Mathematical Modelling of Myeloproliferative Neoplasms and Hematopoietic Stem Cells

Rasmus Kristoffer Pedersen

Mathematical Modelling of Myeloproliferative Neoplasms and Hematopoietic Stem Cells

Bidragets oversatte titel: Matematisk Modellering af Myeloproliferative Sygdomme og Hematopoietiske Stamceller

Publikation: Bog/antologi/afhandling/rapport › Ph.d.-afhandling

Abstract

De Philadelphia-negative myeloproliferative neoplasier (MPN) dækker over en gruppe af blodcancere der inkluderer sygdommene essentiel trombocytose (ET), polycytæmi vera (PV) og primær myelofibrose (PMF). Forskning tyder på at ET, PV og PMF er tæt forbundet og repræsenterer forskellige stadier langs et biologisk kontinuum. I de seneste ˚ar er kronisk inflammation blevet foresl˚aet som afgørende for udviklingen af cancersygdomme generelt og for Philadelphia-negative MPN i særdeleshed. Konceptet om at kronisk inflammation styrer et biologisk kontinuum af sygdomme gør MPN til en oplagt gruppe af sygdomme at undersøge ved hjælp af matematisk modellering, med h˚ab om forbedret patientprognose og videreudvikling af behandling.

I denne afhandling beskrives og analyseres tre mekanisme-baserede matematiske modeller. Alle tre modeller beskrives ved brug af ordinære differentialligninger (ODE). Dette gøres i et forsøg på at opnå indsigt om blodcancere generelt og om en kohorte af patienter fra det kliniske forsøg "DALIAH" specifikt (EudraCT ID: 2011-001919-31). Patienterne der her tages i betragtning, blev alle diagnosticeret med MPN og blev behandlet med pegyleret interferon-α (IFN) over en fem-årig periode.

Den første model blev oprindeligt beskrevet af Andersen et al. (2017) og beskriver forbindelsen mellem kronisk inflammation og udviklingen af MPN. En modeludvidelse gives os muligheden for at relatere patient-specifik timing af IFN-dosis til patient data for leukocyt-tal, thrombocyt-tal samt målinger af JAK2V617F allelbyrde, hvilket viser god overenstemmelse mellem dynamikken af den matematiske model og opførslen der observeres i data.

En ny matematisk model af hematopoitiske stamceller (HSC) præsenteres. Matematisk analyse af modellen leder til et udtryk for HSC fitness, der er afgørende for den lang-sigtede konkurrence mellem HSC-kloner i knoglemarven. Det vurderes at en model reduktion er passende, og HSC fitness viser sig også at være afgørende for opførslen af den reducerede model.

Vi kombinerer de to modeller til en samlet model der beskriver både HSC-mekanismer, blodcelle produktion og effekten af kronisk inflammation på udviklingen af MPN. Denne kombinerede model kan bruges til at undersøge og simulere en bred vifte af scenarier, hvilket giver os mulighed for at komme med nye hypoteser om opførslen af HSC og hele det hematopoietiske system. Dette gør os i stand til at fortolke den effekt IFN-behandling har på MPN-diagnosticerede patienter, ved at relatere modellen til data for individuelle patienter. For de IFN-behandlede patienter fra DALIAH observerede vi en forskel i tidsskalaerne for respons af blodcelletal og JAK2V617F allelbyrde. Baseret på vores model-baserede undersøgelser,
vurderer vi at forskellen på respons skyldes en todelt effekt af IFN: Produktionen af blodceller begrænses på den korte tidsskala for både raske og maligne celler, hvorimod de maligne stamceller bliver særligt begrænset på længere tidsskalaer.

Afslutningsvis præsenterer vi et "proof-of-concept" for hvordan matematisk modellering kalibreret med patient-målinger fra diagnosetidspunkt kan lede til forudsigelser for individuelle patienter. Vi vurderer at matematisk modellering,
kombineret med tilpas indsamling af patientdata og model-kalibration, kan blive et vigtigt værktøj for patient-prognoser i klinikken i den nære fremtid, og vil give anledning til forbedret behandling af MPN ved brug af IFN.

Bidragets oversatte titel	Matematisk Modellering af Myeloproliferative Sygdomme og Hematopoietiske Stamceller
Originalsprog	Engelsk

Udgivelsessted	Roskilde
Forlag	Roskilde Universitet
Antal sider	152
Status	Udgivet - 20 nov. 2020

Navn	IMFUFA-tekst : i, om og med matematik og fysik
Nummer	512
ISSN	0106-6242

Bibliografisk note

Vejledere: Johnny T. Ottesen, Morten Andersen og Hans C. Hasselbalch

Link til dokument

http://milne.ruc.dk/imfufatekster/pdf/512.pdfLicens: CC BY

1 Tidsskriftartikel

Data-driven analysis of JAK2 V617F kinetics during interferon-alpha2 treatment of patients with polycythemia vera and related neoplasms
Pedersen, R. K., Andersen, M., Knudsen, T. A., Sajid, Z., Gudmand-Hoeyer, J., Dam, M. J. B., Skov, V., Kjær, L., Ellervik, C., Larsen, T. S., Hansen, D., Pallisgaard, N., Hasselbalch, H. C. & Ottesen, J. T., 28 jan. 2020, I: Cancer Medicine. 9, 6, s. 2039-2051 13 s.
Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › peer review

Åben adgang
Fil

1 Igangværende

Matematisk Modellering af Myeloproliferative Sygdomme
Pedersen, R. K. (Projektdeltager)
01/09/2017 → …
Projekter: Projekt › Forskning

Citer dette

@phdthesis{184775de8b9247f286b56ad6608e869e,

title = "Mathematical Modelling of Myeloproliferative Neoplasms and Hematopoietic Stem Cells",

abstract = "The Philadelphia-negative myeloproliferative neoplasms (MPNs) is a group of blood cancers which include the diseases essential thrombocythemia (ET), polychytemia vera (PV) and primary myelofibrosis (PMF). Evidence suggests that ET, PV and PMF are closely connected and represent different stages along a biological continuum. In recent years, chronic inflammation has been found to be an important driver of cancers in general and of the Philadelphia-negative MPNs in particular. The concept of inflammation driving a biologically continuum of disease makes MPNs apt for investigations through mathematical modelling, with promise of improved patient prognosis and advancements in treatment.In this thesis, we describe and analyse three mechanism-based mathematical models. All three models are described using systems of ordinary differential equations (ODE). This is done in an effort to gain insight into blood cancers in general and about a cohort of patients enrolled in the clinical trial \DALIAH{"} in particular (EudraCT number: 2011-001919-31). The patients considered here were all diagnosed with MPNs and treated with pegylated interferon-α (IFN) over a five year period. The first model was proposed by Andersen et al. (2017) and describes the connection between chronic inflammation and the progression of MPN. A model extension allows us to relate patientspecific IFN-dose scheduling with patient data of leukocyte-counts, thrombocyte-counts and measurements of the JAK2V617F allele burden, showing good agreement between the dynamics of the mathematical model and the behaviour observed in data.A novel mathematical model of the hematopoietic stem cells (HSC) is proposed. Mathematical analysis of the model suggests a notion of HSC fitness, found to determine long-term competition between HSC clones within the bone-marrow microenvironment. A model reduction is deemed appropriate, and HSC fitness is found to be principal in the behaviour of the reduced model as well.We combine the two models into a single model describing both HSC-mechanisms, blood-cell production and the effect of chronic inflammation on MPN progression. This combined model can be used to investigate and simulate a wide range of scenarios, allowing us to make novel hypotheses about the behaviour of HSC and the entire hematopoietic system. In particular, we are able to interpret the effect IFN-treatment has on MPN-diagnosed patients, by relating the model to data of individual patients. For the IFN-treated MPN patients of the DALIAH trial, we observed a difference in the time-scale of the response of blood-cell counts and of the JAK2V617F allele burden. Based on our model-based investigations, we hypothesize that this difference in response is due to a two-fold effect of IFN: Production of blood-cells is inhibited for both healthy and malignant clones on a short time-scale, while malignant stem cells are substantially inhibited on longer time-scales.Finally, we present a proof-of-concept of how mathematical modelling calibrated to patientmeasurements at diagnosis can make predictions on the level of individual patients. We hypothesize that, with sufficient collection of patient-data and model-calibration, mathematical modelling could be an important prognostic tool in the clinic in the near future, allowing for improved treatment of MPNs using IFN.",

author = "Pedersen, {Rasmus Kristoffer}",

note = "Vejledere: Johnny T. Ottesen, Morten Andersen og Hans C. Hasselbalch",

year = "2020",

month = nov,

day = "20",

language = "English",

series = "IMFUFA-tekst : i, om og med matematik og fysik",

publisher = "Roskilde Universitet",

number = "512",

}

TY - BOOK

T1 - Mathematical Modelling of Myeloproliferative Neoplasms and Hematopoietic Stem Cells

AU - Pedersen, Rasmus Kristoffer

N1 - Vejledere: Johnny T. Ottesen, Morten Andersen og Hans C. Hasselbalch

PY - 2020/11/20

Y1 - 2020/11/20

N2 - The Philadelphia-negative myeloproliferative neoplasms (MPNs) is a group of blood cancers which include the diseases essential thrombocythemia (ET), polychytemia vera (PV) and primary myelofibrosis (PMF). Evidence suggests that ET, PV and PMF are closely connected and represent different stages along a biological continuum. In recent years, chronic inflammation has been found to be an important driver of cancers in general and of the Philadelphia-negative MPNs in particular. The concept of inflammation driving a biologically continuum of disease makes MPNs apt for investigations through mathematical modelling, with promise of improved patient prognosis and advancements in treatment.In this thesis, we describe and analyse three mechanism-based mathematical models. All three models are described using systems of ordinary differential equations (ODE). This is done in an effort to gain insight into blood cancers in general and about a cohort of patients enrolled in the clinical trial \DALIAH" in particular (EudraCT number: 2011-001919-31). The patients considered here were all diagnosed with MPNs and treated with pegylated interferon-α (IFN) over a five year period. The first model was proposed by Andersen et al. (2017) and describes the connection between chronic inflammation and the progression of MPN. A model extension allows us to relate patientspecific IFN-dose scheduling with patient data of leukocyte-counts, thrombocyte-counts and measurements of the JAK2V617F allele burden, showing good agreement between the dynamics of the mathematical model and the behaviour observed in data.A novel mathematical model of the hematopoietic stem cells (HSC) is proposed. Mathematical analysis of the model suggests a notion of HSC fitness, found to determine long-term competition between HSC clones within the bone-marrow microenvironment. A model reduction is deemed appropriate, and HSC fitness is found to be principal in the behaviour of the reduced model as well.We combine the two models into a single model describing both HSC-mechanisms, blood-cell production and the effect of chronic inflammation on MPN progression. This combined model can be used to investigate and simulate a wide range of scenarios, allowing us to make novel hypotheses about the behaviour of HSC and the entire hematopoietic system. In particular, we are able to interpret the effect IFN-treatment has on MPN-diagnosed patients, by relating the model to data of individual patients. For the IFN-treated MPN patients of the DALIAH trial, we observed a difference in the time-scale of the response of blood-cell counts and of the JAK2V617F allele burden. Based on our model-based investigations, we hypothesize that this difference in response is due to a two-fold effect of IFN: Production of blood-cells is inhibited for both healthy and malignant clones on a short time-scale, while malignant stem cells are substantially inhibited on longer time-scales.Finally, we present a proof-of-concept of how mathematical modelling calibrated to patientmeasurements at diagnosis can make predictions on the level of individual patients. We hypothesize that, with sufficient collection of patient-data and model-calibration, mathematical modelling could be an important prognostic tool in the clinic in the near future, allowing for improved treatment of MPNs using IFN.

AB - The Philadelphia-negative myeloproliferative neoplasms (MPNs) is a group of blood cancers which include the diseases essential thrombocythemia (ET), polychytemia vera (PV) and primary myelofibrosis (PMF). Evidence suggests that ET, PV and PMF are closely connected and represent different stages along a biological continuum. In recent years, chronic inflammation has been found to be an important driver of cancers in general and of the Philadelphia-negative MPNs in particular. The concept of inflammation driving a biologically continuum of disease makes MPNs apt for investigations through mathematical modelling, with promise of improved patient prognosis and advancements in treatment.In this thesis, we describe and analyse three mechanism-based mathematical models. All three models are described using systems of ordinary differential equations (ODE). This is done in an effort to gain insight into blood cancers in general and about a cohort of patients enrolled in the clinical trial \DALIAH" in particular (EudraCT number: 2011-001919-31). The patients considered here were all diagnosed with MPNs and treated with pegylated interferon-α (IFN) over a five year period. The first model was proposed by Andersen et al. (2017) and describes the connection between chronic inflammation and the progression of MPN. A model extension allows us to relate patientspecific IFN-dose scheduling with patient data of leukocyte-counts, thrombocyte-counts and measurements of the JAK2V617F allele burden, showing good agreement between the dynamics of the mathematical model and the behaviour observed in data.A novel mathematical model of the hematopoietic stem cells (HSC) is proposed. Mathematical analysis of the model suggests a notion of HSC fitness, found to determine long-term competition between HSC clones within the bone-marrow microenvironment. A model reduction is deemed appropriate, and HSC fitness is found to be principal in the behaviour of the reduced model as well.We combine the two models into a single model describing both HSC-mechanisms, blood-cell production and the effect of chronic inflammation on MPN progression. This combined model can be used to investigate and simulate a wide range of scenarios, allowing us to make novel hypotheses about the behaviour of HSC and the entire hematopoietic system. In particular, we are able to interpret the effect IFN-treatment has on MPN-diagnosed patients, by relating the model to data of individual patients. For the IFN-treated MPN patients of the DALIAH trial, we observed a difference in the time-scale of the response of blood-cell counts and of the JAK2V617F allele burden. Based on our model-based investigations, we hypothesize that this difference in response is due to a two-fold effect of IFN: Production of blood-cells is inhibited for both healthy and malignant clones on a short time-scale, while malignant stem cells are substantially inhibited on longer time-scales.Finally, we present a proof-of-concept of how mathematical modelling calibrated to patientmeasurements at diagnosis can make predictions on the level of individual patients. We hypothesize that, with sufficient collection of patient-data and model-calibration, mathematical modelling could be an important prognostic tool in the clinic in the near future, allowing for improved treatment of MPNs using IFN.

M3 - Ph.D. thesis

T3 - IMFUFA-tekst : i, om og med matematik og fysik

BT - Mathematical Modelling of Myeloproliferative Neoplasms and Hematopoietic Stem Cells

PB - Roskilde Universitet

CY - Roskilde

ER -

Mathematical Modelling of Myeloproliferative Neoplasms and Hematopoietic Stem Cells

Abstract

Bibliografisk note

Link til dokument

Publikation

Data-driven analysis of JAK2 V617F kinetics during interferon-alpha2 treatment of patients with polycythemia vera and related neoplasms

Projekter

Matematisk Modellering af Myeloproliferative Sygdomme

Citer dette