2013. Limitations of the diagnostic criteria for minimally differentiated acute myeloid leukemia (AML-MO). TO THE EDITOR. The FrenchâAmericanâBritish (FAB) ...
Correspondence
2013
Limitations of the diagnostic criteria for minimally differentiated acute myeloid leukemia (AML-MO) TO THE EDITOR The French–American–British (FAB) cooperative group proposed guidelines for diagnosis of AML-M0 in 1991 to standardize the diagnostic criteria.1 The criteria are as follows: negative for myeloperoxidase (MPO) and the Sudan black B reaction, negative for B and T lineage markers, and positive for myeloid markers. These FAB criteria have been used as the standard in diagnosis of AML-M0 since 1991. With advances in molecular biology, the concept of AML-M0 has been changing. Venditti et al2 suggested a new definition of AML-M0 based on the use of monoclonal antibodies specific for cytoplasmic antigens, where they regarded even cases bearing lymphoid antigens as cases of AML-M0, based on positive findings for anti-MPO and negative findings for cytoplasmic CD3 and cytoplasmic CD22. They weighed the expression of cytoplasmic MPO in the diagnosis of AMLM0, and the expression of lymphoid markers is not necessarily evidence against a diagnosis of AML-M0 according to their criteria. However, it remains unknown which of the criteria are most useful for the diagnosis of AML-M0, and any diagnostic criteria of AML-M0 are not comprehensive. We recently experienced a patient who was diagnosed as AML-M0 based upon the FAB criteria.1 A 36-year-old man with high-grade fever was admitted to our hospital in December 1999. The white blood cell count in peripheral blood was elevated to 13.8 × 106/l, of which 71.5% were blastic cells. Bone marrow examination showed that more than 90% of the bone marrow cells were blasts, of which less than 3% were positive for myeloperoxidase (MPO) upon staining. Fluorescence-activated cell sorter (FACS) examination of the bone marrow specimen showed that these cells were positive for CD7, CD13, CD34, CD45, TdT and HLA-DR, but negative for CD2, CD3, CD4, CD5, CD8, CD10, CD14, CD16, CD19, CD20, cCD22, CD33, CD41a and CD56. He was diagnosed as AML-M0 based on the FAB classification.1 He received the standard induction therapy for AML, but to no avail. After the initiation of induction therapy, we further evaluated the characteristics of the leukemic cells. The karyotype of the leukemic cells was normal. Southern blot analysis of the bone marrow showed that neither immunoglobulin heavy chain (IgH) nor T cell receptor (TCR), including the beta, gamma and delta chains, was rearranged. Electron microscopic examination of the leukemic cells showed that the blasts lacked granules with peroxidase activity, but had nuclear pockets, and clustered dense bodies. These findings are characteristic of the pro-thymocytes. FACS analysis using monoclonal antibodies showed that the leukemic cells were negative for cytoplasmic CD22 and cytoplasmic MPO, but cytoplasmic CD3 was highly expressed in these cells. We therefore considered that these leukemic cells might have been committed to the T cell lineage. We initiated anti-ALL reinduction treatment 36 days after the initial chemotherapy. Complete remission was achieved on day 28 of the second induction therapy. He is now receiving consolidation therapy for ALL. Since the proposal of the FAB criteria, nine articles have been published on AML-M0 from six institutions.1–6 The authors modified the FAB criteria in each article (Table 1), and there were some differences in the characteristics of the leukemic cells among the studies (Table 2). For example, the expression of surface lymphoid markers was not necessarily evidence against a diagnosis of AML-M0 in two institutions,2,3 whereas surface lymphoid marker findings had to be negative for diagnosis of AML-M0 in the others.1,4–6 Furthermore, there was a wide variation in the number of surface lymphoid antigens required for the AML-M0 diagnosis in these reports, ranging from four
Correspondence: M Kami, Dept of Hematology, Toranomon Hospital, 2-2-2, Toranomon, Minato-ku, Tokyo 105-8470, Japan; Fax: +81-3-3582-7068 Received 5 June 2000; accepted 21 July 2000
to seven. Consequently, there was a wide variation from 17% to 100% in the incidence of cases positive for cytoplasmic MPO among these studies. Patients who satisfy some of the criteria for diagnosis of AMLM0 may not be diagnosed as having it using other criteria. Although our patient did not satisfy the criteria of Venditti et al,2 he was diagnosed as having AML-M0 based upon the criteria.1 It should be noted that complete remission was achieved only after implementation of anti-ALL therapy. This patient should have been diagnosed and treated as ALL. In general, it may not be significant to determine the lineage of immature leukemic cells, because they frequently express both myeloid and lymphoid antigens due to lineage promiscuity or lineage infidelity. However, our experience has demonstrated that some patients with ALL can be diagnosed as AML-M0 using the FAB criteria. Failure to look for cytoplasmic CD3, especially in a case that is CD7-positive, may involve the risk of missing a case of T-ALL and resulting in a misdiagnosis of AML-MO. Leukemia diagnosis is generally straightforward, but the small minority of cases that do not fit easily into the diagnositc criteria should be worked up as extensively as possible to avoid problems such as illustrated by this case. Examination of the cytoplasmic antigens should be incorporated into a new criteria for AML-M0 such as the WHO criteria to avoid underestimation of cases of AML-M0 and an incorrect diagnosis of ALL. H Iwata1 M Kami1 Y Kishi1 Y Oki1 Y Tanaka2 K Takeuchi3 I Yamazaki4 R Suzuki5 S Morinaga1 Y Mutou1
1
Dept of Hematology, Toranomon Hospital, Tokyo; 2 Dept of Hematology and Oncology and 3 Dept of Pathology, University of Tokyo Hospital, Tokyo; 4 Dept of Clinical Laboratory and Pathology, Hematology, Inoue Memorial Hospital, Chiba; and 5 Laboratory of Chemotherapy, Aichi Cancer Center, Aichi, Japan
References 1 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR, Sultan C: Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML-MO). Br J Haematol 1991; 78: 325–329. 2 Venditti A, Del Poeta G, Buccisano F, Tamburini A, Cox MC, Stasi R, Bruno A, Aronica G, Maffei L, Suppo G, Simone MD, Forte L, Cordero V, Postorino M, Tufilli V, Isacchi G, Masi M, Papa G, Amadori S. Minimally differentiated acute myeloid leukemia (AML-M0): comparison of 25 cases with other French–American– British subtypes. Blood 1997; 89: 621–629. 3 Cohen PL, Hoyer JD, Kurtin PJ, Dewald GW, Hanson CA: Acute myeloid leukemia with minimal differentiation. A multiple parameter study. Am J Clin Pathol 1998; 109: 32–38. 4 Cuneo A, Ferrant A, Michaux JL, Boogaerts M, Demuynck H, Van Orshoven A, Criel A, Stul M, Dal Cin P, Hernandez J et al. Cytogenetic profile of minimally differentiated (FAB M0) acute myeloid leukemia: correlation with clinicobiologic findings. Blood 1995; 85: 3688–3694. 5 Stasi R, Del Poeta G, Venditti A, Masi M, Stipa E, Dentamaro T, Cox C, Dallapiccola B, Papa G. Analysis of treatment failure in patients with minimally differentiated acute myeloid leukemia (AML-M0). Blood 1994; 83: 1619–1625. 6 Yokose N, Ogata K, Ito T, Miyake K, An E, Inokuchi K, Yamada T, Gomi S, Tanabe Y, Ohki I, Kuwabara T, Hasegawa S, Shinohara T, Dan K, Nomura T. Chemotherapy for minimally differentiated acute myeloid leukemia (AML-M0). A report on five cases and review of the literature. Ann Hematol 1993; 66: 67–70. Leukemia
Correspondence
2014
Table 1
Diagnostic criteria of AML-MO in the previous studies since proposal of the FAB criteria
No.
1 2 3 4 5 6
Author/Ref.
Year
No. of patients
Age average (range)
Difference in the diagnostic criteria: lymphoid markers required to be negative
Bennett et al 1 Yokose et al 6 Stasi et al 5 Cuneo et al 4 Venditti et al 2 Cohen et al 3
1991 1993 1994 1995 1997 1998
10 5 20 26 25 17
40.1 (2–79) 43 (32–50) ND ND 60 (27–81) 62 (19–86)
CD2, CD3, CD5, CD10, CD19, cCD22 CD3, CD5, CD10, CD19 CD2, CD3, CD5, CD10, CD19, CD20, CD22 CD2, CD3, CD5, CD10, CD19, CD22 cCD3 and cCD22 none
ND, not described; MPO, myeloperoxidase. CD13 and/or CD33 were required to be positive for the diagnosis of AML-M0 in all the studies. When more than two papers were published from a single institution, we showed the latest one in this table. Table 2
Characteristics of AML-M0 in previous studies since the proposal of the FAB criteria
No. Author/Ref
1 2 3 4 5 6
Bennett et al 1 Yokose et al 6 Stasi et al 5 Cuneo et al 4 Venditti et al 2 Cohen et al 3
Year No. of patients
Karyotype Outcomes abnormal patients who b achieved MPOa CD2 CD3 cCD3 CD7 CD10 CD13 CD14 CD19 CD20 cCD22 CD33 CD34 CD41 HLA-DR TdT (complex)/total remission/total patients
1991 1993 1994 1995 1997 1998
3/10 5/5 9/9 ND 25/25 2/12
10 5 20 26 25 17
Phenotypes
0/10 1/5 0/20 ND 3/25 1/16
0/10 0/5 0/20 ND ND 1/16
ND 5/10 ND 2/5 0/20 9/20 ND 14/26 0/25 8/25 ND 5/16
0/10 7/10 0/5 4/5 0/20 12/20 ND 22/25 3/25 17/25 0/16 16/17
ND 0/5 2/20 0/22 7/25 2/16
0/10 0/5 0/20 ND 2/25 3/17
ND 0/5 ND ND ND 0/17
0/10 ND 0/20 ND 0/25 ND
8/10 2/5 13/20 22/26 16/25 16/17
ND ND 20/20 22/22 24/25 16/16
1/8 0/3 0/20 0/22 ND 0/16
ND 5/5 17/20 ND 22/25 16/16
1/4 4/5 8/15 ND 17/25 14/14
ND 1(1)/5 ND 21(12)/26 16(10)/25 6(6)/14
ND ND 5/14 13/24 7/25 6/17
ND, not described; MPO, myeloperoxidase; CR, complete remission. a Cytoplasmic MPO was identified by either electron microscope or monoclonal antibody. b Patients who showed abnormal karyotype (patients who showed complex karyotype)/total patients examined.
Loss of the TSG101 leucine zipper domain in aggressive non-Hodgkin’s lymphomas
TO THE EDITOR Non-Hodgkin’s lymphomas (NHL) are a heterogeneous group of disease entities with distinctive clinical, morphological, phenotypic, and genetic characteristics. Most aggressive NHLs occur as primary tumors recognized at diagnosis, but they may also develop secondarily from progression of indolent variants. This transformation is associated with a rapidly progressive clinical course and short survival of the patients. The underlying molecular mechanisms involved in the pathogenesis of aggressive lymphomas and the progression of indolent tumors are not well known. The TSG101 tumor susceptibility gene is a new putative tumor suppressor gene involved in different types of murine and human tumors.1 The functional inactivation of the TSG101 gene caused cellular transformation and formation of metastasis in nude mice.1 Alternatively spliced TSG101 non-coding messages, with loss of coding exons and early termination codons, are frequently detected in many carcinomas without alterations at the DNA level. These non-coding alternatively spliced messages are always coexpressed with a normal transcript in tumor cells, and are also detectable in normal cells at a lower frequency.2
Correspondence: PA Lazo, Centro de Investigacio´n del Ca´ncer, CSICUniversidad de Salamanca, Campus Miguel de Unamuno, 37007 Salamanca, Spain; Fax: 34 923 294 743. Received 31 March 2000; accepted 13 July 2000 Leukemia
The normal TSG101 gene codes for a 46-kDa protein with two welldefined domains. The amino terminus has a three-dimensional structure similar to the catalytic domain of the E2 family of ubiquitin conjugating enzyme inhibitors. The E2 family of inhibitors has a blocking effect on the cell cycle in metaphase. This effect on ubiquitination has been recently demonstrated in the case of two TSG101-related proteins, hMMS2 and CROC-1/UEV-1.3 Thus alterations in the ubiquitination process might be of relevance for oncogenesis, since cyclin regulation by ubiquitination has been shown to affect the G1 to S transition and the anaphase block in the cell cycle.4 The TSG101 Cterminus has a coiled-coil domain or leucine zipper1 that permits its binding to transcription factors of the estrogen family of nuclear receptors. Leucine zippers are specific dimerization domains present in many transcription factors. In the case of TSG101, its leucine zipper has been shown to activate the c-fos promoter,3 and inhibits transcription by nuclear receptors of the steroid family.5 All these structural features make the TSG101 protein a candidate to be implicated in oncogenesis. We have recently detected, in 77% of Burkitt’s lymphoma cell lines, the presence of a new alternatively spliced message that codes for a 17-kDa protein (isoform B) that lacks the coiled-coil (leucine zipper) domain, but that partially retains the amino terminus with the ubiquitination inhibitor domain.6 This message has not been detected in normal cells despite the fact that a subpopulation of normal cells is capable of forming alternatively spliced non-coding messages. This isoform B, coded by the new transcript, might have a different function due to the loss of specific protein– protein interactions mediated by the coiled-coil domain (leucine
