Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

P110. Validation of a fast genotyping method for analysis of TPMT polymorphisms

M. Chaparro, M. Román, T. Carbaleiro, J. Novalbos, F. Abad-Santos, J.P. Gisbert

Hospital de la Princesa, Madrid, Spain

Background: Azathioprine (AZA) and mercaptopurine (MP) have shown to be effective in inducing and maintaining remission in IBD patients. The occurrence of adverse events (AE), such as myelossuppression, is a major drawback in the use of AZA/MP. AE might be associated with polimorphisms in the TPMT gene that might decrease TMPT enzyme activity. In clinical practice identification of the TMPT polymorphisms can help prevent thiopurine-induced AE.

Aims: To validate a fast method for TPMT genotyping (LightSNiP, Roche®) compared with the conventional methodology. To calculate the cost of each method. To evaluate the prevalence of TPMT polymorphisms in our population.

Methods: Peripheral blood samples were obtained from each patient. DNA was extracted in a automatic DNA-extractor. The DNA was quantified spectrophotometrically, and purity was tested using the A260/280 ratio. The conventional methodology for genotyping the *2, *3B and *3C polymorphisms (which account for almost 90% of all known polymorphisms) was based on polymerase chain reaction (PCR) and sequencing of exons 5, 7 and 10 of the TPMT gene. Other TPMT variants can also be detected using the conventional PCR and sequencing method. LightSNiP kit carried out real-time PCR in a LightCycler480 (Roche®) and evaluated the presence of the *2, *3B and *3C polymorphisms. The conventional method was considered the gold standard.

Results: 111 patients from our hospital were included, 60% women. Thiopurine-induced myelotoxicity had already been diagnosed in 61% before genotyping, and the other 39% were to undergo genotyping before initiating treatment with thiopurines. The LightSNiP method was performed in 83 patients and the conventional one in all included patients. There were no differences between genotypes or allele frequency obtained by the conventional method and those obtained by LightSNiP. For TPMT, analytical validity was based on the presence/absence of alleles *2, *3B, and *3C. The sensitivity and specificity values were 100% for LightSNiP. Most patients possessed the wild type (81.1%). No patients were homozygous for any mutation. All the patients carrying the *3B polymorphism also carried the *3C polymorphism; therefore, they were considered carriers of *3A, the most common mutation (9.9%), followed by *2 (3.6%) and *3C (2.7%). Furthermore, sequencing revealed the TPMT*8 allele in two patients (1.8%) and the TMPT*19 allele in one patient (0.9%), which could not be detected by the new method. Mutations were more prevalent among patients with thiopurine-induced myelotoxicity (27.3 vs. 7.7%, p = 0.06). The conventional technique took 6 days to provide results, whereas with LightSNiP, results were available in 48 hours. The cost of the LightSNiP and the conventional method were 23€ and 36€ per patient, respectively.

Conclusions: LightSNiP (Roche®) is a reliable, fast, less expensive, and sensitive method for detection of the TPMT polymorphisms. At present, some rare TMPT polymorphisms cannot be detected with this technique. Patients who experience myelotoxicity have a higher frequency of mutant alleles, but they were present only in one-third of these patients, so additional studies are recommended to find new biomarkers of thiopurine-induced AE.