Prognostic Significance of Unbalanced Chromosome Abnormalities Used by 2008 World Health Organization (WHO) Classification to Define “Acute Myeloid Leukemia (AML) with Myelodysplasia-Related Changes” in Adults: a Cancer and Leukemia Group B (CALGB) Study.

One of the major revisions in the 2008 WHO classification of “AML and Related Precursor Neoplasms” is the use of certain cytogenetic abnormalities (abns) as a criterion for inclusion in the “AML with myelodysplasia-related changes” category. These abns include 9 specific reciprocal translocations, 8 unbalanced abns, and complex karyotypes (CK), ie, ≥3 unrelated abns. The clinical features and outcome of patients (pts) with these abns require further study to confirm the appropriateness of their inclusion in this WHO category. Therefore, we evaluated 2,724 consecutive untreated adults meeting criteria for possible inclusion in this WHO category (ie, non-therapy-related AML and not part of the first WHO AML category) in the CALGB cytogenetics database; 516 (19%) pts harbored ≥1 myelodysplasia-related abn and/or CK, and had outcome data available. Their median age was 62 years (y; range, 15–88 y). The 9 reciprocal translocations were very rare, with t(3;5)(q25;q34) found in 6 pts, t(3;21)(q26.2;q22.1) in 2, t(1;3)(p36.3;q21.1) and t(2;11)(p21;q23) in 1 pt each and the remaining 5 translocations not detected. Their rarity precluded further analyses. With the exception of idic(X)(q13), found in 2 pts, the remaining 7 unbalanced abns were more common (Table). Importantly, the abns were not mutually exclusive, ie, ≥2 different abns could co-exist in the same karyotype and/or be part of a CK. Among 453 pts with ≥1 specific unbalanced abn (Table), 62% had CK; the highest % of CK, 89–95%, were seen in the i(17q) or t(17p), −13 or del(13q) and −5 or del(5q) groups and the lowest, only 24%, in del(9q) pts. As a group, non-CK pts had a higher complete remission (CR) rate (P=.002) and longer overall survival (OS; P<.001) than CK pts. This was also the case for OS for most specific abns (Table). Notably, within the non-CK −7 or del(7q) group, −7 pts had worse outcome than del(7q) pts (CR rates, P=.09; OS, P=.002), suggesting that the −7 or 7q- category is not uniform prognostically. Strikingly, pts with del(9q) were younger than pts with other specific myelodysplasia-related unbalanced abns [median age of del(9q) v median age of all others combined, 42 v 63 y, P<.001], and their outcome was better [del(9q) v all others combined: CR rates, 90% v 43%, P<.001; OS, P<.001, 3-y rates, 31% vs. 8%, Figure]. Among 342 pts with CK, those with ≥1 specific myelodysplasia-related unbalanced abn (n=281) had shorter OS than pts without any such abn (n=61; P<.001; 3-y rate, 2% v 20%). We conclude that 1) for most specific myelodysplasia-related unbalanced abns, CK pts do worse than non-CK pts; 2) the presence of specific abns adversely impacts on outcome of CK pts; 3) −7 or 7q- category is not uniform prognostically; 4) del(9q) pts are younger, much less often have CK and their outcome is better than outcome of pts with other unbalanced myelodysplasia-related abns. Consequently, pts with del(9q) appear biologically and clinically different, and their inclusion in the WHO “AML with myelodysplasia-related changes” category should be reconsidered. Future molecular genetic analyses should help characterize each of the cytogenetic subsets within this WHO category further, and might become useful for guiding treatment.

View largeDownload slide

View largeDownload slide

Close modal

No relevant conflicts of interest to declare.