PACSIN2 rs2413739 Polymorphism and Thiopurine Pharmacokinetics: Validation Studies in Pediatric Patients

Thiopurines such as thioguanine (TG), mercaptopurine (MP) and its prodrug azathioprine (AZA) are key antileukemic and immunosuppressive drugs in the treatment of several pediatric diseases such as acute lymphoblastic leukemia (ALL) and inflammatory bowel disease (IBD): their therapeutic action is affected by the enzyme thiopurine-S-methyltransferase (TPMT) whose encoding gene is polymorphic and whose enzymatic activity is also trans -regulated by the intronic single nucleotide polymorphism (SNP) rs2413739 (T>C) in the PACSIN2 gene.

This study was aimed at validating the impact of PACSIN2 rs2413739 on thiopurine pharmacokinetic and pharmacodynamic parameters in two Italian pediatric cohorts: the first cohort was multicentric, consisted of 297 pediatric patients newly diagnosed with ALL (median (IQ) age: 4.8 (3.0-9.2) years; male: 54.9%) treated according to the AIEOP-BFM ALL 2009 protocol; the second consisted of 119 pediatric IBD patients (median age (interquartile range (IQ)): 15.1 (12.3-16.9) years; male: 52.1%) treated continuously with daily weight-based AZA therapy for at least three months before enrollment at the Gastroenterology Unit of the Pediatric Hospital IRCCS "Burlo Garofolo" in Trieste. Moreover, the role of rs2413739 polymorphism on TPMT and PACSIN2 protein expression was evaluated in peripheral blood mononuclear cells (PBMC) of adult healthy donors.

The presence of the PACSIN2 rs2413739 T allele significantly reduced the TPMT activity only in the ALL cohort (p=0.026, linear mixed-effect model, multivariate analysis considering the TPMT SNP rs1142345 and rs1800460), without affecting MP metabolites levels in patients' erythrocytes: the PACSIN2 rs2413739 genetic effect on TPMT activity was particularly evident within TPMT heterozygotes (linear mixed-effect model, p=0.041). Interestingly, TPMT enzymatic function was significantly higher in ALL patients undergoing a long-term thiopurine-based therapy compared to IBD patients (p=0.032, linear mixed-effect model). In IBD patients, treatment response to AZA was significantly affected by PACSIN2 genotype: AZA effectiveness was reduced in patients carrying the T allele of the SNP rs2413739 (linear mixed effects, p=0.018; genotyping results: CC in 19, CT in 29 and TT in 20 patients): indeed, a clinically active disease was observed in ~40% of TT patients, in 30% of those with the CT genotype and in 20% of the CC subjects. For ALL patients, clinical data are still under collection and are not available at the time of preparation of the abstract.

To evaluate the role of PACSIN2 SNP rs2413739 on TPMT and PACSIN2 protein expression, 20 adult healthy donors were genotyped for the SNP of interest, and Western-blot analysis against human TPMT and human PACSIN2 were performed on their PBMCs lysates. No significant contribution of the polymorphism was found on TPMT and PACSIN2 protein expression in this small population. In contrast, a significant direct correlation between the TPMT and PACSIN2 levels was observed (linear mixed effect, p = 0.045), suggesting a putative coordination in their protein expression in PBMCs.

These results support the role of PACSIN2 polymorphism on TPMT activity, although this effect may be tissue-specific or dependent on the patients' clinical condition.