scholarly journals 821 Machine learning models can quantify CD8 positivity in lymphocytes in melanoma clinical trial samples

2021 ◽  
Vol 9 (Suppl 3) ◽  
pp. A859-A859
Author(s):  
Benjamin Glass ◽  
S Adam Stanford-Moore ◽  
Diksha Meghwal ◽  
Nishant Agrawal ◽  
Mary Lin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Trials ◽  
Tumor Microenvironment ◽  
Cell Carcinoma ◽  
Metastatic Melanoma ◽  
Cell Types ◽  
Cell Counts ◽  
Immune Phenotyping ◽  
Cd8 Cell

BackgroundAn accurate histological characterization of immune cells in the tumor microenvironment is essential for developing novel immune oncology targeted therapies and can assist in guiding patient treatment decisions. However, immune phenotyping is subject to challenges of manual scoring and inter-pathologist scoring variability. To support pathologist-scored immune phenotyping across tumor types, we are developing machine learning (ML)-based models that can identify and quantify CD8+ lymphocytes within the stromal and parenchyma regions of tumors from non-small cell lung cancer, renal cell carcinoma, breast cancer, gastric cancer, head and neck squamous cell carcinoma, urothelial carcinoma, and melanoma. Here, we focus on the ML model for melanoma showing recent results for ML-based identification and quantification of CD8+ lymphocytes and concordance with manual pathologic assessment in data derived from clinical trials.MethodsML algorithms were developed to quantify CD8+ lymphocytes in melanoma using 200 samples from a commercial dataset containing both primary and metastatic melanoma cases. Models were trained using the PathAI research platform on digitized whole slide images (WSI) stained for CD8 using clone C8/144b (Dako), and annotations were provided by the PathAI network of expert pathologists. Training included identification of slide artifacts, parenchyma, cancer stroma, and necrosis, as well as CD8+ lymphocytes and other CD8– cell types. Examples of melanin, such as pigmented macrophages, were added to non-CD8+ cell types. To evaluate the performance of the ML model, model-predicted CD8+ counts were compared to a consensus count from five independent pathologists for representative regions (“frames”) using the Pearson correlation. This was done in 112 held-out test frames from 90 WSI baseline samples from three clinical trials of immunotherapy treatment in individuals with metastatic melanoma. Inter-pathologist agreement among the five pathologists was also calculated.ResultsML-based quantitation of CD8 positivity in lymphocytes showed high concordance with manual pathologist consensus counts. In frames validation of CD8+ counts on the test set of WSI, there was high correlation between the ML model and pathologist consensus counts (r=0.92 [95% CI 0.88–0.94]). This correlation was comparable to the agreement among the five expert pathologists (r=0.88 [95% CI 0.85–0.91]).ConclusionsML model-predicted CD8+ cell counts are highly concordant with pathologist scores on WSI samples from melanoma-focused clinical trials. These data demonstrate the capability of AI-powered digital pathology for accurate and reproducible quantitation of CD8+ lymphocytes in clinical trial samples, contributing to improved evaluation of the tumor microenvironment and targeted development of therapeutics.

scholarly journals Understanding and predicting COVID-19 clinical trial completion vs. cessation

PLoS ONE ◽  
2021 ◽  
Vol 16 (7) ◽  
pp. e0253789
Author(s):  
Magdalyn E. Elkin ◽  
Xingquan Zhu
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Trials ◽  
Computational Methods ◽  
Predictive Models ◽  
State Of The Art ◽  
Area Under The Curve ◽  
Information Criteria ◽  
The Other ◽  
Satisfactory Accuracy

As of March 30 2021, over 5,193 COVID-19 clinical trials have been registered through Clinicaltrial.gov. Among them, 191 trials were terminated, suspended, or withdrawn (indicating the cessation of the study). On the other hand, 909 trials have been completed (indicating the completion of the study). In this study, we propose to study underlying factors of COVID-19 trial completion vs. cessation, and design predictive models to accurately predict whether a COVID-19 trial may complete or cease in the future. We collect 4,441 COVID-19 trials from ClinicalTrial.gov to build a testbed, and design four types of features to characterize clinical trial administration, eligibility, study information, criteria, drug types, study keywords, as well as embedding features commonly used in the state-of-the-art machine learning. Our study shows that drug features and study keywords are most informative features, but all four types of features are essential for accurate trial prediction. By using predictive models, our approach achieves more than 0.87 AUC (Area Under the Curve) score and 0.81 balanced accuracy to correctly predict COVID-19 clinical trial completion vs. cessation. Our research shows that computational methods can deliver effective features to understand difference between completed vs. ceased COVID-19 trials. In addition, such models can also predict COVID-19 trial status with satisfactory accuracy, and help stakeholders better plan trials and minimize costs.

Safety and efficacy outcomes with nivolumab plus ipilimumab in patients with advanced renal cell carcinoma and brain metastases: results from the CheckMate 920 trial.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 4515-4515
Author(s):  
Hamid Emamekhoo ◽  
Mark R Olsen ◽  
Bradley Curtis Carthon ◽  
Alexandra Drakaki ◽  
Ivor John Percent ◽  
...  
Keyword(s):  
Clinical Trial ◽  
Renal Cell Carcinoma ◽  
Clinical Trials ◽  
Brain Metastases ◽  
Cell Carcinoma ◽  
Renal Cell ◽  
Safety And Efficacy ◽  
Medical Need ◽  
Grade 3

4515 Background: Combination therapy with nivolumab plus ipilimumab (NIVO+IPI) has demonstrated long-term efficacy and tolerability in patients with previously untreated advanced renal cell carcinoma (aRCC). Previous phase 3 clinical trials of patients with advanced or metastatic cancers have mostly excluded patients with brain metastases. CheckMate 920 is an ongoing, phase 3b/4 clinical trial of NIVO+IPI treatment in patients with aRCC with a high unmet medical need. We present updated safety and efficacy results for the cohort of patients with aRCC of any histology and brain metastases from CheckMate 920 (NCT02982954). Methods: Patients with previously untreated advanced/metastatic aRCC of any histology, with asymptomatic brain metastases (not currently receiving corticosteroids or radiation), and Karnofsky performance status ≥ 70% were assigned to treatment with NIVO 3 mg/kg + IPI 1 mg/kg every 3 weeks × 4 doses followed by NIVO 480 mg every 4 weeks for ≤ 2 years or until disease progression/unacceptable toxicity. The primary endpoint was incidence of grade ≥ 3 immune-mediated adverse events (imAEs) within 100 days of last dose of study drug. Key secondary endpoints included progression-free survival (PFS) and objective response rate (ORR) by RECIST v1.1 (both per investigator). Exploratory endpoints included overall survival (OS). Results: Of 28 treated patients with brain metastases, 85.7% were men; median (range) age was 60 (38–87) years, and 14.3% had sarcomatoid features. With 24.5 months minimum follow-up of the 28 patients enrolled, median duration of therapy (range) was 3.4 (0.0–23.3) months for NIVO and 2.1 (0.0–3.3) months for IPI. No grade 5 imAEs occurred. Grade 3–4 imAEs by category were diarrhea/colitis (7.1%), hypophysitis (3.6%), rash (3.6%), hepatitis (3.6%), and diabetes mellitus (3.6%). Of the 25 patients who were evaluable for ORR, the ORR was 32.0% (95% CI, 14.9–53.5). No patients achieved complete response, 8 achieved partial response, and 10 patients had stable disease. Median time to response (range) was 2.8 (2.4–3.0) months. Median duration (range) of response was 24.0 (3.9–not estimable [NE]) months; 4 of 8 responders remain without reported progression. Of 28 patients, 7 (25%) had intracranial progression. Median PFS (n = 28) was 9.0 (95% CI, 2.9–12.0) months. Median OS (n = 28) was still not reached (95% CI, 14.1 months–NE). Conclusions: In patients with previously untreated aRCC and brain metastases, a population with high unmet medical need that is often underrepresented in clinical trials, the approved treatment regimen of NIVO+IPI followed by NIVO for aRCC showed no new safety signals and continues to show encouraging antitumor activity with longer follow-up. Clinical trial information: NCT02982954.

scholarly journals Predictive modeling of clinical trial terminations using feature engineering and embedding learning

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Magdalyn E. Elkin ◽  
Xingquan Zhu
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Trials ◽  
Area Under The Curve ◽  
Common Factors ◽  
Information Criteria ◽  
Feature Engineering ◽  
Exclusion Criteria ◽  
Direct Estimate ◽  
Satisfactory Prediction

AbstractIn this study, we propose to use machine learning to understand terminated clinical trials. Our goal is to answer two fundamental questions: (1) what are common factors/markers associated to terminated clinical trials? and (2) how to accurately predict whether a clinical trial may be terminated or not? The answer to the first question provides effective ways to understand characteristics of terminated trials for stakeholders to better plan their trials; and the answer to the second question can direct estimate the chance of success of a clinical trial in order to minimize costs. By using 311,260 trials to build a testbed with 68,999 samples, we use feature engineering to create 640 features, reflecting clinical trial administration, eligibility, study information, criteria etc. Using feature ranking, a handful of features, such as trial eligibility, trial inclusion/exclusion criteria, sponsor types etc., are found to be related to the clinical trial termination. By using sampling and ensemble learning, we achieve over 67% Balanced Accuracy and over 0.73 AUC (Area Under the Curve) scores to correctly predict clinical trial termination, indicating that machine learning can help achieve satisfactory prediction results for clinical trial study.

scholarly journals Machine learning enabled subgroup analysis with real-world data to inform better clinical trial design

2021 ◽  
Author(s):  
Jie Xu ◽  
Hao Zhang ◽  
Hansi Zhang ◽  
Jiang Bian ◽  
Fei Wang
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Trials ◽  
Real World ◽  
Clinical Manifestations ◽  
Research Network ◽  
Real World Data ◽  
Eligibility Criteria ◽  
World Data ◽  
Patient Subgroups

Restrictive eligibility criteria for clinical trials may limit the generalizability of treatment effectiveness and safety to real-world patients. In this paper, we propose a machine learning approach to derive patient subgroups from real-world data (RWD), such that the patients within the same subgroup share similar clinical characteristics and safety outcomes. The effectiveness of our approach was validated on two existing clinical trials with the electronic health records (EHRs) from a large clinical research network. One is the donepezil trial for Alzheimer's disease (AD), and the other is the Bevacizumab trial on colon cancer (CRC). The results show that our proposed algorithm can identify patient subgroups with coherent clinical manifestations and similar risk levels of encountering severe adverse events (SAEs). We further exemplify that potential rules for describing the patient subgroups with less SAEs can be derived to inform the design of clinical trial eligibility criteria.

Designing and Evaluating a Real-Time Automated Patient Screening System in an Emergency Department (Preprint)

2019 ◽  
Author(s):  
Yizhao Ni ◽  
Monica Bermudez ◽  
Stephanie Kennebeck ◽  
Stacey Liddy-Hicks ◽  
Judith Dexheimer
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Emergency Department ◽  
Clinical Trials ◽  
Real Time ◽  
Patient Identification ◽  
Post Evaluation ◽  
Patient Screening ◽  
Automated Screening ◽  
Quantitative Assessments

BACKGROUND One critical hurdle for clinical trial recruitment is the lack of an efficient method for identifying subjects who meet eligibility criteria. Given the large volume of data documented in electronic health records (EHRs), it is labor-intensive for the staff to screen relevant information, particularly within the time frame needed. To facilitate subject identification, we developed a natural language processing (NLP) and machine learning-based system, Automated Clinical Trial Eligibility Screener© (ACTES), which analyzed structured data and unstructured narratives automatically to determine patients' suitability for clinical trial enrollment. In this study, we integrated the ACTES into clinical practice to support real-time patient screening. OBJECTIVE Our objective was to evaluate the ACTES's impact on the institutional workflow prospectively and comprehensively. We hypothesized that compared with the manual screening process, using EHR-based automated screening would improve efficiency of patient identification, streamline patient recruitment workflow, and increase enrollment in clinical trials. METHODS The ACTES was fully integrated into the clinical research coordinator (CRC) workflow in the pediatric emergency department (ED) at Cincinnati Children's Hospital Medical Center. The system continuously analyzed EHR information for current ED patients and recommended potential candidates for clinical trials. Relevant patient eligibility information was presented in real-time on a dashboard available to CRCs to facilitate their recruitment. To assess the system's effectiveness, we performed a multidimensional, prospective evaluation for a 12-month period, including a time-and-motion study, quantitative assessments of enrollment, and post-evaluation usability surveys collected from the CRCs. RESULTS Compared to manual screening, use of ACTES reduced the patient screening time by 34% (P<0.0001). The saved time was redirected to other work-related activities that streamlined teamwork among the CRCs (P <0.05). The quantitative assessments showed that automated screening improved the numbers of subjects screened, approached and enrolled by more than 10%, suggesting the potential of ACTES in streamlining recruitment workflow. The post-evaluation surveys indicated that the system was a good computerized solution with satisfactory usability. CONCLUSIONS By leveraging NLP and machine learning technologies, the ACTES demonstrated good capacity for improving efficiency of patient identification. The quantitative assessments demonstrated the potential of ACTES in streamlining recruitment workflow and improving patient approach and enrollment. The post-evaluation surveys suggested that the system was a good computerized solution with satisfactory usability.

Clinical factors associated with outcome in metastatic renal cell carcinoma patients treated with VEGF-targeted therapy

2007 ◽  
Vol 25 (18_suppl) ◽  
pp. 5046-5046 ◽  
Author(s):  
A. Golshayan ◽  
T. K. Choueiri ◽  
P. Elson ◽  
J. A. Garcia ◽  
M. Khaswneh ◽  
...  
Keyword(s):  
Clinical Trial ◽  
Renal Cell Carcinoma ◽  
Clinical Trials ◽  
Prognostic Factors ◽  
Cell Carcinoma ◽  
Clinical Features ◽  
Metastatic Renal Cell Carcinoma ◽  
Renal Cell ◽  
Clinical Trial Results ◽  
Metastatic Rcc

5046 Background: Therapy targeted against the vascular endothelial growth factor (VEGF) pathway is a standard of care in metastatic renal cell carcinoma (RCC). Identification of clinical features of patients more likely to benefit from these agents would aid in patient selection and interpretation of clinical trial results. Methods: We reviewed 120 metastatic RCC patients receiving bevacizumab, sorafenib, sunitinib or axitinib on one of eight prospective clinical trials at the Cleveland Clinic Taussig Cancer Center. Clinical features associated with outcome were identified by univariate analysis, and then a stepwise modeling approach based on Cox proportional hazards regression was used to identify independent prognostic factors and form a model for progression-free survival (PFS). A bootstrap algorithm was used to provide internal validation. Results: Forty-one patients (34%) achieved an objective response by RECIST criteria (95% C.I. 27–44%). The median PFS for the entire group was 13.8 months (m) (95% C.I. 10.7–19.0 m). Multivariate analysis identified the following independent adverse prognostic factors (PF) for PFS: time from diagnosis to current treatment <2 years, baseline platelet count >300 K/μL, baseline neutrophil count >4.5 K/μL, baseline corrected serum calcium <8.5 or >10.0 mg/dL and initial ECOG performance status >0. Using these factors three prognostic subgroups were formed based on the number of adverse PF present . Median PFS in patients with 0 or 1 adverse PF was 20.1 m (95% C.I. 19.0–22.3 m) compared to 13 m (95% C.I. 8.6–17.6 m) in patients with 2 adverse PF and 3.9 m (95% C.I. 1.8–7.2 m) in patients with >2 adverse PF. Conclusions: Five independent prognostic factors for predicting PFS were identified and used to categorize patients with metastatic RCC receiving VEGF-targeted therapies into three risk groups. These factors can be readily incorporated to clinical patient care, stratification schema for clinical trials utilizing these novel agents and for interpretation of clinical trial results using VEGF-targeted agents therapy. No significant financial relationships to disclose.

scholarly journals Application of machine learning methods in clinical trials for precision medicine

2021 ◽  
Author(s):  
Yizhuo Wang ◽  
Bing Z. Carter ◽  
Ziyi Li ◽  
Xuelin Huang
Keyword(s):  
Machine Learning ◽  
Clinical Trial ◽  
Clinical Trials ◽  
Precision Medicine ◽  
Response Rate ◽  
Treatment Options ◽  
Prediction Algorithm ◽  
Good Prediction ◽  
Overall Response ◽  
Trial Participants

AbstractObjectiveA key component for precision medicine is a good prediction algorithm for patients’ response to treatments. We aim to implement machine learning (ML) algorithms into the response-adaptive randomization (RAR) design and improve the treatment outcomes.Materials and MethodsWe incorporated nine ML algorithms to model the relationship of patient responses and biomarkers in clinical trial design. Such a model predicted the response rate of each treatment for each new patient and provide guidance for treatment assignment. Realizing that no single method may fit all trials well, we also built an ensemble of these nine methods. We evaluated their performance through quantifying the benefits for trial participants, such as the overall response rate and the percentage of patients who receive their optimal treatments.ResultsSimulation studies showed that the adoption of ML methods resulted in more personalized optimal treatment assignments and higher overall response rates among trial participants. Compared with each individual ML method, the ensemble approach achieved the highest response rate and assigned the largest percentage of patients to their optimal treatments. For the real-world study, we successfully showed the potential improvements if the proposed design had been implemented in the study.ConclusionIn summary, the ML-based RAR design is a promising approach for assigning more patients to their personalized effective treatments, which makes the clinical trial more ethical and appealing. These features are especially desirable for late-stage cancer patients who have failed all the FDA-approved treatment options and only can get new treatments through clinical trials.

scholarly journals Heterogeneity of tumor microenvironment is associated with clinical prognosis of non-clear cell renal cell carcinoma: a single-cell genomics study

2022 ◽  
Vol 13 (1) ◽  
Author(s):  
Wen-jin Chen ◽  
Hao Cao ◽  
Jian-wei Cao ◽  
Li Zuo ◽  
Fa-jun Qu ◽  
...  
Keyword(s):  
Renal Cell Carcinoma ◽  
T Cells ◽  
Tumor Microenvironment ◽  
Cell Carcinoma ◽  
Cd8 T Cells ◽  
Renal Cell ◽  
Poor Prognosis ◽  
Tumor Biology ◽  
Cell Types ◽  
Duct Carcinoma

AbstractNon-clear renal cell carcinomas (nccRCCs) are less frequent in kidney cancer with histopathological heterogeneity. A better understanding of the tumor biology of nccRCC can provide more effective treatment paradigms for different subtypes. To reveal the heterogeneity of tumor microenvironment (TME) in nccRCC, we performed 10x sing-cell genomics on tumor and normal tissues from patients with papillary renal cell carcinoma (pRCC), chromophobe RCC (chrRCC), collecting duct carcinoma (CDRCC) and sarcomatoid RCC (sarRCC). 15 tissue samples were finally included. 34561 cells were identified as 16 major cell clusters with 34 cell subtypes. Our study presented the sing-cell landscape for four types of nccRCC, and demonstrated that CD8+ T cells exhaustion, tumor-associated macrophages (TAMs) and sarcomatoid process were the pivotal factors in immunosuppression of nccRCC tissues and were closely correlated with poor prognosis. Abnormal metabolic patterns were present in both cancer cells and tumor-infiltrating stromal cells, such as fibroblasts and endothelial cells. Combined with CIBERSORTx tool, the expression data of bulk RNA-seq from TCGA were labeled with cell types of our sing-cell data. Calculation of the relative abundance of cell types revealed that greater proportion of exhausted CD8+ T cells, TAMs and sarRCC derived cells were correlated with poor prognosis in the cohort of 274 nccRCC patients. To the best of our knowledge, this is the first study that provides a more comprehensive sight about the heterogeneity and tumor biology of nccRCC, which may potentially facilitate the development of more effective therapies for nccRCC.

Sign in / Sign up

Export Citation Format

Share Document