The French-American-British (FAB) classification of the acute leukemias was devised in the 1970s and 1980s. The FAB schemas were based largely on morphologic characteristics and relied heavily on examination of routine histologic stain preparations to distinguish lymphoid neoplasms from myeloid neoplasms (Figure 29-1). Although these types of diagnostic criteria have not been abandoned, pathologists are now moving toward more precise classification of many of the leukocyte neoplasms based on recurring chromosomal and genetic lesions found in many patients. These lesions are related to disruptions of oncogenes, tumor suppressor genes, and other regulatory elements that control proliferation, maturation, apoptosis, and other vital cell functions. In 2001 the World Health Organization (WHO) published new classification schemes for nearly all of the tumors of hematopoietic and lymphoid tissues.¹ In some cases WHO melded the older morphologic schemes with the newer schemes. For instance, in the WHO classification scheme for acute myeloid leukemias (AMLs) there are some remnants of the old FAB classification, but new classifications were introduced for leukemias associated with consistently recurring chromosomal translocations. This 2001 WHO classification of hematologic malignancies has undergone a recent revision.²

Figure 29-1 A, Lymphoblasts (bone marrow, Wright stain, ×500). Cells have a diameter two to three times the normal lymphocyte diameter, scant blue cytoplasm, coarse chromatin, deeper staining than myeloblasts, and inconspicuous nucleoli. B, Myeloblasts (bone marrow, Wright stain, ×500). Cells have a diameter three to five times the lymphocyte diameter, moderate gray cytoplasm, uniform fine chromatin, two or more prominent nucleoli, and possibly Aüer rods.

Chromosomal Abnormalities in Hematologic Neoplasms

In part because of the ease of access to bone marrow compared with other organs of the body, researchers have learned a great deal about the biology of cancer by studying hematopoietic neoplasms. The study of chromosomal translocations in hematologic malignancies has taught how a single mutation, or series of mutations in stepwise fashion, can lead to malignant transformation by disrupting the molecular machinery of the cell. The first two genetic lesions found in any kind of human cancer were identified as chromosomal translocations occurring in leukocyte malignancies. These were the t(9;22) translocation in chronic myelogenous leukemia (CML)³ and the t(8;14) translocation in Burkitt lymphoma.⁴

Oncogenes

Oncogenes originally were identified as genes that carried rapidly transforming retroviruses derived from normal cellular homologues, proto-oncogenes. The oncogene definition has evolved to specify genes that cause dominant-acting cancer mutations, regardless of whether they are derived from a retrovirus. The typical proto-oncogene codes for a protein involved in normal cell cycle regulation. Regulation proteins provide the signal transduction that carries messages about cell division and maturation (differentiation) from outside the cell to the nucleus. Mutations of these genes may form oncogenes that disrupt normal cell cycle processes.

Most chromosomal translocations in leukemias involve oncogenes. The “dominant” transforming oncogene is able to alter the gene product and transform the cell into a malignant phenotype, even in the presence of a residual normal allele. Even the afore-mentioned two examples, CML and Burkitt lymphoma, t(9;22) and t(8;14), involve oncogenes that are activated when brought into proximity with their new partners on fusion genes. In the case of CML, the ABL proto-oncogene on chromosome 9 is activated when fused to the BCR component of chromosome 22. In the case of Burkitt lymphoma, the MYC proto-oncogene on chromosome 8 is fused with the immunoglobulin heavy chain locus on chromosome 14. Although oncogenic transformation was first identified by karyotypic analyses, molecular biologic techniques have evolved rapidly over the last three decades, so that more genetic translocations are being identified that create novel fusion genes invisible at the cytogenetic level.

Tumor Suppressor Genes

Tumor suppressor genes are so named because they code for proteins that resist malignant transformation. These genes do not act in a dominant fashion as in the case of oncogenes; rather, cells are transformed into a malignant phenotype only after both alleles of these genes have been lost or otherwise inactivated, the so-called two-hit mechanism proposed by Knudson.⁵ Although tumor suppressor genes are harder to isolate and identify, numerous such genes have now been identified, and many have been found to be associated with autosomal dominant familial cancer predisposition syndromes. Some well-known examples include the RB1 tumor suppressor gene involved in familial retinoblastoma, the TP53 gene in Li-Fraumeni syndrome, the WT1 gene in Wilms tumor, and the NF1 gene in familial neurofibromatosis type 1. Perhaps more importantly, many of these tumor suppressor genes and their protein products are altered in many sporadic cancers, including hematologic cancers.

Other Genetic Changes

Beyond oncogenes and tumor suppressor genes, researchers have found genes that are involved in DNA repair mechanisms. Mutations in these genes do not lead to cellular transformation by altering signaling pathways, but rather lead to genetic instability and increased mutation rates. Examples of such genes are the DNA mismatch repair genes MLH1 and MSH2. Another example is the Fanconi anemia gene, FA, which is important for maintaining genomic stability in hematopoietic tissues.

Molecular Pathways Perturbed by Cellular Transformation

Regardless of the type of chromosomal or genetic abnormality, clinicians understand the molecular consequences that the formation of oncogenes or loss of tumor suppressor genes has on the proliferation and differentiation of hematologic tissues. Specialists recognize several mechanisms, including blocked differentiation, transcriptional repression, disruptions of cell signaling, progression, and apoptosis. The t(15;17) translocation found in acute promyelocytic leukemia (APL), which fuses the PML gene to the RARA (retinoic acid receptor alpha) gene, clearly results in a state of arrested differentiation, because RARA-induced differentiation is inhibited. Treating patients with APL with pharmacologically high doses of all-trans retinoic acid can overcome this block and permit APL cells to differentiate into normal neutrophils. In so doing, the APL cells lose their leukemic potential.

Other chromosomal abnormalities involve transcriptional repression of DNA and condensation abnormalities of chromatin, such as those involved in the core-binding factor leukemic subtypes of AML. A similar example on the lymphoid side of hematopoietic neoplasms are the chromosomal translocations involving the BCL6 gene. Normal BCL6 encodes for a transcriptional repressor responsible for recruiting the histone deacetylase complex, which regulates germinal center formation in lymph nodes. The mutation of BCL6 leads to overexpression of this normal protein so that DNA is excessively repressed, which in this case prevents lymphocytes from progressing beyond the germinal center stage of development.

Since the initial identification of the BCR/ABL fusion gene in CML, there are now many other examples of genetic abnormalities in myeloid malignancies in which the abnormality leads to disruption of cell signaling, often by way of activation of kinase cascades. FLT3 codes for a tyrosine kinase receptor preferentially expressed on hematopoietic stem cells that mediates proliferation and differentiation. A unique mutation resulting in an internal tandem duplication leads to constitutive activation of this pathway (i.e., always turned on) in many forms of AML and other hematopoietic malignancies. Other examples of gene mutations that alter kinase cascades in myeloid or lymphoid cells are c-KIT, NOTCH, JAK2, and RAS.

Many cyclin-dependent kinases are altered in lymphoid malignancies. The cyclin-dependent kinases tightly regulate cell cycle progression through the synthesis, proteolysis, and phosphorylation of cyclins.

Finally, another important molecular signaling pathway in cells involves programmed cell death, or apoptosis. This vital process allows organisms to eliminate redundant, damaged, aged, or infected cells. In the hematopoietic environment, apoptosis is essential to contain and control the massive expansion that the hematopoietic system is capable of generating at times of stress, infection, or hemorrhage. Caspases are a family of proteases that participate in the apoptotic cascade triggered in response to proapoptotic signals. The culmination of this apoptotic cascade is cellular disassembly. The BCL family contains many genes, some of which are proapoptotic and some of which are antiapoptotic. Many of the various forms of non-Hodgkin lymphoma seem to involve disruptions of BCL2, BCL6, BCL10, or other members of the caspase and BCL family of genes comprising the apoptotic cascade.

The list of chromosomal and molecular aberrations known to occur in the various leukocyte neoplasms continues to grow on an almost daily basis. Indexing this list is far beyond the scope of this chapter, but some condensed lists are provided in Chapter 31. More complete indices can be found in other publications such as the WHO reclassification scheme,¹ its 2008 revision, or other hematology textbooks.^6–10

Therapy for Leukocyte Neoplasms

The various forms of therapy available today for leukocyte neoplasms can be roughly divided into the following categories: chemotherapy, radiation therapy, supportive therapy, targeted therapy, and stem cell transplantation. In contrast to many solid tumors, numerous hematologic malignancies now have cure rates that are substantially higher than they were two or three decades ago. Many new and exciting therapies that are less toxic are now under development or are already employed in patient settings. These therapies are bringing more optimism to the care of patients with leukocyte neoplasms than ever before. Selection of the best therapy must start, however, with an accurate diagnosis. Even the most effective therapies do not work if they are applied in the wrong circumstances.

Curative treatment strategies are a realistic goal for patients with Hodgkin lymphoma, CML, hairy cell leukemia, and some forms of non-Hodgkin lymphoma, and for children with acute lymphoblastic leukemia. Cure may be attainable in other patients with acute lymphoblastic or myeloid leukemia, and long-term remissions may be achievable in adults with multiple myeloma. For patients with other leukocyte neoplasms such as mantle cell lymphoma, chronic lymphocytic leukemia, or a therapy-related leukemia, cure remains elusive, and therapy must be directed more toward attaining remissions or providing supportive care.

Chemotherapy

Chemotherapy is oral or parenteral cancer treatment with compounds that possess antitumor properties. The methods of action of the chemotherapy drugs vary considerably. Chemotherapy agents can be classified in two ways: by their effects on the cell cycle and by their biochemical mechanism of action. Some chemotherapy drugs can affect cells only in specific phases of the cell cycle (phase specific), whereas other drugs act without regard to the cell cycle (phase nonspecific) and affect cells during any phase of the cell cycle. Agents in this latter category usually have a linear dose-response curve (i.e., the higher the dose, the more cells are killed). There are two subgroups:

1. Cycle-specific agents, which kill cells that are moving through the cell cycle, regardless of whether the cells are in G₁, G₂, S, or M phase (e.g., alkylating agents, cisplatin).

2. Cycle-nonspecific agents, which kill nondividing cells or cells in the resting state (e.g., steroids and antitumor antibiotics).

Phase-specific agents are effective only if present during a certain phase in the cell cycle (see Chapters 31 and 32). Within a certain dosage range, agents in this category show no increase in killing of cells with a further increase in dosage. Examples are l-asparagine amidohydrolase (G₁ phase), antimetabolites such as methotrexate (S phase), and vinca alkaloids (M phase).

Chemotherapeutic agents affect both neoplastic and normal cells. The effect is most pronounced on rapidly dividing cells, such as those of the mucosa of the gastrointestinal tract and the bone marrow. This limits the dosage and usually determines the maximum tolerated dose for a patient. Chemotherapy agents are categorized in Table 29-1.

TABLE 29-1

Chemotherapy Agents

Agent	Other Names	Uses	Toxic Effects
Alkylating Agents
Busulfan	Myleran	CML, pretransplantation	Myelosuppression, infertility
Cyclophosphamide	Cytoxan, Neosar	Lymphoma, MM, ALL, pretransplantation	Marrow suppression, N&V, cystitis
Nitrogen mustard	Mechlorethamine	Hodgkin lymphoma, NHL	Myelosuppression, N&V, infertility
Chlorambucil	Leukeran	CLL, Waldenström macroglobulinemia, NHL, Hodgkin disease	Myelosuppression, hair loss
Melphalan	Alkeran	MM	Myelosuppression
Carmustine	BCNU	Hodgkin disease, NHL, MM	Myelosuppression
Dacarbazine	DTIC	Hodgkin disease	Myelosuppression, N&V
Plant Alkaloids
Vincristine	Oncovin	ALL, NHL, Hodgkin disease, CLL, MM	Neurotoxicity, hair loss
Vinblastine	Velban	ALL, NHL, Hodgkin disease, CLL, MM	Myelosuppression
Etoposide	VP-16	NHL, pretransplantation	Myelosuppression, hair loss
Antitumor Antibiotics
Daunorubicin	Daunomycin	AML, ALL	Myelosuppression, cardiotoxicity, N&V, hair loss
Doxorubicin	Adriamycin	ALL, AML, Hodgkin disease, NHL, CLL, MM	Myelosuppression, cardiotoxicity, N&V, hair loss
Bleomycin	Bleo	Hodgkin lymphoma, NHL	Lung toxicity, gastrointestinal toxicity
Idarubicin	Idamycin	AML	Myelosuppression
Antimetabolites
Methotrexate	Amethopterin, MTX	ALL, NHL	Myelosuppression, gastrointestinal toxicity
Ara-C	Cytosine arabinoside, cytarabine	AML, NHL, Hodgkin disease	Myelosuppression, gastrointestinal toxicity, hair loss
Mercaptopurine	6-MP	ALL, CML	Hepatotoxicity, myelosuppression
Thioguanine	6-TG	AML	Myelosuppression
Pentostatin	2′-Deoxycoformycin	Hairy cell leukemia, CLL, lymphomas	Neurotoxicity, myelosuppression
Fludarabine	—	CLL, lymphomas, Waldenström macroglobulinemia	Neurotoxicity, myelosuppression
2-CDA	CDA, 2-chlorodeoxyadenosine, 2-cladribine	Hairy cell leukemia, CLL, lymphomas, AML, Waldenström macroglobulinemia	Neurotoxicity, myelosuppression
Glucocorticoids
Prednisone	—	ALL, CLL	Fluid retention, muscle weakness
Methylprednisolone	—	Hodgkin disease, NHL, AMM	Fluid retention, muscle weakness
Hydrocortisone	—	Hodgkin disease, NHL, AMM	Fluid retention, muscle weakness
Dexamethasone	Decadron	MM	Fluid retention, muscle weakness
Others
Hydroxyurea	Hydrea	CML, PV, AMM	Leukopenia, N&V
Asparaginase	l-Asparagine amidohydrolase	ALL, refractory NHL	Nephrotoxicity, N&V
Cisplatin	Cis-platinum	Solid tumors, Hodgkin disease, NHL	Nephrotoxicity, ototoxicity
Procarbazine	—	Hodgkin disease, NHL	Myelosuppression

ALL, Acute lymphoblastic leukemia; AML, acute myeloid leukemia; AMM, agnogenic myeloid metaplasia; CLL, chronic lymphocytic leukemia; CML, chronic myelogenous leukemia; MM, multiple myeloma; NHL, non-Hodgkin lymphoma; N&V, nausea and vomiting; PV, polycythemia vera.

Targeted Therapy

As more has been learned about the specific genetic lesions that cause cancers, researchers have worked to develop targeted therapies that act specifically on the tumor cell and leave normal cells untouched.¹¹ As a result of these advances in translational research, cancer therapy is realizing the dream of targeted therapeutics and is moving away from nonspecific therapies such as chemotherapy and radiation. The Philadelphia chromosomal abnormality, t(9;22) associated with CML was first reported in 1960. The Philadelphia chromosome results from the juxtaposition of two genes, BCR and ABLl, and the resulting fusion protein has elevated tyrosine kinase activity. It took until almost the year 2000 to develop an agent that specifically targets the product of the Philadelphia chromosome. Imatinib mesylate (Gleevec) is a tyrosine kinase inhibitor that inhibits the kinase activity of all proteins that contain ABL. Imatinib is not specific solely for the BCR/ABLl tyrosine kinase; it also inhibits the tyrosine kinase activity of the c-KIT and the platelet-derived growth factor receptor. It is through its activity on these other receptors that imatinib has found usefulness against other tumors as a targeted therapeutic. In CML, however, imatinib has quickly become the standard treatment for patients newly diagnosed with the disease who are not candidates for immediate transplant (see Chapter 34). Resistance to imatinib has developed in some CML cells, and second-generation, more powerful kinase inhibitors, such as dasatinib and nilotinib, have now been cleared by the FDA. Small-molecule inhibitors such as imatinib are typically available in oral form and have relatively few side effects.

Rather than utilizing small-molecule inhibitors, treatment of lymphoid neoplasms has relied more on monoclonal antibodies for targeted therapeutic strategies. Rituximab specifically targets the CD20 antigen present on many non-Hodgkin lymphoma cells. When the antibody binds the lymphoma cell, complement is activated, and the cell lyses. Other possible scenarios leading to cell death with the use of monoclonal antibodies include antibody-mediated cellular cytotoxicity and stimulation of apoptosis.¹⁰ In contrast to small-molecule inhibitors, monoclonal antibodies generally must be delivered intravenously or subcutaneously. Monoclonal antibodies such as rituximab have evolved from the original monoclonal antibodies that were derived from mice. Rituximab represents a humanized form of an antibody designed or modified to be “more human” so that the patient’s immune system will not raise an antibody against the monoclonal antibody. Additional modifications are now being developed, such as radioactively labeled monoclonal antibodies that can carry a killing dose of radioactivity directly to the mutated cell.