Integration of Decision Support Algorithm and Genomic Mutational Analysis for Enhanced Diagnosis and Characterization of Hematological Diseases

Abstract

The bone marrow can be a window into human disease, providing valuable insights into the underlying mechanisms of various conditions that disrupt normal hematopoiesis. Accompanying the bone marrow biopsy are a wealth of ancillary tests performed on the sample and other clinical and laboratory data. This thesis encompasses two studies on the utilization of molecular genetic data in hematologic malignancies. While both projects sit under this unifying theme, the two primary studies of this thesis encompass different types of data analyses required to make the various clinical correlations and recommendations about best practices. The first study focuses on test utilization management. In clinical practice, providers order multiple ancillary tests based on suspicion before confirming a diagnosis, leading to unnecessary costs and potentially missing important tests. While standardized reflex testing based on pathologist evaluation and flow cytometry has been shown to minimize unnecessary testing, the specific impact of using decision support application (DSA) in optimizing test ordering and reducing over-ordering and omitted testing in cases across a range of hematologic conditions and neoplasms has not been explored or demonstrated. This study aims to fill this gap by investigating the effectiveness of a DSA-supported test ordering approach and its impact on improving the cost-effectiveness and diagnostic relevance of ancillary BM testing, in particular molecular and cytogenetic studies. The second study performs an in-depth analysis of one specific molecular assay, a large next generation sequencing (NGS) panel, in the evaluation of patients with chronic lymphocytic leukemia (CLL), a hematologic neoplasm accounting for one-third of adult leukemias that is characterized by the accumulation of abnormal lymphocytes in the blood, bone marrow, lymph nodes, and other organs. Although the underlying cause of CLL is not fully understood, it is thought to involve a combination of genetic and environmental factors. Certain genetic abnormalities, such as deletions or mutations in specific genes, are commonly associated with CLL. This study aims to extend the utilization of this NGS panel from mutation detection to mutational signature identification in CLL and the application of those signatures to disease prognosis. Specifically, this study asks if a single NGS panel test can replace the need for separate testing of somatic hypermutation (SHM) status, a prognostic marker used in standard clinical care for the risk stratification of CLL patients. In addition, this study examines whether other mutational signature(s) provide added prognostic value to the clinical care of these patients. Together, both studies examine the role of molecular diagnostic testing in hematologic disease. The first study focuses upon the pattern of testing utilization and mechanisms to support the optimal use of testing resources. The study culminates with an estimation of the financial benefit of decision-supported testing algorithms. The second study examines the extended analysis of a single molecular test to improve clinical risk stratification. This study focuses on the impact of advanced molecular testing on patient care and biologic understanding of disease. Both studies emphasize conclusions that impact our healthcare system.