Marta Bogowicz

Comparison of PET and CT radiomics for prediction of local tumor control in head and neck squamous cell carcinoma

Abstract Purpose: An association between radiomic features extracted from CT and local tumor control in the head and neck squamous cell carcinoma (HNSCC) has been shown. This study investigated the value of pretreatment functional imaging (18F-FDG PET) radiomics for modeling of local tumor control. Material and Methods: Data from HNSCC patients (n = 121) treated with definitive radiochemotherapy were used for model training. In total, 569 radiomic features were extracted from both contrast-enhanced CT and 18F-FDG PET images in the primary tumor region. CT, PET and combined PET/CT radiomic models to assess local tumor control were trained separately. Five feature selection and three classification methods were implemented. The performance of the models was quantified using concordance index (CI) in 5-fold cross validation in the training cohort. The best models, per image modality, were compared and verified in the independent validation cohort (n = 51). The difference in CI was investigated using bootstrapping. Additionally, the observed and radiomics-based estimated probabilities of local tumor control were compared between two risk groups. Results: The feature selection using principal component analysis and the classification based on the multivariabale Cox regression with backward selection of the variables resulted in the best models for all image modalities (CICT = 0.72, CIPET = 0.74, CIPET/CT = 0.77). Tumors more homogenous in CT density (decreased GLSZMsize_zone_entropy) and with a focused region of high FDG uptake (higher GLSZMSZLGE) indicated better prognosis. No significant difference in the performance of the models in the validation cohort was observed (CICT = 0.73, CIPET = 0.71, CIPET/CT = 0.73). However, the CT radiomics-based model overestimated the probability of tumor control in the poor prognostic group (predicted  = 68%, observed  = 56%). Conclusions: Both CT and PET radiomics showed equally good discriminative power for local tumor control modeling in HNSCC. However, CT-based predictions overestimated the local control rate in the poor prognostic validation cohort, and thus, we recommend to base the local control modeling on the 18F-FDG PET.

Stability of radiomic features in CT perfusion maps

This study aimed to identify a set of stable radiomic parameters in CT perfusion (CTP) maps with respect to CTP calculation factors and image discretization, as an input for future prognostic models for local tumor response to chemo-radiotherapy. Pre-treatment CTP images of eleven patients with oropharyngeal carcinoma and eleven patients with non-small cell lung cancer (NSCLC) were analyzed. 315 radiomic parameters were studied per perfusion map (blood volume, blood flow and mean transit time). Radiomics robustness was investigated regarding the potentially standardizable (image discretization method, Hounsfield unit (HU) threshold, voxel size and temporal resolution) and non-standardizable (artery contouring and noise threshold) perfusion calculation factors using the intraclass correlation (ICC). To gain added value for our model radiomic parameters correlated with tumor volume, a well-known predictive factor for local tumor response to chemo-radiotherapy, were excluded from the analysis. The remaining stable radiomic parameters were grouped according to inter-parameter Spearman correlations and for each group the parameter with the highest ICC was included in the final set. The acceptance level was 0.9 and 0.7 for the ICC and correlation, respectively. The image discretization method using fixed number of bins or fixed intervals gave a similar number of stable radiomic parameters (around 40%). The potentially standardizable factors introduced more variability into radiomic parameters than the non-standardizable ones with 56-98% and 43-58% instability rates, respectively. The highest variability was observed for voxel size (instability rate  >97% for both patient cohorts). Without standardization of CTP calculation factors none of the studied radiomic parameters were stable. After standardization with respect to non-standardizable factors ten radiomic parameters were stable for both patient cohorts after correction for inter-parameter correlations. Voxel size, image discretization, HU threshold and temporal resolution have to be standardized to build a reliable predictive model based on CTP radiomics analysis.

Post-radiochemotherapy PET radiomics in head and neck cancer - The influence of radiomics implementation on the reproducibility of local control tumor models

PURPOSE This study investigated an association of post-radiochemotherapy (RCT) PET radiomics with local tumor control in head and neck squamous cell carcinoma (HNSCC) and evaluated the models against two radiomics software implementations. MATERIALS AND METHODS 649 features, available in two radiomics implementations and based on the same definitions, were extracted from HNSCC primary tumor region in 18F-FDG PET scans 3 months post definitive RCT (training cohort n = 128, validation cohort n = 50) and compared using the intraclass correlation coefficient (ICC). Local recurrence models were trained, separately for both implementations, using principal component analysis (PCA) and the least absolute shrinkage and selection operator. The reproducibility of the concordance indexes (CI) in univariable Cox regression for features preselected in PCA and the final multivariable models was investigated using respective features from the other implementation. RESULTS Only 80 PET radiomic features yielded ICC > 0.8 in the comparison between the implementations. The change of implementation caused high variability of CI in the univariable analysis. However, both final multivariable models performed equally well in the training and validation cohorts (CI > 0.7) independent of radiomics implementation. CONCLUSION The two post-RCT PET radiomic models, based on two different software implementations, were prognostic for local tumor control in HNSCC. However, 88% of the features was not reproducible between the implementations.

Development and validation of a radiomic signature to predict HPV (p16) status from standard CT imaging: a multicenter study

Objectives: Human papillomavirus (HPV) positive oropharyngeal cancer (oropharyngeal squamous cell carcinoma, OPSCC) is biologically and clinically different from HPV negative OPSCC. Here, we evaluate the use of a radiomic approach to identify the HPV status of OPSCC. Methods: Four independent cohorts, totaling 778 OPSCC patients with HPV determined by p16 were collected. We randomly assigned 80% of all data for model training (N = 628) and 20% for validation (N = 150). On the pre-treatment CT images, 902 radiomic features were calculated from the gross tumor volume. Multivariable modeling was performed using least absolute shrinkage and selection operator. To assess the impact of CT artifacts in predicting HPV (p16), a model was developed on all training data (Mall) and on the artifact-free subset of training data (Mno art). Models were validated on all validation data (Vall), and the subgroups with (Vart) and without (Vno art) artifacts. Kaplan–Meier survival analysis was performed to compare HPV status based on p16 and radiomic model predictions. Results: The area under the receiver operator curve for Mall and Mno art ranged between 0.70 and 0.80 and was not significantly different for all validation data sets. There was a consistent and significant split between survival curves with HPV status determined by p16 [p = 0.007; hazard ratio (HR): 0.46], Mall (p = 0.036; HR: 0.55) and Mno art (p = 0.027; HR: 0.49). Conclusion: This study provides proof of concept that molecular information can be derived from standard medical images and shows potential for radiomics as imaging biomarker of HPV status. Advances in knowledge: Radiomics has the potential to identify clinically relevant molecular phenotypes.

Exploratory Radiomics in Computed Tomography Perfusion of Prostate Cancer

BACKGROUND/AIM An evaluation if radiomic features of CT perfusion (CTP) can predict tumor grade and aggressiveness in prostate cancer was performed. MATERIALS AND METHODS Forty-seven patients had biopsy-confirmed prostate cancer, and received a CTP. Blood volume (BV), blood flow (BF) and mean transit time (MTT) maps were derived and 1,701 radiomic features were determined per patient. Regression models were built to estimate post-surgical Gleason score (GS), microvessel density (MVD) and distinguish between the different risk groups. RESULTS Six out of the 47 patients had to be excluded from further analysis. A weak relationship between postsurgical GS and one radiomic parameter was found (R2=0.21, p=0.01). The same parameter combined with MTT inter-quartile range was prognostic for the risk group categorisation (AUC=0.81). Two different radiomic parameters were able to distinguish between low-intermediate risk and high-intermediate risk (AUC=0.77). Four parameters correlated with MVD (R2=0.53, p<0.02). CONCLUSION This exploratory study shows the potential of radiomics to classify prostate cancer.

Influence of inter-observer delineation variability on radiomics stability in different tumor sites

Abstract Background: Radiomics is a promising methodology for quantitative analysis and description of radiological images using advanced mathematics and statistics. Tumor delineation, which is still often done manually, is an essential step in radiomics, however, inter-observer variability is a well-known uncertainty in radiation oncology. This study investigated the impact of inter-observer variability (IOV) in manual tumor delineation on the reliability of radiomic features (RF). Methods: Three different tumor types (head and neck squamous cell carcinoma (HNSCC), malignant pleural mesothelioma (MPM) and non-small cell lung cancer (NSCLC)) were included. For each site, eleven individual tumors were contoured on CT scans by three experienced radiation oncologists. Dice coefficients (DC) were calculated for quantification of delineation variability. RF were calculated with an in-house developed software implementation, which comprises 1404 features: shape (n = 18), histogram (n = 17), texture (n = 137) and wavelet (n = 1232). The IOV of RF was studied using the intraclass correlation coefficient (ICC). An ICC >0.8 indicates a good reproducibility. For the stable RF, an average linkage hierarchical clustering was performed to identify classes of uncorrelated features. Results: Median DC was high for NSCLC (0.86, range 0.57–0.90) and HNSCC (0.72, 0.21–0.89), whereas it was low for MPM (0.26, 0–0.9) indicating substantial IOV. Stability rate of RF correlated with DC and depended on tumor site, showing a high stability in NSCLC (90% of total parameters), acceptable stability in HNSCC (59% of total parameters) and low stability in MPM (36% of total parameters). Shape features showed the weakest stability across all tumor types. Hierarchical clustering revealed 14 groups of correlated and stable features for NSCLC and 6 groups for both HNSCC and MPM. Conclusion: Inter-observer delineation variability has a relevant influence on radiomics analysis and is strongly influenced by tumor type. This leads to a reduced number of suitable imaging features.

76P Robustness of radiomic features in [18F]-FDG PET/CT and [18F]-FDG PET/MR

Background: Radiomics is a promising tool for identification of new prognostic biomarkers. However, image reconstruction settings may affect the absolute values of radiomic features, which reduces their value as reliable biomarkers. PET/MR is becoming more available and often replaces PET/CT. The aim of this study was to quantify to what extend [18F]-FDG PET/CT radiomics models can be transferred to [18F]-FDG PET/MR. Methods: Nine patients with non-small cell lung cancer underwent first an [18F]-FDG PET/MR scan followed by an [18F]-FDG PET/CT scan (SIGNA PET/MR and Discovery PET/CT 690, GE Healthcare) with a delay time of 38 min +/−5 min. Patients had one single FDG injection for both scans. The primary tumors were segmented independently on the PET scans from PET/CT and PET/ MR with two semi-automated methods (gradient-based and threshold-based). Resampling was performed to the lowest resolution. In total, 1358 radiomic features were calculated, i.e. shape (18), intensity (17), texture (137), wavelets (1186). The intra-class correlation coefficient was used to compare the radiomic features in both image modalities. An ICC >0.9was considered stable among both types of PET scans. Results: The median relative volume difference of the tumors segmented on PET/CT and PET/MR was 4.8% (range 0.4–39.9%) for the gradient-based and 18.0% (range 0.7–71.2%) for the threshold-based method. A larger number of radiomic features was stable using the gradient-based method compared to the threshold-based method, which concurs with the improved reproducibility of tumor volume using gradient-basedmethod. More than 70.6% of shape and intensity features yielded an ICC >0.9 among both segmentation methods. However, only 51.5% of texture and 27.2% of wavelet features reached this criterion (for gradient-based and even less in thresholdbased method). In the wavelet features analysis, more features were robust in smoothed images (low-pass filtering) in comparison to images with emphasized heterogeneity (high-pass filtering). Conclusions: Shape and intensity radiomic features were robust comparing the two types of [18F]-FDG PET scans (PET/CT and PET/MR). In contrast, texture and wavelet features showed reduced stability, which needs to be considered for their use in prognostic modelling. Legal entity responsible for the study: Dr. Stephanie Tanadini-Lang Funding: Has not received any funding Disclosure: M. Guckenberger: Committee Member EORTC, ESTRO Board of Directors, Head of Working Group DEGRO. All other authors have declared no conflicts of interest.

Unconscious physiological response of healthy volunteers to dynamic respiration-synchronized couch motion

BackgroundIntrafractional motion can be a substantial uncertainty in precision radiotherapy. Conventionally, the target volume is expanded to account for the motion. Couch-tracking is an alternative, where the patient is moved to compensate for the tumor motion. However, the couch motion may influence the patient’s stress and respiration behavior decreasing the couch-tracking effectiveness.MethodsIn total, 100 volunteers were positioned supine on a robotic couch, which moved dynamically and respiration synchronized. During the measurement, the skin conductivity, the heartrate, and the gaze location were measured indicating the volunteer’s stress. Volunteers rated the subjective motion sickness using a questionnaire. The measurement alternated between static and tracking segments (three cycles), each 1 min long.ResultsThe respiration amplitude showed no significant difference between tracking and static segments, but decreased significantly from the first to the last tracking segment (p < 0.0001). The respiration frequency differed significantly between tracking and static segments (p < 0.0001), but not between the first and the last tracking segment. The physiological parameters and the questionnaire showed mild signals of stress and motion sickness.ConclusionGenerally, people tolerated the couch motions. The interaction between couch motion and the patient’s breathing pattern should be considered for a clinical implementation.Trial registrationThe study was registered at ClinicalTrials.gov (NCT02820532) and the Swiss national clinical trials portal (SNCTP000001878) on June 20, 2016.

Optimizing a perfusion CT protocol for head and neck cancer

Abstract Perfusion computed tomography (CTP) images tumor angiogenesis and can assess tumor aggressiveness. However, the CTP examinations are dose intensive. This study aimed to optimize a routinely used CTP protocol for the head and neck region in oncology in order to reduce the effective dose to the patient and simultaneously achieve the same image quality. The Alderson phantom was scanned on a GE Revolution CT scanner. A scan with our standard protocol for head and neck cancer patients was used (100kV, 80mAs, 5mm slice thickness and backprojection algorithm) and in seven predefined regions (ROI) the signal to noise ratio (SNR) was measured. For the dose optimized protocol, the tube voltage was lowered and the mAs adaptation protocol was used. To improve image quality different percentage of an adaptive statistical iterative reconstruction (ASiR) was applied. For a better resolution we set the slice thickness to 2.5 mm. The mAs adaption range and the percentage of the ASiR reconstruction were varied until we found a combination with the same median SNR in the seven defined ROIs as for our old protocol. For the old and the optimized protocol dose measurements were performed using 25 LiF-TLDs. Organ doses were calculated and the effective dose was determined based on the weighting factors of ICRP103. The optimized scanning protocol used a voltage of 80kV, a mAs range between 15 and 80, a noise level of 10%, and 50% ASiR reconstruction. The median SNR ratio was slightly better (14% better SNR) with the new protocol. An effective dose of 8 mSv was measured with the original protocol and 4 mSv with the optimized scanning protocol. For organs in the scanning field the dose was reduced by a factor of 2 and outside the field by a factor of 2.2. Advanced reconstruction algorithms allow a significant dose reduction and an improvement of image resolution, while maintaining the image quality.