Literature Review of Statistical Methods Comparisons through Simulations When Using External Control Arm for Regulatory or HTA Submissions

Objectives: Using a historical control or external control arm (ECA) to augment or replace a concurrent control arm in a randomized trial is a hot topic given the challenge of patient recruitment in rare diseases or during COVID-19 pandemic. The FDA released draft guidance in 2021 on effectiveness and safety submissions using real-world evidence. While the guidance focuses mainly on elements of study design and data source selection, there is a lack of consensus in the selection of appropriate statistical methods when constructing an ECA. This study discusses rigorous statistical methodology for ECA-supported trials in regulatory or HTA submissions. Method(s): Targeted literature reviews of statistical simulations comparing methods for ECA in statistical journals were performed. The articles compared commonly used ECA-construction and analysis methods were selected and summarized, including but not limited to propensity score (PS)-based matching, weighting, and stratification, and PS plus Bayesian integrated approaches. Result(s): Type I error, power, bias, and coverage probability are common criteria used to compare different methods. When imbalances only exist in known baseline covariates and the outcome distributions are the same between the trial concurrent control and ECA, the PS method alone or paired with commensurate prior yield almost unbiased estimates, good Type I errors, and coverage probability. PS plus Bayesian approaches have wider interval width and lower power compared with PS-only methods. When there is a change in the outcome distribution over time, the PS (matching or IPTW) and commensurate prior integrated methods yield the smallest biases among all methods. Conclusion(s): PS and Bayesian integrated methods outperformed the PS-only methods in terms of bias and Type I error when outcome distribution changed with current trial control. A "sweet spot" that balances all criteria through trial-specific simulations could provide the ideal setting of trial analyses plan based on specific trial design and scenarios.Copyright © 2023

Association of interleukin-6 and CD4+ T cells and two-week prognosis of patients with COVID-19: a predictive role.

OBJECTIVE:  The aim of this study was to determine the association of inflammation and immune responses with the outcomes of patients at various stages, and to develop risk stratification for improving clinical practice and reducing mortality. PATIENTS AND METHODS: We included 77 patients with primary outcomes of either death or survival. Demographics, clinical features, comorbidities, and laboratory tests were compared. Linear, logistic, and Cox regression analyses were performed to determine prognostic factors. RESULTS: The average age was 59 years (35-87 years). There were 12 moderate cases (16.2%), 42 severe cases (54.5%), and 23 critical cases (29.9%); and 41 were male (53.2%). Until March 20, 68 cases were discharged (88.3%), and nine critically ill males (11.7%) died. Interleukin-6 (IL-6) levels on the 1st day were compared with IL-6 values on the 14th day in the severe and the critically ill surviving patients (F=4.90, p=0.034, ß=0.35, 95% CI: 0.00-0.10), and predicted death in the critically ill patients (p=0.028, ß=0.05, OR: 1.05, 95% CI: 1.01-1.10). CD4+ T-cell counts at admission decreased the hazard ratio of death (p=0.039, ß=-0.01, hazard ratio=0.99, 95% CI: 0.98-1.00, and median survival time 13.5 days). CONCLUSIONS: The present study demonstrated that IL-6 levels and CD4+ T-cell count at admission played key roles of predictors in the prognosis, especially for critically ill patients. High levels of IL-6 and impaired CD4+t cells are seen in severe and critically ill patients with COVID-19.

Updated efficacy and safety of taletrectinib in patients (pts) with ROS1+ non-small cell lung cancer (NSCLC)

Background Taletrectinib is a potent, next-generation, CNS-active, ROS1 tyrosine kinase inhibitor (TKI) with selectivity over TRKB. In previous reports from TRUST-I, taletrectinib showed meaningful clinical efficacy and was well tolerated in pts with ROS1+ NSCLC (n = 109) regardless of crizotinib (CRZ) pretreatment status. We report updated efficacy and safety data with ~1.5 yr follow-up. Methods TRUST-I is a multicenter, open-label, single-arm study with two cohorts: ROS1 TKI-naive and CRZ-pretreated. Pts in both cohorts received taletrectinib 600 mg QD. Key study endpoints included IRC-confirmed ORR (cORR), DoR, disease control rate (DCR), PFS, and safety. A pooled analysis of ORR, PFS, and safety including pts from additional clinical trials was also conducted. Results In the 109 pts from TRUST-I (enrolled prior to Feb 2022) the median follow-up was 18.0 mo in TKI-naive (n = 67) and 16.9 mo in CRZ-pretreated pts (n = 42). cORR was 92.5% in TKI-naive and 52.6% in CRZ-pretreated pts (table). Median DoR (mDoR) and mPFS were not reached. Intracranial-ORR was 91.6%;ORR in pts with G2032R was 80.0%. In a pooled analysis with phase I studies, ORR was 89.5% and 50.0% for TKI-naive and CRZ-pretreated pts, respectively;mPFS was 33.2 mo and 9.8 mo. In 178 pts treated at 600 mg QD, treatment-emergent adverse events (TEAEs) were 92.7%;most (64.0%) were grade 1-2. The most common TEAEs were increased AST (60.7%), increased ALT (55.6%), and diarrhea (55.6%). Neurological TEAEs (dizziness, 18.5%;dysgeusia, 12.4%) and discontinuations due to TEAEs (3.4%) were low. Further updated results will be presented. [Formula presented] Conclusions With additional follow-up, taletrectinib continued to demonstrate meaningful efficacy outcomes including high response rates, prolonged PFS, robust intracranial activity, activity against G2032R, and tolerable safety with low incidence of neurological AEs. Clinical trial identification NCT04395677. Editorial acknowledgement Medical writing and editorial assistance were provided by Arpita Kulshrestha of Peloton Advantage, LLC, an OPEN Health company, and funded by AnHeart Therapeutics, Inc Legal entity responsible for the study AnHeart Therapeutics, Inc. Funding AnHeart Therapeutics, Inc. Disclosure S. He: Financial Interests, Personal, Other, Employment: AnHeart Therapeutics. T. Seto: Financial Interests, Institutional, Research Grant: AbbVie, Chugai Pharmaceutical, Daiichi Sankyo, Eli Lilly Japan, Kissei Pharmaceutical, MSD, Novartis Pharma, Pfizer Japan, Takeda Pharmaceutical;Financial Interests, Personal, Other, Employment: Precision Medicine Asia;Financial Interests, Personal, Speaker's Bureau, Honoraria for lectures: AstraZeneca, Bristol-Myers Squibb, Chugai Pharmaceutical, Covidien Japan, Daiichi Sankyo, Eli Lilly Japan, Kyowa Hakko Kirin, MSD, Mochida Pharmaceutical, Nippon Boehringer Ingelheim, Novartis Pharma, Ono Pharmaceutical, Pfizer Japan, Taiho Pharmaceutical, Takeda Pharmaceutical, Towa Pharmaceutical. C. Zhou: Financial Interests, Personal, Other, Consulting fees: Innovent Biologics Qilu, Hengrui, TopAlliance Biosciences Inc;Financial Interests, Personal, Speaker's Bureau, Payment or honoraria: Eli Lilly China, Sanofi, BI, Roche, MSD, Qilu, Hengrui, Innovent Biologics, C-Stone LUYE Pharma, TopAlliance Biosciences Inc, Amoy Diagnositics, AnHeart. All other authors have declared no conflicts of interest.Copyright © 2023 International Association for the Study of Lung Cancer. Published by Elsevier Inc.

Jing Si Herbal Drink as a prospective adjunctive therapy for COVID-19 treatment: Molecular evidence and mechanisms

Background: SARS-CoV-2 has led to a sharp increase in the number of hospitalizations and deaths from pneumonia and multiorgan disease worldwide;therefore, SARS-CoV-2 has become a global health problem. Supportive therapies remain the mainstay treatments against COVID-19, such as oxygen inhalation, antiviral drugs, and antibiotics. Traditional Chinese medicine (TCM) has been shown clinically to relieve the symptoms of COVID-19 infection, and TCMs can affect the pathogenesis of SARS-CoV-2 infection in vitro. Jing Si Herbal Drink (JSHD), an eight herb formula jointly developed by Tzu Chi University and Tzu Chi Hospital, has shown potential as an adjuvant treatment for COVID-19 infection. A randomized controlled trial (RCT) of JSHD as an adjuvant treatment in patients with COVID-19 infection is underway Objectives: This article aims to explore the efficacy of the herbs in JSHD against COVID-19 infection from a mechanistic standpoint and provide a reference for the rational utilization of JSHD in the treatment of COVID-19. Method(s): We compiled evidence of the herbs in JSHD to treat COVID-19 in vivo and in vitro. Result(s): We described the efficacy and mechanism of action of the active ingredients in JSHD to treat COVID-19 based on experimental evidence. JSHD includes 5 antiviral herbs, 7 antioxidant herbs, and 7 anti-inflammatory herbs. In addition, 2 herbs inhibit the overactive immune system, 1 herb reduces cell apoptosis, and 1 herb possesses antithrombotic ability. Conclusion(s): Although experimental data have confirmed that the ingredients in JSHD are effective against COVID-19, more rigorously designed studies are required to confirm the efficacy and safety of JSHD as a COVID-19 treatment.Copyright © 2021

Free energy perturbation–based large-scale virtual screening for effective drug discovery against COVID-19

As a theoretically rigorous and accurate method, FEP-ABFE (Free Energy Perturbation-Absolute Binding Free Energy) calculations showed great potential in drug discovery, but its practical application was difficult due to high computational cost. To rapidly discover antiviral drugs targeting SARS-CoV-2 Mpro and TMPRSS2, we performed FEP-ABFE–based virtual screening for ∼12,000 protein-ligand binding systems on a new generation of Tianhe supercomputer. A task management tool was specifically developed for automating the whole process involving more than 500,000 MD tasks. In further experimental validation, 50 out of 98 tested compounds showed significant inhibitory activity towards Mpro, and one representative inhibitor, dipyridamole, showed remarkable outcomes in subsequent clinical trials. This work not only demonstrates the potential of FEP-ABFE in drug discovery but also provides an excellent starting point for further development of anti-SARS-CoV-2 drugs. Besides, ∼500 TB of data generated in this work will also accelerate the further development of FEP-related methods. © The Author(s) 2022.

Literature Review of Statistical Methods Comparisons Through Simulations When Using External Control Arm for Regulatory or HTA Submissions

Objectives: Using a historical control or external control arm (ECA) to augment or replace a concurrent control arm in a randomized trial is a hot topic given the challenge of patient recruitment in rare diseases or during COVID-19 pandemic. FDA released several draft guidance in 2021 on effectiveness and safety submissions using real-world evidence. While the guidance focuses mainly on elements of study design and data source selection, there is a lack of consensus in the selection of appropriate statistical methods when constructing an ECA. This study aims to discuss rigorous statistical methodology for ECA-supported trial in regulatory or HTA submissions. Method(s): Targeted literature reviews of statistical simulations comparing methods for ECA in statistical journals were performed. The articles compare commonly used ECA-construction and analysis methods were selected and summarized, including but not limited to propensity-score (PS) based- matching, weighting, stratification, and PS plus Bayesian integrated approaches. Result(s): Type I error, power, bias, and coverage probability are common criteria used to compare different methods. When imbalances only exist in known baseline covariates and the outcome distribution are the same between the trial concurrent control and ECA, PS method alone or paired with commensurate prior yield almost unbiased estimates, good Type I errors, and coverage probability. PS plus Bayesian approaches have wider interval width and lower power compared with PS only methods. When there is a change in the outcome distribution over time, PS (matching or IPTW) and commensurate prior integrated method yield smallest biases among all methods. Conclusion(s): PS and Bayesian integrated methods outperformed the PS only methods in terms of bias and type I error when outcome distribution changed with current trial control. A "sweet spot" that balances all criteria through trial-specific simulations could provide the ideal setting of trial analyses plan based on specific trial design and scenarios. Copyright © 2022

Phase Ib study of AVB-S6-500 (axl inhibition) in combination with durvalumab (MEDI4736) in patients with platinum-resistant, recurrent epithelial ovarian cancer (NCT04019288) (320)

Objectives: Immune checkpoint blockade (ICB) has demonstrated efficacy in a small fraction of patients with platinum-resistant ovarian cancer (PROC), some with durable responses. The receptor tyrosine kinase AXL and its sole ligand, GAS6, are possible mediators of T cell exclusion and an attractive target due to the expected synergy between AXL inhibition and immune targeting agents. The recommended phase II dose (RP2D), safety, and efficacy of the combination of AXL inhibition via AVB-S6-500 with durvalumab (MEDI4736) were evaluated in patients with PROC. Methods: In this open-label Phase Ib open-label study, patients with PROC received AVB-S6-500 and durvalumab therapy in escalating dosing regimens guided by a Bayesian optimal interval (BOIN) design: durvalumab (1500 mg Q4W) and AVB-S6-500 (10mg/kg Q2W, 15mg/kg Q2W, 20mg/kg Q2W) with durvalumab infused prior to AVB-S6-500. The response was evaluated using modified RECIST v1.1. Pharmacokinetic/pharmacodynamic (PK/PD) studies were collected, and PD-L1 status and tumor/tumor microenvironment AXL and GAS6 staining pre and on-treatment were assessed. Results: Eleven patients with epithelial ovarian cancer (six clear cells [55%], four high-grade serous [36%], one endometrioid histology [1%]) received treatment per protocol. The median number of prior lines of therapy was 3 (range: 1-5);73% (8/11) of patients had received prior bevacizumab. There were no DLTs noted over the 6-week period and no grade ≥3 adverse events attributed to study drugs. Five patients experienced an immune-related AE, most commonly liver enzyme elevations (36%). Infusion reaction with AVB-S6- 500 was noted in the first two subjects, prompting the institution of a premedication regimen, after which only one of the nine additional patients experienced an infusion reaction. Dose delays greater than one week occurred in six (55%) patients;three patients experienced delays for cancer-related complications (small bowel obstruction, pneumonia, severe fatigue), while three patients experienced delays for non-medical causes (COVID/travel, weather). Patients received therapy for a median of two cycles (range: 1-6), and there were no responses noted across all dosing levels. One patient had stable disease, with a duration of response of three months. Only two patients had strong (2+) AXLstaining on pretreatment biopsy, both with high-grade serous histology. The majority of serum AXL levels were within previously demonstrated ranges (range: 5.6-112ng/mL), though two patients had comparatively high levels (102, 112ng/mL). PK/PD analysis revealed expected AVB-S6-500 levels at initial postdose (C1D1), but low levels at trough (C2D1 predose) when compared to prior AVB-S6-500 data [1]. Conclusions: The combination of AVB-S6-500 and durvalumab was tolerable in this PROC patient population at all dosing levels tested. Exploratory studies to correlate lack of response to AXL-GAS6 pathway alterations, tumor microenvironment, and clinical characteristics, such as prior treatment, dosing delays, burden of disease, and ascites, are ongoing.

COVID-19 Classification from Chest X-rays Based on Attention and Knowledge Distillation

The Coronavirus Disease 2019 (COVID-19) is the pandemic that has had the greatest impact on world economic development in recent years. Early detection is critical to identify patients with COVID-19, chest x-ray is used for early detection is a rapid, extensive and cost-effective method. The existing technology use deep learning methods, and have achieved very good results. However, the training time of deep learning method is long, and the model size makes it difficult to deploy on hardware system. In this work, we have proposed an attention-based ResNet50v2 network, and taken the network as the teacher network to transfer the knowledge to the student network by knowledge distillation. Thus, the student network has higher accuracy and sensitivity to the positive samples of COVID-19 under the condition of low model parameters, high training speed. The experimental results show that our network of teacher and student have achieved 100% accuracy and sensitivity in both COVID-19 and Normal binary classification. In addition, the accuracy rate of teacher network is 98.20%, the sensitivity is 99.58%, the accuracy rate of student network is 97.68%, the sensitivity is 99.17% in the COVID-19, Viral pneumonia and Normal multiple classification, and the parameters of the student network are only 0.269M. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Free energy perturbation-based large-scale virtual screening for effective drug discovery against COVID-19

As a theoretically rigorous and accurate method, FEP-ABFE (Free Energy Perturbation-Absolute Binding Free Energy) calculations showed great potential in drug discovery, but its practical application was difficult due to high computational cost. To rapidly discover antiviral drugs targeting SARS-CoV-2 M- pro and TMPRSS2, we performed FEP-ABFE-based virtual screening for similar to 12,000 protein-ligand binding systems on a new generation of Tianhe supercomputer. A task management tool was specifically developed for automating the whole process involving more than 500,000 MD tasks. In further experimental validation, 50 out of 98 tested compounds showed significant inhibitory activity towards M- pro , and one representative inhibitor, dipyridamole, showed remarkable outcomes in subsequent clinical trials. This work not only demonstrates the potential of FEP-ABFE in drug discovery but also provides an excellent starting point for further development of anti-SARS-CoV-2 drugs. Besides, similar to 500 TB of data generated in this work will also accelerate the further development of FEP-related methods.

An AI-Enabled Virtual Hands-On Teaching Tool for Treatment Planning: A Pancreas SBRT Pilot Study

Purpose/Objective(s): To develop a tutoring program to help physician and physics residents to learn pancreas stereotactic body radiation therapy (SBRT) treatment planning via carefully collected cases and a series of specially designed knowledge models as teaching aid. Such programs are especially important during the Covid pandemic when most traditional hands-on medical teaching had to be moved online and asynchronous. Materials/Methods: The pilot tutoring program was composed of 5 teaching cases (1 benchmark case and 4 teaching cases), an interactive knowledge module (IKM) and a performance grading system. The tutoring program started with the benchmark case completed by each trainee independently to benchmark the baseline skill level. This same case was re-tested to evaluate the performance and clinical planning readiness after the trainee completes the tutoring program. 4 teaching cases were included: simple (1), intermediate (2) and complex (1). All 5 cases have simultaneous integrated boost prescription of 25Gy to the elective volume and 33Gy to the gross tumor volume. The IKM included a dose-volume prediction knowledge model for pancreas SBRT and is integrated via a graphic user interface in the treatment planning system. The trainee can seek reference and guidance from the IKM during learning – mimicking the hands-on tutoring of human expert. The grading system summarize the plan quality by weighing key dosimetric endpoints and their relative importance. Grade point average (GPA) was introduced to qualitatively appraise the plan quality into A, B and C (within 3%, 3-10% and > 10% difference of clinical plan score, corresponding to 4, 3, 2 point respectively). 5 trainees with minimum planning experience completed the teaching course. Results: Trainees achieved an average of 65.1% of total points (3.6 GPA) with 84 minutes planning time for the benchmark case pre-teaching, and improved to an average of 75.7% (4.2 GPA) using 48 minutes post-teaching. The clinical plan scored 72.7% of total points. All trainees improved their teaching plans’ scores after taking the virtual tutoring program. Post-teaching, all trainees received the GPA of A (clinical quality level) on the benchmark case planning. The total teaching time for each trainee ranged between 5 and 7 hours. Conclusion: The tutoring program with knowledge support modules provides encouraging learning outcomes in pancreas SBRT planning for inexperienced planners. This AI-enabled virtual teaching tool could provide valuable addition to the traditional human resource heavy in-person teaching of IMRT and SBRT treatment planning.

Online and Offline Hybrid Teaching Mode on Computer Course under Medical Background

Involved in the impending era of 'Internet + ' and big data, educational informatization has been driving the modernization of education and promoting the deep integration of information technology and teaching. The 'Fundamentals of College Computer' course is a public compulsory course for medical colleges. In order to meet the professional requirements of medical colleges, benefitting from the massive online teaching resources during the novel coronavirus epidemic, this paper applies the teaching paradigm of 'online preview + offline teaching + online afterschool test + offline computer assessment' to exploit the advantages of online and offline hybrid teaching. The teaching practice has showed the establishment of the compound hybrid teaching model under medical big data background is an effective exploration to facilitate the development of medical information. © 2021 IEEE.