Image-derived biomarkers and multimodal imaging strategies for lung cancer management

FDG PET texture indices as imaging biomarkers for epidermal growth factor receptor mutation status in lung adenocarcinoma

Identifying the epidermal growth factor receptor (EGFR) mutation status is important for the optimal treatment of patients with EGFR mutations. We investigated the relationship between ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography (PET) texture indices and EGFR mutation status in patients with newly diagnosed lung adenocarcinoma. We retrospectively analyzed data of patients with newly diagnosed lung adenocarcinoma who underwent pretreatment FDG PET/computed tomography and EGFR mutation testing between August 2014 and November 2020. Patients were divided into mutated EGFR and wild-type EGFR groups. The maximum standardized uptake value (SUVmax) and 31 texture indices for the primary tumor were calculated from PET images and compared between the two groups. Of the 66 patients included, 22 had mutated EGFR and 44 had wild-type EGFR. The SUVmax did not significantly differ between the two groups. Among the 31 evaluated texture indices, the following five showed a statistically significant difference between the groups: correlation (P = 0.003), gray-level nonuniformity for run (P = 0.042), run length nonuniformity (P = 0.02), coarseness (P = 0.006), and gray-level nonuniformity for zone (P = 0.04). Based on the preliminary results of this study in a small patient population, FDG PET texture indices may be potential imaging biomarkers for the EGFR mutation status in patients with newly diagnosed lung adenocarcinoma.

Integrating a dynamic central metabolism model of cancer cells with a hybrid 3D multiscale model for vascular hepatocellular carcinoma growth

We develop here a novel modelling approach with the aim of closing the conceptual gap between tumour-level metabolic processes and the metabolic processes occurring in individual cancer cells. In particular, the metabolism in hepatocellular carcinoma derived cell lines (HEPG2 cells) has been well characterized but implementations of multiscale models integrating this known metabolism have not been previously reported. We therefore extend a previously published multiscale model of vascular tumour growth, and integrate it with an experimentally verified network of central metabolism in HEPG2 cells. This resultant combined model links spatially heterogeneous vascular tumour growth with known metabolic networks within tumour cells and accounts for blood flow, angiogenesis, vascular remodelling and nutrient/growth factor transport within a growing tumour, as well as the movement of, and interactions between normal and cancer cells. Model simulations report for the first time, predictions of spatially resolved time courses of core metabolites in HEPG2 cells. These simulations can be performed at a sufficient scale to incorporate clinically relevant features of different tumour systems using reasonable computational resources. Our results predict larger than expected temporal and spatial heterogeneity in the intracellular concentrations of glucose, oxygen, lactate pyruvate, f16bp and Acetyl-CoA. The integrated multiscale model developed here provides an ideal quantitative framework in which to study the relationship between dosage, timing, and scheduling of anti-neoplastic agents and the physiological effects of tumour metabolism at the cellular level. Such models, therefore, have the potential to inform treatment decisions when drug response is dependent on the metabolic state of individual cancer cells.

Correlation of epidermal growth factor receptor mutation status and PD-L1 expression with [18F]FDG PET using volume-based parameters in non-small cell lung cancer

OBJECTIVE

We investigated the relationship between 2-deoxy-2-[18F]fluoro-D-glucose (FDG) PET using volume-based parameters and epidermal growth factor receptor (EGFR) mutation status, programmed death-ligand-1 (PD-L1) expression level, and their combination, in pretreated non-small cell lung cancer (NSCLC).

METHODS

FDG PET findings and EGFR mutation status and PD-L1 expression level were investigated retrospectively in 93 patients with newly diagnosed NSCLC (77 adenocarcinomas, 16 squamous cell carcinomas). Tumors were divided into six groups: EGFR mutant/negative PD-L1, EGFR mutant/low PD-L1, EGFR mutant/high PD-L1, EGFR wild/negative PD-L1, EGFR wild/low PD-L1, and EGFR wild/high PD-L1. The maximum standardized uptake value (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) for primary tumor were measured from PET images. The EGFR mutation status and PD-L1 expression level were estimated in tumor tissue specimens and compared with the PET parameters.

RESULTS

None of the PET parameters differed significantly between EGFR-mutated and wild-type EGFR. According to the PD-L1 level, significant differences were detected in SUVmax (P = 0.001) and TLG (P = 0.016), but not MTV. Comparing all six groups, significant difference was detected in only SUVmax (P = 0.011).

CONCLUSION

Based on the preliminary results of this study, FDG PET may help in the prediction of PD-L1 expression level, but not EGFR mutation status, in patients with newly diagnosed NSCLC. The SUVmax rather than MTV or TLG, may be of value in predicting the six groups according to the combination of EGFR mutation status and PD-L1 expression level.

Accurate Tumor Delineation vs. Rough Volume of Interest Analysis for 18F-FDG PET/CT Radiomics-Based Prognostic Modeling inNon-Small Cell Lung Cancer

Background

The aim of this work was to investigate the ability of building prognostic models in non-small cell lung cancer (NSCLC) using radiomic features from positron emission tomography and computed tomography with 2-deoxy-2-[fluorine-18]fluoro-d-glucose (¹⁸F-FDG PET/CT) images based on a "rough" volume of interest (VOI) containing the tumor instead of its accurate delineation, which is a significant time-consuming bottleneck of radiomics analyses.

Methods

A cohort of 138 patients with stage II-III NSCLC treated with radiochemotherapy recruited retrospectively (n = 87) and prospectively (n = 51) was used. Two approaches were compared: firstly, the radiomic features were extracted from the delineated primary tumor volumes in both PET (using the automated fuzzy locally adaptive Bayesian, FLAB) and CT (using a semi-automated approach with 3D Slicer™) components. Both delineations were carried out within previously manually defined "rough" VOIs containing the tumor and the surrounding tissues, which were exploited for the second approach: the same features were extracted from this alternative VOI. Both sets for features were then combined with the clinical variables and processed through the same machine learning (ML) pipelines using the retrospectively recruited patients as the training set and the prospectively recruited patients as the testing set. Logistic regression (LR), random forest (RF), and support vector machine (SVM), as well as their consensus through averaging the output probabilities, were considered for feature selection and modeling for overall survival (OS) prediction as a binary classification (either median OS or 6 months OS). The resulting models were compared in terms of balanced accuracy, sensitivity, and specificity.

Results

Overall, better performance was achieved using the features from delineated tumor volumes. This was observed consistently across ML algorithms and for the two clinical endpoints. However, the loss of performance was not significant, especially when a consensus of the three ML algorithms was considered (0.89 vs. 0.88 and 0.78 vs. 0.77).

Conclusion

Our findings suggest that it is feasible to achieve similar levels of prognostic accuracy in radiomics-based modeling by relying on a faster and easier VOI definition, skipping a time-consuming tumor delineation step, thus facilitating automation of the whole radiomics workflow. The associated cost is a loss of performance in the resulting models, although this loss can be greatly mitigated when a consensus of several models is relied upon.

Comparison and Fusion of Machine Learning Algorithms for Prospective Validation of PET/CT Radiomic Features Prognostic Value in Stage II-III Non-Small Cell Lung Cancer

Machine learning (ML) algorithms for selecting and combining radiomic features into multiparametric prediction models have become popular; however, it has been shown that large variations in performance can be obtained by relying on different approaches. The purpose of this study was to evaluate the potential benefit of combining different algorithms into an improved consensus for the final prediction, as it has been shown in other fields. Methods: The evaluation was carried out in the context of the use of radiomics from ¹⁸F-FDG PET/CT images for predicting outcome in stage II-III Non-Small Cell Lung Cancer. A cohort of 138 patients was exploited for the present analysis. Eighty-seven patients had been previously recruited retrospectively for another study and were used here for training and internal validation. We also used data from prospectively recruited patients (n = 51) for testing. Three different machine learning pipelines relying on embedded feature selection were trained to predict overall survival (OS) as a binary classification: Support Vector machines (SVMs), Random Forests (RFs), and Logistic Regression (LR). Two different clinical endpoints were investigated: median OS or OS shorter than 6 months. The fusion of the three approaches was implemented using two different strategies: majority voting on the binary outputs or averaging of the output probabilities. Results: Our results confirm previous findings, highlighting that different ML pipelines select different sets of features and reach different classification performances (accuracy in the testing set ranging between 63% and 67% for median OS, and between 75% and 80% for OS < 6 months). Generating a consensus improved the performance for both endpoints; with the probabilities averaging strategy outperforming the majority voting (accuracy of 78% vs. 71% for median OS and 89 vs. 84% for OS < 6 months). Overall, the performance of these radiomic-based models outperformed the standard clinical staging in both endpoints (accuracy of 58% and 53% accuracy in the testing set for each endpoint). Conclusion: Although obtained in a small cohort of patients, our results suggest that a consensus of machine learning algorithms can improve performance in the context of radiomics. The resulting prognostic stratification in the prospective testing cohort is higher than when relying on the clinical stage. This could be of interest for clinical practice as it could help to identify patients with higher risk amongst stage II and III patients, who could benefit from intensified treatment and/or more frequent follow-up after treatment.

Modern radiopharmaceuticals for lung cancer imaging with positron emission tomography/computed tomography scan: A systematic review

Introduction:

In this study, we evaluated the use and the contribution of radiopharmaceuticals to the field of lung neoplasms imaging using positron emission tomography/computed tomography.

Methods:

We conducted review of the current literature at PubMed/MEDLINE until February 2020. The search language was English.

Results:

The most widely used radiopharmaceuticals are the following:

Experimental/pre-clinical approaches: (18)F-Misonidazole (18F-MISO) under clinical development, D(18)F-Fluoro-Methyl-Tyrosine (18F-FMT), 18F-FAMT (L-[3-18F] (18)F-Fluorothymidine (18F-FLT)), (18)F-Fluoro-Azomycin-Arabinoside (18F-FAZA), (68)Ga-Neomannosylated-Human-Serum-Albumin (68Ga-MSA) (23), (68)Ga-Tetraazacyclododecane (68Ga-DOTA) (as theranostic agent), (11)C-Methionine (11C-MET), 18F-FPDOPA, α_νβ₃ integrin, ⁶⁸Ga-RGD₂, ⁶⁴Cu-DOTA-RGD, ¹⁸F-Alfatide, Folate Radio tracers, and immuno-positron emission tomography radiopharmaceutical agents.

Clinically approved procedures/radiopharmaceuticals agents: (18)F-Fluoro-Deoxy-Glucose (18F-FDG), (18)F-sodium fluoride (18F-NaF) (bone metastases), and (68)Ga-Tetraazacyclododecane (68Ga-DOTA). The quantitative determination and the change in radiopharmaceutical uptake parameters such as standard uptake value, metabolic tumor volume, total lesion glycolysis, FAZA tumor to muscle ratio, standard uptake value tumor to liver ratio, standard uptake value tumor to spleen ratio, standard uptake value maximum ratio, and the degree of hypoxia have prognostic and predictive (concerning the therapeutic outcome) value. They have been associated with the assessment of overall survival and disease free survival. With the positron emission tomography/computed tomography radiopharmaceuticals, the sensitivity and the specificity of the method have increased.

Conclusion:

In terms of lung cancer, positron emission tomography/computed tomography may have clinical application and utility (a) in personalizing treatment, (b) as a biomarker for the estimation of overall survival, disease free survival, and (c) apply a cost-effective patient approach because it reveals focuses of the disease, which are not found with the other imaging methods.

Quantitative PET/MR imaging of lung cancer in the presence of artifacts in the MR-based attenuation correction maps

Background

Positron emission tomography (PET)/magnetic resonance (MR) imaging may become increasingly important for assessing tumor therapy response. A prerequisite for quantitative PET/MR imaging is reliable and repeatable MR-based attenuation correction (AC).

Purpose

To investigate the frequency and test–retest reproducibility of artifacts in MR-AC maps in a lung cancer patient cohort and to study the impact of artifact corrections on PET-based tumor quantification.

Material and Methods

Twenty-five lung cancer patients underwent single-day, test–retest, 18F-fluorodeoxyglucose (FDG) PET/MR imaging. The acquired MR-AC maps were inspected for truncation, susceptibility, and tissue inversion artifacts. An anatomy-based bone template and a PET-based estimation of truncated arms were employed, while susceptibility artifacts were corrected manually. We report the frequencies of artifacts and the relative difference (RD) on standardized uptake value (SUV) based quantification in PET images reconstructed with the corrected AC maps.

Results

Truncation artifacts were found in all 50 acquisitions (100%), while susceptibility and tissue inversion artifacts were observed in six (12%) and 26 (52%) of the scans, respectively. The RD in lung tumor SUV was < 5% from bone and truncation corrections, while up to 20% RD was introduced after susceptibility artifact correction, with large inconsistencies between test–retest scans.

Conclusion

The absence of bone and truncation artifacts have limited effect on the PET quantification of lung lesions. In contrast, susceptibility artifacts caused significant and inconsistent underestimations of the lung tumor SUVs, between test–retest scans. This may have clinical implications for patients undergoing serial imaging for tumor therapy response assessment.

Total metabolic tumor volume by 18F-FDG PET/CT for the prediction of outcome in patients with non-small cell lung cancer

OBJECTIVE

Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) are imaging parameters derived from 18F-FDG PET/CT that have been proposed for risk stratification of cancer patients. The aim of our study was to test whether these whole-body volumetric imaging parameters may predict outcome in patients with non-small cell lung cancer (NSCLC).

METHODS

Sixty-five patients (45 men, 20 women; mean age ± SD, 65 ± 12 years), with histologically proven NSCLC who had undergone 18F-FDG PET/CT scan before any therapy, were included in the study. Imaging parameters including SUV_max, SUV_mean, total MTV (MTV_TOT) and whole-body TLG (TLG_WB) were determined. Univariate and multivariate analyses of clinical and imaging variables were performed using Cox proportional hazards regression. Survival analysis was performed using Kaplan-Meier method and log-rank tests.

RESULTS

A total of 298 lesions were analyzed including 65 primary tumors, 114 metastatic lymph nodes and 119 distant metastases. MTV_TOT and TLG_WB could be determined in 276 lesions. Mean value of MTV_TOT was 81.83 ml ± 14.63 ml (SE) whereas mean value of TLG_WB was 459.88 g ± 77.02 g (SE). Univariate analysis showed that, among the variables tested, primary tumor diameter (p = 0.0470), MTV of primary tumor (p = 0.0299), stage (p < 0.0001), treatment (p < 0.0001), MTV_TOT (p = 0.0003) and TLG_WB (p = 0.0002) predicted progression-free survival in NSCLC patients, while age (p = 0.0550), MTV of primary tumor (p = 0.0375), stage (p < 0.0001), treatment (p < 0.0001), MTV_TOT (p = 0.0001) and TLG_WB (p = 0.0008) predicted overall survival. At multivariate analysis age, TLG_WB and stage were retained in the model for prediction of progression-free survival (p < 0.0001), while age, MTV_TOT and stage were retained in the model for prediction of overall survival (p < 0.0001). Survival analysis showed that patients with TLG_WB ≤ 54.7 g had a significantly prolonged progression-free survival as compared to patients with TLG_WB > 54.7 g (p < 0.0001). Moreover, overall survival was significantly better in patients showing a MTV_TOT ≤ 9.5 ml as compared to those having MTV_TOT > 9.5 ml (p < 0.0001). Similar results were obtained in a subgroup of 43 patients with advanced disease (stages III and IV).

CONCLUSIONS

Whole-body PET-based volumetric imaging parameters are able to predict outcome in NSCLC patients.

PET Radiomics in NSCLC: state of the art and a proposal for harmonization of methodology

Imaging with positron emission tomography (PET)/computed tomography (CT) is crucial in the management of cancer because of its value in tumor staging, response assessment, restaging, prognosis and treatment responsiveness prediction. In the last years, interest has grown in texture analysis which provides an "in-vivo" lesion characterization, and predictive information in several malignances including NSCLC; however several drawbacks and limitations affect these studies, especially because of lack of standardization in features calculation, definitions and methodology reporting. The present paper provides a comprehensive review of literature describing the state-of-the-art of FDG-PET/CT texture analysis in NSCLC, suggesting a proposal for harmonization of methodology.

Reliability of PET/CT Shape and Heterogeneity Features in Functional and Morphologic Components of Non-Small Cell Lung Cancer Tumors: A Repeatability Analysis in a Prospective Multicenter Cohort

The main purpose of this study was to assess the reliability of shape and heterogeneity features in both the PET and the low-dose CT components of PET/CT. A secondary objective was to investigate the impact of image quantization. Methods: A Health Insurance Portability and Accountability Act-compliant secondary analysis of deidentified prospectively acquired PET/CT test-retest datasets of 74 patients from multicenter Merck and American College of Radiology Imaging Network trials was performed. Metabolically active volumes were automatically delineated on PET with a fuzzy locally adaptive bayesian algorithm. Software was used to semiautomatically delineate the anatomic volumes on the low-dose CT component. Two quantization methods were considered: a quantization into a set number of bins (quantization B) and an alternative quantization with bins of fixed width (quantization W). Four shape descriptors, 10 first-order metrics, and 26 textural features were evaluated. Bland-Altman analysis was used to quantify repeatability. Features were subsequently categorized as very reliable, reliable, moderately reliable, or poorly reliable with respect to the corresponding volume variability. Results: Repeatability was highly variable among features. Numerous metrics were identified as poorly or moderately reliable. Others were reliable or very reliable in both modalities and in all categories (shape and first-, second-, and third-order metrics). Image quantization played a major role in feature repeatability. Features were more reliable in PET with quantization B, whereas quantization W showed better results in CT. Conclusion: The test-retest repeatability of shape and heterogeneity features in PET and low-dose CT varied greatly among metrics. The level of repeatability also depended strongly on the quantization step, with different optimal choices for each modality. The repeatability of PET and low-dose CT features should be carefully considered when selecting metrics to build multiparametric models.

Development of a nomogram combining clinical staging with (18)F-FDG PET/CT image features in non-small-cell lung cancer stage I-III

PURPOSE

Our goal was to develop a nomogram by exploiting intratumour heterogeneity on CT and PET images from routine (18)F-FDG PET/CT acquisitions to identify patients with the poorest prognosis.

METHODS

This retrospective study included 116 patients with NSCLC stage I, II or III and with staging (18)F-FDG PET/CT imaging. Primary tumour volumes were delineated using the FLAB algorithm and 3D Slicer™ on PET and CT images, respectively. PET and CT heterogeneities were quantified using texture analysis. The reproducibility of the CT features was assessed on a separate test-retest dataset. The stratification power of the PET/CT features was evaluated using the Kaplan-Meier method and the log-rank test. The best standard metric (functional volume) was combined with the least redundant and most prognostic PET/CT heterogeneity features to build the nomogram.

RESULTS

PET entropy and CT zone percentage had the highest complementary values with clinical stage and functional volume. The nomogram improved stratification amongst patients with stage II and III disease, allowing identification of patients with the poorest prognosis (clinical stage III, large tumour volume, high PET heterogeneity and low CT heterogeneity).

CONCLUSION

Intratumour heterogeneity quantified using textural features on both CT and PET images from routine staging (18)F-FDG PET/CT acquisitions can be used to create a nomogram with higher stratification power than staging alone.

A Population-Based Gaussian Mixture Model Incorporating 18F-FDG PET and Diffusion-Weighted MRI Quantifies Tumor Tissue Classes

The aim of our study was to create a novel Gaussian mixture modeling (GMM) pipeline to model the complementary information derived from(18)F-FDG PET and diffusion-weighted MRI (DW-MRI) to separate the tumor microenvironment into relevant tissue compartments and follow the development of these compartments longitudinally.

METHODS

Serial (18)F-FDG PET and apparent diffusion coefficient (ADC) maps derived from DW-MR images of NCI-H460 xenograft tumors were coregistered, and a population-based GMM was implemented on the complementary imaging data. The tumor microenvironment was segmented into 3 distinct regions and correlated with histology. ANCOVA was applied to gauge how well the total tumor volume was a predictor for the ADC and (18)F-FDG, or if ADC was a good predictor of (18)F-FDG for average values in the whole tumor or average necrotic and viable tissues.

RESULTS

The coregistered PET/MR images were in excellent agreement with histology, both visually and quantitatively, and allowed for validation of the last-time-point measurements. Strong correlations were found for the necrotic (r = 0.88) and viable fractions (r = 0.87) between histology and clustering. The GMM provided probabilities for each compartment with uncertainties expressed as a mixture of tissues in which the resolution of scans was inadequate to accurately separate tissues. The ANCOVA suggested that both ADC and (18)F-FDG in the whole tumor (P = 0.0009, P = 0.02) as well as necrotic (P = 0.008, P = 0.02) and viable (P = 0.003, P = 0.01) tissues were a positive, linear function of total tumor volume. ADC proved to be a positive predictor of (18)F-FDG in the whole tumor (P = 0.001) and necrotic (P = 0.02) and viable (P = 0.0001) tissues.

CONCLUSION

The complementary information of (18)F-FDG and ADC longitudinal measurements in xenograft tumors allows for segmentation into distinct tissues when using the novel GMM pipeline. Leveraging the power of multiparametric PET/MRI in this manner has the potential to take the assessment of disease outcome beyond RECIST and could provide an important impact to the field of precision medicine.

Multiparametric imaging of patient and tumour heterogeneity in non-small-cell lung cancer: quantification of tumour hypoxia, metabolism and perfusion

Purpose

Multiple imaging techniques are nowadays available for clinical in-vivo visualization of tumour biology. FDG PET/CT identifies increased tumour metabolism, hypoxia PET visualizes tumour oxygenation and dynamic contrast-enhanced (DCE) CT characterizes vasculature and morphology. We explored the relationships among these biological features in patients with non-small-cell lung cancer (NSCLC) at both the patient level and the tumour subvolume level.

Methods

A group of 14 NSCLC patients from two ongoing clinical trials (NCT01024829 and NCT01210378) were scanned using FDG PET/CT, HX4 PET/CT and DCE CT prior to chemoradiotherapy. Standardized uptake values (SUV) in the primary tumour were calculated for the FDG and hypoxia HX4 PET/CT scans. For hypoxia imaging, the hypoxic volume, fraction and tumour-to-blood ratio (TBR) were also defined. Blood flow and blood volume were obtained from DCE CT imaging. A tumour subvolume analysis was used to quantify the spatial overlap between subvolumes.

Results

At the patient level, negative correlations were observed between blood flow and the hypoxia parameters (TBR >1.2): hypoxic volume (−0.65, p = 0.014), hypoxic fraction (−0.60, p = 0.025) and TBR (−0.56, p = 0.042). At the tumour subvolume level, hypoxic and metabolically active subvolumes showed an overlap of 53 ± 36 %. Overlap between hypoxic sub-volumes and those with high blood flow and blood volume was smaller: 15 ± 17 % and 28 ± 28 %, respectively. Half of the patients showed a spatial mismatch (overlap <5 %) between increased blood flow and hypoxia.

Conclusion

The biological imaging features defined in NSCLC tumours showed large interpatient and intratumour variability. There was overlap between hypoxic and metabolically active subvolumes in the majority of tumours, there was spatial mismatch between regions with high blood flow and those with increased hypoxia.

Electronic supplementary material

The online version of this article (doi:10.1007/s00259-015-3169-4) contains supplementary material, which is available to authorized users.