Janus ACSP Nanoparticle for Synergistic Chemodynamic Therapy and Radiosensitization.

Protective autophagy and DNA damage repair lead to tumor radio-resistance. Some hypoxic tumors exhibit a low radiation energy absorption coefficient in radiation therapy. High doses of X-rays may lead to side effects in the surrounding normal tissues. In order to overcome the radio-resistance and improve the efficacy of radiotherapy based on the characteristics of the tumor microenvironment, the development of radiosensitizers has attracted much attention. In this study, a Janus ACSP nanoparticle (NP) was developed for chemodynamic therapy and radiosensitization. The reactive oxygen species generated by the Fenton-like reaction regulated the distribution of cell cycles from a radioresistant phase to a radio-sensitive phase. The high-Z element, Au, enhanced the production of hydroxyl radicals (•OH) under X-ray radiation, promoting DNA damage and cell apoptosis. The NP delayed DNA damage repair by interfering with certain proteins involved in the DNA repair signaling pathway. In vivo experiments demonstrated that the combination of the copper-ion-based Fenton-like reaction and low-dose X-ray radiation enhanced the effectiveness of radiotherapy, providing a novel approach for synergistic chemodynamic and radiosensitization therapy. This study provides valuable insights and strategies for the development and application of NPs in cancer treatment.

Hypoxic Exosomal circPLEKHM1-Mediated Crosstalk between Tumor Cells and Macrophages Drives Lung Cancer Metastasis.

Intercellular communication often relies on exosomes as messengers and is critical for cancer metastasis in hypoxic tumor microenvironment. Some circular RNAs (circRNAs) are enriched in cancer cell-derived exosomes, but little is known about their ability to regulate intercellular communication and cancer metastasis. Here, by systematically analyzing exosomes secreted by non-small cell lung cancer (NSCLC) cells, a hypoxia-induced exosomal circPLEKHM1 is identified that drives NSCLC metastasis through polarizing macrophages toward to M2 type. Mechanistically, exosomal circPLEKHM1 promoted PABPC1-eIF4G interaction to facilitate the translation of the oncostatin M receptor (OSMR), thereby promoting macrophage polarization for cancer metastasis. Importantly, circPLEKHM1-targeted therapy significantly reduces NSCLC metastasis in vivo. circPLEKHM1 serves as a prognostic biomarker for metastasis and poor survival in NSCLC patients. This study unveils a new circRNA-mediated mechanism underlying how cancer cells crosstalk with macrophages within the hypoxic tumor microenvironment to promote metastasis, highlighting the importance of exosomal circPLEKHM1 as a prognostic biomarker and therapeutic target for lung cancer metastasis.

Synthesis of easily-modified and useful dibenzo-[b,d]azepines by palladium(II)-catalyzed cyclization/addition with a green solvent.

A novel strategy in which palladium(II)-catalyzed tandem cyclization is used to obtain N-heterocyclic architectures containing a seven-membered ring has been developed and used to synthesize a series of derivatives. The reaction uses an eco-friendly mixed solvent (water : EtOH = 2 : 1) instead of DMSO and maintains a high yield (91%). Its potential application value and reaction mechanism have also been explored.

Can 18F-FES PET Improve the Evaluation of 18F-FDG PET in Patients With Metastatic Invasive Lobular Carcinoma?

PURPOSE: Invasive lobular carcinoma (ILC) exhibits a low affinity for 18F-FDG. The estrogen receptor (ER) is commonly expressed in ILCs, suggesting a potential benefit of targeting with the ER probe 18F-FES in this patient population. The objective of this study was to evaluate the diagnostic performance of 18F-FES imaging in patients with metastatic ILC and compare it with that of 18F-FDG. METHODS: We conducted a retrospective analysis of 20 ILC patients who underwent concurrent 18F-FES and 18F-FDG PET/CT examinations in our center. 18F-FES and 18F-FDG imaging were analyzed to determine the total count of tracer-avid lesions in nonbone sites and their corresponding organ systems, assess the extent of anatomical regions involved in bone metastases, and measure the SUVmax values for both tracers. RESULTS: Among 20 ILC patients, 65 nonbone lesions were found to be distributed in 13 patients, and 16 patients were diagnosed with bone metastasis, which was distributed in 54 skeletal anatomical regions. The detection rate of 18F-FDG in nonbone lesions was higher than that of 18F-FES (57 vs 37, P < 0.001). 18F-FES demonstrated a superior ability to detect nonbone lesions in 4 patients, whereas 18F-FDG was superior in 5 patients (P > 0.05). Among 9/16 patients with bone metastasis, 18F-FES demonstrated a significant advantage in the detection of bone lesions compared with 18F-FDG (P = 0.05). Furthermore, patients with only 18F-FES-positive lesions (12/12) were administered endocrine regimens, whereas patients lacking 18F-FES uptake (2/3) predominantly received chemotherapy. CONCLUSIONS: 18F-FES is more effective than 18F-FDG in detecting bone metastasis in ILC, but it does not demonstrate a significant advantage in nonbone lesions. Additionally, the results of examination with 18F-FES have the potential to guide patient treatment plans.

Heterogeneity of Multiple Pancreatic Neuroendocrine Tumors Identified by 68Ga-DOTANOC and 68Ga-exendin-4 PET/CT in a Patient with Endogenous Hyperinsulinemic Hypoglycemia and Multiple Endocrine Neoplasia 1.

Insulinomas are the most frequent functional neuroendocrine tumors of pancreas. In about 10% cases, insulinomas are associated with hereditary syndrome including multiple endocrine neoplasia 1 (MEN1). Herein, we presented a case of 44-year-old female with recurrent hypoglycemia. In December 1998, this patient has undergone resection of two pancreatic lesions due to hypoglycemia and diagnosed as insulinoma. After operation, the symptom of hypoglycemia disappeared. However, from 2021, hypoglycemic symptoms reappeared frequently and even coma. In June 2023, enhanced CT showed multiple pancreatic lesions abundant with blood supply. Fasting serum blood glucose and insulin were 1.73mmol/L and 15.2U/L (2.6-11.8U/L). Germline genes suggested MEN1 pathogenic mutations. 68Ga-DOTANOC PET/CT indicated there were multiple lesions located in the pancreas and duodenum with high expression of somatostatin receptor (SSTR). 68Ga-exendin-4 PET/CT were added to localize the insulinoma. Most lesions with high expression of SSTR in body and tail of pancreas manifested part of them with high uptake of 68Ga-exendin-4, and an additional lesion with high expression of glucagon-like peptide 1 receptor was only detected by 68Ga-exendin-4 PET/CT. It showed highly heterogeneity. From the distal pancreatectomy, a total 5 tumors were found in the body and tail of pancreas, which were diagnosed as neuroendocrine tumors (NETs). After the operation, all the symptoms related to hypoglycemia disappeared. Immunohistochemical results of SSTR2 and insulin were consistent with the imaging finding of dual tracer PET/CT. From this case, combination of 68Ga-DOTANOC and 68Ga-exendin-4 PET/CT was recommended in the patients of MEN1 and insulinoma to estimate the heterogeneity of multiple neuroendocrine tumors that contributing to detect all the NET lesions and locate the tumors with secretion of insulin.

68Ga-FAPI PET imaging monitors response to combined TGF-ßR inhibition and immunotherapy in metastatic colorectal cancer.

BACKGROUNDImproving and predicting tumor response to immunotherapy remains challenging. Combination therapy with a transforming growth factor-ß receptor (TGF-ßR) inhibitor that targets cancer-associated fibroblasts (CAFs) is promising for the enhancement of efficacy of immunotherapies. However, the effect of this approach in clinical trials is limited, requiring in vivo methods to better assess tumor responses to combination therapy.METHODSWe measured CAFs in vivo using the 68Ga-labeled fibroblast activation protein inhibitor-04 (68Ga-FAPI-04) for PET/CT imaging to guide the combination of TGF-ß inhibition and immunotherapy. One hundred thirty-one patients with metastatic colorectal cancer (CRC) underwent 68Ga-FAPI and 18F-fluorodeoxyglucose (18F-FDG) PET/CT imaging. The relationship between uptake of 68Ga-FAPI and tumor immunity was analyzed in patients. Mouse cohorts of metastatic CRC were treated with the TGF-ßR inhibitor combined with KN046, which blocks programmed death ligand 1 (PD-L1) and CTLA-4, followed by 68Ga-FAPI and 18F-FDG micro-PET/CT imaging to assess tumor responses.RESULTSPatients with metastatic CRC demonstrated high uptake rates of 68Ga-FAPI, along with suppressive tumor immunity and poor prognosis. The TGF-ßR inhibitor enhanced tumor-infiltrating T cells and significantly sensitized metastatic CRC to KN046. 68Ga-FAPI PET/CT imaging accurately monitored the dynamic changes of CAFs and tumor response to combined the TGF-ßR inhibitor with immunotherapy.CONCLUSION68Ga-FAPI PET/CT imaging is powerful in assessing tumor immunity and the response to immunotherapy in metastatic CRC. This study supports future clinical application of 68Ga-FAPI PET/CT to guide precise TGF-ß inhibition plus immunotherapy in CRC patients, recommending 68Ga-FAPI and 18F-FDG dual PET/CT for CRC management.TRIAL REGISTRATIONCFFSTS Trial, ChiCTR2100053984, Chinese Clinical Trial Registry.FUNDINGNational Natural Science Foundation of China (82072695, 32270767, 82272035, 81972260).

Nomogram to predict the presence of PSMA-negative but FDG-positive lesion in castration-resistant prostate cancer: a multicenter cohort study.

Background: PSMA-negative but FDG-positive (PSMA-/FDG+) lesion in dual-tracer (68Ga-PSMA and 18F-FDG) positron emission tomography/computed tomography (PET/CT) is associated with an unfavorable response to Lutetium-177 (177Lu)-PSMA-617. This study sought to develop both radiomics and clinical models for the precise prediction of the presence of PSMA-/FDG+ lesions in patients with castration-resistant prostate cancer (CPRC). Methods: A cohort of 298 patients who underwent dual-tracer PET/CT with a less than 5-day interval was included. The evaluation of the prognostic performance of the radiomics model drew upon the survival data derived from 40 patients with CRPC treated with 177Lu-PSMA-617 in an external cohort. Two endpoints were evaluated: (a) prostate-specific antigen (PSA) response rate, defined as a reduction exceeding 50% from baseline and (b) overall survival (OS), measured from the initiation of 177Lu-PSMA-617 to death from any cause. Results: PSMA-/FDG+ lesions were identified in 56 (18.8%) CRPC patients. Both radiomics (area under the curve [AUC], 0.83) and clinical models (AUC, 0.78) demonstrated robust performance in PSMA-/FDG+ lesion prediction. Decision curve analysis revealed that the radiomics model yielded a net benefit over the 'screen all' strategy at a threshold probability of ⩾4%. At a 5% probability threshold, the radiomics model facilitated a 21% reduction in 18F-FDG PET/CT scans while only missing 2% of PSMA-/FDG+ cases. Patients with a low estimated score exhibited significantly prolonged OS (hazard ratio = 0.49, p = 0.029) and a higher PSA response rate (75% versus 35%, p = 0.011) compared to those with a high estimated score. Conclusion: This study successfully developed two models with accurate estimations of the risk associated with PSMA-/FDG+ lesions in CRPC patients. These models held potential utility in aiding the selection of candidates for 177Lu-PSMA-617 treatment and guiding 68Ga-PSMA PET/CT-directed radiotherapy.

Predictive nomogram for PSMA-/FDG+ lesion This study developed two models with accurate estimations of the risk associated with specific lesions in prostate cancer.

18F-FAPI PET/CT performs better in evaluating mediastinal and hilar lymph nodes in patients with lung cancer: comparison with 18F-FDG PET/CT.

BACKGROUND: The aim of this study was to evaluate the efficacy of fluorine 18 (18F) labeled fibroblast activation protein inhibitor (FAPI) in identifying mediastinal and hilar lymph node metastases and to develop a model to quantitatively and repeatedly identify lymph node status. METHODS: Twenty-seven patients with 137 lymph nodes were identified by two PET/CT images. The sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value (NPV) of lymph node status were analyzed, and the optimal cut-off value was identified by ROC analysis. RESULTS: The SUVmax of metastatic lymph nodes on 18F-FAPI was higher than that on 18F-FDG PET/CT (10.87 ± 7.29 vs 6.08 ± 5.37, p < 0.001). 18F-FAPI presented much greater lymph node detection sensitivity, specificity, accuracy, PPV and NPV than 18F-FDG PET/CT (84% vs. 71%; 92% vs. 67%; 90% vs. 69%, 84% vs. 52%, and 92% vs. 83%, respectively). Additionally, the diagnostic effectiveness of 18F-FAPI in small lymph nodes was greater than that of 18F-FDG PET/CT (specificity: 96% vs. 72%; accuracy: 93% vs. 73%; PPV: 77% vs. 33%, respectively). Notably, the optimal cut-off value for specificity and PPV of 18F-FAPI SUVmax was 5.3; the optimal cut-off value for sensitivity and NPV was 2.5. CONCLUSION: 18F-FAPI showed promising diagnostic efficacy in metastatic mediastinal and hilar lymph nodes from lung cancer patients, with a higher SUVmax, especially in small metastatic nodes, compared with 18F-FDG. In addition, this exploratory work recommended optimal SUVmax cutoff values to distinguish between nonmetastatic and metastatic lymph nodes, thereby advancing the development of image-guided radiation. Trial registration ClinicalTrials.gov identifier: ChiCTR2000036091.

Investigation of PET image quality with acquisition time/bed and enhancement of lesion quantification accuracy through deep progressive learning.

OBJECTIVE: To improve the PET image quality by a deep progressive learning (DPL) reconstruction algorithm and evaluate the DPL performance in lesion quantification. METHODS: We reconstructed PET images from 48 oncological patients using ordered subset expectation maximization (OSEM) and deep progressive learning (DPL) methods. The patients were enrolled into three overlapped studies: 11 patients for image quality assessment (study 1), 34 patients for sub-centimeter lesion quantification (study 2), and 28 patients for imaging of overweight or obese individuals (study 3). In study 1, we evaluated the image quality visually based on four criteria: overall score, image sharpness, image noise, and diagnostic confidence. We also measured the image quality quantitatively using the signal-to-background ratio (SBR), signal-to-noise ratio (SNR), contrast-to-background ratio (CBR), and contrast-to-noise ratio (CNR). To evaluate the performance of the DPL algorithm in quantifying lesions, we compared the maximum standardized uptake values (SUVmax), SBR, CBR, SNR and CNR of 63 sub-centimeter lesions in study 2 and 44 lesions in study 3. RESULTS: DPL produced better PET image quality than OSEM did based on the visual evaluation methods when the acquisition time was 0.5, 1.0 and 1.5 min/bed. However, no discernible differences were found between the two methods when the acquisition time was 2.0, 2.5 and 3.0 min/bed. Quantitative results showed that DPL had significantly higher values of SBR, CBR, SNR, and CNR than OSEM did for each acquisition time. For sub-centimeter lesion quantification, the SUVmax, SBR, CBR, SNR, and CNR of DPL were significantly enhanced, compared with OSEM. Similarly, for lesion quantification in overweight and obese patients, DPL significantly increased these parameters compared with OSEM. CONCLUSION: The DPL algorithm dramatically enhanced the quality of PET images and enabled more accurate quantification of sub-centimeters lesions in patients and lesions in overweight or obese patients. This is particularly beneficial for overweight or obese patients who usually have lower image quality due to the increased attenuation.

Imaging of Tumor Stroma Using 68Ga-FAPI PET/CT to Improve Diagnostic Accuracy of Primary Tumors in Head and Neck Cancer of Unknown Primary: A Comparative Imaging Trial.

The low detection rate of primary tumors by current diagnostic techniques remains a major concern for patients with head and neck cancer of unknown primary (HNCUP). Therefore, in this study, we aimed to investigate the potential role of 68Ga-labeled fibroblast activation protein inhibitor (68Ga-FAPI) PET/CT compared with 18F-FDG PET/CT for the detection of primary tumors of HNCUP. Methods: In this prospective comparative imaging trial conducted at Fudan University Shanghai Cancer Center, 91 patients with negative or equivocal findings of a primary tumor by comprehensive clinical examination and conventional imaging were enrolled from June 2020 to September 2022. The presence of a primary tumor was recorded by 3 experienced nuclear medicine physicians. Primary lesions were validated by histopathologic analysis and a composite reference standard. Results: Of the 91 patients (18 women, 73 men; median age, 60 y; age range, 24-76 y), primary tumors were detected in 46 (51%) patients after a thorough diagnostic work-up. 68Ga-FAPI PET/CT detected more primary lesions than 18F-FDG PET/CT (46 vs. 17, P < 0.001) and showed better sensitivity, positive predictive value, and accuracy in locating primary tumors (51% vs. 25%, 98% vs. 43%, and 51% vs. 19%, respectively). Furthermore, 68Ga-FAPI PET/CT led to treatment changes in 22 of 91 (24%) patients compared with 18F-FDG PET/CT. The Kaplan-Meier curve illustrated that patients with unidentified primary tumors had a significantly worse prognosis than patients with identified primary tumors (hazard ratio, 5.77; 95% CI, 1.86-17.94; P = 0.0097). Conclusion: 68Ga-FAPI PET/CT outperforms 18F-FDG PET/CT in detecting primary lesions and could serve as a sensitive, reliable, and reproducible imaging modality for HNCUP patients.

Single-cell profiling of the microenvironment in human bone metastatic renal cell carcinoma.

Bone metastasis is of common occurrence in renal cell carcinoma with poor prognosis, but no optimal treatment approach has been established for bone metastatic renal cell carcinoma. To explore the potential therapeutic targets for bone metastatic renal cell carcinoma, we profile single cell transcriptomes of 6 primary renal cell carcinoma and 9 bone metastatic renal cell carcinoma. We also include scRNA-seq data of early-stage renal cell carcinoma, late-stage renal cell carcinoma, normal kidneys and healthy bone marrow samples in the study to better understand the bone metastasis niche. The molecular properties and dynamic changes of major cell lineages in bone metastatic environment of renal cell carcinoma are characterized. Bone metastatic renal cell carcinoma is associated with multifaceted immune deficiency together with cancer-associated fibroblasts, specifically appearance of macrophages exhibiting malignant and pro-angiogenic features. We also reveal the dominance of immune inhibitory T cells in the bone metastatic renal cell carcinoma which can be partially restored by the treatment. Trajectory analysis showes that myeloid-derived suppressor cells are progenitors of macrophages in the bone metastatic renal cell carcinoma while monocytes are their progenitors in primary tumors and healthy bone marrows. Additionally, the infiltration of immune inhibitory CD47+ T cells is observed in bone metastatic tumors, which may be a result of reduced phagocytosis by SIRPA-expressing macrophages in the bone microenvironment. Together, our results provide a systematic view of various cell types in bone metastatic renal cell carcinoma and suggest avenues for therapeutic solutions.

Value of multi-center 18 F-FDG PET/CT radiomics in predicting EGFR mutation status in lung adenocarcinoma.

BACKGROUND: Accurate, noninvasive, and reliable assessment of epidermal growth factor receptor (EGFR) mutation status and EGFR molecular subtypes is essential for treatment plan selection and individualized therapy in lung adenocarcinoma (LUAD). Radiomics models based on 18 F-FDG PET/CT have great potential in identifying EGFR mutation status and EGFR subtypes in patients with LUAD. The validation of multi-center data, model visualization, and interpretation are significantly important for the management, application and trust of machine learning predictive models. However, few EGFR-related research involved model visualization and interpretation, and multi-center trial. PURPOSE: To develop explainable optimal predictive models based on handcrafted radiomics features (HRFs) extracted from multi-center 18 F-FDG PET/CT to predict EGFR mutation status and molecular subtypes in LUAD. METHODS: Baseline 18 F-FDG PET/CT images of 383 LUAD patients from three hospitals and one public data set were collected. Further, 1808 HRFs were extracted from the primary tumor regions using Pyradiomics. Predictive models were built based on cross-combination of seven feature selection methods and seven machine learning algorithms. Yellowbrick and explainable artificial intelligence technology were used for model visualization and interpretation. Receiver operating characteristic curve, classification report and confusion matrix were used for model performance evaluation. Clinical applicability of the optimal models was assessed by decision curve analysis. RESULTS: STACK feature selection method combined with light gradient boosting machine (LGBM) reached optimal performance in identifying EGFR mutation status ([area under the curve] AUC = 0.81 in the internal test cohort; AUC = 0.62 in the external test cohort). Random forest feature selection method combined with LGBM reached optimal performance in predicting EGFR mutation molecular subtypes (AUC = 0.89 in the internal test cohort; AUC = 0.61 in the external test cohort). CONCLUSIONS: Explainable machine learning models combined with radiomics features extracted from multi-center/scanner 18 F-FDG PET/CT have certain potential to identify EGFR mutation status and subtypes in LUAD, which might be helpful to the treatment of LUAD.

Preclinical Evaluation and a Pilot Clinical Positron Emission Tomography Imaging Study of [68Ga]Ga-FAPI-FUSCC-II.

Fibroblast activation protein (FAP), a type II integral membrane serine protease, is a promising target for tumor diagnosis and therapy. OncoFAP has been recently discovered for PET imaging procedures for various solid malignancies. In this study, we presented the development of manual radiolabeling procedures for the preparation of OncoFAP-based radiopharmaceuticals for cancer imaging. A novel series of [68Ga/177Lu]Ga/Lu-FAPI-FUSCC-I/II were produced with high radiochemical yields. [68Ga]Ga-FAPI-FUSCC-I/II and [177Lu]Lu-FAPI-FUSCC-I/II were stable in phosphate-buffered saline, fetal bovine serum, and human serum for at least 3 h. In vitro cellular uptake and blocking experiments implied that they had specificity to FAP. Additionally, the low nanomolar IC50 values of FAPI-FUSCC-II indicated that it had a high target affinity to FAP. The in vivo biodistribution and blocking study in mice bearing HT-1080-FAP tumors showed that both exhibited specific tumor uptake. [68Ga]Ga-FAPI-FUSCC-II showed a higher tumor uptake and a higher tumor/nontarget ratio than [68Ga]Ga-FAPI-FUSCC-I and [68Ga]Ga-FAPI-04. The results of ex vivo biodistribution were in accordance with the biodistribution results. Clinical [68Ga]Ga-FAPI-FUSCC-II-PET/CT imaging further demonstrated its favorable biodistribution and kinetics with elevated and reliable uptake by primary tumors (maximum standardized uptake value (SUVmax), 12.17 ± 6.67) and distant metastases (SUVmax, 9.24 ± 4.28). In summary, [68Ga]Ga-FAPI-FUSCC-II displayed increased tumor uptake and retention compared to [68Ga]Ga-FAPI-04, giving it potential as a promising tracer for the diagnostic imaging of malignant tumors with positive FAP expression.

Deep learning-based whole-body characterization of prostate cancer lesions on [68Ga]Ga-PSMA-11 PET/CT in patients with post-prostatectomy recurrence.

PURPOSE: The automatic segmentation and detection of prostate cancer (PC) lesions throughout the body are extremely challenging due to the lesions' complexity and variability in appearance, shape, and location. In this study, we investigated the performance of a three-dimensional (3D) convolutional neural network (CNN) to automatically characterize metastatic lesions throughout the body in a dataset of PC patients with recurrence after radical prostatectomy. METHODS: We retrospectively collected [68 Ga]Ga-PSMA-11 PET/CT images from 116 patients with metastatic PC at two centers: center 1 provided the data for fivefold cross validation (n = 78) and internal testing (n = 19), and center 2 provided the data for external testing (n = 19). PET and CT data were jointly input into a 3D U-Net to achieve whole-body segmentation and detection of PC lesions. The performance in both the segmentation and the detection of lesions throughout the body was evaluated using established metrics, including the Dice similarity coefficient (DSC) for segmentation and the recall, precision, and F1-score for detection. The correlation and consistency between tumor burdens (PSMA-TV and TL-PSMA) calculated from automatic segmentation and artificial ground truth were assessed by linear regression and Bland‒Altman plots. RESULTS: On the internal test set, the DSC, precision, recall, and F1-score values were 0.631, 0.961, 0.721, and 0.824, respectively. On the external test set, the corresponding values were 0.596, 0.888, 0.792, and 0.837, respectively. Our approach outperformed previous studies in segmenting and detecting metastatic lesions throughout the body. Tumor burden indicators derived from deep learning and ground truth showed strong correlation (R2 ≥ 0.991, all P < 0.05) and consistency. CONCLUSION: Our 3D CNN accurately characterizes whole-body tumors in relapsed PC patients; its results are highly consistent with those of manual contouring. This automatic method is expected to improve work efficiency and to aid in the assessment of tumor burden.

Differentiation of cancer of unknown primary and lymphoma in head and neck metastatic poorly differentiated cancer using 18 F-FDG PET/CT tumor metabolic heterogeneity index.

OBJECTIVE: To explore the value of 18 F-FDG PET/CT tumor metabolic heterogeneity index (HI) and establish and validate a nomogram model for distinguishing head and neck cancer of unknown primary (HNCUP) from lymphoma with head and neck metastatic poorly differentiated cancer. METHODS: This retrospective analysis was conducted on 1242 patients with cervical metastatic poorly differentiated cancer. 108 patients, who were clinically and pathologically confirmed as HNCUP or lymphoma, were finally enrolled. Two independent sample t-tests and χ 2 test were used to compare the clinical and imaging features. Binary logistic regression was used to screen for independent predictive factors. RESULTS: Among the 108 patients), 65 patients were diagnosed with HNCUP and 43 were lymphoma. Gender ( P  = 0.001), SUV max ( P  < 0.001), SUV mean ( P  < 0.001), TLG ( P  = 0.012), and HI ( P  < 0.001) had statistical significance in distinguishing HNCUP and lymphoma. Female ( OR  = 4.546, P  = 0.003) and patients with HI ≥ 2.37 ( OR  = 3.461, P  = 0.047) were more likely to be diagnosed as lymphoma. CONCLUSION: For patients with cervical metastatic poorly differentiated cancer, gender and HI were independent predictors of pathological type. For such patients, clinical attention should be paid to avoid misdiagnosing lymphoma as HNCUP, which may delay treatment.

Clinico-biological-radiomics (CBR) based machine learning for improving the diagnostic accuracy of FDG-PET false-positive lymph nodes in lung cancer.

BACKGROUND: The main problem of positron emission tomography/computed tomography (PET/CT) for lymph node (LN) staging is the high false positive rate (FPR). Thus, we aimed to explore a clinico-biological-radiomics (CBR) model via machine learning (ML) to reduce FPR and improve the accuracy for predicting the hypermetabolic mediastinal-hilar LNs status in lung cancer than conventional PET/CT. METHODS: A total of 260 lung cancer patients with hypermetabolic mediastinal-hilar LNs (SUVmax ≥ 2.5) were retrospectively reviewed. Patients were treated with surgery with systematic LN resection and pathologically divided into the LN negative (LN-) and positive (LN +) groups, and randomly assigned into the training (n = 182) and test (n = 78) sets. Preoperative CBR dataset containing 1738 multi-scale features was constructed for all patients. Prediction models for hypermetabolic LNs status were developed using the features selected by the supervised ML algorithms, and evaluated using the classical diagnostic indicators. Then, a nomogram was developed based on the model with the highest area under the curve (AUC) and the lowest FPR, and validated by the calibration plots. RESULTS: In total, 109 LN- and 151 LN + patients were enrolled in this study. 6 independent prediction models were developed to differentiate LN- from LN + patients using the selected features from clinico-biological-image dataset, radiomics dataset, and their combined CBR dataset, respectively. The DeLong test showed that the CBR Model containing all-scale features held the highest predictive efficiency and the lowest FPR among all of established models (p < 0.05) in both the training and test sets (AUCs of 0.90 and 0.89, FPRs of 12.82% and 6.45%, respectively) (p < 0.05). The quantitative nomogram based on CBR Model was validated to have a good consistency with actual observations. CONCLUSION: This study presents an integrated CBR nomogram that can further reduce the FPR and improve the accuracy of hypermetabolic mediastinal-hilar LNs evaluation than conventional PET/CT in lung cancer, thereby greatly reducing the risk of overestimation and assisting for precision treatment.

Hyper-Connected Transformer Network for Multi-Modality PET-CT Segmentation.

[18F]-Fluorodeoxyglucose (FDG) positron emission tomography - computed tomography (PET-CT) has become the imaging modality of choice for diagnosing many cancers. Co-learning complementary PET-CT imaging features is a fundamental requirement for automatic tumor segmentation and for developing computer aided cancer diagnosis systems. In this study, we propose a hyper-connected transformer (HCT) network that integrates a transformer network (TN) with a hyper connected fusion for multi-modality PET-CT images. The TN was leveraged for its ability to provide global dependencies in image feature learning, which was achieved by using image patch embeddings with a self-attention mechanism to capture image-wide contextual information. We extended the single-modality definition of TN with multiple TN based branches to separately extract image features. We also introduced a hyper connected fusion to fuse the contextual and complementary image features across multiple transformers in an iterative manner. Our results with two clinical datasets show that HCT achieved better performance in segmentation accuracy when compared to the existing methods.Clinical Relevance-We anticipate that our approach can be an effective and supportive tool to aid physicians in tumor quantification and in identifying image biomarkers for cancer treatment.

Condition control training-based ConVMLP-ResU-Net for semantic segmentation of esophageal cancer in 18F-FDG PET/CT images.

The precise delineation of esophageal gross tumor volume (GTV) on medical images can promote the radiotherapy effect of esophagus cancer. This work is intended to explore effective learning-based methods to tackle the challenging auto-segmentation problem of esophageal GTV. By employing the progressive hierarchical reasoning mechanism (PHRM), we devised a simple yet effective two-stage deep framework, ConVMLP-ResU-Net. Thereinto, the front-end ConVMLP integrates convolution (ConV) and multi-layer perceptrons (MLP) to capture localized and long-range spatial information, thus making ConVMLP excel in the location and coarse shape prediction of esophageal GTV. According to the PHRM, the front-end ConVMLP should have a strong generalization ability to ensure that the back-end ResU-Net has correct and valid reasoning. Therefore, a condition control training algorithm was proposed to control the training process of ConVMLP for a robust front end. Afterward, the back-end ResU-Net benefits from the yielded mask by ConVMLP to conduct a finer expansive segmentation to output the final result. Extensive experiments were carried out on a clinical cohort, which included 1138 pairs of 18F-FDG positron emission tomography/computed tomography (PET/CT) images. We report the Dice similarity coefficient, Hausdorff distance, and Mean surface distance as 0.82 ± 0.13, 4.31 ± 7.91 mm, and 1.42 ± 3.69 mm, respectively. The predicted contours visually have good agreements with the ground truths. The devised ConVMLP is apt at locating the esophageal GTV with correct initial shape prediction and hence facilitates the finer segmentation of the back-end ResU-Net. Both the qualitative and quantitative results validate the effectiveness of the proposed method.

End-to-end deep learning radiomics: development and validation of a novel attention-based aggregate convolutional neural network to distinguish breast diffuse large B-cell lymphoma from breast invasive ductal carcinoma.

Background: Apart from invasive pathological examination, there is no effective method to differentiate breast diffuse large B-cell lymphoma (DLBCL) from breast invasive ductal carcinoma (IDC). In this study, we aimed to develop and validate an effective deep learning radiomics model to discriminate between DLBCL and IDC. Methods: A total of 324 breast nodules from 236 patients with baseline 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) were retrospectively analyzed. After grouping breast DLBCL and breast IDC patients, external and internal datasets were divided according to the data collected by different centers. Preprocessing was then used to process the original PET/CT images and an attention-based aggregate convolutional neural network (AACNN) model was designed. The AACNN model was trained using patches of CT or PET tumor images and optimized with an improved loss function. The final ensemble predictive model was built using distance weight voting. Finally, the model performance was evaluated and statistically verified. Results: A total of 249 breast nodules from Fudan University Shanghai Cancer Center (FUSCC) and 75 breast nodules from Shanghai Proton and Heavy Ion Center (SPHIC) were selected as internal and external datasets, respectively. On the internal testing, our method yielded an area under the curve (AUC), accuracy (ACC), sensitivity (SEN), specificity (SPE), positive predictive value (PPV), negative predictive value (NPV), and harmonic mean of precision and sensitivity (F1) of 0.886, 83.0%, 80.9%, 85.0%, 84.8%, 81.2%, and 0.828, respectively. Meanwhile on the external testing, the results were 0.788, 71.6%, 61.4%, 84.7%, 84.0%, 62.6%, and 0.709, respectively. Conclusions: Our study outlines a deep learning radiomics method which can automatically, noninvasively, and accurately differentiate breast DLBCL from breast IDC, which will be more in line with the needs and strategies of precision medicine, individualized diagnosis, and treatment.