1130-0108/2015/107/2/72-78 REVISTA ESPAÑOLA DE ENFERMEDADES DIGESTIVAS REV ESP ENFERM DIG (Madrid COPYRIGHT © 2015 ARÁN EDICIONES, S. L. Vol. 107, N.º 2, pp. 72-78, 2015 ORIGINAL PAPERS Mutational profile of KIT and PDGFRA genes in gastrointestinal stromal tumors in Peruvian samples José Buleje1, Óscar Acosta1, María Guevara-Fujita1, Yanina Enriquez2, Luis Taxa2,3, Enrique Machicado4, Frank Lizaraso-Caparó1 and Ricardo Fujita1 1Centro de Genética y Biología Molecular. Facultad de Medicina Humana.Universidad de San Martín de Porres. Lima, Perú. 2Laboratorio de Patología. Facultad de Medicina Humana. Universidad de San Martín de Porres. Lima, Perú. 3Departamento de Patología. Instituto Nacional de Enfermedades Neoplásicas- INEN. Lima, Perú. 4Department of General and Digestive Surgery. Hospital A. Loayza. Lima, Perú ABSTRACT INTRODUCTION Introduction: Gastrointestinal stromal tumors (GISTs) are Gastrointestinal stromal tumors (GIST) are the most mesenchymal neoplasms usually caused by somatic mutations in the genes KIT (c-kit) or PDGFRA. Mutation characterization has become common mesenchymal tumors of the gastrointestinal an important exam for GIST patients because it is useful in predicting tract (1). The estimated incidence of GIST is 15 cases per the response to the inhibitors of receptor tyrosine kinase (RTK). million in the general population, and the median age at Objectives: The aim of this study was to determine the presentation is 60 years (2). The most common sites of frequency of KIT and PDGFRA mutations in 25 GIST samples origin for GIST are the stomach (39-70 %) and small intes- collected over two years at two national reference hospitals in Peru. There were 21 samples collected from the Instituto Nacional tine (31-45 %), but GISTs may appear anywhere along the de Enfermedades Neoplásicas (INEN, national cancer center) and gastrointestinal tract or within the abdomen as extra-gas- 4 samples collected from Hospital A. Loayza. trointestinal tumors (3). Methods and materials: In this retrospective study, we Approximately 95 % of GIST tumors express the tyro- performed polymerase chain reaction (PCR) amplification and deoxyribonucleic acid (DNA) sequencing of KIT (exons 9, 11, sine kinase KIT receptor, also called CD117, which is 13, and 17) and PDGFRA (exons 12 and 18) genes in 20 FFPE now accepted as the most specific immunohistochemical (formalin-fixed, paraffin-embedded) and 5 frozen GIST samples. marker for GIST (4,5). Other markers, such as CD34 and Results: We report 21 mutations, including deletions, Desmin that show variable expression (70 % and 5 %, duplications, and missense, no mutations in 2 samples, and 2 respectively) are also used to confirm the diagnosis of samples with no useful DNA for further analysis. Eighty-six percent of these mutations were located in exon 11 of KIT, and 14 % were GIST tumors (6-8). located in exon 18 of PDGFRA. In recent years, it has been established that 75 % to Conclusions: Our study identified mutations in 21 out of 80 % of GISTs harbor mutations in the KIT gene (c-kit). 25 GIST samples from 2 referential national hospitals in Peru, Activating mutations of the KIT gene in GIST occur in and the mutation proportion follows a global tendency observed exons 9, 11, 13, and 17 corresponding to the juxtamem- from previous studies (i.e., the majority of samples presented KIT mutations followed by a minor percentage of PDGFRA mutations). brane (JM) intracellular regulatory domain, the extram- This study presents the first mutation data of the KIT and PDGFRA embrane domain, and the two intracytoplasmic tyrosine genes from Peruvian individuals with GIST. kinase domains, respectively (9). Mutations in the JM domain affect its autoregulatory function and promote Key words: KIT. PDGFRA. Gastrointestinal stromal tumors. spontaneous kinase activation (10). The location of KIT Mutational analysis. Received: 31-07-2013 Accepted: 10-11-2014 Correspondence: José Luis Buleje Sono. Centro de Genética y Biología Buleje J, Acosta Ó, Guevara-Fujita M, Enriquez Y, Taxa L, Molecular. Facultad de Medicina Humana. Universidad de San Martín de Machicado E, Lizaraso-Caparó F, Fujita R. Mutational profile Porres. Av. Alameda del Corregidor 1531. Urb Los Sirius, Las Viñas. La of KIT and PDGFRA genes in gastrointestinal stromal tumors in Molina, Lima 12. Perú Peruvian samples. Rev Esp Enferm Dig 2015;107:72-78. e-mail: [email protected] Vol. 107, N.º 2, 2015 MUTATIONAL PROFILE OF KIT AND PDGFRA GENES IN GASTROINTESTINAL STROMAL TUMORS IN PERUVIAN SAMPLES 73 mutations is important for pharmacological treatment; for Human Subjects Committee policy of the National Health example, it is well known that a mutation in exon 11 of the Institute of Peru. KIT gene is associated with a better response to treatment with the inhibitor imatinib (Gleevec®), the potent inhib- itor of receptor tyrosine kinase (RTK), and a decreasing Deoxyribonucleic acid (DNA) extraction response for mutations in exons 9, 13, 17, and wild-type tumors (9). Genomic DNA was extracted using the High Pure PCR Another member of the RTK family of genes, PDG- template Preparation Kit (ROCHE) following the manu- FRA, is also associated with the pathogenesis of GIST, facturer’s protocol. A QubitTM fluorometer (InvitrogenTM) and the mutations in KIT are mutually exclusive with was used to quantify the DNA. In 23 samples in which the those in PDGFRA (11). Mutations in the PDGFRA gene quality and the quantity of DNA were appropriate, the target are observed in 7-12 % of cases (12), and the most fre- exons were amplified. In two of the FFPE tissue samples, the quent mutations are observed in exons 12 (regulatory amplification was not successful due to poor DNA quality. juxtamembrane domain) and 18 (second tyrosine kinase The DNA was resuspended in a TE10:1 buffer, coded, and domain). kept at 4 °C until use. The presence, nature, and location of the KIT/PDG- FRA oncogenic mutations may translate into differenc- es in tumor aggressiveness and influence the likelihood Mutational analysis of a clinical response to imatinib, a selective tyrosine kinase inhibitor able to interfere with the activation of Amplification of target exons and mutational analysis KIT and PDGFR receptors by competing with ATP in of KIT and PDGFRA genes the ATP-binding pocket (13,14). Generally, patients with tumors carrying KIT exon 11 mutations respond much The coding sequence and intron-exon boundaries for better to treatment in comparison with tumors carrying exons 9, 11, 13, and 17 of the KIT gene and exons 12 and exon 9 mutations (15). However, GISTs with mutations 18 of the PDGFRA gene were amplified by PCR using in exon 11 could develop resistance to imatinib treatment primers and the conditions that were optimized in others if subsequent secondary point mutations appear, which studies (26,38). The PCRs were performed in a 25 μl total suggests an important escape mechanism for tumor cells reaction volume consisting of 50 ng of the DNA template. with a KIT-dependent proliferation mechanism temporar- The PCR mix contained 10 pmol of each primer, 0.5 U of ily inhibited by imatinib (15). Taq polymerase (Thermo Scientific®), buffer 1X, 0.2 mM In this retrospective study, we attempted to determine of each dNTP, and 1.5 mM of MgCl2. The cycling condi- the frequency of KIT and PDGFRA mutations in GIST tions for the PCR amplification were 95 °C for 5 minutes, samples collected from two national referential hospitals 35 cycles of 95 °C for 45 seconds, annealing temperatures in Peru (INEN and A. Loayza Hospital). This is the first of 55 °C to 57 °C for 45 seconds (Table I), and 72 °C study of this type conducted in Peru, and the aim is to for 1 minute followed by a final elongation step at 72 °C encourage a clinical approach on the molecular aspects for 10 minutes. The PCR products were then subjected to of GISTs in order to provide adequate treatment for lon- electrophoresis on 2 % agarose gels and observed under an ger survival and a better quality of life for patients with ultraviolet light after ethidium bromide staining. these tumors. DNA sequencing MATERIAL AND METHODS PCR amplicons were purified using a QIAquick PCR Samples Purification Kit (QIAGEN) and sequenced in both directions using the BigDye Terminator version 3.1 Cycle sequencing A total of 25 GIST samples collected over a two-year kit and the 3500 Genetic Analyzer (Applied Biosystems). The period were used for this retrospective study. All samples generated DNA sequences were analyzed with a Sequencing were determined by immunohistochemistry as CD117 pos- Analysis software 5.1 (Applied Biosystems) and then aligned itive. Five samples were obtained from frozen tissue, and using the Basic Local Alignment Search Tool (BLAST) twenty were derived from formalin-fixed, paraffin-embed- (http://blast.ncbi.nlm.nih.gov/Blast.cgi). All mutations were ded (FFPE) blocks. Coded tumor samples were obtained identified based on the National Center for Biotechnology from two different national reference hospitals (21 from Information database of genetic variation. The reference INEN and 4 from A. Loayza Hospital). This study was sequences used to describe the mutations were NM_000222 conducted with an ethical clearance of IRBs of partici- and NM_0062056 for the KIT and PDGFRA genes, respec- pating hospitals and according to the guidelines of surgi- tively. The numbering of specific mutations and SNPs was cal procedures for tumor removal from the Research on referenced from http://www.ensembl.org as of February 2014. REV ESP ENFERM