Pathology of Therapy-Related Myeloid Neoplasms (t-MNs)

Therapy-related myeloid neoplasms (t-MNs) are defined by the World Health Organization (WHO) as a distinct heterogeneous group of clonal hematopoietic stem cell disorders that are directly related to previous cytotoxic chemotherapy and/or radiation therapy.[1, 2] As with other myeloid neoplasms, therapy-related myeloid neoplasms (t-MNs) demonstrate at least 50% erythroid precursors in the bone marrow.[2]

On the basis of clinical behavior and morphologic features, two predominant and clinically significant types of therapy-related myeloid neoplasms (t-MNs) have been defined, distinguished principally on the basis of the nature of cytotoxic therapy: those arising after treatment with alkylating chemotherapy (eg, cyclophosphamide, chlorambucil, cisplatin) and/or ionizing radiation therapy and those arising after therapy with topoisomerase II inhibitors. These have a relatively long latency period (5-10 y) after the primary exposure. Other drug classes (ie, antimetabolites/immunosuppressants) have been implicated in the development of these neoplasms, but in these cases, the clinical course is less distinct.

On bone marrow aspiration, the morphologic features are those of myelodysplasia. Patients who are exposed to topoisomerase II inhibitors (eg, etoposide, doxorubicin) tend to present with frank leukemia within 1 year of the time primary therapy was initiated (see the following image).

Pathology of therapy-related myeloid neoplasms. Acute myeloid leukemia from a patient who presented 10 months after therapy with topoisomerase II inhibitor.

Subtypes of therapy-related myeloid neoplasms (t-MNs) share diagnostic features of conventionally defined myeloid malignancies and include the following[2] :

• Myelodysplastic syndromes (t-MDS)

• Acute myeloid leukemia (t-AML)

However, the therapy-related myeloid neoplasms neoplasms (t-MNs) progress quickly regardless of their morphologic appearance at presentation and are considered to be a single diagnostic entity. The associated cytogenetic abnormality is important for determining therapy and prognosis and should be identified in the final diagnosis.[2]

Note that there have been reports of therapy-related myeloid neoplasms neoplasms (t-MNs) associated with germ-line mutations in cancer susceptibility genes; therefore, clinicians should carefully review family histories to identify cancer susceptibility in individuals with therapy-related myeloid neoplasms neoplasms (t-MNs).[2]

See also Pathology of Acute Myeloid Leukemia With Myelodysplasia-Related Features, Pathology of Other Myeloid Related Precursor Neoplasms, and Pathology of Acute Myeloid Leukemia Not Otherwise Categorized.

Epidemiology

In general, myelodysplastic syndrome (MDS) has an incidence of 5 new MDS diagnoses per 100,000 people, with men affected more than women.[3]  In Western nations, the incidence is 22-45 per 100,000 people older than 70 years, whereas Asian countries (eg, Japan, China, Korea, India, Thailand,Turkey) report MDS at younger ages (median: 40-50 years).[3] The differences between these regions may be due in part to environmental pollutions and/or other factors (eg, uncontrolled pesticide use).

The incidence of therapy-related myeloid neoplasms (t-MNs) is dependent on the type, dose, and intensity of the therapeutic intervention and on the nature of the underlying primary malignancy/disease process. All age groups are affected; however, the risk of developing secondary myeloid neoplasms following alkylating chemotherapy or radiation therapy seems to increase with age. The risk of developing therapy-related myeloid neoplasms (t-MNs) dramatically decreases after 10 years.[4]

An extensive review and analysis of previously published data highlighted vague trends, with the greatest likelihood of developing therapy-related myeloid neoplasms (t-MNs) following treatment of hematopoietic malignancies.[5] However, there was also significant variability between studies, and the precise incidence is not clear. Approximately 30% of therapy-related myeloid neoplasm (t-MN) cases involve patients treated for nonneoplastic disorders, and those treated with high-dose chemotherapy followed by autologous stem cell transplantation.[6, 7] Significantly, cases of therapy-related myeloid neoplasms (t-MNs) represent approximately 10%-30% of all confirmed cases of (MDSs), acute myeloid leukemia (AML), and myelodysplastic/myeloproliferative neoplasms (MDS/MPN).[5, 6]

The heritable risk factors predisposing to the development of therapy-related myeloid neoplasms (t-MNs) are the topic of intense study. Defects in the RAS-BRAF signal-transduction pathway, point mutations of AML1 and p53, and polymorphisms affecting drug metabolism have been implicated.[8, 9]

The clinical presentation of therapy-related myeloid neoplasms (t-MNs) is largely dependent on the nature of the antecedent therapeutic regimen. Patients treated with alkylating agents and/or radiation therapy typically present 5-10 years after exposure; the latency period for patients treated with topoisomerase inhibitors is on the order of 1-5 years.

Despite the archetypal and often distinguishing morphologic features of these two groups, the clinical presentations are nonspecific and are related to the degree of bone marrow failure (see Morphologic Features). Clinical features are generally related to the corresponding cytopenia (eg, bleeding with thrombocytopenia; fatigue and exercise intolerance with anemia; fever and infection with neutropenia).

Most commonly, the morphologic features of alkylating agent/radiotherapy–associated therapy-related myeloid neoplasms (t-MNs) are those of myelodysplasia, in particular, myelodysplastic syndromes (MDS) with multilineage dysplasia (MLD) (MDS-MLD) (formerly refractory cytopenia with multilineage dysplasia [RCMD]). Evaluation of the peripheral blood smear demonstrates one or more cytopenias, often including a macrocytic anemia. Platelets may show nonspecific variation in size; giant forms may be present. Dysplasia of the granulocyte lineage may be assessed on the peripheral blood smear; hypolobated nuclei and hypogranularity are common features.

The bone marrow is most frequently normocellular to hypercellular with a variable blast percentage. The erythroid series may show the full breadth of dysplasia, including nuclear and cytoplasmic atypia (see the first image below), as well as megaloblastoid features. Ring sideroblasts are often present and may be identified with iron stains (eg, Prussian blue) (see the second image below).

Pathology of therapy-related myeloid neoplasms. Nucleated erythroid precursors with cytoplasmic vacuolization.

Pathology of therapy-related myeloid neoplasms. Ring sideroblast.

Megakaryocyte dysplasia is manifest as hypolobate or hyperlobate forms, megakaryocytes with nuclear segregation (see the first image below), and/or microcytic forms with abnormal paratrabecular localization and clustering (see the second and third images below). As with peripheral blood, the bone marrow granulocyte lineage may show variation and abnormalities of granulation and nuclear hypolobation of the mature forms (ie, pseudo–Pelger-Huet anomaly) (see the fourth image below).

Pathology of therapy-related myeloid neoplasms. Abnormal megakaryocytes with segregated nuclei.

Pathology of therapy-related myeloid neoplasms. Micromegakaryocyte.

Pathology of therapy-related myeloid neoplasms. Bone marrow biopsy specimen showing erythroid hyperplasia and clusters of abnormal megakaryocytes.

Pathology of therapy-related myeloid neoplasms. Pseudo–Pelger-Huet cell: granulocyte with bilobed nucleus attached by thin band of chromatin/nuclear membrane.

Cases of therapy-related myeloid neoplasms (t-MNs) related to treatment with topoisomerase II inhibitors typically present as overt acute myeloid leukemia (see the following image). Morphologic features are not unique in therapy-related cases. Although monocytic and monoblastic differentiation is often present, the appearance may be that of de novo cases, including those with recurrent cytogenetic abnormalities. Cases of therapy-related acute lymphoblastic leukemia (ALL) have also been described.[10]

Pathology of therapy-related myeloid neoplasms. Acute myeloid leukemia from a patient who presented 10 months after therapy with topoisomerase II inhibitor.

Importantly, the two treatment-related groups cannot always be distinguished on the basis of histologic findings; a complete clinical history, in addition to cytogenetic studies, is necessary to accurately diagnose and prognosticate these neoplasms.[2] The morphologic findings are not entirely specific and are rarely distinguishable from those of de novo cases of MDS, acute myeloid leukemia (AML), or myelodysplastic/myeloproliferative neoplasms (MDS/MPN). In contrast to de novo myeloid neoplasms, however, there is no significant clinical or prognostic benefit to further subclassification.[11]

Immunophenotypic studies are not used to distinguish therapy-related myeloid neoplasms (t-MNs) from de novo cases of myeloid neoplasms but rather to clarify abnormal populations. The myeloblasts are characteristically CD34-positive; their distribution and burden may be demonstrated with CD34 immunohistochemical studies.

Flow cytometric analysis may be helpful in assessing the proportion of myeloid blasts, as well as aberrant antigenic expression. Aberrant expression of CD7 (a T-cell–associated antigen), CD56 (a natural killer [NK]-cell–associated antigen), and/or CD19 (a B-cell–associated antigen) may be seen in therapy-related myeloid neoplasm (t-MN) blasts. The relevance of such findings and of associated cytogenetic changes is similar to that in de novo cases. Immunophenotypic maturation patterns of the myeloid and erythroid lineages may also be evaluated; they may be particularly helpful when morphologic and cytogenetic evaluations of the bone marrow are inconclusive (see Morphologic Features and Cytogenetic Features).[12, 13]

Therapeutic advances have dramatically influenced the clinical course of numerous malignant, reactive, and autoimmune disorders. However, the therapies themselves can be mutagenic.

Cytogenetic abnormalities are present in the majority of cases of therapy-related myeloid neoplasms (t-MNs) and are related to the therapeutic regimen.[6, 7] Conventional cytogenetic analysis (ie, karyotype) and fluorescence in situ hybridization (FISH) studies for specific abnormalities may be helpful in the diagnosis and prognostication of disease.

Alkylating agents and radiation therapy

Cases of therapy-related myeloid neoplasms (t-MNs) due to treatment with alkylating agents and radiation characteristically show unbalanced translocations and partial or complete loss of chromosomes 7 and/or 5. Monosomy 7 is the most common finding associated with alkylating agents; loss of the long arm of chromosome 5 (5q-) is the most common finding associated with ionizing radiation. Additional chromosomal abnormalities are common, yielding a complex karyotype.[6]

Topoisomerase II inhibitors

Cases of therapy-related myeloid neoplasms (t-MNs) arising after use of topoisomerase II inhibitors are associated with a short latency to presentation. Balanced translocations are also seen in these cases. Frequent rearrangements of 11q23(MLL) and 21q22(AML1) are identified. Balanced translocations seen in de novo cases of acute myeloid leukemia (AML) are also present, including t(15;17)(q22;q21), inv(16)(p13.q22), and t(8;21)(q22;q22), although they are less common than in de novo cases.[6, 7]

Although therapy-related myelodysplastic syndromes (t-MDS), acute myeloid leukemia (t-AML), and myelodysplastic/myeloproliferative neoplasms (t-MDS/MPN) have morphologically disparate phases, rapid progression and shortened survival are common in all cases, irrespective of subclassification.[11] Overall survival is poor (< 10% at 5 y).[1, 14]  Therapy-related myeloid neoplasms (t-MNs) account for 10%-20% of all the cases of MDS, AML, and MDS/MPN.[3]

As with other myeloid neoplasms, the cytogenetic features of therapy-related myeloid neoplasms (t-MNs) are predictive of outcome. However, despite the morphologic and cytogenetic similarities of some therapy-related myeloid neoplasms (t-MNs) and de novo myeloid neoplasms, therapy-induced disease typically carries a worse prognosis than its de novo counterpart.[15] Balanced translocations are associated with a relatively better prognosis, whereas loss of the entire or partial chromosomes 5 or 7 is associated with a particularly dismal prognosis.[16]

As seen in de novo cases of AML, favorable cytogenetic findings, including t(15;17)(q22;q12) and inv(16)(p13.1q22), confer a relatively better prognosis; however, this is not universal. A study that compared t(8;21) t-AML to de novo t(8;21) AML demonstrated that despite virtually indistinguishable morphologic, immunophenotypic, and cytogenetic features, the median overall survival was dramatically worse in therapy-related cases.[17]

More recently, a study has found that although the mutational characteristics of therapy-related myeloid neoplasms (t-MNs) are similar to that of primary MDS, there is a distinct pattern to the former, as follows[18] :

• Therapy-related myeloid neoplasms (t-MNs): TP53 mutations were more frequent in T-MN (29.5 vs 7%), a poor prognostic factor, and detected in 92% of cases with at least 15% of ring sideroblasts (RS) and SF3B1 wild-type cases; RS phenotype not associated with better survival and  SF3B1 was mutated in 32% of cases with at least 15% RS
• MDS: Spliceosomal complex mutations were more common in P-MDS (56.5 vs 25.6%); ring sideroblasts phenotype associated with better survival and SF3B1 was mutated in 96% of primary MDS with at least 15% RS

The revised  International Prognostic Scoring System (IPSS-R) for  "very low" and "low" was similar in biological and clinical characteristics for both groups.[18]