Evaluation of 3-l- and 3-d-[18F]Fluorophenylalanines as PET Tracers for Tumor Imaging

Purpose: The preclinical evaluation of 3-l- and 3-d-[18F]FPhe in comparison to [18F]FET, an established tracer for tumor imaging. Methods: In vitro studies were conducted with MCF-7, PC-3, and U87 MG human tumor cell lines. In vivo µPET studies were conducted in healthy rats with/without the inhibition of peripheral aromatic l-amino acid decarboxylase by benserazide pretreatment (n = 3 each), in mice bearing subcutaneous MCF-7 or PC-3 tumor xenografts (n = 10), and in rats bearing orthotopic U87 MG tumor xenografts (n = 14). Tracer accumulation was quantified by SUVmax, SUVmean and tumor-to-brain ratios (TBrR). Results: The uptake of 3-l-[18F]FPhe in MCF-7 and PC-3 cells was significantly higher relative to [18F]FET. The uptake of all three tracers was significantly reduced by the suppression of amino acid transport systems L or ASC. 3-l-[18F]FPhe but not 3-d-[18F]FPhe exhibited protein incorporation. In benserazide-treated healthy rats, brain uptake after 42–120 min was significantly higher for 3-d-[18F]FPhe vs. 3-l-[18F]FPhe. [18F]FET showed significantly higher uptake into subcutaneous MCF-7 tumors (52–60 min p.i.), while early uptake into orthotopic U87 MG tumors was significantly higher for 3-l-[18F]FPhe (SUVmax: 3-l-[18F]FPhe, 107.6 ± 11.3; 3-d-[18F]FPhe, 86.0 ± 4.3; [18F]FET, 90.2 ± 7.7). Increased tumoral expression of LAT1 and ASCT2 was confirmed immunohistologically. Conclusion: Both novel tracers enable accurate tumor delineation with an imaging quality comparable to [18F]FET.

Predicting distant metastases in soft-tissue sarcomas from PET-CT scans using constrained hierarchical multi-modality feature learning

Objective: Positron emission tomography-computed tomography (PET-CT) is regarded as the imaging modality of choice for the management of soft-tissue sarcomas (STSs). Distant metastases (DM) are the leading cause of death in STS patients and early detection is important to effectively manage tumors with surgery, radiotherapy and chemotherapy. In this study, we aim to early detect DM in patients with STS using their PET-CT data. Approach: We derive a new convolutional neural network (CNN) method for early DM detection. The novelty of our method is the introduction of a constrained hierarchical multi-modality feature learning approach to integrate functional imaging (PET) features with anatomical imaging (CT) features. In addition, we removed the reliance on manual input, e.g., tumor delineation, for extracting imaging features. Main results: Our experimental results on a well-established benchmark PET-CT dataset show that our method achieved the highest accuracy (0.896) and AUC (0.903) scores when compared to the state-of-the-art methods (unpaired student’s t-test p-value < 0.05). Significance: Our method could be an effective and supportive tool to aid physicians in tumor quantification and in identifying image biomarkers for cancer treatment.

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

BackgroundThe 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.MethodsA 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.ResultsOverall, 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).ConclusionOur 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.

Recent Advances in Second Near-Infrared Region (NIR-II) Fluorophores and Biomedical Applications

Fluorescence imaging technique, characterized by high sensitivity, non-invasiveness and no radiation hazard, has been widely applicated in the biomedical field. However, the depth of tissue penetration is limited in the traditional (400–700 nm) and NIR-I (the first near-infrared region, 700–900 nm) imaging, which urges researchers to explore novel bioimaging modalities with high imaging performance. Prominent progress in the second near-infrared region (NIR-II, 1000–1700 nm) has greatly promoted the development of biomedical imaging. The NIR-II fluorescence imaging significantly overcomes the strong tissue absorption, auto-fluorescence as well as photon scattering, and has deep tissue penetration, micron-level spatial resolution, and high signal-to-background ratio. NIR-II bioimaging has been regarded as the most promising in vivo fluorescence imaging technology. High brightness and biocompatible fluorescent probes are crucial important for NIR-II in vivo imaging. Herein, we focus on the recently developed NIR-II fluorescent cores and their applications in the field of biomedicine, especially in tumor delineation and image-guided surgery, vascular imaging, NIR-II-based photothermal therapy and photodynamic therapy, drug delivery. Besides, the challenges and potential future developments of NIR-II fluorescence imaging are further discussed. It is expected that our review will lay a foundation for clinical translation of NIR-II biological imaging, and inspire new ideas and more researches in this field.

Subtraction iodine imaging with area detector CT to improve tumor delineation and measurability of tumor size and depth of invasion in tongue squamous cell carcinoma

Purpose Tumor size and depth of invasion (DOI) are mandatory assessments for tumor classification in tongue cancer but are often non-assessable on CT due to dental artifacts. This study investigated whether subtraction iodine imaging (SII) would improve tumor delineation and measurability. Materials and methods Fifty-seven consecutive patients with tongue cancer, who underwent scanning with a 320-row area detector CT with contrast administration and were treated with surgical resection, were retrospectively evaluated. CT was reconstructed with single-energy projection-based metallic artifact reduction (sCT). SII was generated by subtracting the pre-contrast volume scans from the post-contrast volume scans using a high-resolution deformable registration algorithm. MRI scans were also evaluated for comparing the ability of measurements. Two radiologists visually graded the tumor delineation using a 5-point scale. Tumor size and DOI were measured wherever possible. The tumor delineation score was compared using the Wilcoxon signed-rank method. Spearman’s correlations between imaging and pathological measurements were calculated. Intraclass correlation coefficients of measurements between readers were estimated. Results The tumor delineation score was greater on sCT-plus-SII than on sCT alone (medians: 3 and 1, respectively; p < 0.001), with higher number of detectable cases observed with sCT-plus-SII (36/57 [63.2%]) than sCT alone (21/57 [36.8%]). Tumor size and DOI measurability were higher with sCT-plus-SII (29/57 [50.9%]) than with sCT alone (17/57 [29.8%]). MRI had the highest detectability (52/57 [91.2%]) and measurability (46/57 [80.7%]). Correlation coefficients between radiological and pathological tumor size and DOI were similar for sCT (0.83–0.88), sCT-plus-SII (0.78–0.84), and MRI (0.78–0.90). Intraclass correlation coefficients were higher than 0.95 for each modality. Conclusions SII improves detectability and measurability of tumor size and DOI in patients with oral tongue squamous cell carcinoma, thus increasing the diagnostic potential. SII may also be beneficial for cases unevaluable on MRI due to artifacts or for patients with contraindications to MRI.

Automatic segmentation of lung tumors on CT images based on a 2D & 3D hybrid convolutional neural network

Objective: A stable and accurate automatic tumor delineation method has been developed to facilitate the intelligent design of lung cancer radiotherapy process. The purpose of this paper is to introduce an automatic tumor segmentation network for lung cancer on CT images based on deep learning. Methods: In this paper, a hybrid convolution neural network (CNN) combining 2D CNN and 3D CNN was implemented for the automatic lung tumor delineation using CT images. 3D CNN used V-Net model for the extraction of tumor context information from CT sequence images. 2D CNN used an encoder–decoder structure based on dense connection scheme, which could expand information flow and promote feature propagation. Next, 2D features and 3D features were fused through a hybrid module. Meanwhile, the hybrid CNN was compared with the individual 3D CNN and 2D CNN, and three evaluation metrics, Dice, Jaccard and Hausdorff distance (HD), were used for quantitative evaluation. The relationship between the segmentation performance of hybrid network and the GTV volume size was also explored. Results: The newly introduced hybrid CNN was trained and tested on a dataset of 260 cases, and could achieve a median value of 0.73, with mean and stand deviation of 0.72 ± 0.10 for the Dice metric, 0.58 ± 0.13 and 21.73 ± 13.30 mm for the Jaccard and HD metrics, respectively. The hybrid network significantly outperformed the individual 3D CNN and 2D CNN in the three examined evaluation metrics (p < 0.001). A larger GTV present a higher value for the Dice metric, but its delineation at the tumor boundary is unstable. Conclusions: The implemented hybrid CNN was able to achieve good lung tumor segmentation performance on CT images. Advances in knowledge: The hybrid CNN has valuable prospect with the ability to segment lung tumor.

Radiomics in hepatic metastasis by colorectal cancer

Background Radiomics is an emerging field and has a keen interest, especially in the oncology field. The process of a radiomics study consists of lesion segmentation, feature extraction, consistency analysis of features, feature selection, and model building. Manual segmentation is one of the most critical parts of radiomics. It can be time-consuming and suffers from variability in tumor delineation, which leads to the reproducibility problem of calculating parameters and assessing spatial tumor heterogeneity, particularly in large or multiple tumors. Radiomic features provides data on tumor phenotype as well as cancer microenvironment. Radiomics derived parameters, when associated with other pertinent data and correlated with outcomes data, can produce accurate robust evidence based clinical decision support systems. The principal challenge is the optimal collection and integration of diverse multimodal data sources in a quantitative manner that delivers unambiguous clinical predictions that accurately and robustly enable outcome prediction as a function of the impending decisions. Methods The search covered the years from January 2010 to January 2021. The inclusion criterion was: clinical study evaluating radiomics of liver colorectal metastases. Exclusion criteria were studies with no sufficient reported data, case report, review or editorial letter. Results We recognized 38 studies that assessed radiomics in mCRC from January 2010 to January 2021. Twenty were on different tpics, 5 corresponded to most criteria; 3 are review, or letter to editors; so 10 articles were included. Conclusions In colorectal liver metastases radiomics should be a valid tool for the characterization of lesions, in the stratification of patients based on the risk of relapse after surgical treatment and in the prediction of response to chemotherapy treatment.

Feasibility of Different Tumor Delineation Approaches for 18F-PSMA-1007 PET/CT Imaging in Prostate Cancer Patients

BackgroundDelineation of PSMA-positive tumor volume on PET using PSMA-ligands is of highest clinical interest as changes of PSMA-PET/CT-derived whole tumor volume (WTV) have shown to correlate with treatment response in metastatic prostate cancer patients. So far, WTV estimation was performed on PET using 68Ga-labeled ligands; nonetheless, 18F-labeled PET ligands are gaining increasing importance due to advantages over 68Ga-labeled compounds. However, standardized tumor delineation methods for 18F-labeled PET ligands have not been established so far. As correlation of PET-based information and morphological extent in osseous and visceral metastases is hampered by morphological delineation, low contrast in liver tissue and movement artefacts, we correlated CT-based volume of lymph node metastases (LNM) and different PET-based delineation approaches for thresholding on 18F-PSMA-1007 PET.MethodsFifty patients with metastatic prostate cancer, 18F-PSMA-1007 PET/CT and non-bulky LNM (short-axis diameter ≥10mm) were included. Fifty LNM were volumetrically assessed on contrast-enhanced CT (volumetric reference standard). Different approaches for tumor volume delineation were applied and correlated with the reference standard: I) fixed SUV threshold, II) isocontour thresholding relative to SUVmax (SUV%), and thresholds relative to III) liver (SUVliver), IV) parotis (SUVparotis) and V) spleen (SUVspleen).ResultsA fixed SUV of 4.0 (r=0.807, r2 = 0.651, p&lt;0.001) showed the best overall association with the volumetric reference. 55% SUVmax (r=0.627, r2 = 0.393, p&lt;0.001) showed highest association using an isocontour-based threshold. Best background-based approaches were 60% SUVliver (r=0.715, r2 = 0.511, p&lt;0.001), 80% SUVparotis (r=0.762, r2 = 0.581, p&lt;0.001) and 60% SUVspleen (r=0.645, r2 = 0.416, p&lt;0.001). Background tissues SUVliver, SUVparotis &amp; SUVspleen did not correlate (p&gt;0.05 each). Recently reported cut-offs for intraprostatic tumor delineation (isocontour 44% SUVmax, 42% SUVmax and 20% SUVmax) revealed inferior association for LNM delineation.ConclusionsA threshold of SUV 4.0 for tumor delineation showed highest association with volumetric reference standard irrespective of potential changes in PSMA-avidity of background tissues (e. g. parotis). This approach is easily applicable in clinical routine without specific software requirements. Further studies applying this approach for total tumor volume delineation are initiated.

massNet: integrated processing and classification of spatially resolved mass spectrometry data using deep learning for rapid tumor delineation

Motivation: Mass spectrometry imaging (MSI) provides rich biochemical information in a label-free manner and therefore holds promise to substantially impact current practice in disease diagnosis. However, the complex nature of MSI data poses computational challenges in its analysis. The complexity of the data arises from its large size, high dimensionality, and spectral non-linearity. Preprocessing, including peak picking, has been used to reduce raw data complexity, however peak picking is sensitive to parameter selection that, perhaps prematurely, shapes the downstream analysis for tissue classification and ensuing biological interpretation. Results: We propose a deep learning model, massNet, that provides the desired qualities of scalability, non-linearity, and speed in MSI data analysis. This deep learning model was used, without prior preprocessing and peak picking, to classify MSI data from a mouse brain harboring a patient-derived tumor. The massNet architecture established automatically learning of predictive features, and automated methods were incorporated to identify peaks with potential for tumor delineation. The model's performance was assessed using cross-validation, and the results demonstrate higher accuracy and a 174-fold gain in speed compared to the established classical machine learning method, support vector machine. Availability and Implementation: The code is publicly available on GitHub.