A5014969426- Cancer Immunotherapy and Biomarkers
- Immunotherapy and Immune Responses
- Cancer Genomics and Diagnostics
- Radiomics and Machine Learning in Medical Imaging
- Monoclonal and Polyclonal Antibodies Research
- Lung Cancer Diagnosis and Treatment
- Lymphoma Diagnosis and Treatment
- Hepatocellular Carcinoma Treatment and Prognosis
- Lung Cancer Treatments and Mutations
- Lung Cancer Research Studies
- Colorectal Cancer Treatments and Studies
- Venous Thromboembolism Diagnosis and Management
- Medical Imaging and Pathology Studies
- Colorectal and Anal Carcinomas
- Acute Ischemic Stroke Management
- Bladder and Urothelial Cancer Treatments
- Ultrasound in Clinical Applications
Institut Gustave Roussy
2022-2024
Université Paris-Saclay
2023
Abstract Purpose: The objective of the study is to propose immunotherapy progression decision (iPD) score, a practical tool based on patient features that are available at first evaluation treatment, help oncologists decide whether continue treatment or switch rapidly another therapeutic line when facing progressive disease evaluation. Experimental Design: This retrospective included 107 patients with according RECIST 1.1. Clinical, radiological, and biological data baseline were analyzed....
The objective of our study is to propose fast, cost-effective, convenient, and effective biomarkers using the perfusion parameters from dynamic contrast-enhanced ultrasound (DCE-US) for evaluation immune checkpoint inhibitors (ICI) early response. retrospective cohort used in this included 63 patients with metastatic cancer eligible immunotherapy. DCE-US was performed at baseline, day 8 (D8), 21 (D21) after treatment onset. A tumor curve modeled on these three dates, change seven measured...
<div>AbstractPurpose:<p>The objective of the study is to propose immunotherapy progression decision (iPD) score, a practical tool based on patient features that are available at first evaluation treatment, help oncologists decide whether continue treatment or switch rapidly another therapeutic line when facing progressive disease evaluation.</p>Experimental Design:<p>This retrospective included 107 patients with according RECIST 1.1. Clinical, radiological, and...
<p>Distribution of primary cancer types in the development set (n=107).</p>
<p>InterpretML Overall Importance using Explainable Boosting Machines (EBM). The mean absolute score reflects the overall importance assigned by model to predict each patient's category. All features appear important, with number of organs affected metastasis and emergence new lesions as most important ones</p>
<p>Log hazard ratios (HR) computed by cox model. The variables studied were: GRIm, RMH, NLR BL & the iPD score. only significant variable was score (HR = 2, p < 0.005).</p>
<p>Survival of pseudo-PD vs confirmed PD in our cohort.</p>
<p>InterpretML Overall Importance using Explainable Boosting Machines (EBM). The mean absolute score reflects the overall importance assigned by model to predict each patient's category. All features appear important, with number of organs affected metastasis and emergence new lesions as most important ones.</p>
<p>Organ involvement's impact on survival. The organs selected were the ones most affected in our study's patients. tests are for survival of patients whether an organ has a tumor or not. were: A) Liver; B) Lung; C) Subdiaphragm lymph nodes; D) Supra Diaphragm E) Peritoneal Carcinosis; and F) Bone.</p>
<p>Survival of pseudo-PD vs confirmed PD in our cohort.</p>
<p>Effect of the removal each criterion on accuracy a multinomial logistic regression fit to predict category patient. The minimal losses are underlined, and maximum written in bold.</p>
<p>Radiological feature impact on survival. Radiological features include progression of target lesions > 20%, non-target 50% and the emergence new lesions.</p>
<p>Distribution of primary cancer types in the development set (n=107).</p>
<p>Radiological feature impact on survival. Radiological features include progression of target lesions > 20%, non-target 50% and the emergence new lesions.</p>
<p>Correlation tables of our criteria. Bottom left are spearman coefficients, top right Phi_K coefficients.</p>
<p>Organ involvement's impact on survival. The organs selected were the ones most affected in our study's patients. tests are for survival of patients whether an organ has a tumor or not. were: A) Liver; B) Lung; C) Subdiaphragm lymph nodes; D) Supra Diaphragm E) Peritoneal Carcinosis; and F) Bone.</p>
<p>InterpretML Overall Importance using Explainable Boosting Machines (EBM). The mean absolute score reflects the overall importance assigned by model to predict each patient's category. All features appear important, with number of organs affected metastasis and emergence new lesions as most important ones</p>
<p>Correlation tables of our criteria. Bottom left are spearman coefficients, top right Phi_K coefficients.</p>
<p>Log Hazard ratio (HR) of each parameter obtained from a cox model fitted to predict survival.</p>
<p>Log hazard ratios (HR) computed by cox model. The variables studied were: GRIm, RMH, NLR BL & the iPD score. only significant variable was score (HR = 2, p < 0.005).</p>
<p>InterpretML Overall Importance using Explainable Boosting Machines (EBM). The mean absolute score reflects the overall importance assigned by model to predict each patient's category. All features appear important, with number of organs affected metastasis and emergence new lesions as most important ones.</p>
<p>Log Hazard ratio (HR) of each parameter obtained from a cox model fitted to predict survival.</p>