Validation of variants in SLC28A3 and UGT1A6 as genetic markers predictive of anthracycline‐induced cardiotoxicity in children. Issue 8 (25th February 2013)

• Record Type: Journal Article
• Authors: Visscher, H.; Ross, C.J.D.; Rassekh, S.R.; Sandor, G.S.S.; Caron, H.N.; van Dalen, E.C.; Kremer, L.C.; van der Pal, H.J.; Rogers, P.C.; Rieder, M.J.; Carleton, B.C.; Hayden, M.R.
• Is Part Of: Pediatric blood & cancer. Volume 60:Issue 8(2013:Aug.)
• Page Start: 1375
• Page End: 1381
• Publication Date: 2013-02-25
• Subjects: Tumors in children -- Periodicals; Blood -- Diseases -- Periodicals; Cancer in children -- Periodicals; 618.92
• DOI: 10.1002/pbc.24505
• Languages: English
• ISSNs: 1545-5009

Abstract

<abstract abstract-type="main" xml:lang="en"> <title>Abstract</title> <sec id="pbc24505-sec-0001" sec-type="section"> <title>Background</title> <p> <bold>.</bold> The use of anthracyclines as effective antineoplastic drugs is limited by the occurrence of cardiotoxicity. Multiple genetic variants predictive of anthracycline‐induced cardiotoxicity (ACT) in children were recently identified. The current study was aimed to assess replication of these findings in an independent cohort of children.</p> </sec> <sec id="pbc24505-sec-0002" sec-type="section"> <title>Procedure</title> <p> <bold>.</bold> Twenty‐three variants were tested for association with ACT in an independent cohort of 218 patients. Predictive models including genetic and clinical risk factors were constructed in the original cohort and assessed in the current replication cohort.</p> </sec> <sec id="pbc24505-sec-0003" sec-type="section"> <title>Results</title> <p> <bold>.</bold> We confirmed the association of rs17863783 in <italic>UGT1A6</italic> and ACT in the replication cohort (<italic>P</italic> = 0.0062, odds ratio (OR) 7.98). Additional evidence for association of rs7853758 (<italic>P</italic> = 0.058, OR 0.46) and rs885004 (<italic>P</italic> = 0.058, OR 0.42) in <italic>SLC28A3</italic> was found (combined <italic>P</italic> = 1.6 × 10<sup>−5</sup> and <italic>P</italic> = 3.0 × 10<sup>−5</sup>, respectively). A previously constructed prediction model did not significantly improve risk prediction in the replication cohort over clinical factors alone. However, an improved prediction model constructed using replicated genetic variants as well as clinical factors discriminated significantly better between cases and controls than clinical factors alone in both original (AUC 0.77 vs. 0.68, <italic>P</italic> = 0.0031) and replication cohort (AUC 0.77 vs. 0.69, <italic>P</italic> = 0.060).</p> </sec> <sec id="pbc24505-sec-0004" sec-type="section"> <title>Conclusions</title> <p> <bold>.</bold> We validated genetic variants in two genes predictive of ACT in an independent cohort. A prediction model combining replicated genetic variants as well as clinical risk factors might be able to identify high‐ and low‐risk patients who could benefit from alternative treatment options. Pediatr Blood Cancer 2013;601375‐1381. © 2013 Wiley Periodicals, Inc.</p> </sec> </abstract>