Genetic and metabolic studies towards personalized conditioning regimen prior to stem cell transplantation

Abstract

Hematopoietic stem cell transplantation (HSCT) is a curative treatment for several malignant and non- malignant diseases. The busulphan (Bu)/cyclophosphamide (Cy) combination is one of the most common conditioning regimens given prior to HSCT.

The general aim of the present thesis is to investigate the molecular mechanisms underlying the metabolism of the Bu/Cy conditioning regimen in order to personalize the treatment and improve the clinical outcome.

To follow the metabolic pathway of busulphan, a new gas chromatography-mass spectrometry (GC-MS) method was developed and validated for the quantification and detection of busulphan and its four major metabolites.

Incubation of the first core metabolite of busulphan, tetrahydrothiophene (THT), with human liver microsomes or recombinant enzymes has resulted in the formation of subsequent metabolites. The highest initial THT disappearance rate and the highest CLint value were observed with FMO3 followed by several CYPs indicating that FMO3 and, to a lesser extent, CYPs are involved in the metabolic pathway of busulphan. Moreover, FMO3 inhibition significantly (P < 0.05) affected Bu and THT kinetics in mice. In patients, FMO3 expression was significantly (P < 0.05) up-regulated during Bu treatment.

In order to personalize oral Bu dosage, a reliable limited sampling model was developed and evaluated in both adult and pediatric patients.

To understand the role of cyclophosphamide in the conditioning regimen, the gene expression profile over two days of Cy treatment was investigated, where 299 genes were found to be specifically affected by the treatment. Cyclophosphamide down-regulated the expression of several genes mapped to immune/autoimmune activation and graft rejection including CD3, CD28, CTLA4 and IL2R, and up-regulated immune-related receptor genes, e.g. IL1R2, IL18R1, and FLT3.

Significant (P < 0.01) up-regulation, with high inter-individual variation, of the cytochrome P450 oxidoreductase (POR) gene was also observed during Cy treatment. In vitro, different batches of CYP2B6.1, with different ratios of POR/CYP, showed positive correlation between the intrinsic clearance (Vmax/Km) and the POR/CYP ratio for the Cy 4-hydroxylation.

Further analysis of the above mentioned patients, prior to Cy treatment, revealed that CYP2J2 mRNA expression was significantly (P < 0.01) higher compared to healthy controls. CYP2J2 expression was further up-regulated during Cy treatment, with high inter-individual variation. Repeated treatment with Cy resulted in an increased 4-OH-Cy/Cy ratio, indicating auto-induction of Cy-metabolism.

The viability of HL-60 cells, lacking CYP2B6 but expressing CYP2J2, was reduced after incubation with Cy. Inhibition of CYP2J2 reduced 4-OH-Cy formation and improved HL-60 cell survival. Cy incubation with recombinant CYP2J2 confirmed that CYP2J2 is involved in Cy bioactivation.

In summary, the present results have improved our understanding of the Bu/Cy metabolism. This knowledge may help to interpret several interactions, high inter-individual variability, adverse effects and unexpected toxicity observed during and/or after the conditioning regimen. This certainly will help in developing new strategies for personalized medicine and thus improve clinical outcome.