An Automated Deep Learning-Based Framework for Uptake Segmentation and Classification on PSMA PET/CT Imaging of Patients with Prostate Cancer

Uptake segmentation and classification on PSMA PET/CT are important for automating whole-body tumor burden determinations. We developed and evaluated an automated deep learning (DL)-based framework that segments and classifies uptake on PSMA PET/CT. We identified 193 [18F] DCFPyL PET/CT scans of patients with biochemically recurrent prostate cancer from two institutions, including 137 [18F] DCFPyL PET/CT scans for training and internally testing, and 56 scans from another institution for external testing. Two radiologists segmented and labelled foci as suspicious or non-suspicious for malignancy. A DL-based segmentation was developed with two independent CNNs. An anatomical prior guidance was applied to make the DL framework focus on PSMA-avid lesions. Segmentation performance was evaluated by Dice, IoU, precision, and recall. Classification model was constructed with multi-modal decision fusion framework evaluated by accuracy, AUC, F1 score, precision, and recall. Automatic segmentation of suspicious lesions was improved under prior guidance, with mean Dice, IoU, precision, and recall of 0.700, 0.566, 0.809, and 0.660 on the internal test set and 0.680, 0.548, 0.749, and 0.740 on the external test set. Our multi-modal decision fusion framework outperformed single-modal and multi-modal CNNs with accuracy, AUC, F1 score, precision, and recall of 0.764, 0.863, 0.844, 0.841, and 0.847 in distinguishing suspicious and non-suspicious foci on the internal test set and 0.796, 0.851, 0.865, 0.814, and 0.923 on the external test set. DL-based lesion segmentation on PSMA PET is facilitated through our anatomical prior guidance strategy. Our classification framework differentiates suspicious foci from those not suspicious for cancer with good accuracy.

Psychosocial characteristics of potential and actual living kidney donors

The psychosocial assessment is an essential component of the living kidney donor (LKD) evaluation. However, it remains uncertain how specific psychosocial factors impact LKD eligibility. We performed a retrospective chart review of LKD candidates who initiated the evaluation process and who had completed a required, in-person licensed social work (LSW) visit. LSW notes were reviewed for frequency of psychosocial factors that may impact the success of LKD candidate approval by the selection committee. 325 LKD candidates were included in the study: 104 not-approved and 221 approved. Not-approved LKD candidates were more likely to receive a negative family reaction to wanting to donate than approved LKD candidates (8.7% vs 1.4%, p < 0.01). On multivariate analysis, Black race, history of psychiatric illness, highest level of education being high school, and high psychosocial risk score assignment were all associated with a lower odds ratio of being approved. The majority of not-approved LKD candidates were disqualified for medical reasons (N = 76, 73.1%). In conclusion, psychosocial factors impact donation even after LKD candidates make it to an in-person evaluation.

Phase 1 study of cabozantinib in combination with topotecan-cyclophosphamide for patients with relapsed Ewing sarcoma or osteosarcoma

PURPOSE

Phase 1 study assessing the safety and toxicity of cabozantinib in combination with topotecan and cyclophosphamide for relapsed osteosarcoma and Ewing sarcoma.

METHODS

Oral cabozantinib (25 mg/m2 ) was administered daily for 21 (dose level 1) or 14 (dose level -1B) days. Topotecan (0.75 mg/m2 ) and cyclophosphamide (250 mg/m2 ) were administered intravenously (IV) on days 1-5. A modified 3+3 design based upon first cycle dose-limiting toxicities (DLT) was used for dose escalation.

RESULTS

Twelve patients with a median age of 15 (12.9-33.2) years were enrolled (seven with Ewing sarcoma; five with osteosarcoma); all were evaluable for toxicity. At dose level 1, three of six patients developed first cycle DLT: grade 3 epistaxis, grade 3 transaminitis, and prolonged grade 2 thrombocytopenia. Six patients were enrolled on dose level -1B (interrupted cabozantinib, given days 8-21), with one first cycle DLT (grade 3 pneumothorax) observed. Of the 10 response evaluable patients, one had partial response (Ewing sarcoma), seven had stable disease, and two had progressive disease.

CONCLUSIONS

The recommended phase 2 doses and schedules for this combination are topotecan 0.75 mg/m2 IV days 1-5, cyclophosphamide 250 mg/m2 IV days 1-5, and cabozantinib 25 mg/m2 days 8-21. Non-concomitant administration of cabozantinib with cytotoxic therapy in this population has acceptable toxicity, while allowing for potential disease control.

Visualizing genomic characteristics across an RNA-Seq based reference landscape of normal and neoplastic brain

In order to better understand the relationship between normal and neoplastic brain, we combined five publicly available large-scale datasets, correcting for batch effects and applying Uniform Manifold Approximation and Projection (UMAP) to RNA-Seq data. We assembled a reference Brain-UMAP including 702 adult gliomas, 802 pediatric tumors and 1409 healthy normal brain samples, which can be utilized to investigate the wealth of information obtained from combining several publicly available datasets to study a single organ site. Normal brain regions and tumor types create distinct clusters and because the landscape is generated by RNA-Seq, comparative gene expression profiles and gene ontology patterns are readily evident. To our knowledge, this is the first meta-analysis that allows for comparison of gene expression and pathways of interest across adult gliomas, pediatric brain tumors, and normal brain regions. We provide access to this resource via the open source, interactive online tool Oncoscape, where the scientific community can readily visualize clinical metadata, gene expression patterns, gene fusions, mutations, and copy number patterns for individual genes and pathway over this reference landscape.

Anxiety, depression, and substance experimentation in childhood

Previous research has demonstrated that adults with comorbid depressive and anxiety disorders are significantly more likely to show pathological use of drugs or alcohol. Few studies, however, have examined associations of this type in children. A better understanding of the relationships between affective disorders and substance experimentation in childhood could help clarify the complex ways in which pathological substance use symptoms develop early in life. The present study included 11,785 children (Mage = 9.9) participating in the Adolescent Brain Cognitive Development (ABCD) study. Depressive and anxiety disorder diagnoses were evaluated as concurrent predictors of experimentation with alcohol and tobacco. A series of linear regressions revealed that children with either depressive or anxiety disorders were significantly more likely to experiment with alcohol or tobacco. However, children with both depressive and anxiety diagnoses were not more likely to experiment than children without a diagnosis. These results suggest that anxiety or depressive diagnoses in childhood may be associated with a greater likelihood of substance experimentation, but severe psychological distress may suppress these effects.

Emerging therapies for relapsed/refractory multiple myeloma: CAR-T and beyond

The pace of innovation of multiple myeloma therapy in recent years is remarkable with the advent of monoclonal antibodies and the approval of novel agents with new mechanisms of action. Emerging therapies are on the horizon for clinical approval with significant implications in extending patient survival and advancing closer to the goal of a cure, especially in areas of immunotherapy such as chimeric antigen receptor T cells, bispecific T cell engager antibodies, antibody drug conjugates, newer generations of monoclonal antibodies, and small molecule inhibitor and modulators. This review provides an update of current myeloma therapeutics in active preclinical and early clinical development and discusses the mechanism of action of several classes of novel therapeutics.

Characterization of an intratracheal aerosol challenge model of Brucella melitensis in guinea pigs

B. melitensis is considered the most virulent of the Brucella species, and a need exists for an improved laboratory animal model of infection that mimics natural transmission and disease. Guinea pigs are highly susceptible to infection with Brucella spp. and develop a disease syndrome that mimics natural disease after aerosol inoculation. Intratracheal inoculation is a targeted means of generating aerosols that offer advantages over aerosol chamber delivery. To establish this delivery method, female, Hartley guinea pigs were infected via intratracheal inoculation with PBS or 16M B. melitensis at low dose (101 to 103) or high dose (106 to 108) and monitored for 30 days for signs of disease. Guinea pigs in the high dose groups developed fever between 12-17 days post-inoculation. Bacteria were recovered from the spleen, liver, lymph nodes, lung, and uterus at 30-days post-inoculation and demonstrated dose dependent mean increases in colonization and pathologic changes consistent with human brucellosis. To study the kinetics of extrapulmonary dissemination, guinea pigs were inoculated with 107 CFU and euthanized at 2-hours post inoculation and at weekly intervals for 3 weeks. 5.8x105 to 4.2x106 CFU were recovered from the lung 2 hours post-inoculation indicating intratracheal inoculation is an efficient means of infecting guinea pigs. Starting at 1-week post inoculation bacteria were recovered from the aforementioned organs with time dependent mean increases in colonization. This data demonstrates that guinea pigs develop a disease syndrome that models the human manifestation of brucellosis, which makes the guinea pig a valuable model for pathogenesis studies.

Identification of an Efficient Gene Expression Panel for Glioblastoma Classification

We present here a novel genetic algorithm-based random forest (GARF) modeling technique that enables a reduction in the complexity of large gene disease signatures to highly accurate, greatly simplified gene panels. When applied to 803 glioblastoma multiforme samples, this method allowed the 840-gene Verhaak et al. gene panel (the standard in the field) to be reduced to a 48-gene classifier, while retaining 90.91% classification accuracy, and outperforming the best available alternative methods. Additionally, using this approach we produced a 32-gene panel which allows for better consistency between RNA-seq and microarray-based classifications, improving cross-platform classification retention from 69.67% to 86.07%. A webpage producing these classifications is available at http://simplegbm.semel.ucla.edu.