Within this chapter, secondary acute myeloid leukemia (AML) and secondary "advanced myelodysplasia (MDS)" (International Prognostic Scoring System categories intermediate-2 and high) are combined based on the likelihood that these illnesses have similar natural histories12; the somewhat arbitrary criteria used to distinguish them3; and, after account is made for other covariates such as cytogenetics and age, their similar response to therapy.4
Secondary AML/MDS can mean disease that is diagnosed only after a period of abnormal blood counts (antecedent hematologic disorder, or AHD) or that has arisen following cytotoxic therapy for another illness. In the vast majority of cases, these illnesses are other malignancies, although cases developing after treatment of a "nonmalignant" disease (such as rheumatoid arthritis) with drugs such as methotrexate would also qualify. Using the above definition and using more than one month as the criterion for an AHD, one-half of the 1990 patients treated at M.D. Anderson Hospital from 1991 to present for AML/advanced MDS were secondary cases. Of these cases, 708 had an AHD but no prior chemotherapy (PCH), 108 had PCH without an AHD, and 162 had both an AHD and a PCH. The great majority of these patients received ara-C at 1-2 g/m2 either daily for 4 or 5 days or 1.5 g/m2 daily by continuous infusion for 3-4 days combined with idarubicin, fludarabine, topotecan, or troxacitabine. The differences among these regimens are medically insignificant,56 and they are considered together as "standard chemotherapy" (SCH).
Table 7.1 depicts results of SCH in the aforementioned M.D. Anderson patients. Note that the differ ence in compete remission (CR) rate between de novo patients (66%) and the highest CR rate in the three groups of secondary patients (47%) is greater than the difference between the highest (47%) and lowest CR (40%) rates in the three secondary groups. In general, the same is true when considering probabilities of relapse-free survival (RFS) or survival (Table 7.1, Figure 7.1). These observations provide a rationale for considering patients with a history of PCH or an AHD as secondary AML, to be contrasted with de novo cases (Figure 7.2).
Whether the poor results in secondary AML/MDS are due to some inherent, not currently identified, characteristic of this type of AML/MDS or rather to the presence of other covariates is unclear. The principal predictors of outcome with SCH in AML/MDS are performance status, age, and cytogenetics; the former two are associated with treatment-related mortality (TRM) and the latter with resistance to therapy. Of note, each of these is unevenly distributed between de novo and secondary cases (Table 7.2). Patients with de novo disease are on average 10 years younger, but are twice as likely to have a poor performance status (Zubrod 3 or 4). The association between de novo AML/MDS and poor performance status likely reflects the higher white blood cell count and the correspondingly greater risk for organ infiltration in de novo disease. The tendency of secondary cases to have a relatively high proportion of prognostically unfavorable cytogenetic abnormalities (-5/-7) and a relatively low proportion of better prognosis abnormalities [inv(16), t(8;21)] is well known.7 In light of these inequalities, it is reasonable to compare outcome in de novo and secondary patients who share a given karyotype, age range, and performance status. Tables 7.3 and 7.4 do so for
Was this article helpful?