Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Agent-based simulations of sequential and combinatorial treatment for osteoporosis
Adapt a model of treatment of osteoporosis with denosumab, PTH, bisphosphonates or romosozumab to explore sequential and branching simulations and identify optimal sequential and combinatorial drug treatments
Keywords: agent-based, multiphysics, in silico, bone, osteoporosis, clinical trials, drug treatment, sequential, optimization
Approximately 10 million osteoporotic fragility fractures occur every year and lead to over 400,000 deaths and health care costs in excess of US$80 billion. We at the laboratory for bone biomechanics at the ETH Zurich have developed a computational model to predict how the bone microstructure and strength will change as a result of bone diseases and their treatments.
The aim of this project is to apply in silico modelling to analyze the effects of sequential or combinatorial treatment with the most widespread therapeutics for osteoporosis on bone. Specifically the project will apply a micro-multiphysics agent-based (micro-MPA) model which was already used successfully to simulate the effect of treatment with denosumab, bisphosphonates, PTH and romosozumab on iliac crest biopsies from a group of post-menopausal women (age: 72±5 years, scanned with micro-CT at a resolution of 10.5 microns, article: https://doi.org/10.1002/jbm4.10494 + conferences). In this project we intend to run digital twin simulations of the effect of sequential and combinatorial treatment with these 4 drugs using the same baseline images.
Both bisphosphonates and denosumab are anti-resorptives, meaning their mechanism of action most immediately leads to a decrease in bone resorption by osteoclasts. Due to osteoclast-osteoblast coupling bone formation also decreases and the net effect of denosumab and bisphosphonates is to decrease rates of bone remodelling by 50 to 90% with a net shift in the formation resorption balance in favour of formation.Both bisphosphonates and denosumab are anti-resorptive medications but they differ in their long-term effect on bone mineral density (BMD). Denosumab treatment results in a steady rise in BMD over a period of 10 years while with bisphosphonate treatment BMD only increases over a period of approximately 3 years then remains relatively constant. Bisphosphonates bind to bone mineral with high affinity, inhibit osteoclastic resorption and promote osteoclast apoptosis. Denosumab is an antibody to RANKL, a cytokine which promotes differentiation of osteoclast precursors to osteoclasts. Binding of denosumab to RANKL thus results in decreased osteoclastogenesis.
By contrast romosozumab and PTH are not anti-resorptives. Romosozumab has a mixed effect, it is the only drug that both decreases resorption and increases formation at least for a limited time period following injection. Romosozumab does this by binding sclerostin, a cytokine which reduces osteoblast formation from its precursors and from lining cells. PTH injections primarily increase formation by reducing osteoblast apoptosis.
In the first phase of this project, the changes in densitometric, static and morphometric parameters in the simulation output will be compared to literature data from large randomized controlled sequential and combinatorial clinical trials. Combinatorial clinical trials of interest include bisphosphonates and PTH (10.1056/NEJMoa031975), denosumab and PTH (NCT01750086, DATA-HD). Sequential clinical trials of interest include bisphosphonates after PTH (PaTH, PEAK, ACTIVExtend), PTH after bisphosphonates (STRUCTURE, OPTAMISE), denosumab after romosozumab (FRAME), romosozumab after denosumab (NCT00896532), denosumab after bisphosphonates (NCT00919711). The aim of this phase of the project will be to be as exhaustive as possible in our comparison of simulation outputs with literature and if necessary to design optimizer loops to adjust the ~150 degrees of freedom of the micro-MPA model to obtain a physiological model whose output matches the bone mineral density, bone turnover markers and morphometric parameters reported for the sequential and combinatorial treatments in the literature.
In the second phase of this project, the adapted micro-MPA model designed in the first step will be used to test new sequences and timelines and new drug combinations that have yet to be explored in clinical trials and that may be of interest to clinical trial designers. We will build a library of simulation outputs that would have taken years of experimentation and millions of dollars in investments to generate clinically.
Approximately 10 million osteoporotic fragility fractures occur every year and lead to over 400,000 deaths and health care costs in excess of US$80 billion. We at the laboratory for bone biomechanics at the ETH Zurich have developed a computational model to predict how the bone microstructure and strength will change as a result of bone diseases and their treatments.
The aim of this project is to apply in silico modelling to analyze the effects of sequential or combinatorial treatment with the most widespread therapeutics for osteoporosis on bone. Specifically the project will apply a micro-multiphysics agent-based (micro-MPA) model which was already used successfully to simulate the effect of treatment with denosumab, bisphosphonates, PTH and romosozumab on iliac crest biopsies from a group of post-menopausal women (age: 72±5 years, scanned with micro-CT at a resolution of 10.5 microns, article: https://doi.org/10.1002/jbm4.10494 + conferences). In this project we intend to run digital twin simulations of the effect of sequential and combinatorial treatment with these 4 drugs using the same baseline images.
Both bisphosphonates and denosumab are anti-resorptives, meaning their mechanism of action most immediately leads to a decrease in bone resorption by osteoclasts. Due to osteoclast-osteoblast coupling bone formation also decreases and the net effect of denosumab and bisphosphonates is to decrease rates of bone remodelling by 50 to 90% with a net shift in the formation resorption balance in favour of formation.Both bisphosphonates and denosumab are anti-resorptive medications but they differ in their long-term effect on bone mineral density (BMD). Denosumab treatment results in a steady rise in BMD over a period of 10 years while with bisphosphonate treatment BMD only increases over a period of approximately 3 years then remains relatively constant. Bisphosphonates bind to bone mineral with high affinity, inhibit osteoclastic resorption and promote osteoclast apoptosis. Denosumab is an antibody to RANKL, a cytokine which promotes differentiation of osteoclast precursors to osteoclasts. Binding of denosumab to RANKL thus results in decreased osteoclastogenesis.
By contrast romosozumab and PTH are not anti-resorptives. Romosozumab has a mixed effect, it is the only drug that both decreases resorption and increases formation at least for a limited time period following injection. Romosozumab does this by binding sclerostin, a cytokine which reduces osteoblast formation from its precursors and from lining cells. PTH injections primarily increase formation by reducing osteoblast apoptosis.
In the first phase of this project, the changes in densitometric, static and morphometric parameters in the simulation output will be compared to literature data from large randomized controlled sequential and combinatorial clinical trials. Combinatorial clinical trials of interest include bisphosphonates and PTH (10.1056/NEJMoa031975), denosumab and PTH (NCT01750086, DATA-HD). Sequential clinical trials of interest include bisphosphonates after PTH (PaTH, PEAK, ACTIVExtend), PTH after bisphosphonates (STRUCTURE, OPTAMISE), denosumab after romosozumab (FRAME), romosozumab after denosumab (NCT00896532), denosumab after bisphosphonates (NCT00919711). The aim of this phase of the project will be to be as exhaustive as possible in our comparison of simulation outputs with literature and if necessary to design optimizer loops to adjust the ~150 degrees of freedom of the micro-MPA model to obtain a physiological model whose output matches the bone mineral density, bone turnover markers and morphometric parameters reported for the sequential and combinatorial treatments in the literature.
In the second phase of this project, the adapted micro-MPA model designed in the first step will be used to test new sequences and timelines and new drug combinations that have yet to be explored in clinical trials and that may be of interest to clinical trial designers. We will build a library of simulation outputs that would have taken years of experimentation and millions of dollars in investments to generate clinically.
Review the literature for clinical studies reporting changes in densitometric, static morphometric and dynamic morphometric parameters of bone during sequential and/or combinatorial treatments.
Critically evaluate potential candidates for validation of the simulation output of the micro-MPA model
For the chosen set of combinatorial and sequential treatments run micro-MPA based simulations on the Swiss National Supercomputer and critically compare the simulation outputs with the bone mineral density at the total hip and bone turnover markers reported in the selected clinical trials. Adjust the ~150 degrees of freedom of the micro-MPA model to design an optimized physiological model whose output matches the bone mineral density, bone turnover markers, trends in cell behaviour and cytokine concentrations and morphometric parameters reported for the sequential and combinatorial treatments in the literature.
Identify new sequences and timelines and new drug combinations of interest that have yet to be explored in clinical trials. Build a library and database of simulation outputs for these and critically analyze the results to highlight potentially effective therapy sequences and combinations.
Prepare a presentation for the entire laboratory summarizing your results (30 minutes) and a report for your master thesis/semester project/bachelor thesis.
Review the literature for clinical studies reporting changes in densitometric, static morphometric and dynamic morphometric parameters of bone during sequential and/or combinatorial treatments. Critically evaluate potential candidates for validation of the simulation output of the micro-MPA model
For the chosen set of combinatorial and sequential treatments run micro-MPA based simulations on the Swiss National Supercomputer and critically compare the simulation outputs with the bone mineral density at the total hip and bone turnover markers reported in the selected clinical trials. Adjust the ~150 degrees of freedom of the micro-MPA model to design an optimized physiological model whose output matches the bone mineral density, bone turnover markers, trends in cell behaviour and cytokine concentrations and morphometric parameters reported for the sequential and combinatorial treatments in the literature.
Identify new sequences and timelines and new drug combinations of interest that have yet to be explored in clinical trials. Build a library and database of simulation outputs for these and critically analyze the results to highlight potentially effective therapy sequences and combinations.
Prepare a presentation for the entire laboratory summarizing your results (30 minutes) and a report for your master thesis/semester project/bachelor thesis.
Email: charles.ledoux@hest.ethz.ch
Website: https://www.bone.ethz.ch/research/clin-mech/multiphysics.html
Charles Ledoux PhD Student
ETH Zürich, HCP H 18.2
Leopold-Ruzicka-Weg 4 CH - 8093 Zürich