Impact of a bronchial genomic classifier on clinical decision making in patients undergoing diagnostic evaluation for lung cancer

Decision impact studies, evidence of clinical utility for genomic assays in cancer: A scoping review

BACKGROUND

Decision impact studies have become increasingly prevalent in cancer prognostic research in recent years. These studies aim to evaluate the impact of a genomic test on decision-making and appear to be a new form of evidence of clinical utility. The objectives of this review were to identify and characterize decision impact studies in genomic medicine in cancer care and categorize the types of clinical utility outcomes reported.

METHODS

We conducted a search of four databases, Medline, Embase, Scopus and Web of Science, from inception to June 2022. Empirical studies that reported a "decision impact" assessment of a genomic assay on treatment decisions or recommendations for cancer patients were included. We followed scoping review methodology and adapted the Fryback and Thornbury Model to collect and analyze data on clinical utility. The database searches identified 1803 unique articles for title/abstract screening; 269 articles moved to full-text review.

RESULTS

87 studies met inclusion criteria. All studies were published in the last 12 years with the majority for breast cancer (72%); followed by other cancers (28%) (lung, prostate, colon). Studies reported on the impact of 19 different proprietary (18) and generic (1) assays. Across all four levels of clinical utility, outcomes were reported for 22 discrete measures, including the impact on provider/team decision-making (100%), provider confidence (31%); change in treatment received (46%); patient psychological impacts (17%); and costing or savings impacts (21%). Based on the data synthesis, we created a comprehensive table of outcomes reported for clinical utility.

CONCLUSIONS

This scoping review is a first step in understanding the evolution and uses of decision impact studies and their influence on the integration of emerging genomic technologies in cancer care. The results imply that DIS are positioned to provide evidence of clinical utility and impact clinical practice and reimbursement decision-making in cancer care. Systematic review registration: Open Science Framework osf.io/hm3jr.

Percepta Genomic Sequencing Classifier and decision-making in patients with high-risk lung nodules: a decision impact study

BACKGROUND

Incidental and screening-identified lung nodules are common, and a bronchoscopic evaluation is frequently nondiagnostic. The Percepta Genomic Sequencing Classifier (GSC) is a genomic classifier developed in current and former smokers which can be used for further risk stratification in these patients. Percepta GSC has the capability of up-classifying patients with a pre-bronchoscopy risk that is high (> 60%) to "very high risk" with a positive predictive value of 91.5%. This prospective, randomized decision impact survey was designed to test the hypothesis that an up-classification of risk of malignancy from high to very high will increase the rate of referral for surgical or ablative therapy without additional intervening procedures while increasing physician confidence.

METHODS

Data were collected from 37 cases from the Percepta GSC validation cohort in which the pre-bronchoscopy risk of malignancy was high (> 60%), the bronchoscopy was nondiagnostic, and the patient was up-classified to very high risk by Percepta GSC. The cases were randomly presented to U.S pulmonologists in three formats: a pre-post cohort where each case is presented initially without and then with a GSG result, and two independent cohorts where each case is presented either with or without with a GSC result. Physicians were surveyed with respect to subsequent management steps and confidence in that decision.

RESULTS

One hundred and one survey takers provided a total of 1341 evaluations of the 37 patient cases across the three different cohorts. The rate of recommendation for surgical resection was significantly higher in the independent cohort with a GSC result compared to the independent cohort without a GSC result (45% vs. 17%, p < 0.001) In the pre-post cross-over cohort, the rate increased from 17 to 56% (p < 0.001) following the review of the GSC result. A GSC up-classification from high to very high risk of malignancy increased Pulmonologists' confidence in decision-making following a nondiagnostic bronchoscopy.

CONCLUSIONS

Use of the Percepta GSC classifier will allow more patients with early lung cancer to proceed more rapidly to potentially curative therapy while decreasing unnecessary intervening diagnostic procedures following a nondiagnostic bronchoscopy.

Evaluating the Use of Circulating MicroRNA Profiles for Lung Cancer Detection in Symptomatic Patients

Importance

The overall low survival rate of patients with lung cancer calls for improved detection tools to enable better treatment options and improved patient outcomes. Multivariable molecular signatures, such as blood-borne microRNA (miRNA) signatures, may have high rates of sensitivity and specificity but require additional studies with large cohorts and standardized measurements to confirm the generalizability of miRNA signatures.

Objective

To investigate the use of blood-borne miRNAs as potential circulating markers for detecting lung cancer in an extended cohort of symptomatic patients and control participants.

Design, Setting, and Participants

This multicenter, cohort study included patients from case-control and cohort studies (TREND and COSYCONET) with 3102 patients being enrolled by convenience sampling between March 3, 2009, and March 19, 2018. For the cohort study TREND, population sampling was performed. Clinical diagnoses were obtained for 3046 patients (606 patients with non-small cell and small cell lung cancer, 593 patients with nontumor lung diseases, 883 patients with diseases not affecting the lung, and 964 unaffected control participants). No samples were removed because of experimental issues. The collected data were analyzed between April 2018 and November 2019.

Main Outcomes and Measures

Sensitivity and specificity of liquid biopsy using miRNA signatures for detection of lung cancer.

Results

A total of 3102 patients with a mean (SD) age of 61.1 (16.2) years were enrolled. Data on the sex of the participants were available for 2856 participants; 1727 (60.5%) were men. Genome-wide miRNA profiles of blood samples from 3046 individuals were evaluated by machine-learning methods. Three classification scenarios were investigated by splitting the samples equally into training and validation sets. First, a 15-miRNA signature from the training set was used to distinguish patients diagnosed with lung cancer from all other individuals in the validation set with an accuracy of 91.4% (95% CI, 91.0%-91.9%), a sensitivity of 82.8% (95% CI, 81.5%-84.1%), and a specificity of 93.5% (95% CI, 93.2%-93.8%). Second, a 14-miRNA signature from the training set was used to distinguish patients with lung cancer from patients with nontumor lung diseases in the validation set with an accuracy of 92.5% (95% CI, 92.1%-92.9%), sensitivity of 96.4% (95% CI, 95.9%-96.9%), and specificity of 88.6% (95% CI, 88.1%-89.2%). Third, a 14-miRNA signature from the training set was used to distinguish patients with early-stage lung cancer from all individuals without lung cancer in the validation set with an accuracy of 95.9% (95% CI, 95.7%-96.2%), sensitivity of 76.3% (95% CI, 74.5%-78.0%), and specificity of 97.5% (95% CI, 97.2%-97.7%).

Conclusions and Relevance

The findings of the study suggest that the identified patterns of miRNAs may be used as a component of a minimally invasive lung cancer test, complementing imaging, sputum cytology, and biopsy tests.

Impact of the Percepta Genomic Classifier on Clinical Management Decisions in a Multicenter Prospective Study

BACKGROUND

The Percepta genomic classifier has been clinically validated as a complement to bronchoscopy for lung nodule evaluation.

RESEARCH QUESTION

The goal of this study was to examine the impact on clinical management decisions of the Percepta result in patients with low- and intermediate-risk lung nodules.

STUDY DESIGN AND METHODS

A prospective "real world" registry was instituted across 35 US centers to observe physician management of pulmonary nodules following a nondiagnostic bronchoscopy. To assess the impact on management decisions of the Percepta genomic classifier, a subset of patients was analyzed who had an inconclusive bronchoscopy for a pulmonary nodule, a Percepta result, and an adjudicated lung diagnosis with at least 1 year of follow-up. In this cohort, change in the decision to pursue additional invasive procedures following Percepta results was assessed.

RESULTS

A total of 283 patients met the study eligibility criteria. In patients with a low/intermediate risk of malignancy for whom the clinician had designated a plan for a subsequent invasive procedure, a negative Percepta result down-classified the risk of malignancy in 34.3% of cases. Of these down-classified patients, 73.9% had a change in their management plan from an invasive procedure to surveillance, and the majority avoided a procedure up to 12 months following the initial evaluation. In patients with confirmed lung cancers, the time to diagnosis was not significantly delayed when comparing Percepta down-classified patients vs patients who were not down-classified (P = .58).

INTERPRETATION

The down-classification of nodule malignancy risk with the Percepta test decreased additional invasive procedures without a delay in time to diagnosis among those with lung cancer.

Low Dose CT for Lung Cancer Screening: The Background, the Guidelines, and a Tailored Approach to Patient Care

Despite lung cancer's prevalence and high burden of mortality, a useful screening program has taken a long time to develop. Now, most of the associated national organizations recommend low dose CT screening in the appropriate population. However, since the USPSTF guidelines were published, implementation has been slow. This article outlines the current evidence and provides additional resources to help physicians tailor a screening program for their patients.

Survival prediction in intensive-care units based on aggregation of long-term disease history and acute physiology: a retrospective study of the Danish National Patient Registry and electronic patient records

BACKGROUND

Intensive-care units (ICUs) treat the most critically ill patients, which is complicated by the heterogeneity of the diseases that they encounter. Severity scores based mainly on acute physiology measures collected at ICU admission are used to predict mortality, but are non-specific, and predictions for individual patients can be inaccurate. We investigated whether inclusion of long-term disease history before ICU admission improves mortality predictions.

METHODS

Registry data for long-term disease histories for more than 230 000 Danish ICU patients were used in a neural network to develop an ICU mortality prediction model. Long-term disease histories and acute physiology measures were aggregated to predict mortality risk for patients for whom both registry and ICU electronic patient record data were available. We compared mortality predictions with admission scores on the Simplified Acute Physiology Score (SAPS) II, the Acute Physiologic Assessment and Chronic Health Evaluation (APACHE) II, and the best available multimorbidity score, the Multimorbidity Index. An external validation set from an additional hospital was acquired after model construction to confirm the validity of our model. During initial model development data were split into a training set (85%) and an independent test set (15%), and a five-fold cross-validation was done during training to avoid overfitting. Neural networks were trained for datasets with disease history of 1 month, 3 months, 6 months, 1 year, 2·5 years, 5 years, 7·5 years, 10 years, and 23 years before ICU admission.

FINDINGS

Mortality predictions with a model based solely on disease history outperformed the Multimorbidity Index (Matthews correlation coefficient 0·265 vs 0·065), and performed similarly to SAPS II and APACHE II (Matthews correlation coefficient with disease history, age, and sex 0·326 vs 0·347 and 0·300 for SAPS II and APACHE II, respectively). Diagnoses up to 10 years before ICU admission affected current mortality prediction. Aggregation of previous disease history and acute physiology measures in a neural network yielded the most precise predictions of in-hospital mortality (Matthews correlation coefficient 0·391 for in-hospital mortality compared with 0·347 with SAPS II and 0·300 with APACHE II). These results for the aggregated model were validated in an external independent dataset of 1528 patients (Matthews correlation coefficient for prediction of in-hospital mortality 0·341).

INTERPRETATION

Longitudinal disease-spectrum-wide data available before ICU admission are useful for mortality prediction. Disease history can be used to differentiate mortality risk between patients with similar vital signs with more precision than SAPS II and APACHE II scores. Machine learning models can be deconvoluted to generate novel understandings of how ICU patient features from long-term and short-term events interact with each other. Explainable machine learning models are key in clinical settings, and our results emphasise how to progress towards the transformation of advanced models into actionable, transparent, and trustworthy clinical tools.

FUNDING

Novo Nordisk Foundation and Innovation Fund Denmark.

A bronchial-airway gene-expression classifier to improve the diagnosis of lung cancer: Clinical outcomes and cost-effectiveness analysis

Bronchoscopy is the safest procedure for lung cancer diagnosis when an invasive evaluation is required after imaging procedures. However, its sensitivity is relatively low, especially for small and peripheral lesions. We assessed benefits and costs of introducing a bronchial gene-expression classifier (BGC) to improve the performance of bronchoscopy and the overall diagnostic process for early detection of lung cancer. We used discrete-event simulation to compare clinical and economic outcomes of two different strategies with the standard practice in former and current smokers with indeterminate nodules: (i) location-based strategy-integrated the BGC to the bronchoscopy indication; (ii) simplified strategy-extended use of bronchoscopy plus BGC also on small and peripheral lesions. Outcomes modeled were rate of invasive procedures, quality-adjusted-life-years (QALYs), costs and incremental cost-effectiveness ratios. Compared to the standard practice, the location-based strategy (i) reduced absolute rate of invasive procedures by 3.3% without increasing costs at the current BGC market price. It resulted in savings when the BGC price was less than $3,000. The simplified strategy (ii) reduced absolute rate of invasive procedures by 10% and improved quality-adjusted life expectancy, producing an incremental cost-effectiveness ratio of $10,109 per QALY. In patients with indeterminate nodules, both BGC strategies reduced unnecessary invasive procedures at high risk of adverse events. Moreover, compared to the standard practice, the simplified use of BGC for central and peripheral lesions resulted in larger QALYs gains at acceptable cost. The location-based is cost-saving if the price of classifier declines.

Guidance of sentinel lymph node biopsy decisions in patients with T1–T2 melanoma using gene expression profiling

Aim: Can gene expression profiling be used to identify patients with T1–T2 melanoma at low risk for sentinel lymph node (SLN) positivity? Patients & methods: Bioinformatics modeling determined a population in which a 31-gene expression profile test predicted <5% SLN positivity. Multicenter, prospectively-tested (n = 1421) and retrospective (n = 690) cohorts were used for validation and outcomes, respectively. Results: Patients 55–64 years and ≥65 years with a class 1A (low-risk) profile had SLN positivity rates of 4.9% and 1.6%. Class 2B (high-risk) patients had SLN positivity rates of 30.8% and 11.9%. Melanoma-specific survival was 99.3% for patients ≥55 years with class 1A, T1–T2 tumors and 55.0% for class 2B, SLN-positive, T1–T2 tumors. Conclusion: The 31-gene expression profile test identifies patients who could potentially avoid SLN biopsy.

Novel high-resolution computed tomography-based radiomic classifier for screen-identified pulmonary nodules in the National Lung Screening Trial

Purpose

Optimization of the clinical management of screen-detected lung nodules is needed to avoid unnecessary diagnostic interventions. Herein we demonstrate the potential value of a novel radiomics-based approach for the classification of screen-detected indeterminate nodules.

Material and methods

Independent quantitative variables assessing various radiologic nodule features such as sphericity, flatness, elongation, spiculation, lobulation and curvature were developed from the NLST dataset using 726 indeterminate nodules (all ≥ 7 mm, benign, n = 318 and malignant, n = 408). Multivariate analysis was performed using least absolute shrinkage and selection operator (LASSO) method for variable selection and regularization in order to enhance the prediction accuracy and interpretability of the multivariate model. The bootstrapping method was then applied for the internal validation and the optimism-corrected AUC was reported for the final model.

Results

Eight of the originally considered 57 quantitative radiologic features were selected by LASSO multivariate modeling. These 8 features include variables capturing Location: vertical location (Offset carina centroid z), Size: volume estimate (Minimum enclosing brick), Shape: flatness, Density: texture analysis (Score Indicative of Lesion/Lung Aggression/Abnormality (SILA) texture), and surface characteristics: surface complexity (Maximum shape index and Average shape index), and estimates of surface curvature (Average positive mean curvature and Minimum mean curvature), all with P<0.01. The optimism-corrected AUC for these 8 features is 0.939.

Conclusions

Our novel radiomic LDCT-based approach for indeterminate screen-detected nodule characterization appears extremely promising however independent external validation is needed.

The Airway Transcriptome as a Biomarker for Early Lung Cancer Detection

Lung cancer remains the leading cause of cancer-related death due to its advanced stage at diagnosis. Early detection of lung cancer can be improved by better defining who should be screened radiographically and determining which imaging-detected pulmonary nodules are malignant. Gene expression biomarkers measured in normal-appearing airway epithelium provide an opportunity to use lung cancer-associated molecular changes in this tissue for early detection of lung cancer. Molecular changes in the airway may result from an etiologic field of injury and/or field cancerization. The etiologic field of injury reflects the aberrant physiologic response to carcinogen exposure that creates a susceptible microenvironment for cancer initiation. In contrast, field cancerization reflects effects of "first-hit" mutations in a clone of cells from which the tumor ultimately arises or the effects of the tumor on the surrounding tissue. These fields might have value both for assessing lung cancer risk and diagnosis. Cancer-associated gene expression changes in the bronchial airway have recently been used to develop and validate a 23-gene classifier that improves the diagnostic yield of bronchoscopy for lung cancer among intermediate-risk patients. Recent studies have demonstrated that these lung cancer-related gene expression changes extend to nasal epithelial cells that can be sampled noninvasively. While the bronchial gene expression biomarker is being adopted clinically, further work is necessary to explore the potential clinical utility of gene expression profiling in the nasal epithelium for lung cancer diagnosis, lung cancer risk assessment, and precision medicine for lung cancer treatment and chemoprevention. Clin Cancer Res; 24(13); 2984-92. ©2018 AACR.

Cost-Effectiveness of a Bronchial Genomic Classifier for the Diagnostic Evaluation of Lung Cancer

INTRODUCTION

The use of a bronchial genomic classifier has been shown to improve the diagnostic accuracy of bronchoscopy for suspected lung cancer by identifying patients who may be more suitable for radiographic surveillance as opposed to invasive procedures. Our objective was to assess the cost-effectiveness of bronchoscopy plus a genomic classifier versus bronchoscopy alone in the diagnostic work-up of patients at intermediate risk for lung cancer.

METHODS

A decision-analytic Markov model was developed to project the costs and effects of two competing strategies by using test performance from the Airway Epithelial Gene Expression in the Diagnosis of Lung Cancer-1 and Airway Epithelial Gene Expression in the Diagnosis of Lung Cancer-2 studies. The diagnostic accuracy of noninvasive and invasive follow-up, as well as associated adverse event rates, were derived from published literature. Procedure costs were based on claims data and 2016 inpatient and outpatient reimbursement amounts. The model projected the number of invasive follow-up procedures, 2-year costs and quality-adjusted life-years (QALYs) by strategy, and resulting incremental cost-effectiveness ratio discounted at 3% per annum.

RESULTS

Use of the genomic classifier reduced invasive procedures by 28% at 1 month and 18% at 2 years, respectively. Total costs and QALY gain were similar with classifier use ($27,221 versus $27,183 and 1.512 versus 1.509, respectively), resulting in an incremental cost-effectiveness ratio of $15,052 per QALY.

CONCLUSIONS

Our analysis suggests that the use of a genomic classifier is associated with meaningful reductions in invasive procedures at about equal costs and is therefore a high-value strategy in the diagnostic work-up of patients at intermediate risk of lung cancer.

Improving accuracy of RNA-based diagnosis and prognosis of oral cancer by using noninvasive methods

RNA-based diagnosis and prognosis of squamous cell carcinoma has been slow to come to the clinic. Improvements in RNA measurement, statistical evaluation, and sample preservation, along with increased sample numbers, have not made these methods reproducible enough to be used clinically. We propose that, in the case of squamous cell carcinoma of the oral cavity, a chief source of variability is sample dissection, which leads to variable amounts of stroma mixed in with tumor epithelium. This heterogeneity of the samples, which requires great care to avoid, makes it difficult to see changes in RNA levels specific to tumor cells. An evaluation of the data suggests that, paradoxically, brush biopsy samples of oral lesions may provide a more reproducible method than surgical acquisition of samples for miRNA measurement. The evidence also indicates that body fluid samples can show similar changes in miRNAs with oral squamous cell carcinoma (OSCC) as those seen in tumor brush biopsy samples - suggesting much of the miRNA in these samples is coming from the same source: tumor epithelium. We conclude that brush biopsy or body fluid samples may be superior to surgical samples in allowing miRNA-based diagnosis and prognosis of OSCC in that they feature a rapid method to obtain homogeneous tumor cells and/or RNA.