Principles of Classification of Myeloid Neoplasms

The 2016 revision of the fourth edition of the World Health Organization (WHO) classification of myeloid neoplasms, WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues , is used in this book. The principles of the WHO classification have been described elsewhere, and the process for developing this consensus classification is summarized in Chapter 13 . Briefly, the classification relies on a combination of clinical, morphologic, immunophenotypic, genetic, and other biologic features to define specific disease entities—a logical approach similar to that followed by a clinician and a pathologist working together to reach a diagnosis in a patient suspected of having a myeloid neoplasm. The relative contribution of each feature varies, depending on the case. Only through familiarity with the classification system and with the criteria for each entity can the appropriate studies be chosen to arrive at an accurate diagnosis in an expedient manner. Although perhaps overused as an example of the prototype for the classification of myeloid neoplasms, chronic myeloid leukemia (CML) symbolizes the utility of the WHO approach. This leukemia is recognized mainly by its clinical and morphologic features and is consistently associated with a specific genetic defect, the BCR-ABL1 fusion gene. This abnormality leads to the production of a constitutively activated protein tyrosine kinase that interacts with a number of different cellular pathways to influence the proliferation, survival, and differentiation of neoplastic cells. The protein is sufficient to cause the leukemia, but it also provides a target for therapy that has prolonged the lives of thousands of patients with this disease. The diagnosis of CML, however, is not based on any single parameter. There are other myeloid leukemias that mimic its clinical presentation and morphology, and the BCR-ABL1 fusion is seen not only in CML but also in some cases of de novo acute lymphoblastic leukemia, acute myeloid leukemia (AML), and mixed phenotype acute leukemia. Thus, CML is a perfect model for the integration of all pieces of relevant information to define an entity in a classification scheme. Furthermore, there are still mysteries regarding CML, so there is still more to learn (see Chapter 47 ).

As the focus in all neoplasms turns increasingly to the genetic infrastructure of malignant cells and to molecular abnormalities that may be targets for therapeutic agents, it is only natural that more genetic and molecular data are incorporated into the diagnostic algorithms or nomenclature of classification schemes. The 2001 third edition of the WHO classification included, for the first time in any widely used system, genetic information as criteria for the diagnosis of not only CML but also some subtypes of AML. By the time the 2008 fourth edition of the WHO classification was published, a number of significant genetic abnormalities were discovered that are associated with subgroups of myeloid neoplasms or with specific disease entities within the subgroups. Since that time, an even larger number of genetic and epigenetic events associated with myeloid neoplasms have been described, making an approach that uses these data for both classification and prognostication more challenging. In some instances, such as malignant eosinophilia associated with rearrangements involving PDGFRA or PDGFRB , the genetic defect (coupled with the morphology and clinical findings) is the major criterion for naming the disease and for selecting specific targeted therapy (see Chapter 50 ). In other instances, such as the BCR-ABL1 –negative myeloproliferative neoplasms (MPNs) that are often but not invariably associated with the JAK2 V617F mutation, the presence of the genetic defect is an objective criterion that identifies the myeloid proliferation as neoplastic. Additional criteria are necessary to define the specific disease associated with the mutated JAK2 and to distinguish it from other MPNs that share the same mutation (see Chapter 47 ). Therefore, although the 2016 WHO classification scheme incorporates an increasing number of genetic abnormalities, a multidisciplinary approach is still required for the classification of myeloid neoplasms. This multidisciplinary approach succeeds in defining many distinct disease entities that cannot be adequately identified by relying on morphology or clinical features alone. Such a limited approach to the myeloid neoplasms is no longer adequate, and a diagnosis is not complete in many cases until the results of all studies have been correlated, often requiring amended pathology reports.

New entities and new diagnostic criteria for old entities in the WHO classification scheme are based mainly on published clinical and scientific studies that have been widely quoted and their significance widely acknowledged. However, to accommodate recent data that have not yet “matured,” the classification continues to include a number of “provisional entities.” These are newly described or characterized disorders that are clinically or scientifically important and should be considered in the classification, but additional studies are needed to clarify their significance. Some previous provisional entities have been refined in the current classification and are now incorporated as full entities, and their presence emphasizes that the classification is ever-changing.