, Volume 73, Issue 1, pp 93-99

Morphological diagnoses of the Japan Adult Leukemia Study Group acute myeloid leukemia protocols: Central review

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


A morphological review system of the Japan Adult Leukemia Study Group has developed from the AML-87 through the AML-92 experience.We reviewed 1427 (90%) of 1592 cases enrolled in the AML-87, -89, or -92 protocols for morphology; 1408 (88%) were eligible. The rate of diagnostic concordance between each institute and the Committee on Morphological Diagnosis ranged from 76% to 80%. Acute myeloid leukemia (AML) subtypes were as follows: AML M0, 27 (2%); M1, 179 (13%); M2, 472 (34%); M3, 358 (25%); M4, 265 (19%); M5, 57 (4%); M6, 39 (3%); and M7, 11 (1%). The reason for the high number of patients with AML M3 is that many M3 patients were enrolled in the AML-92 protocol, which contained all-trans-retinoic acid. AML M0, M6 and M7 belonged to the poor prognostic groups. Auer bodies were found in 284 (53%) of 538 patients who survived significantly longer than those without Auer bodies in AML-87/-89. In AML-92 except for AML M3, 259 (43%) of 602 cases were Auer+ and also showed better survival rates. The survival of patients with >50% myeloperoxidase (MPO)-positive blast cells was better than those with ⩽50% MPO+ blast cells in AML-87/-89. This trend was also seen in AML-92 excluding M3. AML with trilineage dysplasia (AML/TLD) is characterized as a subtype of de novo AML that shows morphological dysplasia of mature hematopoietic cells on a background of leukemic blast cells.The number of patients with AML/TLD was 89 (16.5%) of 545 patients reviewed in AML-87/-89. AML-92, except for M3, showed a higher rate of patients with TLD (161 cases; 27.6%) because there were no patients with TLD in the AML M3 group. Survival rates for AML/TLD were worse than those for AML/non-TLD in both the AML-87/-89 and -92 protocols. Eighty percent of all cases (793/986) entered in AML-92 were analyzed cytogenetically. Fifty-one cases were not available for karyotyping because of a lack of mitoses or inappropriate preparations.The most frequent karyotype was normal, which accounted for 34.2%.The t(15;17), t(8;21), and inv(16) karyotypes, which are regarded as good risk factors, were 23.8%, 9.2%, and 1.6%, respectively. Abnormal chromosomes 5, 7, t(9;22), and t(6;9) were considered to be poor or intermediate risk factors. As a new system of karyotyping begins in the ongoing AML protocol, useful chromosomal data will be obtained in the near future.