Cytidine deaminase enzyme activity is a predictive biomarker in gemcitabine-treated cancer patients

Expression of gemcitabine metabolizing enzymes and stromal components reveal complexities of preclinical pancreatic cancer models for therapeutic testing

BACKGROUND

Pancreatic ductal adenocarcinoma (PDAC) poorly responds to antineoplastic agents. Discrepancies between preclinical success and clinical failure of compounds has been a continuous challenge and major obstacle in PDAC research.

AIM

To investigate the association of the tumor microenvironment (TME) composition and gemcitabine metabolizing enzyme (GME) expression in vitro and several in vivo models.

METHODS

mRNA expression and protein levels of GME (cytosolic 5'-nucleotidase 1 A; NT5C1A, cytidine deaminase; CDA, deoxycytidine kinase; DCK), gemcitabine transporters (ENT1, ENT2, RRM1, RRM2) and stromal components (hyaluroninc acid, podoplanin, masson trichrome, picrosirius) were assessed by qRT-PCR and immunohistochemistry in murine LSL-KrasG12D/+;LSL-Trp53R172 H/+; Pdx-1-Cre (KPC), orthotopically transplanted mice (OTM), human primary resected PDAC tissue (hPRT), corresponding patient-derived xenograft (PDX) mice, and KPC-SPARC-/- mice. mRNA expression of GME was analyzed in PDAC cell lines (Panc-1, MIA PaCa, BXPC3 and L3.6) upon incubation on collagen or pancreatic stellate cell (PSC) conditioned media by qRT-PCR.

RESULTS

Endogenous KPC tumors exhibited significantly higher levels of GME compared to OTM. However, GME levels did not differ between hPRT and corresponding PDX mice. Using Kendalls Tau correlation coefficient we did not show a significant correlation of GME and components of the TME except for NT5C1A and hyaluronic acid in PDX mice (p=0.029). GME were not significantly altered upon SPARC depletion in vivo, and upon treatment with PSC-conditioned media or incubation on collagen plated dishes in vitro.

CONCLUSIONS

Our findings suggest that the expression of GME is independent from the deposition of stromal components. KPC mice are most appropriate to study stromal composition whereas PDX mice maintain GME expression of the corresponding hPRT and could be best suited for pharmacokinetic studies.

Exploratory focused pharmacogenetic testing reveals novel markers associated with risperidone pharmacokinetics in Saudi children with autism

Background: Autism spectrum disorders (ASDs) encompass a broad range of phenotypes characterized by diverse neurological alterations. Genomic studies have revealed considerable overlap between the molecular mechanisms implicated in the etiology of ASD and genes involved in the pharmacokinetic (PK) and pharmacodynamic (PD) pathways of antipsychotic drugs employed in ASD management. Given the conflicting data originating from candidate PK or PD gene association studies in diverse ethnogeographic ASD populations, dosage individualization based on "actionable" pharmacogenetic (PGx) markers has limited application in clinical practice. Additionally, off-label use of different antipsychotics is an ongoing practice, which is justified given the shortage of approved cures, despite the lack of satisfactory evidence for its safety according to precision medicine. This exploratory study aimed to identify PGx markers predictive of risperidone (RIS) exposure in autistic Saudi children. Methods: This prospective cohort study enrolled 89 Saudi children with ASD treated with RIS-based antipsychotic therapy. Plasma levels of RIS and 9-OH-RIS were measured using a liquid chromatography-tandem mass spectrometry system. To enable focused exploratory testing, genotyping was performed with the Axiom PharmacoFocus Array, which included a collection of probe sets targeting PK/PD genes. A total of 720 PGx markers were included in the association analysis. Results: A total of 27 PGx variants were found to have a prominent impact on various RIS PK parameters; most were not located within the genes involved in the classical RIS PK pathway. Specifically, 8 markers in 7 genes were identified as the PGx markers with the strongest impact on RIS levels (p < 0.01). Four PGx variants in 3 genes were strongly associated with 9-OH-RIS levels, while 5 markers in 5 different genes explained the interindividual variability in the total active moiety. Notably, 6 CYP2D6 variants exhibited strong linkage disequilibrium; however, they significantly influenced only the metabolic ratio and had no considerable effects on the individual estimates of RIS, 9-OH-RIS, or the total active moiety. After correction for multiple testing, rs78998153 in UGT2B17 (which is highly expressed in the brain) remained the most significant PGx marker positively adjusting the metabolic ratio. For the first time, certain human leukocyte antigen (HLA) markers were found to enhance various RIS exposure parameters, which reinforces the gut-brain axis theory of ASD etiology and its suggested inflammatory impacts on drug bioavailability through modulation of the brain, gastrointestinal tract and/or hepatic expression of metabolizing enzymes and transporters. Conclusion: Our hypothesis-generating approach identified a broad spectrum of PGx markers that interactively influence RIS exposure in ASD children, which indicated the need for further validation in population PK modeling studies to define polygenic scores for antipsychotic efficacy and safety, which could facilitate personalized therapeutic decision-making in this complex neurodevelopmental condition.