ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 13, No. 6 Copyright © 1983, Institute for Clinical Science, Inc.

Acute Presenting with Myeloid and Lymphoid Cell Markers* DAVID O. SCAMURRA, M.D.,t FREDERICK R. DAVEY, M.D.,f:j: DOUGLAS A. NELSON, M.D.,t ANTHONY S. KUREC, B.S.,t and JACK GOLDBERG, M.D4 Departments of Pathology f and Medicine, $ SUNY Upstate Medical Center, Syracuse, NY 13210

ABSTRACT A case is described of acute leukemia whose neoplastic cells possessed myeloid and lymphoid characteristics. Neoplastic cells possessed cyto­ plasmic granules containing Sudan black B material and diaminobenzidine myeloperoxidase. In addition, these leukemic cells were positive for ter­ minal deoxynucleotidyl transferase, la antigen, and the common acute lymphocytic leukemia antigen. These findings indicate that biphenotypic cell markers may exist in cases of acute leukemia. It is our belief that these results are best explained as either a mixed myeloid-lymphoid leukemia or a stem cell leukemia capable of differentiating into myeloid and lym­ phoid cells.

Introduction and/or immunologic characteristics of both myeloid and lymphoid cell lines.16 The appearance of acute myeloid leu­ The purpose of this report is to present kemia (AML) as a second malignancy has morphologic, cytochemical, and immu­ been reported with increased frequency nologic studies on the leukemic cells in patients with ovarian carcinoma, poly­ from a patient with acute leukemia. In cythemia vera, Hodgkin’s disease, and this case, the leukemic cells exhibited multiple myeloma.1,12,24’25’30 Occasion­ characteristics of both acute myeloid and ally, cases of AML following non-Hodg­ acute lymphoid leukemia at the time of kin’s , chronic lymphocytic presentation before any chemotherapy leukemia, and acute lymphocytic leu­ was given to the patient. kemia have occurred.21,22,27 Even more rare are cases of acute leukemia with neoplastic cells exhibiting cytochemical Case Report A 59 year old woman was admitted to the hospital because of marked fatigue, a 2 0 pound weight loss, * Mailing Address for reprints: Frederick R. and bleeding from the nose and mouth. Her physical Davey, M.D., Division of Clinical Pathology, SUNY examination revealed a left facial nerve palsy, pallor, Upstate Medical Center, 750 East Adams Street, epistaxis, and gingival bleeding. There was no Syracuse, NY 13210. lymphadenopathy or organomegally. A roentgeno­ 496 0091-7370/83/1100-0496 $01.20 © Institute for Clinical Science, Inc. ACUTE LEUKEMIA 497 graph of the chest was normal. Pertinent laboratory nonimmune rosettes with sheep eryth­ values were as follows: leukocyte count 31.6 x 103 per (jlI with a differential count of 75 percent blasts, rocytes (E-rosettes)17 and to contain sur­ rare promyelocyte and myelocyte, six percent bands, face and cytoplasmic immunoglobulins.26 nine percent neutrophils and 1 0 percent lympho­ cytes. There were five nucleated red blood cells per Leukemic cells were examined for the 100 leukocytes. The bone marrow aspirate films con­ presence of Ia-like antigen by indirect tained 92 percent blasts and the biopsy was hyper- immunofluorescence.34 To detect Fc cellular. A diagnosis of acute leukemia was made and the patient received induction therapy consisting of receptors, the malignant cells were first daunorubicin, cytosine arabinoside, and vincristine incubated with aggregated human gam­ administered intravenously, and prednisone given maglobulin and subsequently with fluo­ orally. Leukemic cells were also identified in the spinal fluid samples, and methotrexate was adminis­ rescein labeled antihuman IgG.7 The tered intrathecally. Following therapy, the patient presence of terminal deoxynucleotidyl developed pancytopenia and the bone marrow aspi­ transferase (TdT) was determined by in­ rate and biopsy were hypocellular. Unfortunately, the patient developed bacterial septicemia and, de­ direct immunofluorescence using fluo­ spite antibiotic therapy, died 16 days after the start rescein-labeled goat anti-rabbit IgG after of chemotherapy. previous incubation of the cells with rabbit anti-calf TdT serum.32 The tumor Materials and Methods cells were also examined for the presence of human T-cell antigen using a direct A 500 cell differential count was per­ immunofluorescence technique. formed on each bone marrow aspirate previously stained with Wright-Giemsa Results reagents. H istopathology a n d C ytology C ytochemical S tains The bone marrow aspirate and biopsy Bone marrow aspirates and peripheral showed blasts measuring 12 to 20 |x in blood air dried smears were stained for diameter and possessing irregularly the presence of 3,3'-diaminobenzidine shaped nuclei. The nuclear to cyto­ (DAB) peroxidse,5 neutral lipids (Sudan plasmic ratio was moderately high re­ black B),31 alpha-naphthyl acetate es­ vealing only a slight amount of light blue terase,19 alpha-naphthyl butyrate es­ cytoplasm. The nuclear chromatin was terase,35 acid phosphatase,20 and periodic fine and one to two small nucleoli were acid-Schiff (PAS) material.13 The per­ present in the majority of leukemic cells centage of positive cells was calculated (figure 1). from a 200 cell count. C ytochemical S t u d ie s I m m u n o l o g ic S t u d ie s Thirty-nine percent of leukemic cells Ten milliliters of peripheral blood from the bone marrow aspirate were pos­ were obtained in preservative-free hep­ itive for DAB-peroxidase reaction and 31 arin. Mononuclear cells were separated percent were positive for Sudan black B from other blood elements by centrifu­ (figure 2). Aüer rod like structures were gation and a gradient composed of Ficoll observed in both the DAB peroxidase and diatrizoate salts.* Leukemic cells and Sudan black B stained preparations. were evaluated for their ability to form Ninety percent of the cells were positive for acid phosphatase. Leukemic cells * Separation Medium, Litton Bio- were negative for PAS and alpha-naph­ netics, Kensington, MD. thyl acetate esterase (table I). 498 SCAMURRA, DAVEY, NELSON, KUREC, AND GOLDBERG

F i g u r e 1. Bone mar­ row aspirate smear re­ vealing numerous blasts. Some cells have mod­ erate amount of cytoplasm while others have only a thin rim of cytoplasm. Nu­ cleoli are present in sev­ eral cells (Wright-Giemsa x 1 0 0 0 ).

I m m u n o l o g ic M a rk er S t u d ie s Discussion Ninety percent of leukemic cells were In the current case, leukemic cells positive for TdT (figure 3), 84 percent were positive for DAB myeloperoxidase, positive for the common ALL antigen Sudan black B as well as for TdT, and 96 percent positive for Ia-like an­ common ALL antigen, and la antigen. tigen. Leukemic cells were negative for Myeloperoxidase activity is present at all other immunologic markers (table I). stages of neutrophil development and is

F i g u r e 2. Blasts stained for the presence of DAB myeloperoxidase. Note the positive reaction product at the end of pointer (x 1 0 0 0 ). ACUTE LEUKEMIA 499 TABLE I acute megakaryocytic leukemia.9 Myelo­ Summary of Cytochemical and Immunologic peroxidase activity is negative in blasts Marker Studies of Leukemic Cells from cases of ALL. Flandrin and Daniel8 indicate that complete absence of myelo­ Positive Leukemic Assay peroxidase activity is the best practical Cells (Percent) criterion for diagnosis of ALL. Sudan black B 31 Sudan black B stains a variety of lipids DAB-Myeloperoxidase 39 Periodic-acid Schiff negative including neutral fats, phospholipids, Alpha-naphthyl negative and sterols.31 In leukocytes, Sudan black butyrate esterase Alpha-naphthyl negative B positive material is usually present in acetate esterase granulocyte precursors and immature Acid phosphatase 90 (diffuse) E rosettes 4 monocytes but not in lymphoid cells. To Surface immunoglobulins 3 our knowledge, there are only two re­ Cytoplasmic immunoglobulins 1 Ia-like antigen 96 ported cases of sudanophilia in lympho­ Fc receptors 9 blasts collected from cases of ALL.14,33 Terminal deoxynucleotidyl 90 transferase from these two cases were antigen 6 Common ALL antigen 84 positive for common ALL antigen, la an­ Monocyte antigen 2 tigen, and TdT. In normal tissues, TdT positive cells are observed in approximately 80 per­ cent of cortical thymocytes and two to localized to the nonspecific granules in five percent of bone marrow cells.3 Also, the myeloid cell cytoplasm.918 Myelope­ TdT is present in the majority of neo­ roxidase activity is best visualized by plastic cells in 90 percent of cases of ALL using 3,3'-diaminobenzidine (DAB) as and in most cases of lymphoblastic lym­ substrate5 and DAB myeloperoxidase ac­ phoma.3 In addition, TdT is positive in tivity is usually present in blasts from blasts from approximately one-third of cases of acute myelogenous leukemia, the cases of chronic myelogenous leu­ acute myelomonocytic leukemia, and kemia in blastic crisis.15 Approximately

F i g u r e 3 . Immuno­ fluorescence of TdT posi­ tive cell in bone marrow aspirate smear (X 1 0 0 0 ). 500 SCAMURRA, DAVEY, NELSON, KUREC, AND GOLDBERG six percent of cases of acute myelogenous possessed characteristics of both myeloid leukemia have an elevated number of and lymphoid cells. The exact interpre­ TdT positive cells. In some of these tation of these findings suggests at least cases,15 features of both myeloid and four explanations. lymphoid (i.e., mixed my- The first possibility is that the patient eloid-lymphoid leukemias) are present may have been in the blastic phase of with TdT negativity occurring in myeloid chronic myelogenous leukemia. Al­ cells and TdT positivity in lymphoid though no karyotype was performed in cells. However, in a few cases, TdT ac­ this patient, the presentation and past tivity was demonstrated in cells which history make this an unlikely possibility. were myeloid by other criteria4 sug­ In addition, blasts from patients in blastic gesting an aberrant TdT expression in phase of chronic myelogenous leukemia malignant myeloid cells. are not simultaneously positive for DAB The la (HLA-D, DR) antigens are myeloperoxidase, Sudan black B, TdT, present on stem cells, B , and the common ALL antigen. monocytes, and activated T lympho­ As a second possibility, derepression of cytes.36 However, the la antigen is a marker genome of leukemic cells from present on neoplastic cells from most either myeloid or lymphoid origin could cases of acute leukemia, chronic lym­ allow the expression of an antigen or en­ phoid leukemias and non-Hodgkin’s lym­ zyme of the opposite line. Bettelheim et phomas of lineage.23 Thus the al2 demonstrated the expression of a my­ presence of la antigen on the neoplastic eloid-specific antigen (VIM-DS) on TdT cells from the current case is of little help positive blastic cell populations from two in determining the lineage of these leu­ cases of childhood ALL. Furthermore, kemic cells. leukemic cells from one of their cases In normal adult individuals the com­ possessed la and common ALL antigens. mon ALL antigen is present on two per­ These investigators2 suggested that the cent of bone marrow cells and normal manifestation of the VIM-DS antigen was neutrophils.11,29 In the fetal liver, five to the result of a derepression of the ge­ 10 percent of cells are positive for the nome in ALL blasts. Genome derepres­ common ALL antigen.10 Normal periph­ sion is probably not operative in our case eral blood T and B lymphocytes, null because two myeloid (SBB and DAB cells, immature granulocytes, mono­ myeloperoxidase) and two lymphoid cytes, or platelets do not possess the (TdT and common ALL antigen) markers common ALL antigen.6 The common exist on these leukemic blasts. Thus, ALL antigen is usually present on blasts more than one genome would need to be from non B, non T ALL, pre B cell ALL, derepressed and this is an unlikely some cases of undifferentiated acute leu­ event. kemias, 40 percent of cases of chronic Thirdly, this case may represent a true myleogenous leukemia in blastic crisis, mixed myeloid-lymphoid leukemia with and a few cases of non-Hodgkin’s lym­ two separate malignant clones, a smaller phoma. However, the common ALL an­ (30 to 40 percent) population of myelob­ tigen is virtually absent in cases of acute lasts and a larger (60 to 70 percent) group myeloid leukemia.11,23 Thus, the pres­ of lymphoblasts. The apparent overlap of ence of the common ALL antigen on the myeloid and lymphoid characteristics on neoplastic cells from our patient suggests 20 to 30 percent of the blasts could be a lymphoid origin to at least a subpopu­ an artifact secondary to the counting of lation of the leukemic cells. only 200 cells for each marker assay. In the current report, neoplastic cells Finally, it is possible that multiple my­ ACUTE LEUKEMIA 501 eloid and lymphoid markers exist on a Absence of common ALL on normal bipotent myeloid, erythroid, and granulocyte progeni­ common myeloid-lymphoid stem cell tors. Blood 58:333-336, 1981. which has not yet been fully character­ 7. D ic k l e r , H. B. and K u n k e l , H. G . : Interaction ized. Perentesis et al28 described a five with the aggregated 7 globulin with B lympho­ cytes. J. Exp. Med. ¿36:191-196, 1972. year old girl with acute leukemia. Her 8. F l a n d r in , G . and D a n ie l , M. T.: Practical value blasts initially appeared lymphoid and of cytochemical studies for the classification of they contained numerous surface anti­ acute leukemias. Recent Results in Cancer Res. 43:43-56, 1973. gens typical of ALL. However, after six 9. F l a n d r in , G . and D a n ie l , M. T.: Cytochemistry days of therapy, the malignant cell had a in the classification of leukemias. The Leukemia myelomonocytic morphology. Although Cell. Catovsky, D ., ed. Edinburgh, Churchill Livingstone, 1981, pp. 2 9 -4 8 . new surface antigens were demonstrated 10. G rea v es, M ., D e l ia , D ., J anossy, G ., e t a l .: on the myelomonoblasts, some of the Acute lymphoblastic leukaemia associated an­ monoclonal which reacted tigen. IV. Expression on non-leukaemic “lym­ phoid” cells. Leukemia Res. 4:15-32, 1980. with the original lymphoid blasts con­ 11. G rea v es, M . F., H a r ir i, G ., N e w m a n , R. A., et tinued to react with new myelomono­ a l . : Selective expression of the common acute cytic blasts. These findings argue in favor lymphoblastic leukemia (gplOO) antigen on im­ mature lymphoid cells and their malignant coun­ of the existence of a stem cell leukemia terparts. Blood 61:628—639, 1983. with multiple myeloid and lymphoid cell 12. H a rris, C. C.: A delayed complication of cancer characteristics. The present authors be­ therapy. J. Natl. Cancer Inst. 63:275-277, 1979. 13. H a y h o e, F. G . J. and F l em a n s, R. J. : An Atlas lieve that the immunologic and cyto- of Haematology Cytology. London, Wolf Med­ chemical data from the current case ical Books, 1969, p. 316. could best be explained by either of the 14. Ho, F. C . S., C h a n , G . T. C ., and T o d d , D.: Non-specificity of Sudan black B in the diagnosis latter two hypotheses. However, it of acute myeloid leukaemia. Brit. J. Haematol. seems evident that further ontogenic, 53:171-172, 1983. immunologic and cytochemical studies 15. H o f f b r a n d , A. V., G a n e s h a g u r u , K ., L l e w - e l l in , P., and J anossy, G .: Biochemical markers on additional cases are needed to estab­ in leukemia and lymphoma. Recent Results in lish the exact mechanism of the biphen- Cancer Res. 69:25-39, 1979. otypic markers in these cases of acute 16. H u l l , M. T. and G r ie p , J. A.: Mixed leukemia: Lymphatic and myelomonocytic. Am. J. Clin. leukemia. Path. 74:47 3, 1980. 17. J o n d a l , M., H o l m , G ., and W ig z e l l , H.: Sur­ face markers on human T and B lymphocytes. I: References A large population of lymphocytes forming non- immune rosettes with sheep red blood cells. J. 1. A u c l e r c , G., J a c q u il l a t , C., A u c l e r c , M . F., Exp. Med. ¿36:207-215, 1972. e t a l . : Post-therapeutic acute leukemia. Cancer 18. K a pl o w , L. S.: Substitute for benzidine in my­ 44:2017-2025, 1979. eloperoxidase stain. Am. J. Clin. Path. 63 :45 1, 2. B e t t e l h e im , P ., P a ie t t a , D., M a jo h e , O., e t a l . : 1975. Expression of a myeloid marker on TdT-positive 19. L i, C. Y. and L a m , L . T.: Esterase in human leu­ acute lymphocytic leukemia cells: Evidence by kocytes. J. Histochem. Cytochem. 2 7 :1 - 1 3 , double-fluorescence staining. Blood 60:1392- 1973. 1396, 1982. 20. L i, C. Y., Y a m , L . T., and L a m , K. W: Acid 3. B e u t l e r , E. and B l u m e , K. G.: Terminal deoxy- phosphatase isoenzyme in human leukocytes in nucleotidyl transferase: Biochemical properties, normal and pathological conditions. J. Histo­ cellular distribution, and hematologic signifi­ chem. Cytochem. ¿8:473—481, 1970. cance. Prog. Hematol. 11:47-63, 1979. 21. M a d o f f , L ., D avey, F. R ., G o r d o n , G . B., et 4. Bradstock, K. F., Hoffbrand, A. V., G a n e s h a - a l . : The development of acute myelomonocytic g u r u , K., e t a l .: Terminal deoxynucleotidyl leukemia in a patient with acute lymphocytic transferase expression in acute non-lymphoid leukemia. Cancer 48:1157-1163, 1981. leukaemia: An analysis by immunofluorescence. 22. M c P h e d r a n , P. and H e a t h , C. W. : Acute leu­ B rit. J. H a e m a to l. 47:133-143, 1981. kemia occurring during chronic lymphocytic leu­ 5. C a to v s k y , D., de Salvo Cardullo, L., O ’B r ie n , kemia. Blood 35:7-11, 1970. M ., e t a l . : Cytochemical markers of differentia­ 23. N a d l e r , L. M., R it z , J., G r if f in , J. D., e t a l . : tion in acute leukemia. Cancer Res. 4i:4824- Diagnosis and treatment of human leukemias 4832, 1981. and utilizing monoclonal antibodies. 6 . C l a v e l l , L. A., L ip t o n , J. M., a n d B a st , R. C.: Prog. Hematol. 12:187-225, 1981. 502 SCAMURRA, DAVEY, NELSON, KUREC, AND GOLDBERG

24. N e w m a n , D. R., M a ld o n a d o , J. E., H a rriso n , kemia. Report of eight cases and review of the E. G., e t a l . : Myelomonocytic leukemia in literature. Am. J. Med. 58:339-353, 1975. Hodgkin’s disease. Cancer 25:128-134, 1970. 31. S h e e h a n , H. L. and S t o r e y , G. W . : An im­ 2 5 . O s s e r m a n , E. F.: Monocytic and monomyelo- proved method of staining leukocyte granules cytic leukaemia with increased serum and urine with Sudan black B. J. Path. Bact. 59:336-337, lysozyme as a late complication of plasma cell 1947. myeloma. Brit. Med. J. 2:327, 1971. 32. S t a s s , S . A., S c l m a c h e r , H. R., K e n e k l is , 26. P a p a m ic h a il , M., B r o w n , J. C., H o l b o r o w , T. P., e t a l . : Terminal deoxynucleotidyl trans­ E. J.: Immunoglobulins on the surface of human ferase immunofluorescence of bone marrow lymphocytes. Lancet 2:850—852, 1971. smears. Am. J. Clin. Pathol. 72:898—903, 1979. 27. P e n n , I.: Leukemias and lymphomas associated 33. T r ic o t , G., B r o e c k a e r t -V a n O r s h o v e n , A., V an with the use of cytotoxic and immunosuppressive H o o f , A., e t a l . : Sudan black B positivity in drugs. Recent Results in Cancer Res. 69:7—13, acute lymphoblastic leukaemia. Brit. J. Hae­ 1979. matol. 57:615-621, 1982. 28. Perentesis, J., R am say , N. K. C., Brunning, R., 34. W o l o s , J. A. and D a v ey , F. R.: B lymphocyte et al. : Biphenotypic leukemia: Immunologic and function in B cell chronic lymphocytic leukaemia. morphologic evidence for a common lymphoid- Brit. J. Haematol. 49:395-403, 1981. myeloid progenitor in humans. J. P e d i a t r . 35. Y a m , L. T., Li, C . Y ., and C r o sb y , W . H.: Cy- 102:63- 67, 1983. tochemical identification of monocytes and gran­ 29. R it z , J., P e s a n d o , J. M ., N o t is -M c C o n a r t y , J., ulocytes. Am. J. Clin. Pathol. 55:283—290, 1971. e t a l .: A monoclonal to human acute 36. Z m ije w s k i, C. M.: HLA antigens and their de­ lymphoblastic leukaemia antigen. Nature tection. Theoretical Aspects of HLA. Hackel, E. 283:583-585, 1980. ed. A technical workshop. Anaheim, CA, Amer­ 30. R o s n e r , F. and G r u n w a l d , H.: For acute leu­ ican Association of Blood Banks, 1982, pp. kemia group B: Hodgkin’s disease and acute leu­ 33-50.