Patient-Centered Genomic Diagnostic Testing for AML: A Quality Improvement Project.

Abstract

The classification of AML and therapeutic options are now largely driven by genetically defined subtypes. Personalization of treatment relies on timely completion and reporting of cytogenetic and molecular tests, creating challenges in clinical practice. We initiated a quality improvement study with the aim to optimize the process for ordering of genomic diagnostic tests and to reduce test turnaround times (TATs). A multidisciplinary working group consisting of hematologists, laboratory scientists, technologists, and hematopathologists was formed and identified the following tests as necessary for expedited testing in patients with AML younger than 75 years: next-generation sequencing (NGS) myeloid panel, karyotype analysis, <i>FLT3</i> PCR, <i>NPM1</i> PCR, <i>CBFB::MYH11</i> PCR, and <i>RUNX1::RUNX1T1</i> PCR, and proposed a reflexive flow cytometry-triggered genomic diagnostic testing algorithm for newly diagnosed AML (ND-AML). We used the model of improvement and implemented three Plan-Do-Study-Act (PDSA) cycles: education and guidelines for management of ND-AML, implementation of the reflex laboratory-triggered diagnostic testing algorithm for ND-AML, and automation of NGS workflow. We assessed compliance with test ordering according to prescribed guidelines and TAT. After PDSA 2, test ordering improved significantly to more than 90% of relevant tests being initiated at AML diagnosis; and TAT was reduced by 27.6% for NGS and by 54.8% for <i>NPM1</i> PCR. After PDSA 3, TAT for NGS was overall reduced by 63.3% to 11.4 days and within our 14-day target. We were able to also meet our target TAT of 5 days or less for <i>FLT3</i> and <i>NPM1</i> PCRs. A multidisciplinary approach with shared decision making between hematologists and laboratory practitioners was essential in the development of an algorithm for reflex testing in AML that resulted in improved test ordering and TAT.