Bayesian Methods for Joint Longitudinal and Survival Data
Home > Our Services >
Confidence Beyond Compliance
PharmaLex is recognized as a global leader in the application of Bayesian methodology consulting and programming in the Biopharma industry, from pre-clinical to late phase studies and in a wide range of therapeutic areas.
Bayesian thinking is at the heart of our approach. While our team is comfortable handling classical or frequentist statistics as well as advanced Bayesian statistics, our problem-solving strategy always puts new problems into a Bayesian perspective. This provides our customers with a crystal-clear understanding of the solutions proposed, its impact and the long term opportunities.
Providing both statistical training and consulting services to pharmaceutical and biopharmaceutical companies, our goal is the same as the companies we support – to efficiently and consistently develop and deliver quality medicine to patients.
PharmaLex has significant expertise in supporting discovery laboratories in pharmaceutical sciences with great emphasis on in-vitro and in-vivo animal models to ensure the studies are adequately reproducible and predictive to support decision making. PharmaLex applies a lifecycle vision and Quality by Design (QbD) approach for in-vitro and in-vivo assays that starts by the definition of the objectives and aims and ends with a control strategy. In between, best practice of Design of Experiments (DoE), advanced modeling and Bayesian statistics are applied to ensure reproducibility and predictability of results.
Our services include:
Statistics for the development and qualification of bioassays
- HTS, LTS
- Design of experiments
- Advanced signal processing
- Optimization and qualification
- Parallelism assessment and determination
- Multicriteria decision making and ranking of compounds
- Z score, Uncertainty
- Reproducibility evaluation
- Transfer and comparisons
- Control strategies
Statistics for the development of in-vivo pharmacology models
- Design and optimization
- Powering for reproducibility, Z score
- Development of accurate end-points
- Modeling to support translation to human
- PK/PD, non-linear and Bayesian models
- 3R methodologies and comparison
- Historical control groups
- Control strategies
Development of biomarkers
- Identification of biomarkers
- Control of FDR (False Discovery rates)
- Evaluation of performances (Sensitivity, Specificity, ROC, NPV, PPV)
- Development and validation of kits
- Optimization and design of clinical trial to qualify biomarkers
- Multiplex
Therapeutic areas
- Immunology
- Oncology (Xenograft models, …)
- Metabolic disorders
- CNS
- Cardiology
PharmaLex is specialized in the development and analyses of innovative clinical trials to maximize the probability of success and the outcome. We use a blend of expertise ranging from Bayesian statistics, pharmacometry and scientific programming to develop by means of intensive simulations cost-effective clinical trials, including adaptive designs and analyses methods. We also produced the highest quality reports (TFLs) and contribute to the study reports and publications.
Our services include:
Protocol and SAP development
- Methodological support and contribution
- Evaluation and optimization of clinical end-points
- Power and optimization using Clinical trial simulations
- Development of wide range of Adaptive designs
- Advanced Bayesian statistics
- Meta-analyses and MAP approaches for prior elicitation
Contribution to CSR and submission
- Writing and review of CSR
- Participation to meetings with regulatory bodies
Pediatric development, rare and orphan disease
- Specialized in rare disease
- Use of historical data
- Development of optimized designs with limited data
Biomarker qualification
- Expertise in bioassays
- Advanced simulations for design optimization
Analyses of complex data and signals
- Modeling of PET-scan
- Modeling of EEG and ECG data
Pharmacovigilance and recruitment
- Signal detection
- Bayesian models for efficient Signal detection methodologies
- Modeling and prediction of recruitment of patients
Programming
- SAS Programming, CDISC/SDTM/ADAM
- Production of TFLs for CSR
- Including NCA PK reporting
JMP-Clinical
- Training for use and development using JMP-Clinical
- Applied Risk Based Monitoring using JMP-Clinical
Bayesian thinking is at the heart of our approach
While our team is comfortable handling classical or frequentist statistics as well as advanced Bayesian statistics, our problem solving strategy always puts new problems into a Bayesian perspective. This provides our customers with a crystal clear understanding of the solutions proposed, its impact and the long term opportunities.
PharmaLex has applied and successfully implemented Bayesian statistics to a wide range of topics in discovery, pre-clinical studies, clinical trials, manufacturing and pharmacovigilance. Our scientists have also trained hundreds of statisticians in Bayesian statistics using several programming languages such as SAS Proc MCMC, STAN, JAGS, MCMCPack, Bugs. Some of our statisticians are also the founders of the conference dedicated to Applied Bayesian statistics in Pharmaceutical Sciences (www.bayes-pharma.org).
Fields of applications of Bayesian statistics
- Bioassays optimization and validation
- Analytical procedure optimization and validation
- Design space for processes, formulation and assays
- CPV or Process validation
- Stability modeling, shelf-life, IRL
- Definition of specifications
- Predictive capability of processes
- SPC and control strategies
- Analytical biosimilarity
- Adaptive design in clinical trials
- PK/PD models and Bayesian ODE
- Prediction of Clinical trials
- Biomarker validation and qualification
- Sampling in continuous manufacturing
- Evaluation of cosmetic products
- Transfer of assays
- Comparison of analytical procedures
- Dissolution testing
- Content Uniformity assessment
- Reproducibility of preclinical model
- Elicitation of priors for early clinical trials
Languages in Bayesian statistics
- SAS Proc MCMC
- STAN
- JAGS
- BUGS
- MCMCPack
Training in Bayesian statistics
- Theory
- Modeling
- Sampler
- Languages
- Diagnostics
- Application
- Advanced topics
PharmaLex has a recognized and innovative expertise in the development, validation and control of analytical procedures and bioassays. We have been a pioneer in applying the Quality by Design (QbD) and lifecycle thinking to analytical procedures and bioassays with a vision: achieving confidence and compliance. Our team is composed of chemists, biologists and pharmacists that are able to provide the most advanced statistical support to ensure your analytical procedures and bioassays are reliable and compliant. Our scientific and quality team can also provide you with the automated analyses tools to maximize your productivity in accordance with regulations. Finally we also have a tremendous expertise in PAT procedure development and validation.
Our services include:
Bioassays
- Expertise in a wide range of bioassays (ELISA, qPCR, FACS, MNA, …)
- Development by means of DoE
- Optimization for precision and range
- Strategies and methods for parallelism evaluation
- Advanced signal processing, multiplex assays
- Non-linear modeling
- Format optimization and justification
- Evaluation of capability and uncertainty
- Control strategy and transfer
Analytical procedures
- Modeling of wide range of procedures (HPLC, MS, NIR, RAMAN, ….)
- Development by means of DoE
- Optimization of procedures by means of advanced modeling
- Advanced signal processing (UV, NIR, MIR, RAMAN)
- Validation of PAT procedures NIR and RAMAN based
- Format optimization and justification
- Evaluation of capability and uncertainty
- Control strategy
PharmaLex has many years experience in proposing robust and innovative statistical solutions for the development, validation and control of processes, in particular bio-processes for mABs, vaccines and cell cultures. Combined with expertise in analytics, PharmaLex ensures that the robustness of (bio) processes is guaranteed.
Our services include:
Lifecycle and Quality by Design methodology
- Strategies for stage 1, stage 2 and stage 3
- Advanced Design of Experiments (DoE)
- Computation of multi criteria Design Space
- CPV
- Robustness metrics and predictive capability
- Statistical Process control
- Strategies and report of transfer
- Optimization and modeling of scale-up scale-down
- Definition of specifications
- Development and implementation of PAT methodologies
Continuous manufacturing
- Setup and justification of sampling plan
- Development and implementation of PAT
- Methodologies for release testing
- Assessment of content uniformity
Stability
- Design of optimized stability studies
- Computation of shelf-life
- Internal Release Limits for linear and non-linear degradation
- Accelerated modeling and prediction
Formulation
- Specialized Design such as mixtures design
- Advanced Modeling for optimization
- Combined prediction with stability assessment
PharmaLex uses state-of-the-art Artificial Intelligence (AI) and its long experience in data science to solve our specific customer’s problems, as an addition to our strategical statistical services. Our team, composed of experts from various scientific fields such as mathematics, Bayesian statistics, chemistry, bioengineering, etc., is working in close collaboration with our customer to meet their needs. At PharmaLex, we believe a multi-disciplinary team is the best way to make effective data science and support effective decision making in the pharmaceutical world.
Our healthcare expertise combined with our multidisciplinary team are used to deeply understand our customer’s very questions and envisage top-notch AI solution wherever needed. For instance, it is common to apply AI and machine learning tools on big and unstructured data contained in data lakes and other LIMS.
The PharmaLex team masters most advanced programming languages that are used in statistics and Data Science such as SAS, R, STAN, JAGS, Python, Java, C++, but also modern operating environments such as Hadoop or Tensorflow (Keras).
Our key value-added deliverable has always been in delivering fit-for-purpose modeling on which solutions are build. To ensure compliance, our statistical experts are working closely with our IT and Quality departments. It allows for example the creation of AI-powered software, that is completely compliant with regulatory environment (GAMP5, 21CFR part 11). This accumulated sum of competencies makes us one of the few being able to develop solutions in full accordance with all the (Bio-)Pharmaceutical regulation (ICH-USP-EMA). In addition, the software validation ensures that the right diagnostic or answer is provided with high confidence. For validating AI solutions, specific trials are developed to ensure fair comparison with human gold standard or other algorithmic solutions. For instance, our clinical experts can evaluate the impact of modeling/prediction uncertainty directly on the outcome of the clinical study.
As most data scientists, we started applying machine learning methodologies in applied pharmaceutical development (process, assay, etc.) using classical tools such as SVM, Random Forest, Lasso, PLS, etc. We developed predictive methodologies to evaluate model quality as the classical metrics may fail at providing the right conclusion. We stay upfront of the technology by using most recent algorithms such as Boosted Trees or by handling uncertainty in Neural Network (similar to what we do in Bayesian statistics). Computational power for handling big data in short time frame is also available.
Examples of applications
- Survival time predictions using big cytometry database: comparison of the performance of various AI algorithms
- Biomarkers detection on large tox datasets
- Root cause analysis based on variable importance metrices from gradient boosting model
- Detection of epileptic seizures
- Automatic quantification of substances from NMR signals or full DAD-chromatogram (even unresolved !)
- Bioprocess control based on RAMAN or NIR spectra
- Kinetic and Dynamic models based on EEG signals
- Chemical process control using NIR spectra
- Statistical deconvolution from cEIF and UV data
- Fraud detection for counterfeit drugs (but also for bank transactions…)
PharmaLex provides a broad array of clinical data management services to support this critical phase in clinical research. We help our clients generate high-quality, accurate, and statistically sound data from clinical trials. Our advanced technologies are combined with innovative solutions to reduce complexities and accelerate the drug development to marketing.
Our experts have proven experience in the management of eCRF, ePRO, IWRS, testing & validation covering all therapeutic areas. The end-to-end solution we offer can be adapted to the needs of our clients in order to create true value.
Our services include:
Study design
- Input in CRF design
- Creation of annotated CRF
- Data Management Plan (DMP)
- Data Cleaning Plan (DCP) & Programming
- Data Validation Plan (DVP)
- E-Platform setup (incl. eCRF/ePRO/IWRS), testing & validation
- Data exchange with external data sources and systems (in/out)
- ECRF completion guidelines
Study conduct
- E-Platform and database hosting
- Handling reference ranges
- Data integration
- Data Cleaning Process
- (Risk-based) Data Monitoring through vertical and longitudinal analysis
- Data exports
Study close-out
- Data reconciliation across multiple sources and systems
- Database lock, transfer & archiving
- Data Management Report
Preparing and generating FDA-compliant reports for Method Validation, Method Transfer and Product Stability can be a time-consuming and difficult exercise.
PharmaLex statistical software is a complete solution for a more streamlined and productive laboratory.
The benefits:
- User friendly: generate an FDA/EMA compliant report in 10 minutes instead of 10 days
- Complies to global regulatory standards, including ICH and USP
- Focused on decision making: one graph, one decision, understandable by non-statisticians
- Suitable for GxP use: GAMP5 validated and 21 CRF Part 11 compliant
- Go beyond the possibilities of Excel (e.g. REML, quantile computation, etc.) to always have exact statistical results
Related Resources
Related Services
What our clients say
There was a close relationship between the regulatory strategy and the writing activities. You developed a fine-tuned regulatory strategy which will be a key factor to smooth the approval of our dossier.
Large Pharma, Global
VP Regulatory Strategy
VP Regulatory Strategy
In these unprecedented times, PharmaLex continues to support all clients by meeting virtually and providing experienced, remote-based consultants.
Thomas Dobmeyer, CEO PharmaLex
CEO PharmaLex
CEO PharmaLex