This document discusses applying a common grading system for laboratory data from local laboratories involved in clinical trials for early drug development in oncology. It notes the challenges of standardizing data from multiple local labs with different normal ranges and methods. It proposes capturing lab results and normal ranges directly on case report forms to facilitate data management and integration across sites.
Web-Triage An Application for patient registration in phase I dose escalation...Angelo Tinazzi
This document describes an application called Web-Triage for patient registration in phase I dose escalation oncology studies. SENDO is a non-profit academic research organization that coordinates phase I-II oncology clinical trials across sites in Italy and Switzerland. Phase I studies test new drugs in cancer patients to determine safety and optimal dosing. Web-Triage streamlines patient screening and registration to accelerate trial enrollment. The application was designed to meet the specific needs of phase I oncology trials, including rapid sorting of eligible patients to optimize treatment.
The document discusses the complexity of operating an expert center for cancer research as a private-public partnership. It describes the Europroteome experience managing a colon cancer biobank to discover biomarkers and targets using multiple technologies while integrating clinical data. Challenges included maintaining the biobank, meeting investor and industry demands, and managing expectations. The needs of the pharmaceutical industry are also reviewed, focusing on reducing development risks and costs through validated targets linked to pathways.
The document summarizes the Medfiber project, which aims to develop a non-invasive fiber optic probe for urinary incontinence diagnosis. The current methods are invasive, inconclusive, and expensive. Medfiber would be catheter-free, have high accuracy, and reduce hospitalization costs. It combines interferometric fiber optic sensor technology with a disposable all-in-one probe. The founders describe the market opportunity, business model, sales and marketing plan, team, and project milestones. Their projections estimate the company could reach over 3 million Euros in net income within 5 years if scalability goals are met.
A Systematic Review of ADaM IG InterpretationAngelo Tinazzi
The document summarizes a systematic review of publications about the implementation of the ADaM model. Over 100 papers were identified that discussed ADaM implementation, with the majority coming from CRO authors. Several areas of interpretation in the ADaM guidelines were identified from the literature, including how to classify parameters in BDS, derive rows versus columns, and determine what constitutes an "analysis-ready" dataset. The review concluded that feedback from users would help the CDISC team further develop and clarify the ADaM guidelines.
This document describes a presentation by Angelo Tinazzi of Cytel Inc. looking for an SDTM specialist to work at the PhUSE conference in London from October 12-15, 2014. It provides examples of challenges faced in legacy data conversion, including a case study of performing a gap analysis for a small biotech company and discussing issues in modeling oncology study data to SDTM.
The use of Adaptive designs is becoming quite popular and well-perceived by the regulatory agencies such as the FDA in the US. “Adaptation” can occur in different fashion and potentially make studies more efficient (e.g. shorter duration, fewer patients) more likely to demonstrate an effect of the drug if one exists, or more informative (see “Adaptive Design Clinical Trials for Drugs and Biologics” FDA guidance).
The aim of this presentation is to illustrate a case where an adaptive design was used in a Phase III oncology pivotal study having Overall Survival as a primary end-point. The particular adaptation implemented was an un-blinded SSR that applied a promising zone approach.
The main focus will be how the adaptive design impacted the SDTM modelling, the design of some ADaM datasets (e.g. those containing the time-to-event endpoints and therefore using ADTTE ADaM model) and later on how some mapping and analysis decisions were described in both the study and analysis reviewer guide.
INTERPRETING CDISC ADaM IG THROUGH USERS INTERPRETATIONAngelo Tinazzi
The document appears to be a presentation about interpreting the CDISC ADaM Implementation Guide. It discusses conducting a systematic review of papers discussing ADaM implementation to understand areas of ambiguity and different interpretations. The review found over 100 papers focused on ADaM implementation, with the most discussions at PharmaSUG and PhUSE conferences. Common topics of interpretation included traceability, non-ADaM analysis datasets, validation, and issues with the basic data structure. The presentation aims to facilitate discussion on resolving areas of ambiguity in the guide.
This document discusses CDISC standards for representing survival data from oncology clinical trials. It provides an overview of CDISC and describes the SDTM and ADaM domains that are useful for capturing efficacy endpoints involving survival, such as overall survival, progression-free survival and tumor response. Examples are given of how survival data from different patients would be represented in an ADTTE (Analysis Dataset for Time to Event) dataset according to CDISC ADaM standards.
Web-Triage An Application for patient registration in phase I dose escalation...Angelo Tinazzi
This document describes an application called Web-Triage for patient registration in phase I dose escalation oncology studies. SENDO is a non-profit academic research organization that coordinates phase I-II oncology clinical trials across sites in Italy and Switzerland. Phase I studies test new drugs in cancer patients to determine safety and optimal dosing. Web-Triage streamlines patient screening and registration to accelerate trial enrollment. The application was designed to meet the specific needs of phase I oncology trials, including rapid sorting of eligible patients to optimize treatment.
The document discusses the complexity of operating an expert center for cancer research as a private-public partnership. It describes the Europroteome experience managing a colon cancer biobank to discover biomarkers and targets using multiple technologies while integrating clinical data. Challenges included maintaining the biobank, meeting investor and industry demands, and managing expectations. The needs of the pharmaceutical industry are also reviewed, focusing on reducing development risks and costs through validated targets linked to pathways.
The document summarizes the Medfiber project, which aims to develop a non-invasive fiber optic probe for urinary incontinence diagnosis. The current methods are invasive, inconclusive, and expensive. Medfiber would be catheter-free, have high accuracy, and reduce hospitalization costs. It combines interferometric fiber optic sensor technology with a disposable all-in-one probe. The founders describe the market opportunity, business model, sales and marketing plan, team, and project milestones. Their projections estimate the company could reach over 3 million Euros in net income within 5 years if scalability goals are met.
A Systematic Review of ADaM IG InterpretationAngelo Tinazzi
The document summarizes a systematic review of publications about the implementation of the ADaM model. Over 100 papers were identified that discussed ADaM implementation, with the majority coming from CRO authors. Several areas of interpretation in the ADaM guidelines were identified from the literature, including how to classify parameters in BDS, derive rows versus columns, and determine what constitutes an "analysis-ready" dataset. The review concluded that feedback from users would help the CDISC team further develop and clarify the ADaM guidelines.
This document describes a presentation by Angelo Tinazzi of Cytel Inc. looking for an SDTM specialist to work at the PhUSE conference in London from October 12-15, 2014. It provides examples of challenges faced in legacy data conversion, including a case study of performing a gap analysis for a small biotech company and discussing issues in modeling oncology study data to SDTM.
The use of Adaptive designs is becoming quite popular and well-perceived by the regulatory agencies such as the FDA in the US. “Adaptation” can occur in different fashion and potentially make studies more efficient (e.g. shorter duration, fewer patients) more likely to demonstrate an effect of the drug if one exists, or more informative (see “Adaptive Design Clinical Trials for Drugs and Biologics” FDA guidance).
The aim of this presentation is to illustrate a case where an adaptive design was used in a Phase III oncology pivotal study having Overall Survival as a primary end-point. The particular adaptation implemented was an un-blinded SSR that applied a promising zone approach.
The main focus will be how the adaptive design impacted the SDTM modelling, the design of some ADaM datasets (e.g. those containing the time-to-event endpoints and therefore using ADTTE ADaM model) and later on how some mapping and analysis decisions were described in both the study and analysis reviewer guide.
INTERPRETING CDISC ADaM IG THROUGH USERS INTERPRETATIONAngelo Tinazzi
The document appears to be a presentation about interpreting the CDISC ADaM Implementation Guide. It discusses conducting a systematic review of papers discussing ADaM implementation to understand areas of ambiguity and different interpretations. The review found over 100 papers focused on ADaM implementation, with the most discussions at PharmaSUG and PhUSE conferences. Common topics of interpretation included traceability, non-ADaM analysis datasets, validation, and issues with the basic data structure. The presentation aims to facilitate discussion on resolving areas of ambiguity in the guide.
This document discusses CDISC standards for representing survival data from oncology clinical trials. It provides an overview of CDISC and describes the SDTM and ADaM domains that are useful for capturing efficacy endpoints involving survival, such as overall survival, progression-free survival and tumor response. Examples are given of how survival data from different patients would be represented in an ADTTE (Analysis Dataset for Time to Event) dataset according to CDISC ADaM standards.
This document discusses Filipino male vanity and grooming habits. It notes that 48% of Filipino men think they are attractive, higher than rates in other Asian countries. Filipino men spend significant time and money on their appearance, with 84% rating their looks as important. They frequently get haircuts, use grooming products like cologne, and visit beauty salons. Some get cosmetic procedures. This vanity is argued to stem from a desire to attract women and feel confident despite economic instability.
Presented at PhUSE 2013
The evaluation of efficacy in oncology studies, in particular for solid tumors, is pretty standard and well defined by several regulatory guidance (e.g. EMA and FDA), including some specific cancer type guidance (e.g. NSCLC from FDA).
Although some references will be also given for non-solid tumors, the paper will mainly focus on solid tumors efficacy
endpoints.
Overall Survival, Best Overall Response as per RECIST criteria, Progression Free Survival (PFS), Time to Progression (TTP), Best Overall Response Rate are some of the key efficacy indicators that will be discussed.
The document discusses various topics related to grading systems, including:
1) Grading systems are designed to provide incentives for achievement and identify student problems.
2) Many papers and theses discuss developing computerized or online grading systems to automate the grading process.
3) Automating grading systems aims to reduce errors, save time, and easily track student performance records.
The document discusses the Combined Data Interchange Standard Consortium (CDISC) and its Standard Data Tabulation Model (SDTM). CDISC develops standards to support clinical research data exchange and submission. SDTM defines a standard structure for study data tabulations submitted to regulators. The document outlines key aspects of SDTM including its implementation guide, fundamentals, observation classes, special purpose domains, trial design model, relationship datasets, metadata, controlled terminology, and date/time variables.
The document summarizes the key aspects of the K-12 education system overhaul in the Philippines including the introduction of a new senior high school program. It outlines the new senior high school tracks, subjects, facilities, schedules, teachers, grading system and components. The grading system introduces ratings of Did Not Meet Expectations to Outstanding based on performance and uses weighted averages of written work, performance tasks and final exams to determine overall grades.
This document defines a data flow diagram (DFD) and its components. A DFD is a graphical representation of how data flows through a system. It shows external entities, processes, data stores, and data flows. External entities interact with the system, processes manipulate data, data stores hold data, and data flows show the movement of data. The document provides examples of DFD symbols and components. It also explains that DFDs can be leveled to show more detail at each level, with level 0 providing an overview and higher levels showing more granular processes.
Thesis in IT Online Grade Encoding and Inquiry System via SMS TechnologyBelLa Bhe
This document provides background information on an online grade encoding and inquiry system via SMS technology for the San Mateo Municipal College. It discusses the college's current manual grading system and the problems with it, such as the long process for students to inquire about their grades. The objectives of developing a new online system are outlined, including allowing instructors to encode grades online and students to inquire about grades via SMS. The scope and limitations of the new system are also defined. Finally, the significance of the study in benefiting instructors, students, administrators, and future researchers is described.
EXAMINING THE EFFECT OF FEATURE SELECTION ON IMPROVING PATIENT DETERIORATION ...IJDKP
This document discusses examining the effect of feature selection on improving patient deterioration prediction in intensive care units. The authors apply feature selection techniques to laboratory test data from the MIMIC-II database to identify the most important laboratory tests for predicting patient deterioration. They find that feature selection can help reduce redundant tests, potentially saving costs and allowing earlier treatment. The selected features provide insights into critical tests without domain expertise. In future work, the authors plan to evaluate additional feature selection methods and classification algorithms on this task.
1. Genomic predictors like Oncotype DX and Mammaprint provide prognostic information about the risk of cancer recurrence in early-stage breast cancer patients. They have been validated in multiple retrospective studies but need further prospective validation.
2. These genomic predictors provide additional information beyond standard clinical parameters like ER and PR status. However, they do not perfectly correlate with these parameters and may capture different biological information.
3. Improving genomic predictors may involve developing them within specific molecular subclasses of breast cancer and using larger datasets to train them. Whole genome analyses could provide more detailed biological insights to guide treatment decisions.
Validazione dei cfDNA Test e Controlli di Qualità Esterni e InterniRoberto Scarafia
BREVE PREMESSA:
Negli ultimi anni si sono sviluppate tecnologie altamente sofisticate che consentono di valutare il rischio per condizioni cromosomiche fetali. L'ampio ventaglio di opzioni oramai disponibili nell'ambito degli screening non invasivi pone numerosi quesiti su quale tecnologia utilizzare e le problematiche specifiche connesse alla tecnologia. Durante questa mezza giornata di aggiornamento verranno dunque presentate in modo semplificato le basi molecolari delle differenti tecnologie coi vantaggi e vantaggi correlati, quali test sono disponibili e il loro livello di certificazione in relazione alla normativa europea inerente alla marchiatura CE-IVD, quali sono le cause di risultati discordanti, dei ‘no results’ e la gestione dei casi con risultato ad alto rischio, no result e discordanze
OBIETTIVI FORMATIVI:
• Descrivere le differenti tecnologie disponibili coi relativi vantaggi e svantaggi;
• Presentare le cause biologiche dei risultati discordanti mediante cfDNA test;
• Illustrare le diverse cause di ‘no result’ e le implicazioni sulle performances del test;
• Descrivere l’utilità delle certificazioni, validazioni dei cfDNA test e dei controlli esterni di
qualità;
• Discutere circa l’utilità clinica dei contenuti aggiuntivi oltre alle trisomie 21,18,13;
• Discutere circa il follow-up e il management dei risultati ad alto rischio, dei no results e dei
risultati discordanti.
DISTANT-CTO: A Zero Cost, Distantly Supervised Approach to Improve Low-Resour...Anjani Dhrangadhariya
PICO recognition is an information extraction task for identifying participant, intervention, comparator, and outcome information from clinical literature.
Manually identifying PICO information is the most time-consuming step for conducting systematic reviews (SR) which is already a labor-intensive process.
A lack of diversified and large, annotated corpora restricts innovation and adoption of automated PICO recognition systems.
The largest-available PICO entity/span corpus is manually annotated which is too expensive for a majority of the scientific community.
To break through the bottleneck, we propose DISTANT-CTO, a novel distantly supervised PICO entity extraction approach using the clinical trials literature, to generate a massive weakly-labeled dataset with more than a million ``Intervention'' and ``Comparator'' entity annotations.
We train distant NER (named-entity recognition) models using this weakly-labeled dataset and demonstrate that it outperforms even the sophisticated models trained on the manually annotated dataset with a 2\% F1 improvement over the Intervention entity of the PICO benchmark and more than 5\% improvement when combined with the manually annotated dataset.
We investigate the generalizability of our approach and gain an impressive F1 score on another domain-specific PICO benchmark.
The approach is not only zero-cost but is also scalable for a constant stream of PICO entity annotations.
Fortis Clinical Research Ltd. is a leading clinical research organization in India that provides various clinical research services to support the global pharmaceutical industry in drug development. The document outlines Fortis' vision, services, leadership team, facilities, capabilities, and experience conducting various types of clinical trials and bioanalytical work.
Validation of an m-Health Solution for the Follow-Up of Post-operative Patien...ITACA-TSB
This document describes a study that validated the use of an m-Health system for monitoring postoperative patients from ambulatory surgery units. The study involved 310 patients split into a pilot group that used the m-Health system and a control group without it. The study aimed to evaluate clinical indicators and patient outcomes. The results showed no significant differences between the groups in pain levels, complications, or calls/visits to the hospital. The m-Health system did not negatively impact postoperative care or decrease quality of care compared to standard phone follow-ups.
UK-Italy dementia workshop, January 2018 - TagliaviniUKinItaly
The document summarizes the Italian Network of Neuroscience and Neurorehabilitation (IRCCS), which consists of 24 research hospitals specializing in neurology, child neurology, neurosurgery, and neurorehabilitation. The network conducts clinical activities and translational research, with funding from the Italian Ministry of Health and regional health systems. It has the capacity to enroll large cohorts of patients in clinical trials and research studies due to its extensive clinical resources and databases. A current research project aims to develop operational criteria for early detection of Alzheimer's disease using multi-factorial protocols combining clinical, imaging, and molecular data across several network hospitals.
The document provides information about a short course on next generation sequencing and analysis of sequence variants. It includes an agenda with sessions on introduction to NGS applications in medical research, data analysis pipelines, interpretation of variants, and tools for predicting pathogenicity. It also provides background on the organizing institutions, the CNAG sequencing center and its projects, and an overview of bioinformatics analysis pipelines and resources.
Recent advances in Tuberculosis diagnosisNishantTawari
This document discusses recent advances in tuberculosis diagnosis. It notes that in 2017 there were over 10 million new TB cases globally, including 2.8 million in India. New diagnostic techniques have been developed to improve detection of both drug-sensitive and drug-resistant TB. These include nucleic acid amplification tests like Xpert MTB/RIF, which can detect TB and rifampin resistance in under 3 hours. Other techniques discussed are line probe assays, automated liquid culture systems, and urine lipoarabinomannan tests. The document examines the advantages and limitations of various methods for directly and indirectly detecting active TB.
Brendan Delany – Chair in Medical Informatics and Decision Making, Imperial...HIMSS UK
The document discusses the EU-funded TRANSFoRm project, which aimed to develop methods and validated architectures to support a learning health system. The project involved 21 partners from 10 EU member states. It sought to enable real-time clinical diagnosis and trials using data from electronic health records. It developed ontologies and standards to maintain meaning across the learning health system. A prototype clinical decision support system integrated into a primary care electronic health record was evaluated in a simulation and found to improve diagnostic accuracy and management without increasing consultation time or test ordering.
Healthcare is undergoing a technological transformation, and it is imperative for the industry to leverage new technologies to generate, collect, and track novel data. Panel chaired by Ralf Reilmann of the George Huntington Institut, Muenster.
This document describes the mass spectrometry laboratory at the University of Liège. It lists the staff members and provides an overview of the laboratory's philosophy and services. The laboratory offers a range of standard services including protein identification using mass spectrometry, glycan analysis, and quantitative proteomics. It also engages in custom projects involving areas like binding interactions, mass spectrometry imaging, and nanoparticle synthesis and functionalization. The goal is to provide high-quality mass spectrometry services and expertise to both internal and external researchers.
Sk microfluidics and lab on-a-chip-ch6stanislas547
This document discusses cancer diagnostics and monitoring using microfluidic lab-on-a-chip technologies. It describes how integrating DNA/protein separation, detection, and analysis into microfluidic chips could allow for frequent, non-invasive testing of cancer biomarkers in blood or other bodily fluids. This would enable more precise monitoring of cancer treatment effectiveness and earlier detection of recurrence compared to standard techniques. The document outlines approaches involving microfluidic separation channels coupled to molecular detection and proposes a credit card-sized disposable chip sensor integrated with a small control unit for point-of-care cancer screening and monitoring.
This document discusses Filipino male vanity and grooming habits. It notes that 48% of Filipino men think they are attractive, higher than rates in other Asian countries. Filipino men spend significant time and money on their appearance, with 84% rating their looks as important. They frequently get haircuts, use grooming products like cologne, and visit beauty salons. Some get cosmetic procedures. This vanity is argued to stem from a desire to attract women and feel confident despite economic instability.
Presented at PhUSE 2013
The evaluation of efficacy in oncology studies, in particular for solid tumors, is pretty standard and well defined by several regulatory guidance (e.g. EMA and FDA), including some specific cancer type guidance (e.g. NSCLC from FDA).
Although some references will be also given for non-solid tumors, the paper will mainly focus on solid tumors efficacy
endpoints.
Overall Survival, Best Overall Response as per RECIST criteria, Progression Free Survival (PFS), Time to Progression (TTP), Best Overall Response Rate are some of the key efficacy indicators that will be discussed.
The document discusses various topics related to grading systems, including:
1) Grading systems are designed to provide incentives for achievement and identify student problems.
2) Many papers and theses discuss developing computerized or online grading systems to automate the grading process.
3) Automating grading systems aims to reduce errors, save time, and easily track student performance records.
The document discusses the Combined Data Interchange Standard Consortium (CDISC) and its Standard Data Tabulation Model (SDTM). CDISC develops standards to support clinical research data exchange and submission. SDTM defines a standard structure for study data tabulations submitted to regulators. The document outlines key aspects of SDTM including its implementation guide, fundamentals, observation classes, special purpose domains, trial design model, relationship datasets, metadata, controlled terminology, and date/time variables.
The document summarizes the key aspects of the K-12 education system overhaul in the Philippines including the introduction of a new senior high school program. It outlines the new senior high school tracks, subjects, facilities, schedules, teachers, grading system and components. The grading system introduces ratings of Did Not Meet Expectations to Outstanding based on performance and uses weighted averages of written work, performance tasks and final exams to determine overall grades.
This document defines a data flow diagram (DFD) and its components. A DFD is a graphical representation of how data flows through a system. It shows external entities, processes, data stores, and data flows. External entities interact with the system, processes manipulate data, data stores hold data, and data flows show the movement of data. The document provides examples of DFD symbols and components. It also explains that DFDs can be leveled to show more detail at each level, with level 0 providing an overview and higher levels showing more granular processes.
Thesis in IT Online Grade Encoding and Inquiry System via SMS TechnologyBelLa Bhe
This document provides background information on an online grade encoding and inquiry system via SMS technology for the San Mateo Municipal College. It discusses the college's current manual grading system and the problems with it, such as the long process for students to inquire about their grades. The objectives of developing a new online system are outlined, including allowing instructors to encode grades online and students to inquire about grades via SMS. The scope and limitations of the new system are also defined. Finally, the significance of the study in benefiting instructors, students, administrators, and future researchers is described.
EXAMINING THE EFFECT OF FEATURE SELECTION ON IMPROVING PATIENT DETERIORATION ...IJDKP
This document discusses examining the effect of feature selection on improving patient deterioration prediction in intensive care units. The authors apply feature selection techniques to laboratory test data from the MIMIC-II database to identify the most important laboratory tests for predicting patient deterioration. They find that feature selection can help reduce redundant tests, potentially saving costs and allowing earlier treatment. The selected features provide insights into critical tests without domain expertise. In future work, the authors plan to evaluate additional feature selection methods and classification algorithms on this task.
1. Genomic predictors like Oncotype DX and Mammaprint provide prognostic information about the risk of cancer recurrence in early-stage breast cancer patients. They have been validated in multiple retrospective studies but need further prospective validation.
2. These genomic predictors provide additional information beyond standard clinical parameters like ER and PR status. However, they do not perfectly correlate with these parameters and may capture different biological information.
3. Improving genomic predictors may involve developing them within specific molecular subclasses of breast cancer and using larger datasets to train them. Whole genome analyses could provide more detailed biological insights to guide treatment decisions.
Validazione dei cfDNA Test e Controlli di Qualità Esterni e InterniRoberto Scarafia
BREVE PREMESSA:
Negli ultimi anni si sono sviluppate tecnologie altamente sofisticate che consentono di valutare il rischio per condizioni cromosomiche fetali. L'ampio ventaglio di opzioni oramai disponibili nell'ambito degli screening non invasivi pone numerosi quesiti su quale tecnologia utilizzare e le problematiche specifiche connesse alla tecnologia. Durante questa mezza giornata di aggiornamento verranno dunque presentate in modo semplificato le basi molecolari delle differenti tecnologie coi vantaggi e vantaggi correlati, quali test sono disponibili e il loro livello di certificazione in relazione alla normativa europea inerente alla marchiatura CE-IVD, quali sono le cause di risultati discordanti, dei ‘no results’ e la gestione dei casi con risultato ad alto rischio, no result e discordanze
OBIETTIVI FORMATIVI:
• Descrivere le differenti tecnologie disponibili coi relativi vantaggi e svantaggi;
• Presentare le cause biologiche dei risultati discordanti mediante cfDNA test;
• Illustrare le diverse cause di ‘no result’ e le implicazioni sulle performances del test;
• Descrivere l’utilità delle certificazioni, validazioni dei cfDNA test e dei controlli esterni di
qualità;
• Discutere circa l’utilità clinica dei contenuti aggiuntivi oltre alle trisomie 21,18,13;
• Discutere circa il follow-up e il management dei risultati ad alto rischio, dei no results e dei
risultati discordanti.
DISTANT-CTO: A Zero Cost, Distantly Supervised Approach to Improve Low-Resour...Anjani Dhrangadhariya
PICO recognition is an information extraction task for identifying participant, intervention, comparator, and outcome information from clinical literature.
Manually identifying PICO information is the most time-consuming step for conducting systematic reviews (SR) which is already a labor-intensive process.
A lack of diversified and large, annotated corpora restricts innovation and adoption of automated PICO recognition systems.
The largest-available PICO entity/span corpus is manually annotated which is too expensive for a majority of the scientific community.
To break through the bottleneck, we propose DISTANT-CTO, a novel distantly supervised PICO entity extraction approach using the clinical trials literature, to generate a massive weakly-labeled dataset with more than a million ``Intervention'' and ``Comparator'' entity annotations.
We train distant NER (named-entity recognition) models using this weakly-labeled dataset and demonstrate that it outperforms even the sophisticated models trained on the manually annotated dataset with a 2\% F1 improvement over the Intervention entity of the PICO benchmark and more than 5\% improvement when combined with the manually annotated dataset.
We investigate the generalizability of our approach and gain an impressive F1 score on another domain-specific PICO benchmark.
The approach is not only zero-cost but is also scalable for a constant stream of PICO entity annotations.
Fortis Clinical Research Ltd. is a leading clinical research organization in India that provides various clinical research services to support the global pharmaceutical industry in drug development. The document outlines Fortis' vision, services, leadership team, facilities, capabilities, and experience conducting various types of clinical trials and bioanalytical work.
Validation of an m-Health Solution for the Follow-Up of Post-operative Patien...ITACA-TSB
This document describes a study that validated the use of an m-Health system for monitoring postoperative patients from ambulatory surgery units. The study involved 310 patients split into a pilot group that used the m-Health system and a control group without it. The study aimed to evaluate clinical indicators and patient outcomes. The results showed no significant differences between the groups in pain levels, complications, or calls/visits to the hospital. The m-Health system did not negatively impact postoperative care or decrease quality of care compared to standard phone follow-ups.
UK-Italy dementia workshop, January 2018 - TagliaviniUKinItaly
The document summarizes the Italian Network of Neuroscience and Neurorehabilitation (IRCCS), which consists of 24 research hospitals specializing in neurology, child neurology, neurosurgery, and neurorehabilitation. The network conducts clinical activities and translational research, with funding from the Italian Ministry of Health and regional health systems. It has the capacity to enroll large cohorts of patients in clinical trials and research studies due to its extensive clinical resources and databases. A current research project aims to develop operational criteria for early detection of Alzheimer's disease using multi-factorial protocols combining clinical, imaging, and molecular data across several network hospitals.
The document provides information about a short course on next generation sequencing and analysis of sequence variants. It includes an agenda with sessions on introduction to NGS applications in medical research, data analysis pipelines, interpretation of variants, and tools for predicting pathogenicity. It also provides background on the organizing institutions, the CNAG sequencing center and its projects, and an overview of bioinformatics analysis pipelines and resources.
Recent advances in Tuberculosis diagnosisNishantTawari
This document discusses recent advances in tuberculosis diagnosis. It notes that in 2017 there were over 10 million new TB cases globally, including 2.8 million in India. New diagnostic techniques have been developed to improve detection of both drug-sensitive and drug-resistant TB. These include nucleic acid amplification tests like Xpert MTB/RIF, which can detect TB and rifampin resistance in under 3 hours. Other techniques discussed are line probe assays, automated liquid culture systems, and urine lipoarabinomannan tests. The document examines the advantages and limitations of various methods for directly and indirectly detecting active TB.
Brendan Delany – Chair in Medical Informatics and Decision Making, Imperial...HIMSS UK
The document discusses the EU-funded TRANSFoRm project, which aimed to develop methods and validated architectures to support a learning health system. The project involved 21 partners from 10 EU member states. It sought to enable real-time clinical diagnosis and trials using data from electronic health records. It developed ontologies and standards to maintain meaning across the learning health system. A prototype clinical decision support system integrated into a primary care electronic health record was evaluated in a simulation and found to improve diagnostic accuracy and management without increasing consultation time or test ordering.
Healthcare is undergoing a technological transformation, and it is imperative for the industry to leverage new technologies to generate, collect, and track novel data. Panel chaired by Ralf Reilmann of the George Huntington Institut, Muenster.
This document describes the mass spectrometry laboratory at the University of Liège. It lists the staff members and provides an overview of the laboratory's philosophy and services. The laboratory offers a range of standard services including protein identification using mass spectrometry, glycan analysis, and quantitative proteomics. It also engages in custom projects involving areas like binding interactions, mass spectrometry imaging, and nanoparticle synthesis and functionalization. The goal is to provide high-quality mass spectrometry services and expertise to both internal and external researchers.
Sk microfluidics and lab on-a-chip-ch6stanislas547
This document discusses cancer diagnostics and monitoring using microfluidic lab-on-a-chip technologies. It describes how integrating DNA/protein separation, detection, and analysis into microfluidic chips could allow for frequent, non-invasive testing of cancer biomarkers in blood or other bodily fluids. This would enable more precise monitoring of cancer treatment effectiveness and earlier detection of recurrence compared to standard techniques. The document outlines approaches involving microfluidic separation channels coupled to molecular detection and proposes a credit card-sized disposable chip sensor integrated with a small control unit for point-of-care cancer screening and monitoring.
Efficiency of Remote Technical Support for the ICD follow-up outpatientsNayaMed
During the EHRA Pacing 2013, NayaMed had the chance to present a poster showing the :"Efficiency of Remote Technical Support for the ICD follow-up outpatients."
PR-246: A deep learning system for differential diagnosis of skin diseasesSunghoon Joo
This document summarizes a study on developing a deep learning system (DLS) for differential skin disease diagnosis using teledermatology data. The DLS was trained on a dataset of 14,000 skin disease cases labeled by 43 dermatologists. It achieved average sensitivity of 80% on validation data, outperforming dermatologists and other medical professionals. Subgroup analysis found the DLS was better at distinguishing malignant, infectious, and non-infectious diseases requiring different treatments. Integrated gradients helped explain the model's decisions. Clinical metadata, like self-reported symptoms, also improved performance. In conclusion, the DLS shows promise as a diagnostic tool for common skin diseases.
A Novel approach for quantitative real-time particle analysis of lentiviral v...Myriade
Lentiviral vectors are efficient vehicles for stable gene transfer in dividing and non-dividing cells. They tend to be increasingly used as a powerful tool to introduce genes into cells ex vivo, for instance in CAR-T cell therapies.
During manufacturing and production of lentiviral vectors, relevant quality control is necessary to allow batch release (1). Among standard quality control methods that can be used, quantification of lentiviral vector particles – or physical titer – is one of the most important. Up to now, this characterization can be achieved either indirectly with p24 protein quantification or with physical methods like Tunable Resistive Pulse Sensing (TRPS) for example, both methods implying prior preparation of samples (lysis, dilution or filtration). These two methods thus show important limitations as they cannot accurately reflect the true nature of the product, in addition to being relatively time-consuming (2).
Myriade, a French company created in 2017, is developing Videodrop, a new optical technique performing real-time, user-friendly, and label-free measurement of lentiviral vector physical titer. This method, based on full-field interferometry (3), was tested on various lentiviral vector samples: in a context of Drug Product (DP) release as well as in-process controls.
We compared three lentiviral physical titration methods on aratinga.bio productions: p24 ELISA, qNano and Videodrop – Myriade instrument. The correlation between Videodrop analysis and the other two methods appeared to be robust, with high R² values. These results suggest that Myriade technology is relevant for DP release as well as in-process controls, offering the ability to be a tool for continuous improvement. It is an easy-touse and fast alternative to the standard more complex and time-consuming physical titration methods.
Lab-on-a-Chip for cancer diagnostics and monitoringstanislas547
This document discusses lab-on-a-chip technology for cancer diagnostics and monitoring. It describes how lab-on-a-chip allows miniaturization of diagnostic tools to fit on a small chip. Examples are given of chips that can detect cancer markers from small samples of blood or other bodily fluids. The document outlines how lab-on-a-chip could provide frequent, non-invasive monitoring of cancer markers to guide treatment and detect recurrence. However, challenges remain in developing control units and integrating all necessary functions like fluid handling and molecular analysis onto a single chip.
Monte Carlo And Ct Interface For Medical Treatment Plansfondas vakalis
The document discusses using Geant4, an open-source Monte Carlo simulation toolkit, to develop a general-purpose dosimetry system for medical treatment planning with brachytherapy applications. Key goals are precision, realistic geometry and material modeling from CT images, and speed for clinical use. The system would provide an alternative to commercial software which uses approximations and is not flexible or affordable for all applications like hadron therapy or niche uses. Geant4 capabilities enable accurate modeling of physics interactions down to low energies needed for medical simulations.
Similar to From Local Laboratory to Standardisation and beyond Applying a common grading system (20)
This document discusses an adaptive clinical trial design that was used in a phase III oncology study. The particular adaptation was an unblinded sample size re-estimation based on interim analysis results. This required changes to the SDTM and ADaM data models to account for the interim analysis cut-off dates. The reviewer guides were also updated to explain how to identify patients in the interim analysis and which analysis datasets to use for re-calculating results based on the interim and final cut-offs.
This document summarizes a paper presented at the PhUSE 2014 conference about migrating clinical trial data from its original format to the CDISC SDTM format. It outlines the key steps in the SDTM migration process, including performing a gap analysis, understanding the source datasets, modeling the migration, performing the migration, and final validation. It emphasizes that SDTM migration can be challenging, especially for studies with complex designs, and requires careful planning by a specialized SDTM migration expert.
Therapeutic Area Standards –Reflections on Oncology standards and what is ne...Angelo Tinazzi
This document discusses CDISC standards for oncology clinical trials and opportunities for improvement. It notes that CDISC has incorporated domains for tumor response assessment but questions remain around implementing RECIST criteria and capturing prior cancer therapies. The document suggests areas where CDISC oncology standards could provide more guidance, such as trial design elements like unlimited treatment cycles, capturing exposure modifications, and providing more support for non-solid tumor response assessments. Overall, it examines current CDISC oncology standards and identifies examples of additional clarity and domains that could better meet the needs of oncology clinical research.
While the evolution of information technology is bringing the data closer to customers for their own exploration, the need of a comprehensive understanding of the therapeutic area knowledge for programmers in clinical development is increasing. Starting with a basic understanding on the medical background, special assessment methods, ways of statistically analyzing and displaying the data, to name a few essential ones enables programmers to interact with partners (e.g. scientist, statisticians etc.) on equal par.
In this intent, activities to collect and provide comprehensive information around the Oncology and Rheumatoid Arthritis Therapeutic Areas (TA) via the PhUSE Wiki had started in February 2013 and continued throughout the year. Various PhUSE members have spent time and energy to provide and expand their knowledge and make it available to the entire community.
Today, although there is still much to do to complete and maintain the collected material, the two TA Wikis are a useful tool for Statistical Programmers approaching these TA for the first time or who want to improve their knowledge. Moreover the PhUSE Wiki can be seen as a basic tool for future developments to improve the way professionals in the different TA work. An established working relationship across organizations, pharmaceutical companies or external service providers, will help to support implementation of TA-specific standards from mapping raw data in SDTM, data analysis using ADaM and finally data presentation in standardized outputs. The PhUSE Wiki can be the central place to share important updates such as new CDISC TA standards or the availability of new TA regulatory guidance. On the other hand we see the Wiki as a place to discuss, to stimulate and inspire new initiatives among the “SAS-Programming Community”, be it Statisticians, Programmers, Data Managers or everyone else involved; this may include specific TA working related white papers and/or scripts being part of the FDA Working Groups WG5 “Development of Standard Scripts for Analysis and Programming” Project 08 “Create white papers providing recommended display and analysis including Table, List and Figure shells”.
Presented at PhUSE/FDA CSS 2014 in Silver Spring (US)
Interpreting CDISC ADaM IG through Users InterpretationAngelo Tinazzi
This document summarizes a presentation given at the PhUSE 2013 conference titled "Interpreting CDISC ADaM IG through Users Interpretation". The presentation aimed to systematically review publications on the CDISC Analysis Data Model (ADaM) standard in order to evaluate how different organizations have implemented and interpreted ADaM. Over 100 presentations focused on ADaM implementation were identified from conferences like PharmaSUG and PhUSE. Key topics of discussion included how to map non-standard clinical domains to ADaM, determining how "analysis-ready" datasets should be, and handling listings/derived values not supported in ADaM. The presentation provided qualitative summaries of user interpretations and applications of various ADaM guidelines and
This document summarizes key efficacy endpoints used in oncology clinical trials, including for solid tumors and non-solid tumors like acute myeloid leukemia. For solid tumors, the best overall response (BOR) is assessed using RECIST criteria to evaluate tumor shrinkage or progression based on target and non-target lesion measurements. Key time-to-event endpoints discussed include overall survival (OS), progression-free survival (PFS), and time to progression (TTP). For acute myeloid leukemia, response is assessed based on blood counts and bone marrow blast percentage according to International Working Group criteria, with endpoints like complete remission rate and event-free survival. Surrogate endpoints are also discussed.
A gentle introduction to survival analysisAngelo Tinazzi
This document provides an introduction to survival analysis techniques for statistical programmers. It discusses key concepts in survival analysis including censoring, the Kaplan-Meier method for estimating survival probabilities, and assumptions of survival models. Programming aspects like creating time-to-event datasets and using SAS procedures for survival analysis are also covered.
This document provides an overview of meta-analysis and summarizes its key aspects and statistical methods. It discusses how meta-analysis can combine results from multiple studies to obtain a single estimate of treatment effect. It also summarizes the steps involved in planning and conducting a meta-analysis, including defining the question, inclusion criteria, searching strategies, and statistical methods for analyzing different types of outcomes. Finally, it reviews several software options available for performing meta-analyses.
CLINICAL STUDY REPORT - IN-TEXT TABLES, TABLES FIGURES AND GRAPHS, PATIENT AN...Angelo Tinazzi
This document discusses technical requirements and solutions for producing statistical outputs for clinical study reports according to ICH E3 guidelines. It provides an overview of key points in ICH E3 related to in-text tables, post-text tables and figures, narratives, and patient data listings. It also discusses considerations for formatting outputs, including paper size and style guidelines. Potential solutions for automating output generation using SAS are presented.
The Implementation of ICH Development Safety Update Report (DSUR) GuidanceAngelo Tinazzi
The document summarizes a presentation about implementing the ICH Development Safety Update Report (DSUR) guidance. It discusses:
- The background and overview of the new ICH DSUR guidance, including its history and key points.
- Implications for study sponsors, including streamlining multiple reports into a single DSUR and establishing standardized processes.
- Components of DSUR implementation, such as developing a clinical trials inventory, extracting and managing data, and mapping data for analysis outputs.
The document discusses standards for clinical trial data submission, including:
- Regulatory agencies are pushing for standardized data submission to improve review efficiency under PDUFA. CDISC standards including SDTM and ADaM are becoming required.
- CDISC regularly updates and expands clinical data standards to include new domains and enhancements.
- A properly constructed define.xml file documenting the metadata is important for reviewers to understand the data.
This document provides an overview of oncology and cancer clinical trials from a data standards and programming perspective. It begins with basic cancer definitions and epidemiology. Key aspects of clinical trials in oncology are then discussed, including complex efficacy endpoints, safety evaluations, and exposure assessments. Standardization efforts through CDISC are summarized, including SDTM and ADaM domains for oncology. Regulatory guidelines from the FDA and EMA are also covered. Throughout, challenges specific to oncology trials from a data and programming standpoint are highlighted. The aim of the PhUSE oncology wiki is also introduced as a resource for further information.
The application of STDM in a no-profit and disease specific organisation - CD...Angelo Tinazzi
This document summarizes a presentation given by Angelo Tinazzi at the 2008 CDISC Italian-Speaking User Group Meeting in Milan. The presentation discusses the application of SDTM at SENDO Tech, a non-profit clinical research organization focused on oncology drug development. SENDO implements SDTM using a hybrid approach, applying some SDTM standards within their clinical database and performing additional transformations and mappings in SAS post-processing to fully comply with SDTM. The presentation outlines SENDO's SDTM implementation challenges due to the heterogeneity of oncology data and describes their methods for mapping clinical data to SDTM domains and variables. It also provides examples of new domains SENDO has
The application of STDM in a no-profit and disease specific organisation - CD...
From Local Laboratory to Standardisation and beyond Applying a common grading system
1. PhUSE
Lisbon, Portugal – 8-10 October 2007
From Local Laboratory to
Standardisation and beyond:
Applying a commong
Early Drug Development
In Oncology grading system
Data Management Stream (DM01)
Angelo Tinazzi
Data Management and Programming Unit
SENDO Tech S.r.l. – Milan (ITALY)
co-authors
Irene Corradino, Enrica Paschetto, Sonia Colombini
2. SENDO (Southern Europe New Drug Organisation)
Non profit Academic Research Organisation (ARO)
Early Drug Development in Oncology
Coordinating a Network of oncology-hospitals
5 phase I (2 in Italy, 3 in Switzerland)
~ 30 phase II (Italy, Switzerland, Spain)
Pre-clinical Laboratory (PK, PD)
Head Quarter based in Milan
Clinical Development
Clinical Operations
Data-Management
Biostatistics
Medical Writing
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 2
3. SENDO (Southern Europe New Drug Organisation) - Partners
HQ-Activities
Clinical development
Clinical Operation
Data Center
Regulatory
INT, Milano
Monitoring
IOSI, Bellinzona, Luca Gianni
Logistic
Cristiana Sessa
Core activities
Trial design
Selected Screening & MoA
Clinical trials
Pharmacokinetics
Pharmacodynamics
and also ....
CHUV Lausanne, KSSG S Gallen,
Istituto Mario Negri Milano
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 3
4. Previous Discuss at PhUSE about Lab Data Management
Szilagyi B, Binder C.
Complex Laboratory Data Management, Strategies
and Tools for a Way out of the Maze.
PhUSE 2005; DM05
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 4
5. Laboratory Data
Use of Laboratory Tests in Clinical Trials
For safety
Activity
Categories
Pharmacodynamic
Pharmacokynetic
Microbiology
Immunology
Cytology
Pharmacogenomic
They are also used to make immediate clinical
decision for patient’s care and to define the drug
profile….focus on
Haematology
Chemistry
Urinalysis
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 5
6. Laboratory Data Characteristics
Qualitative vs Quantitative vs Semi-Quantitative
Quantitative
Most chemistry/hematology
They are expressed in a specific unit
They refere to a range (minimum-maximum)
Semi-Quantitative (i.e. trace)
Qualitative (i.e. +/-)
Clinical Interpretation
Not clinically significant
Clinically Significant (Adverse Event)
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 6
7. Laboratory Data Characteristics
Focus on hematology data
White Blood Count (WBC)
Hemoglobin
Neutrophils
Monocytes
Basophils
Eosinophils
Band
Lymphocytes
Platelets
Red Blood Cells
Hematocrit
Hemoglobin
Coagulation tests (i.e. PTT, PT)
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 7
8. Laboratory Data Characteristics
Focus on chemistry data
Electrolytes
Sodium
Potassium
Chloride
Bicarbonate
Carbon Dioxide
They maintain body fluid and blood pressure
essential for the function of most body systems
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 8
9. Laboratory Data Characteristics
Focus on chemistry data
Enzymes
Aspartate Aminotransferase (AST/SGOT)
Alanine Aminotransferase (ALT/SGPT)
Gamma Glutamyl Transferase (γGT)
Alkaline Phosphatase
Troponin I
Creatine Phosphokinase (CPK)
Lactate Dehydrogenase (LDH)
They help diagnose liver and heart diseases
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 9
10. Laboratory Data Characteristics
Focus on urinalysis data
Protein
Cells
Hormone
They tests the health of organ and body process
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 10
11. Laboratory Data Characteristics
Normal Ranges
Normal ranges, or reference ranges, are used to determine if a
person’s value is “normal”. The ‘normal range’ for a given
constituent of clinical interest is considered to be the
concentrations of the constituent which are found in the body
fluid or excretions of a group of clinically normal persons.
by gender
by age
fasting / non-fasting
analysis method / kit used by laboratory may change over
time, and so the normal ranges
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 11
12. Local Laboratory vs Central Laboratories
Central Labs
Lab samples are analysed (and taken) in the same lab center
Standard methods (and machine calibration)
Unique normal ranges for each sample
Electronic data transfer (no data-transcription errors)
Local Labs
Lab samples are analysed (and taken) in different lab centers
Sample can be taken anywhere / anytime
Multiple normal ranges, so different methods applied
No transport issue, but data need to re-keyed
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 12
13. Local laboratory data-management
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 13
14. Laboratory CRF – Option 1
Normal Ranges and Unit
Collected directly onto patient CRF
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 14
15. Laboratory CRF – Option 2
120 0 100 0 350 0
120 109/L (100-350)
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 15
16. The SENDO Experience with 4 Trials
Study Nr Nr. of Nr. of samples Nr. of Average Nr. of
Patients collected Different Local Labs Used
(average nr by Local by each
Patient, min-max) Labs Used patient (min-max)
1 (ph I) 34 37 (2-90) 44 3.0 (1-7)
2 (ph I) 32 27 (2-79) 33 2.7 (1-6)
3 (ph I) 20 24 (2-61) 31 2.8 (1-5)
4 (ph II) 38 14 (2-33) 11 1.4 (1-3)
Overall in 97
SENDO
Repository
High heterogeneity in
unit reported (the example
is for Platelets count only)
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 16
17. Quality Control
Missing data (unit, range, interpretation)
Hand writing legibility
Unit and value incosistencies
Normal Range Validity
Outliers detection
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 17
18. Statistical Analysis Process
Main Analysis
Univariate (mean, std, min, max, etc)
Shift Tables / Change From Baseline (absolute, %, log)
Correlations
Time to Event (i.e. Time to lowest observation, or time to nadir)
Worst toxic effect observed
Data must be ‘manipulated’ so that results obtained
from different labs can be summed, weighted and
compared
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 18
19. Statistical Analysis Process
Standardisation
Ensures that all laboratory values are expressed in the
same unit (Système International d’Unités - SI)
It consists in the adoption of a standard unit by applying
conversion factors
Multiply 0.2558
Potassium 13.7 mg/dL SI unit is mmol/L 3.5 mmol/L
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 19
20. Statistical Analysis Process
Normalisation 5-29
12-35
The application of a normalisation method to ensure 10-45
homogenization of results obtained from different local 15-40
labs 12-30
U s − Ls Reference/Standard Range
s = Ls + ( x − L x ) 5
U x − Lx Local Labs Range 10
Observed Value 12
12
The standard reference can be taken from the literature or from a sample of
15
normal ranges by taking the 10°and the 90°percentiles 29
30
Assume an observed value of 10 measured in the lab with normal range 5-25, if
35
our standard range has been determined to be 10-35…..
40
35 − 10 45
s = 10 + (10 − 5) = 16.25 The normalised value
−5
25 Normalization. Karvanen J. DIA, Vol. 37, pp. 101-107; 2004
The statistical basis of Laboratory
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 20
21. Applying a common grading system: the CTCAE
NCI Common Terminology Criteria for Adverse
Events – CTCAE (v3.0)
A standard in oncology for classifying Adverse Events
Severity
A Grading system ranging from ‘0’ (no toxic effect) to ‘4’
(severe toxic effect), with the addition of ‘5’ (death)
An event has unique representation
Events are organised in categories
Link with MedDRA Term
CTCAE Event Categories
Allergy / Immunology Gastrointestinal Ocular / Visual
Auditor / Ear Growth and Development Pain
Blood / Bone Marrow Hemorrhage / Bleeding Pulmonary / Upper Respiratory
Cardiac Arrhythmia Hepatobiliary / Pancreas Renal / Genitourinary
Coagulation Infection Secondary Malignancy
Constitutional Symptoms Lymphatic Sexual / Reproductive Function
Death Metabolic / Laboratory Surgery / Intra-Operative Injury
Dermatology / Skin Musculoskeletal / Soft Tissue Syndromes
Endocrine Neurology Vascular
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 21
22. Applying a common grading system: the CTCAE
Qualitative Definition (Arthritis)
a grade 1 is defined as “Mild pain with inflammation, erythema, or
joint swelling, but not interfering with function”
a grade 4, is defined as “Disabling”
Quantitative Definition (Diarrhea)
a grade 1, is defined as “Increase of <4 stools per day over baseline;
mild increase in ostomy output compared to baseline”
a grade 4, is defined….
Quantitative Definition based on Lab Data Results (Platelets Count)
Grade 1 Grade 2 Grade 3 Grade 4
<LLN – 75,000/mm3 <75,000 – 50,000/mm3 <50,000 – 25,000/mm3 <25,000/mm3
<LLN – 75.0 x 10^9 /L <75.0 – 50.0 x 10^9 /L <50.0 – 25.0 x 10^9 /L <25.0 x 10^9 /L
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 22
23. Example of Platelets Count from Local Lab to CTCAE Calculation
CTCAE Platelets Definition
Grade 1 Grade 2 Grade 3 Grade 4
<LLN – 75,000/mm3 <75,000 – 50,000/mm3 <50,000 – 25,000/mm3 <25,000/mm3
<LLN – 75.0 x 10^9 /L <75.0 – 50.0 x 10^9 /L <50.0 – 25.0 x 10^9 /L <25.0 x 10^9 /L
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 23
24. Quality Control – Additional tips in identifying Invalid Values
For each parameter, sort the converted SI
value in ascending order
Review the lowest and highest values when
are different from the expected/normal values
by a factor of 10,100,1000
Look for jumps in values
Look for values that are substantially above or
below typical normal ranges values
Review grade 3-4 CTCAE
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 24
25. Conclusions – 1
In many studies, laboratory data represent
50-80% of the data to be collected
Central laboratory are not always applicable,
however electronical data-transfer from main
individual laboratory used may help
Tools (i.e. SAS macro routines), are required to
manage and control the various steps of Local
Laboratory Data collection and analysis
Specialist in laboratory data-management
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 25
26. Conclusions – 2
Estabilish a central data-repository of local labs
How small are the differences / abnormalities
that need to be defined?
Choice between a more or less sophisticated
method of harmonization of laboratory results (e.g.
Normalization vs SI Standardization)
CDISC LAB Team
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 26
27. Questions
Tinazzi A Corradino I Paschetto E Colombini S: From Local Laboratory to Standardisation and
beyond: Applying a common grading system PhUSE 2007, DM01 (Lisbon, Portugal – 8-10 October 2007) 27