scans to develop more generalisable radiomic classifiers for distinguishing granulomas from adenocarcinomas. METHODS:). To mitigate the variation of CT acquisition parameters, we defined 'stable' radiomic features as those for which the feature expression remains relatively unchanged between different sites, as assessed using a Wilcoxon rank-sum test. These stable features were used to develop more generalisable radiomic classifiers that were more resilient to variations in lung CT scans. Features were ranked based on two criteria, firstly based on discriminability (i.e. maximising AUC) alone and subsequently based on maximising both feature stability and discriminability. Different machine-learning classifiers (Linear discriminant analysis, Quadratic discriminant analysis, Support vector machines and random forest) were trained with features selected using the two different criteria and then compared on the two independent test sets for distinguishing granulomas from adenocarcinomas, in terms of area under the receiver operating characteristic curve. RESULTS:[n = 62]: maximum AUCs of 0.87 versus. 0.79; p-value = 0.021). These differences held for features extracted from scans with <3 mm slice thickness (AUC = 0.88 versus. 0.80; p-value = 0.039, n = 100) and for the ≥3 mm cases (AUC = 0.81 versus. 0.76; p-value = 0.034, n = 105). In both experiments, shape and peritumoural texture features had a higher stability compared with intratumoural texture features. CONCLUSIONS:Our study suggests that explicitly accounting for both stability and discriminability results in more generalisable radiomic classifiers to distinguish adenocarcinomas from granulomas on non-contrast CT scans. Our results also showed that peritumoural texture and shape features were less affected by the scanner parameters compared with intratumoural texture features; however, they were also less discriminating compared with intratumoural features.

Introduction: At present, there is no accurate and validated way to predict which patients would have disease recurrence following definitive therapy in ES-NSCLC patients. NSCLC mortality and recurrence risk has recently been shown to be different among different genders. In this project, in addition to creating a unified Radiomic based model, we have developed and validated Gender-Specific Radiomics models which can better to predict Disease free survival (DFS) in ES-NSCLC.
Method(s): This study comprised a total of 312 ES-NSCLC patients from 3 different institutions. A total of 757 intratumoral and peritumoral radiomic textural features were extracted from a pre-treatment diagnostic non-contrast CT scan for every patient. The three models were constructed using training cohort D1- Mall for all combined all patients(N=173), MM for Male population-specific model(N=83), and MF for model specific to Females(N=89) using the most stable, significant and uncorrelated features. Based on these three models, the three Radiomic Risk Scores were constructed using a Lasso-regularized multivariate Cox-regression model. The patients were divided into High and Low-risk groups using an optimal threshold, giving maximum hazard ratio(HR) within the training cohorts. The models were validated and compared within each other using DVAL(D 2+D3).
Result(s): All three models included three features (Table-1). The MALL could not predict DFS within any specific gender subtype but had HR of 2.17 [1.15-4.08] for the entire DVAL. The MM model explicitly constructed for the male population had HR of 2.84 [1.05-7.70] within the male-specific DVAL, increasing it by ~30.87% over overall HR. Similarly, the MF model constructed specifically for females increased the HR to 12.76[2.36-68.9] in the DVAL within specific Female population. [Formula presented] [Formula presented]
Conclusion(s): Gender-specific Radiomics based models are better at predicting DFS in ES-NSCLC than radiomic models which do not explicitly account for gender. These might be capturing the underlying differences in tumor biology and characteristics between males and females. Keywords: Gender-Specific, Radiomics, CT-Scans
Copyright

INTRODUCTION/BACKGROUND:Thromboembolism (TE) is common in patients with non-small cell lung cancer (NSCLC) and is associated with worse outcomes. Recent advances in the understanding of NSCLC have led to the identification of molecular subtypes such as ALK and EGFR mutations. The association of these subtypes with risk of TE has not been fully explored. METHODS:We conducted a retrospective cohort study of consecutive NSCLC patients seen at the Cleveland Clinic from July 2002 through July 2017 for whom molecular classification and follow-up were available. TE events included deep-vein thrombosis (DVT), pulmonary embolism (PE), visceral vein thrombosis (VVT) and arterial events. TE-free survival and overall survival rates for each of the molecular subtype (wild-type, ALK-mutant and EGFR-mutant) were estimated by the Kaplan-Meier method. Cox proportional hazard regression analysis was used to identify factors associated with the endpoints TE and overall survival. TE was analyzed as a conditional, time-dependent covariate to assess its impact with respect to overall survival. RESULTS:The study population consisted of 461 patients. Approximately half were females (n=263, 57%) and 58% (n=270) were older than 65 years. TE occurred in 98 of 461 patients (21.3%) during a median follow-up of 33.1 months. The highest cumulative rates of TE were observed in patients with ALK-mutant NSCLC (N=20/46, 43.5%) followed by patients with EGFR-mutant cancers (N=35/165, 21.2%) and wild-type cancers (N=43/250, 17.2%) p<0.05. Cumulative incidence of TE at six months of follow-up was 15.7% (95% CI: 5.0-26.4%) for ALK-mutant cancers, 8.8% (95% CI: 4.4-13.2%) for EGFR-mutant cancers, and 9.2% (95% CI: 5.4-12.9%) for wild-type cancers. Patients who experienced TE had worse overall survival [all patients: HR=2.8 95% CI 2.1-3.6, p<0.001]. CONCLUSIONS:Patients with ALK-mutant advanced lung adenocarcinoma have the highest rate of TE. TE is associated with worse survival across molecular subtypes. These findings should be taken into consideration in decision-making regarding thromboprophylaxis.