Radiomics prognostic analysis of PET/CT images in a multicenter head and neck cancer cohort: investigating ComBat strategies, sub-volume characterization, and automatic segmentation.

PURPOSE: This study aimed to investigate the impact of several ComBat harmonization strategies, intra-tumoral sub-volume characterization, and automatic segmentations for progression-free survival (PFS) prediction through radiomics modeling for patients with head and neck cancer (HNC) in PET/CT images. METHODS: The HECKTOR MICCAI 2021 challenge set containing PET/CT images and clinical data of 325 oropharynx HNC patients was exploited. A total of 346 IBSI-compliant radiomic features were extracted for each patient's primary tumor volume defined by the reference manual contours. Modeling relied on least absolute shrinkage Cox regression (Lasso-Cox) for feature selection (FS) and Cox proportional-hazards (CoxPH) models were built to predict PFS. Within this methodological framework, 8 different strategies for ComBat harmonization were compared, including before or after FS, in feature groups separately or all features directly, and with center or clustering-determined labels. Features extracted from tumor sub-volume clustering were also investigated for their prognostic additional value. Finally, 3 automatic segmentations (2 threshold-based and a 3D U-Net) were also compared. All results were evaluated with the concordance index (C-index). RESULTS: Radiomics features without harmonization, combined with clinical factors, led to models with C-index values of 0.69 in the testing set. The best version of ComBat harmonization, i.e., after FS, for feature groups separately and relying on clustering-determined labels, achieved a C-index of 0.71. The use of features extracted from tumor sub-volumes further improved the C-index to 0.72. Models that relied on the automatic segmentations yielded close but slightly lower prognostic performance (0.67-0.70) compared to reference contours. CONCLUSION: A standard radiomics pipeline allowed for prediction of PFS in a multicenter HNC cohort. Applying a specific strategy of ComBat harmonization improved the performance. The extraction of intra-tumoral sub-volume features and automatic segmentation could contribute to the improvement and automation of prognosis modeling, respectively.

Hybrid simultaneous whole-body 2-[18F]FDG-PET/MRI imaging in newly diagnosed multiple myeloma: first diagnostic performance and clinical added value results.

OBJECTIVES: Mixing diagnostic and prognostic data provided by whole-body MRI (WB-MRI) and 2-18F-fluorodeoxyglucose (2-[18F]FDG) positron emission tomography (2-[18F]FDG-PET) from a single simultaneous imaging technique for newly diagnosed multiple myeloma (NDMM) initial workup seems attractive. However, to date, the published data are scarce and this possibility has not been fully explored. In this prospective study, we aimed to explore the diagnostic performance and added clinical value of WB-2-[18F]FDG-PET/MRI imaging in NDMM. METHODS: All patients with confirmed NDMM at the Nantes University Hospital were prospectively enrolled in this study and underwent WB-2-[18F]FDG-PET/MRI imaging on a 3-T Biograph mMR before receiving treatment. Before imaging, they were considered either as symptomatic or as smoldering MM (SMM). Diagnostic performance of global WB-2-[18F]FDG-PET/MRI imaging, as well as PET and MRI separately for FL and diffuse BMI detection, was assessed and compared in each group. PET-based (maximal standardized uptake value, SUVmax) and MRI-based (mean apparent diffusion coefficient value, ADCmean) quantitative features were collected for FL/para-medullary disease (PMD)/bone marrow and were compared. RESULTS: A total of 52 patients were included in this study. PET and MRI were equally effective at detecting patients with FL (69% vs. 75%) and with diffuse BMI (62% for both) in the symptomatic MM group. WB-2-[18F]FDG-PET/MRI imaging detected FL in 22% of patients with SMM (with a higher diagnostic performance for MRI), resulting in a significant impact on clinical management in this population. SUVmax and ADCmean quantitative features were weakly or not correlated. CONCLUSIONS: WB-2-[18F]FDG-PET/MRI could represent the next-generation imaging modality for MM. KEY POINTS: • Whole-body 2-[18F]FDG-PET/MRI imaging detected at least one focal bone lesion in 75% of patients with symptomatic multiple myeloma, and PET and MRI were equally effective at identifying patients with a focal bone lesion. • Whole-body 2-[18F]FDG-PET/MRI imaging detected a focal bone lesion in 22% of patients with smoldering multiple myeloma (with a higher diagnostic performance for MRI). • MRI had a significant impact on clinical management of smoldering multiple myeloma.

Random survival forest to predict transplant-eligible newly diagnosed multiple myeloma outcome including FDG-PET radiomics: a combined analysis of two independent prospective European trials.

PURPOSE: Fluorodeoxyglucose-positron emission tomography/computed tomography (FDG-PET/CT) is included in the International Myeloma Working Group (IMWG) imaging guidelines for the work-up at diagnosis and the follow-up of multiple myeloma (MM) notably because it is a reliable tool as a predictor of prognosis. Nevertheless, none of the published studies focusing on the prognostic value of PET-derived features at baseline consider tumor heterogeneity, which could be of high importance in MM. The aim of this study was to evaluate the prognostic value of baseline PET-derived features in transplant-eligible newly diagnosed (TEND) MM patients enrolled in two prospective independent European randomized phase III trials using an innovative statistical random survival forest (RSF) approach. METHODS: Imaging ancillary studies of IFM/DFCI2009 and EMN02/HO95 trials formed part of the present analysis (IMAJEM and EMN02/HO95, respectively). Among all patients initially enrolled in these studies, those with a positive baseline FDG-PET/CT imaging and focal bone lesions (FLs) and/or extramedullary disease (EMD) were included in the present analysis. A total of 17 image features (visual and quantitative, reflecting whole imaging characteristics) and 5 clinical/histopathological parameters were collected. The statistical analysis was conducted using two RSF approaches (train/validation + test and additional nested cross-validation) to predict progression-free survival (PFS). RESULTS: One hundred thirty-nine patients were considered for this study. The final model based on the first RSF (train/validation + test) approach selected 3 features (treatment arm, hemoglobin, and SUVmaxBone Marrow (BM)) among the 22 involved initially, and two risk groups of patients (good and poor prognosis) could be defined with a mean hazard ratio of 4.3 ± 1.5 and a mean log-rank p value of 0.01 ± 0.01. The additional RSF (nested cross-validation) analysis highlighted the robustness of the proposed model across different splits of the dataset. Indeed, the first features selected using the train/validation + test approach remained the first ones over the folds with the nested approach. CONCLUSION: We proposed a new prognosis model for TEND MM patients at diagnosis based on two RSF approaches. TRIAL REGISTRATION: IMAJEM: NCT01309334 and EMN02/HO95: NCT01134484.

Fully automatic segmentation of diffuse large B cell lymphoma lesions on 3D FDG-PET/CT for total metabolic tumour volume prediction using a convolutional neural network.

PURPOSE: Lymphoma lesion detection and segmentation on whole-body FDG-PET/CT are a challenging task because of the diversity of involved nodes, organs or physiological uptakes. We sought to investigate the performances of a three-dimensional (3D) convolutional neural network (CNN) to automatically segment total metabolic tumour volume (TMTV) in large datasets of patients with diffuse large B cell lymphoma (DLBCL). METHODS: The dataset contained pre-therapy FDG-PET/CT from 733 DLBCL patients of 2 prospective LYmphoma Study Association (LYSA) trials. The first cohort (n = 639) was used for training using a 5-fold cross validation scheme. The second cohort (n = 94) was used for external validation of TMTV predictions. Ground truth masks were manually obtained after a 41% SUVmax adaptive thresholding of lymphoma lesions. A 3D U-net architecture with 2 input channels for PET and CT was trained on patches randomly sampled within PET/CTs with a summed cross entropy and Dice similarity coefficient (DSC) loss. Segmentation performance was assessed by the DSC and Jaccard coefficients. Finally, TMTV predictions were validated on the second independent cohort. RESULTS: Mean DSC and Jaccard coefficients (± standard deviation) in the validations set were 0.73 ± 0.20 and 0.68 ± 0.21, respectively. An underestimation of mean TMTV by - 12 mL (2.8%) ± 263 was found in the validation sets of the first cohort (P = 0.27). In the second cohort, an underestimation of mean TMTV by - 116 mL (20.8%) ± 425 was statistically significant (P = 0.01). CONCLUSION: Our CNN is a promising tool for automatic detection and segmentation of lymphoma lesions, despite slight underestimation of TMTV. The fully automatic and open-source features of this CNN will allow to increase both dissemination in routine practice and reproducibility of TMTV assessment in lymphoma patients.

Functional Imaging for Therapeutic Assessment and Minimal Residual Disease Detection in Multiple Myeloma.

Serum markers and bone marrow examination are commonly used for monitoring therapy response in multiple myeloma (MM), but this fails to identify minimal residual disease (MRD), which frequently persists after therapy even in complete response patients, and extra-medullary disease escape. Positron emission tomography with computed tomography using 18F-deoxyglucose (FDG-PET/CT) is the reference imaging technique for therapeutic assessment and MRD detection in MM. To date, all large prospective cohort studies of transplant-eligible newly diagnosed MM patients have shown a strong and independent pejorative prognostic impact of not obtaining complete metabolic response by FDG-PET/CT after therapy, especially before maintenance. The FDG-PET/CT and MRD (evaluated by flow cytometry or next-generation sequencing at 10-5 and 10-6 levels, respectively) results are complementary for MRD detection outside and inside the bone marrow. For patients with at least a complete response, to reach double negativity (FDG-PET/CT and MRD) is a predictive surrogate for patient outcome. Homogenization of FDG-PET/CT interpretation after therapy, especially clarification of complete metabolic response definition, is currently underway. FDG-PET/CT does not allow MRD to be evaluated when it is negative at initial workup of symptomatic MM. New PET tracers such as CXCR4 ligands have shown high diagnostic value and could replace FDG in this setting. New sensitive functional magnetic resonance imaging (MRI) techniques such as diffusion-weighted MRI appear to be complementary to FDG-PET/CT for imaging MRD detection. The goal of this review is to examine the feasibility of functional imaging, especially FDG-PET/CT, for therapeutic assessment and MRD detection in MM.

What is the Best Radionuclide for Immuno-PET of Multiple Myeloma? A Comparison Study Between 89Zr- and 64Cu-Labeled Anti-CD138 in a Preclinical Syngeneic Model.

Although positron emission tomography (PET) imaging with 18-Fluorodeoxyglucose (18F-FDG) is a promising technique in multiple myeloma (MM), the development of other radiopharmaceuticals seems relevant. CD138 is currently used as a standard marker for the identification of myeloma cells and could be used in phenotype tumor imaging. In this study, we used an anti-CD138 murine antibody (9E7.4) radiolabeled with copper-64 (64Cu) or zirconium-89 (89Zr) and compared them in a syngeneic mouse model to select the optimal tracers for MM PET imaging. Then, 9E7.4 was conjugated to TE2A-benzyl isothiocyanate (TE2A) and desferrioxamine (DFO) chelators for 64Cu and 89Zr labeling, respectively. 64Cu-TE2A-9E7.4 and 89Zr-DFO-9E7.4 antibodies were evaluated by PET imaging and biodistribution studies in C57BL/KaLwRij mice bearing either 5T33-MM subcutaneous tumors or bone lesions and were compared to 18F-FDG-PET imaging. In biodistribution and PET studies, 64Cu-TE2A-9E7.4 and 89Zr-DFO-9E7.4 displayed comparable good tumor uptake of subcutaneous tumors. On the bone lesions, PET imaging with 64Cu-TE2A-9E7.4 and 89Zr-DFO-9E7.4 showed higher uptake than with 18F-FDG-PET. Comparison of both 9E7.4 conjugates revealed higher nonspecific bone uptakes of 89Zr-DFO-9E7.4 than 64Cu-TE2A-9E7.4. Because of free 89Zr's tropism for bone when using 89Zr-anti-CD138, 64Cu-anti-CD138 antibody had the most optimal tumor-to-nontarget tissue ratios for translation into humans as a specific new imaging radiopharmaceutical agent in MM.

PET Imaging for Initial Staging and Therapy Assessment in Multiple Myeloma Patients.

Multiple myeloma (MM) is a hematological neoplasm characterized by the clonal proliferation of malignant plasma cells in the bone marrow. MM results in diffuse or focal bone infiltration and extramedullary lesions. Over the past two decades, advances have been made with regard to the diagnosis, staging, treatment, and imaging of MM. Computed tomography (CT) and magnetic resonance imaging (MRI) are currently recommended as the most effective imaging modalities at diagnostic. Yet, recent data from the literature suggest that positron emission tomography combined with computed tomography (PET/CT) using 18F-deoxyglucose (FDG) is a promising technique for initial staging and therapeutic monitoring in this pathology. This paper reviews the recent advances as well as the potential place of a more specific radiopharmaceutical in MM.

Immuno-PET for Clinical Theranostic Approaches.

Recent advances in molecular characterization of tumors have allowed identification of new molecular targets on tumor cells or biomarkers. In medical practice, the identification of these biomarkers slowly but surely becomes a prerequisite before any treatment decision, leading to the concept of personalized medicine. Immuno-positron emission tomography (PET) fits perfectly with this approach. Indeed, monoclonal antibodies (mAbs) labelled with radionuclides represent promising probes for theranostic approaches, offering a non-invasive solution to assess in vivo target expression and distribution. Immuno-PET can potentially provide useful information for patient risk stratification, diagnosis, selection of targeted therapies, evaluation of response to therapy, prediction of adverse effects or for titrating doses for radioimmunotherapy. This paper reviews some aspects and recent developments in labelling methods, biological targets, and clinical data of some novel PET radiopharmaceuticals.

Tumor immunotargeting using innovative radionuclides.

This paper reviews some aspects and recent developments in the use of antibodies to target radionuclides for tumor imaging and therapy. While radiolabeled antibodies have been considered for many years in this context, only a few have reached the level of routine clinical use. However, alternative radionuclides, with more appropriate physical properties, such as lutetium-177 or copper-67, as well as alpha-emitting radionuclides, including astatine-211, bismuth-213, actinium-225, and others are currently reviving hopes in cancer treatments, both in hematological diseases and solid tumors. At the same time, PET imaging, with short-lived radionuclides, such as gallium-68, fluorine-18 or copper-64, or long half-life ones, particularly iodine-124 and zirconium-89 now offers new perspectives in immuno-specific phenotype tumor imaging. New antibody analogues and pretargeting strategies have also considerably improved the performances of tumor immunotargeting and completely renewed the interest in these approaches for imaging and therapy by providing theranostics, companion diagnostics and news tools to make personalized medicine a reality.

Comment on 'A PET reconstruction formulation that enforces non-negativity in projection space for bias reduction in Y-90 imaging'.

We read with great interest the paper by Limet al(2018Phys. Med. Biol.63035042) on bias reduction in Y-90 PET imaging. In particular, they proposed a new formulation of the tomographic reconstruction problem that enforces non-negativity in projection space as opposed to image space. We comment on the algorithm they derived from this formulation and bring some clarifications on the constraint that this algorithm respects.

DCE-MRI to distinguish all monoclonal plasma cell disease stages and correlation with diffusion-weighted MRI/PET-based biomarkers in a hybrid simultaneous whole body-2-[18F]FDG-PET/MRI imaging approach.

BACKGROUND: Dynamic contrast-enhanced-MRI (DCE-MRI) is able to study bone marrow angiogenesis in patients with multiple myeloma (MM) and asymptomatic precursor diseases but its role in the management of MM has not yet been established. The aims of this prospective study was to compare DCE-MRI-based parameters between all monoclonal plasma cell disease stages in order to find out discriminatory parameters and to seek correlations with other diffusion-weighted MRI and positron emission tomography (PET)-based biomarkers in a hybrid simultaneous whole-body-2-[18F]fluorodeoxyglucose (FDG)-PET/MRI (WB-2-[18F]FDG-PET/MRI) imaging approach. METHODS: Patients with newly diagnosed Monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM) or symptomatic MM according to international myeloma working group and underwent WB-2-[18F]FDG-PET/MRI imaging including bone marrow DCE sequences at the Nantes University Hospital were prospectively enrolled in this study before receiving treatment. RESULTS: One hundred and sixty-seven patients (N = 167, mean age: 64 years ± 11 [Standard deviation], 66 males) were considered for the analysis. DCE-MRI-based Peak Enhancement Intensity (PEI), Time to PEI (TPEI) and their maximum intensity time ratio (MITR: PEI/TPEI) values were significantly different between the different monoclonal plasma cell disease stages, PEI values increasing and TPEI values decreasing progressively along the spectrum of plasma cell disorders, from MGUS stage to symptomatic multiple myeloma. PEI values were significantly higher in patients with diffuse bone marrow involvement (either in PET or in MRI images) than in those without diffuse bone marrow involvement, unlike TPEI values. PEI and TPEI values were not significantly different between patients with or without focal bone lesions. CONCLUSION: Different DCE-MRI-based parameters (PEI, TPEI, MITR) could significantly differentiate all monoclonal plasma cell disease stages and complemented conventional MRI and PET-based biomarkers.

Prognostic Value of 18F-FDG PET Radiomics Features at Baseline in PET-Guided Consolidation Strategy in Diffuse Large B-Cell Lymphoma: A Machine-Learning Analysis from the GAINED Study.

The results of the GA in Newly Diagnosed Diffuse Large B-Cell Lymphoma (GAINED) study demonstrated the success of an 18F-FDG PET-driven approach to allow early identification-for intensification therapy-of diffuse large B-cell lymphoma patients with a high risk of relapse. Besides, some works have reported the prognostic value of baseline PET radiomics features (RFs). This work investigated the added value of such biomarkers on survival of patients involved in the GAINED protocol. Methods: Conventional PET features and RFs were computed from 18F-FDG PET at baseline and extracted using different volume definitions (patient level, largest lesion, and hottest lesion). Clinical features and the consolidation treatment information were also considered in the model. Two machine-learning pipelines were trained with 80% of patients and tested on the remaining 20%. The training was repeated 100 times to highlight the test set variability. For the 2-y progression-free survival (PFS) outcome, the pipeline included a data augmentation and an elastic net logistic regression model. Results for different feature groups were compared using the mean area under the curve (AUC). For the survival outcome, the pipeline included a Cox univariate model to select the features. Then, the model included a split between high- and low-risk patients using the median of a regression score based on the coefficients of a penalized Cox multivariate approach. The log-rank test P values over the 100 loops were compared with a Wilcoxon signed-ranked test. Results: In total, 545 patients were included for the 2-y PFS classification and 561 for survival analysis. Clinical features alone, consolidation features alone, conventional PET features, and RFs extracted at patient level achieved an AUC of, respectively, 0.65 ± 0.07, 0.64 ± 0.06, 0.60 ± 0.07, and 0.62 ± 0.07 (0.62 ± 0.07 for the largest lesion and 0.54 ± 0.07 for the hottest). Combining clinical features with the consolidation features led to the best AUC (0.72 ± 0.06). Adding conventional PET features or RFs did not improve the results. For survival, the log-rank P values of the model involving clinical and consolidation features together were significantly smaller than all combined-feature groups (P < 0.007). Conclusion: The results showed that a concatenation of multimodal features coupled with a simple machine-learning model does not seem to improve the results in terms of 2-y PFS classification and PFS prediction for patient treated according to the GAINED protocol.

Enhancing histopathological image classification of invasive ductal carcinoma using hybrid harmonization techniques.

This study aims to develop a robust pipeline for classifying invasive ductal carcinomas and benign tumors in histopathological images, addressing variability within and between centers. We specifically tackle the challenge of detecting atypical data and variability between common clusters within the same database. Our feature engineering-based pipeline comprises a feature extraction step, followed by multiple harmonization techniques to rectify intra- and inter-center batch effects resulting from image acquisition variability and diverse patient clinical characteristics. These harmonization steps facilitate the construction of more robust and efficient models. We assess the proposed pipeline's performance on two public breast cancer databases, BreaKHIS and IDCDB, utilizing recall, precision, and accuracy metrics. Our pipeline outperforms recent models, achieving 90-95% accuracy in classifying benign and malignant tumors. We demonstrate the advantage of harmonization for classifying patches from different databases. Our top model scored 94.7% for IDCDB and 95.2% for BreaKHis, surpassing existing feature engineering-based models (92.1% for IDCDB and 87.7% for BreaKHIS) and attaining comparable performance to deep learning models. The proposed feature-engineering-based pipeline effectively classifies malignant and benign tumors while addressing variability within and between centers through the incorporation of various harmonization techniques. Our findings reveal that harmonizing variabilities between patches from different batches directly impacts the learning and testing performance of classification models. This pipeline has the potential to enhance breast cancer diagnosis and treatment and may be applicable to other diseases.

Performance of baseline FDG-PET/CT radiomics for prediction of bone marrow minimal residual disease status in the LyMa-101 trial.

The prognostic value of 18F-Fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) at baseline or the predictive value of minimal residual disease (MRD) detection appear as potential tools to improve mantle cell lymphoma (MCL) patients' management. The LyMa-101, a phase 2 trial of the LYSA group (ClinicalTrials.gov:NCT02896582) reported induction therapy with obinutuzumab, a CD20 monoclonal antibody. Herein, we investigated the added prognostic value of radiomic features (RF) derived from FDG-PET/CT at diagnosis for MRD value prediction. FDG-PET/CT of 59 MCL patients included in the LyMa-101 trial have been independently, blindly and centrally reviewed. RF were extracted from the disease area with the highest uptake and from the total metabolic tumor volume (TMTV). Two models of machine learning were used to compare several combinations for prediction of MRD before autologous stem cell transplant consolidation (ASCT). Each algorithm was generated with or without constrained feature selections for clinical and laboratory parameters. Both algorithms showed better discrimination performances for negative vs positive MRD in the lesion with the highest uptake than in the TMTV. The constrained use of clinical and biological features showed a clear loss in sensitivity for the prediction of MRD status before ASCT, regardless of the machine learning model. These data plead for the importance of FDG-PET/CT RF compared to clinical and laboratory parameters and also reinforced the previously made hypothesis that the prognosis of the disease in MCL patients is linked to the most aggressive contingent, within the lesion with the highest uptake.