Citations

The SARS-CoV-2 virus has caused a pandemic, infecting nearly 80 million people worldwide, with mortality exceeding six million. The average survival span is just 14 days from the time the symptoms become aggressive. The present study delineates the deep-driven vascular damage in the pulmonary, renal, coronary, and carotid vessels due to SARS-CoV-2. This special report addresses an important gap in the literature in understanding (i) the pathophysiology of vascular damage and the role of medical imaging in the visualization of the damage caused by SARS-CoV-2, and (ii) further understanding the severity of COVID-19 using artificial intelligence (AI)-based tissue characterization (TC). PRISMA was used to select 296 studies for AI-based TC. Radiological imaging techniques such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound were selected for imaging of the vasculature infected by COVID-19. Four kinds of hypotheses are presented for showing the vascular damage in radiological images due to COVID-19. Three kinds of AI models, namely, machine learning, deep learning, and transfer learning, are used for TC. Further, the study presents recommendations for improving AI-based architectures for vascular studies. We conclude that the process of vascular damage due to COVID-19 has similarities across vessel types, even though it results in multi-organ dysfunction. Although the mortality rate is ~2% of those infected, the long-term effect of COVID-19 needs monitoring to avoid deaths. AI seems to be penetrating the health care industry at warp speed, and we expect to see an emerging role in patient care, reduce the mortality and morbidity rate.

6077 Background: E3311 is a phase II randomized study which showed favorable outcomes among intermediate (INT) risk HPV+ OPC patients (pts) who underwent TOS followed by pathology-guided or adapted, deintensified adjuvant treatment. Among HPV+ pts treated with definitive chemoradiation, survival outcomes are worse among those who smoked > 10 pack years (pk-yrs). Methods: We retrospectively analyzed demographics, pathologic results, and efficacy outcomes from E3311 by smoking group (current (C) vs. former (F) and > 10 vs. ≤10 pk-yrs – the latter a pre-specified stratification factor for INT patients). Binary and categorical variables were compared using a chi-square test (or Fisher’s exact test for small sample sizes). Ordinal variables were compared using a Wilcoxon rank sum test. Progression-free survival (PFS) and overall survival (OS) were estimated using the Kaplan-Meier method and compared using a log-rank test. Results: Among 359 evaluable pts, performance status (PS) was significantly worse for pts with > 10 pk-yrs vs. ≤10 pk-yrs (15.4% vs. 7.9% with PS of 1, p = 0.034). Primary site, margin status, histologic grade, stage, and extranodal extension were not significantly different between the groups of > 10 vs. ≤10 pk-yrs. Smoking status (F vs. C) was available for 182 pts with a history of smoking. Slightly more C vs. F smokers had tonsil as primary site (79.5% vs. 65.0%, p = 0.09). Positive margins were significantly more frequent among C smokers (10.3% vs. 2.1%; p = 0.029). Overall, there were no significant differences in PFS (p = 0.55) or OS (p = 0.94), comparing those with > 10 vs. ≤10 pk-yrs, or comparing C vs. F smokers (p = 0.76, p = 0.82, respectively). Similarly, no significant differences were observed within the treatment arms. (Table 1) Conclusions: In this analysis of smoking status in E3311, INT risk HPV+ OPC pts who are C smokers or have a history of > 10 pk-yrs had favorable 3-yr PFS and OS rates that were not significantly worse than those with < 10 pk-yrs history. This data represents the first treatment approach for HPV+ OPC in which outcomes were not influenced by smoking status. Clinical trial information: NCT01898494. [Table: see text]

Background and Motivation: Parkinson’s disease (PD) is one of the most serious, non-curable, and expensive to treat. Recently, machine learning (ML) has shown to be able to predict cardiovascular/stroke risk in PD patients. The presence of COVID-19 causes the ML systems to become severely non-linear and poses challenges in cardiovascular/stroke risk stratification. Further, due to comorbidity, sample size constraints, and poor scientific and clinical validation techniques, there have been no well-explained ML paradigms. Deep neural networks are powerful learning machines that generalize non-linear conditions. This study presents a novel investigation of deep learning (DL) solutions for CVD/stroke risk prediction in PD patients affected by the COVID-19 framework. Method: The PRISMA search strategy was used for the selection of 292 studies closely associated with the effect of PD on CVD risk in the COVID-19 framework. We study the hypothesis that PD in the presence of COVID-19 can cause more harm to the heart and brain than in non-COVID-19 conditions. COVID-19 lung damage severity can be used as a covariate during DL training model designs. We, therefore, propose a DL model for the estimation of, (i) COVID-19 lesions in computed tomography (CT) scans and (ii) combining the covariates of PD, COVID-19 lesions, office and laboratory arterial atherosclerotic image-based biomarkers, and medicine usage for the PD patients for the design of DL point-based models for CVD/stroke risk stratification. Results: We validated the feasibility of CVD/stroke risk stratification in PD patients in the presence of a COVID-19 environment and this was also verified. DL architectures like long short-term memory (LSTM), and recurrent neural network (RNN) were studied for CVD/stroke risk stratification showing powerful designs. Lastly, we examined the artificial intelligence bias and provided recommendations for early detection of CVD/stroke in PD patients in the presence of COVID-19. Conclusion: The DL is a very powerful tool for predicting CVD/stroke risk in PD patients affected by COVID-19.

Abstract Bortezomib, a proteasome inhibitor (PI), has shown efficacy in the treatment of newly diagnosed and relapsed light chain (AL) amyloidosis, and is often used in combination with cyclophosphamide and dexamethasone. Ixazomib is the first oral PI to be approved in routine practice but has not yet been evaluated in the upfront treatment setting. Newly diagnosed AL amyloidosis patients with measurable disease and adequate organ function were enrolled. The primary objective was to determine the hematologic response rate of ixazomib in combination with cyclophosphamide and dexamethasone. Treatment was given for 12 cycles, followed by ixazomib maintenance until progression. Thirty-five patients were included; their median age was 67 years, and 69% were male. Major organ involvement included heart (66%) and kidneys (54%). A median of 4 induction cycles (range, 1-12) were administered. The overall hematologic response to induction was 63% and included complete response in 11.4% and very good partial response in 37.1% of patients. One patient was upstaged to complete response during maintenance. The most common reason for going off study was the institution of alternate therapy (61%). With a median follow-up of 29.7 months for the living patients, the 2-year progression-free survival and overall survival were 74% and 78%, respectively. The median time to alternate therapy was 7.5 months. Grade ≥3 hematologic and nonhematologic adverse events occurred in 23% and 49% of patients. Given ixazomib’s favorable toxicity profile, which is an important advantage for the typically frail AL population, further evaluation of ixazomib in other combinations in the upfront setting is warranted. This trial was registered at www.clinicaltrials.gov as #NCT01864018.

Background: The previous COVID-19 lung diagnosis system lacks both scientific validation and the role of explainable artificial intelligence (AI) for understanding lesion localization. This study presents a cloud-based explainable AI, the “COVLIAS 2.0-cXAI” system using four kinds of class activation maps (CAM) models. Methodology: Our cohort consisted of ~6000 CT slices from two sources (Croatia, 80 COVID-19 patients and Italy, 15 control patients). COVLIAS 2.0-cXAI design consisted of three stages: (i) automated lung segmentation using hybrid deep learning ResNet-UNet model by automatic adjustment of Hounsfield units, hyperparameter optimization, and parallel and distributed training, (ii) classification using three kinds of DenseNet (DN) models (DN-121, DN-169, DN-201), and (iii) validation using four kinds of CAM visualization techniques: gradient-weighted class activation mapping (Grad-CAM), Grad-CAM++, score-weighted CAM (Score-CAM), and FasterScore-CAM. The COVLIAS 2.0-cXAI was validated by three trained senior radiologists for its stability and reliability. The Friedman test was also performed on the scores of the three radiologists. Results: The ResNet-UNet segmentation model resulted in dice similarity of 0.96, Jaccard index of 0.93, a correlation coefficient of 0.99, with a figure-of-merit of 95.99%, while the classifier accuracies for the three DN nets (DN-121, DN-169, and DN-201) were 98%, 98%, and 99% with a loss of ~0.003, ~0.0025, and ~0.002 using 50 epochs, respectively. The mean AUC for all three DN models was 0.99 (p < 0.0001). The COVLIAS 2.0-cXAI showed 80% scans for mean alignment index (MAI) between heatmaps and gold standard, a score of four out of five, establishing the system for clinical settings. Conclusions: The COVLIAS 2.0-cXAI successfully showed a cloud-based explainable AI system for lesion localization in lung CT scans.

e19583 Background: Brexucabtagene autoleucel (BA) is an FDA approved therapy for relapsed/refractory (R/R) mantle cell lymphoma (MCL), based on results from ZUMA-2 study. We report the safety and efficacy of BA in standard of care practice among centers in the US Lymphoma CAR-T Consortium. Methods: 16 centers participated in this retrospective study. Patients (pts) who underwent leukapheresis by 12/31/2021 with an intent to manufacture BA were included. Baseline clinical characteristics, bridging therapy, adverse events after BA infusion, and post-infusion outcome data were collected. Eligibility for ZUMA-2 was retrospectively determined. Survival outcomes were analyzed using the Kaplan-Meier method. Results: At the data cut-off date, 189 pts underwent leukapheresis, among whom 167 (88%) completed BA infusion, 22 (12%) did not receive infusion. The median age was 67 years. 16% had high risk simplified MIPI, 57% had Ki-67≥50%, 41% had aggressive histology, 49% had TP53 alteration, 51% with POD24 and 10% had CNS involvement. The median number of prior lines of therapy was 3 (range 1-10). 86% had prior BTKi treatment (89% refractory). 130 (78%) pts would not have met ZUMA-2 eligibility criteria. 113 (68%) pts received bridging therapy, which included BTKi, venetoclax, chemotherapy. Median time from leukapheresis to lymphodepletion chemotherapy: 28 days (range 17-140). Cytokine release syndrome (CRS) rate was 90% (8% grade ≥3; 1 grade 5), and immune effector cell-associated neurotoxicity syndrome (ICANS) rate was 61% (32% grade ≥3). Grade 5 ICANS (n = 0). Medications used to manage CRS and ICANS were 125 (76%) for tocilizumab, 112 (68%) for corticosteroid, and 27 (16%) for anakinra. 32 (20%) pts required ICU admission, 18 pts required vasopressors, and 5 pts required mechanical ventilation. Day 30 response was evaluable in 155 pts (89% ORR, 70% CR). The best ORR was 89%, with 80% CR, 9% PR. The ORR/CR rates were 88%/79% for aggressive histology, 89%/77% for high Ki-67% (> 50%), 90%/72% for TP53 altered, 81%/75% for CNS involved, 89%/79% for BTKi-exposed, 91%/83% for BTKi-naïve, and 89%/78% for those not meeting ZUMA-2 eligibility criteria. The median duration of response was not reached and at 6-month was 67% (95% CI 57-75). With a median follow-up of 5.6 months (range 0.2-15.3), median PFS was not reached, the 6-month PFS rate was 63% (95% CI 54-71), and median OS was 15.3 months and the 6-month OS rate was 85% (95% CI 77-90). Conclusions: This multicenter retrospective study demonstrated encouraging safety and efficacy data of BA in R/R MCL in the real-world practice. Despite 78% of the pts being ineligible for ZUMA-2, the responses, CRS, ICANS and outcomes were comparable to ZUMA-2 data. Long term safety and efficacy will be reported.

Abstract Plexiform Neurofibromas (PN) are a common manifestation of the genetic disorder neurofibromatosis type 1 (NF1). These benign nerve sheath tumors often cause significant morbidity, with treatment options limited historically to surgery. There have been tremendous advances over the past two decades in our understanding of PN, and the recent regulatory approvals of the MEK inhibitor selumetinib are reshaping the landscape for PN management. At present, there is no agreed upon PN definition, diagnostic evaluation, surveillance strategy, or clear indications for when to initiate treatment and selection of treatment modality. In this review, we address these questions via consensus recommendations from a panel of multidisciplinary NF1 experts.

Background Oncolytic virotherapy (OV) represents a method to treat a variety of solid tumors by inducing antitumor immune responses. While this therapy has been extremely efficacious in preclinical models, translating these successes into human patients has proven challenging. One of the major reasons for these failures is the existence of immune-regulatory mechanisms, which dampen the efficacy of virally induced antitumor immunity. Unfortunately, the full extent of these immune-regulatory pathways remains unclear. Methods To address this issue, we generated a doubly recombinant, oncolytic myxoma virus which expresses both a soluble fragment of programmed cell death protein 1 (PD1) and an interleukin 12 (IL-12) fusion protein (vPD1/IL-12 (virus-expressing PD1 and IL-12)). We then tested the molecular impact and therapeutic efficacy of this construct in multiple models of disseminated disease to identify novel pathways, which are associated with poor therapeutic outcomes. Results Our results demonstrate that vPD1/IL-12 causes robust inflammation during therapy including inducing high levels of tumor necrosis factor (TNF). Surprisingly, although expression of TNF has generally been assumed to be beneficial to OV, the presence of this TNF appears to inhibit therapeutic efficacy by reducing intratumoral T-cell viability. Likely because of this, disruption of the TNF pathway, either through genetic knockout or antibody-based blockade, significantly enhances the overall outcomes of vPD1/IL-12-based therapy that allows for the generation of complete cures in normally non-responsive models. Conclusions These data suggest that some aspects of OV-induced inflammation might represent a double-edged sword during therapy and that specific blockade of TNF might enhance the efficacy of these treatments.