Terminal Deoxynucleotidyl Transferase Staining in Acute Leukemia and Normal Bone Marrow in Routinely Processed Paraffin Sections

Transcription

1 HEMATOPATHOLOGY Original Article Terminal Deoxynucleotidyl Transferase Staining in Acute Leukemia and Normal Bone Marrow in Routinely Processed Paraffin Sections ATTILIO ORAZI, MD, JENNY COTTON, MD, GIORGIO CATTORETTI, MD, PATRICIA K. KOTYLO, MD, KARLA JOHN, MT(ASCP), JOHN T. MANNING, MD, AND RICHARD S. NEIMAN, MD Terminal deoxynucleotidyl transferase (TdT) is a nuclear protein widely used as a marker for the diagnosis and classification of acute leukemia. The usual methods for detecting TdT require smears, imprints, or cryostat sections of unfixed tissue. A polyclonal rabbit anti- TdT serum was used to immunostain 54 routinely processed bone marrow sections from patients with acute leukemic disorders, using a recently described antigen-unmasking technique based on microwave oven heating. The specificity of this method of TdT analysis was confirmed by comparing the results obtained with conventional TdT analysis by indirect immunofluorescence. Terminal deoxynucleotidyl transferase reactivity was also evaluated in 44 nonmalignant and normal bone marrow specimens. All cases that were TdT-positive by immunofluorescence (4 of 4 "pre-b" and T-cell acute lymphoblastic leuke- mia, of 5 acute myeloid leukemia, and of 5 chronic myeloid leukemia in blast crisis) were also positive in paraffin sections. The percentage fluorescence positivity correlated with the percentage of immunoperoxidase stained cells in 44 of 45 cases. The remaining nonneoplastic and normal bone marrow biopsy specimens were TdT-negative. These results show that TdT immunoperoxidase staining of conventionally processed bone marrow specimens can be readily achieved by the use of a simple antigen-unmasking technique and may provide useful diagnostic information particularly in cases in which fresh tissue samples are unavailable. (Key words: Acute leukemia; Bone marrow biopsy; Immunoperoxidase; Terminal deoxynucleotidyl transferase) Am J Clin Pathol 9;0: Terminal deoxynucleotidyl transferase (TdT) is a DNA polymerase located in the cell nucleus that catalyzes the addition of deoxynucleotide triphosphate to the '-OH groups of singlestrand DNA. This enzyme is expressed by a minor population of normal bone marrow lymphocytes, mostly of B-cell lineage, and in acute leukemias, particularly of lymphoblastic type. As such, it is widely used as a marker for the diagnosis and classification of acute leukemia. Terminal deoxynucleotidyl transferase is usually detected biochemically on cell extracts or cytochemically by indirect immunofluorescence techniques using bone marrow slide prep- From the 'Division of Hematopathology, Department of Pathology, Indiana University School of Medicine, Indianapolis. Indiana; the Division of Anatomic Pathology, Istiluto Nazionale dei Tumori, Milan, Italy (current affiliation: Department of Pathology, Division of Surgical Pathology, College ofphysicians and Surgeons of Columbia University, New York, New York): and the ^Department of Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas. Manuscript received June, 9; revision accepted November 9, 9. Address reprint requests to Dr. Orazi: Room UH446, Department of Pathology, Division of Hematopathology, Indiana University School of Medicine, University Hospital, 550 University Boulevard, Indianapolis, IN 460. arations. It may also be measured by flow cytometry.'" Several recent studies have proposed the use of procedures such as immunoperoxidase or immunoalkaline phosphatase labeling of bone marrow smears and cytocentrifuge preparations as an alternative. 4 ' 5 A major limitation of these methods is that they do not allow any correlation between immunologic marker expression and bone marrow tissue architecture. In addition, the presence of fibrosis can prevent marrow aspiration or result in a scanty cellular sample in which blasts are diluted by peripheral blood cells. Immunohistologic studies of bone marrow tissue sections have been hampered because most lymphoid antigens such as TdT appear unable to survive routine fixation and decalcification. 6 ' 7 The frozen-section immunohistologic technique, which can be successfully performed on nondecalcified bone marrow biopsy specimens, offers excellent antigenic preservation. 6 ' 7 The main disadvantage of this technique is the lack of cytologic detail; in addition, it is not routinely used in most laboratories. In this report we describe the use of a polyclonal antiserum raised against calf thymus TdT to detect TdT expression in routinely processed paraffin-embedded bone marrow biopsy specimens. To overcome the problem connected with antigenic deterioration of TdT protein in conventionally processed tissues, we used a microwave oven-based technique that has recently been reported to produce excellent retrieval of masked antigenic epitopes in routinely processed histologic mate- 640 Downloaded from on 5 January 08

2 ORAZI ET AL. 64 Immunoperoxidase Labeling of TdT in Paraffin Sections TABLE. IMMUNOPEROXIDASE STAINING WITH ANTI- TDT IN PARAFFIN BONE MARROW SECTIONS OF ACUTE LEUKEMIC DISORDERS Diagnosis ALL Precursor B cell (common) Tcell B cell (Burkitt type) AML Ml M M4 CML-BC Myeloid Lymphoid Total Staining with Anti-TdT [Cases reactive/total no. of case (%)] /4 (97) 8/8 (00) 0/ (0) 4/44() / (00) 0/ (0) 0/ (0) /5 (40) /4 (5) / (00) /5 (40) 45/54(8) ALL and AML diagnosis according to FAB system. CML = chronic myeloid leukemia: BC = blast crisis. rial. 8 ' 90 " The ability to detect this well-characterized acute leukemia marker in paraffin-embedded bone marrow specimens can provide an opportunity to assess TdT expression when fresh tissue is unavailable. It also can allow retrospective studies in fixed tissue samples of archival material. MATERIALS AND METHODS The cases of pediatric and adult leukemias selected for this study were chosen retrospectively on the basis of the availability of both TdT immunofluorescent assessment and flow cytometric immunophenotypic results on fresh tissue as well as paraffin-embedded bone marrow biopsy specimens or clot sections for immunoperoxidase staining. Fifty-four cases of acute leukemic conditions that met this criteria were selected: all main subtypes of acute lymphoblastic leukemia (ALL) (44 cases) and the more frequent types of acute myeloid leukemia (5 cases) were represented (Table ). In addition, five cases of chronic myeloid leukemia in blast crisis were studied. All leukemia cases were classified according to the FAB system on the basis of bone marrow smear morphology and cytochemistry. To confirm the known distribution of TdT, we also evaluated paraffin sections of 9 cases of nonneoplastic bone marrow. These biopsy specimens were obtained from 7 pediatric and adult patients with malignancies and uninvolved bone marrow (Table ). Five normal bone marrow specimens (two children and three adult allogenic marrow transplant donors) were also studied and used as controls. Frozen bone marrow tissue that had been stored at -90 C was available for comparison in two of these latter cases and in two of the nonneoplastic specimens. The bone marrow trephine-biopsy specimens were fixed in 0% buffered formalin for 4 hours, or in B5 for hours then decalcified in 0% EDTA, or in 0% nitric acid and embedded in paraffin. In the leukemic cases, TdT was detected cytochemically using a polyclonal rabbit anti-tdt antibody (Molecular Genetic Resources, Tampa, FL) by a standard indirect immunofluorescent technique on methanol-fixed smears or cytocentrifuge preparations of hemolyzed whole marrow. Flow cytometric immunophenotyping of the leukemia cases was performed according to previously published methods"' with the following panel of directly conjugated monoclonal antibodies: CDla(OKT6), CD(T), CD(T), CD5(Leu-l), CD4(T4), CD8(T8), CD0(J5), CD9(B4), CD0(Leu-6), CD(Leu-4), CD(My7), CD(My9), CD4(My4), CD lc(leum5), and HLA-DR. Based on the results of these markers the 4 cases of precursor B-cell (pre-b, common) ALL were TdT-positive (one case negative), HLA-DR, CD 9, CD 0 and variably positive for the other B-cell antigens. The two cases of B-cell ALL showed TdT-negative, CD 0, CD 9, CD0, CD, surface immunoglobulin IgM/«blasts. The T-cell ALLs were characterized by TdT, CD7, CD la, CD5, CD, CD positivity, variable expression of CD0, CD4, CD8 and HLA-DR negativity. The five myeloid leukemia cases and the four cases of myeloid blastic transformation of chronic myeloid leukemia were characterized by the expression of myeloperoxidase on more than % of the blasts by cytochemistry. These cases expressed HLA-DR, CD, or CD or a combination of them, with variable expression of CD lc or CD 4. The lymphoid blast transformation studied showed a T-cell phenotype with TdT-positive, CD7, CD5, CD, CD, HLA-DR- blasts. Immunohistochemical staining was performed as previously described. 5 After deparaffinization the slides were transferred to Phosphate Buffered Saline (PBS). Endogenous peroxidase was quenched by a 5-minute incubation in % hydrogen peroxide. After two washes in PBS, slides were transferred to plastic Coplin jars filled with citrate buffer (ph 6.0) and heated in a microwave oven twice for 5 minutes at a power of 0 W 9,M (modified method originally described by Shi and colleagues 8 ). After cooling at room temperature for 5 minutes, slides were covered with normal goat serum for 0 minutes and incubated overnight at 4 C with the anti-tdt antibody at a dilution of :0 (affinity purified rabbit anti-tdt obtained from Supertechs, Bethesda, MD). Negative control slides for each cases TABLE. IMMUNOPEROXIDASE STAINING WITH ANTI- TDT IN PARAFFIN SECTIONS OF NORMAL AND NONNEOPLASTIC BONE MARROWS Normal bone marrows* Pediatric Adult Nonmalignant bone marrows Pediatric postchemotherapy marrows-)- Pediatric staging marrows:): Adult postchemotherapy marrows (HD) Adult staging marrows Adult posttransplant marrows Total * These BM specimens were obtained from allogeneic marrow transplant donors. t This pediatric group includes 4 neuroblastoma (NBL). Hogdkin's disease (HD), and I non-hodgkin's lymphoma (NHL) cases. X This pediatric group includes NBL, NHL, HD, and rhabdomyosarcoma cases. This adult group includes 6 HD. I NHL. and carcinoma of the breast. This adult group includes 7 patients with locally advanced breast cancer with uninvolved marrow and 6 patients with NHL. BM specimens were obtained after chemotherapy and autologus BM transplant. Vol. 0-No. 5 Downloaded from on 5 January 08

3 64 HEMATOPATHOLOGY Original Article received the same immunohistologic treatment with substitution of the primary antibody with PBS. The slides were then stained with a biotin-conjugated goat anti-rabbit antibody (0 minutes; Kirkegaard & Perry Laboratories, Gaithersburg, MD), followed by peroxidase-conjugated streptavidin (0 minutes; Kirkegaard & Perry Laboratories). The enzyme was developed with,'-diaminobenzidine (Sigma, St. Louis, MO). After three washes in tap water the slides were counterstained in diluted hematoxylin for 0 seconds. The criterion fortdt positivity by indirect immunofluorescence in leukemia cases was the expression of that antigen by at least 5% of the blast cell population if the morphologic evaluation of a cytocentrifuge slide showed a high percentage of blasts. 6 The interpretation of anti-tdt immunoperoxidase staining involved both an attempt to quantify the percentage of cells staining as well as an estimate of the intensity of staining. For each sample, an average of 000 cells were independently counted by two investigators (A.O. and J.C.) who were unaware of the TdT results previously obtained by immunofluorescence. A case was considered TdT-positive by immunohistochemistry when the antigen was expressed by at least 5% of the blast cells.the intensity of the staining was semiquantitatively evaluated (-, negative;, weak staining;, strong staining) (Table ). We tested additional anti-tdt antibodies (a pooled mixture of monoclonal anti-tdt antibodies, clone HT-//, marketed by Dako Italy SpA, Milan, Italy and the 6A6.09 monoclonal antibody, from BioGenex (San Ramon, CA): the HT-// clone was ineffective on routinely processed paraffin-embedded material and the 6A6.09 monoclonal antibody failed to show the specificity required. RESULTS Bone Marrow Indirect Immunofluorescence for TdT The distribution of TdT staining in the leukemia cases examined by immunofluorescent technique was similar to that reported in previous studies. " 5,,6 Thirty-three of 4 (97%) precursor B-cell (pre-b common) cases and all eight (00%) cases of T-cell ALL in this series expressed TdT. Two of five cases (40%) of acute myeloid leukemia and one offive (0%) cases of chronic myeloid leukemia in blast crisis (the only case of lymphoid blastic crisis studied), also expressed TdT (Table ). The percentage of positive cells among the leukemia cases considered TdT-positive ranged from % to 98% (median, %) (Table ). Bone Marrow Immunohistochemistry for TdT Results obtained by immunoperoxidase staining with anti- TdT paralleled those obtained by immunofluorescence analysis (Table ). The staining reaction was generally strong and restricted to the nuclei. Cytoplasmic staining was noticed in cell in mitosis and in a few blasts, in combination with nuclear staining. The evaluation of the intensity of the staining reaction showed that approximately % of the leukemic cases showed strong staining of the blast cells (Figs.,, ). Immunoperoxidase staining was stronger in the formalin-fixed than in the B5-fixed specimens: this latter group included all cases with weaker staining (Table ). Decalcification used to process the TABLE. COMPARISON BETWEEN IMMUNOFLUORESCENCE AND IMMUNOPEROXIDASE LABELING TECHNIQUES FOR TdT IN ACUTE LEUKEMIC DISORDERS Case No. Diagnosis IF% Acute lymphoblastic leukemia (ALL) ALL B-cell ALL B-cell Acute myeloid leukemia (AML) Blastic crisis AML-M AML-M AML-M AML-M AML-M < (BC) of chronic myeloid leukemia BC-myeloid BC-myeloid BC-myeloid < BC-myeloid < BC-lymphoid IP /o < 7 IP Intensity = precursor B-cell ALL: IF% = percentage of cells TdT{) by immunofluorescence; I P% = percentage of cells TdT() by immunoperoxidase. IP intensity = intensity oflp staining of marrow TdT() cells (, weak;, strong). AML cases were classified according to the FAB system. A.J.C.P. November 9 Downloaded from on 5 January 08

4 ORAZI ET AL. 64 Immunoperoxidase Labeling of TdT in Paraffin Sections FIG.. Bone marrow biopsy specimen from B-cell precursor ALL (case 0) showing positive () TdT nuclear staining of lymphoblasts, which comprise 95% of nucleated cells (X 500). biopsy specimens did not affect the intensity of TdT staining. No significant intra-observer variation was observed. Only one of 54 acute leukemic conditions studied showed some discrepancy between immunoperoxidase and immunofluorescence results: a case of myeloid chronic myeloid leukemia in blast crisis (case 50) having a negative immunofluorescence reading in which immunoperoxidase showed 7% of blast cell staining with TdT. Results of the TdT immunoperoxidase staining in the nonneoplastic and normal specimens revealed that in the pediatric cases the percentage of cells expressing TdT ranged from.% to 4.0% (Fig. 4) with higher values observed in children less than 7 years of age (children >7 years of age mean TdT 8.%, median 9.0%, range 4% to 4%; children <7 years of age mean ^ FIG.. Bone marrow biopsy specimen from T-cell ALL (case 44) exhibiting positive () nuclear staining of lymphoblasts. Terminal deoxynucleotidyl transferase-positive cells represent 88% of nucleated cells (X 500). TdT.%, median.5%, range % to %). Flow cytometric data available in two patients less than two years of age showed an increased number of B-cell precursors identified by TdT, CD 9, CD 0, HLA-DR positivity. The percentage of cells expressing TdT in adult bone marrow specimens ranged from.% to 4.0% (Fig. 5) with a mean of.7% (median, -.5%). TdT staining of frozen bone marrow samples available in four adult specimens (two normal and two nonneoplastic) showed results that parallelled those obtained by immunoperoxidase (not shown). In the normal and nonneoplastic bone marrow specimens, TdT-positive cells were mostly isolated lymphoid-looking cells of small to medium size, rarely found in aggregates of ««<. Wm*Wm» FIG.. Bone marrow particle section from acute myeloid leukemiam (case 45) with myeloblasts showing positive () nuclear staining with anti-tdt antibody. Terminal deoxynucleotidyl transferase-positive cells represent 86% of nucleated cells (X 500). FIG. 4. Bone marrow biopsy specimen from a nonneoplastic marrow from a -year-old patient (staging for neuroblastoma) demonstrating high frequency of precursor cells stained with TdT. Terminal deoxynucleotidyl transferase-positive cells represent 4% of nucleated cells (X 500). Vol. 0-No. 5 Downloaded from on 5 January 08

5 644 HEMATOPATHOLOGY Original Article FIG. 5. Bone marrow biopsy specimen from a normal 8-year-old adult (allogenic bone marrow transplant donor) demonstating rare TdT-positive precursor cells (arrows). Terminal deoxynucleotidyl transferase frequency equals.4% of nucleated cells (X 500). two to three positive elements in pediatric marrows, frequently adjacent to an adipocyte (Fig. 4). DISCUSSION This study indicates that immunohistochemical staining by immunoperoxidase can be used to reliably demonstrate TdT in paraffin-embedded bone marrow sections of acute leukemia. This study extends previous investigations from this laboratory on the use of the immunoperoxidase procedure to demonstrate a wide range of cellular markers (eg, p5, CD4, PCNA, myeloid- and Iymphoid-cell-associated antigens) in routinely processed bone marrow biopsy specimens. 057 This technique can provide useful diagnostic information in leukemic disorders, especially when marrow fibrosis makes aspiration impossible or causes sample dilution with underestimation of the number of blasts. 8 Several recent investigations have independently confirmed the value of paraffin section antibody studies for phenotyping cases of leukemia. 9,0, An immunoperoxidase method for detecting TdT was used to identify TdT-positive cells in formalin-fixed and paraffin-embedded normal thymus and within a testis involved by ALL., Another group using a similar immunoperoxidase technique demonstrated TdT-positive cells in paraffin sections of agar-embedded peripheral blood lymphocytes from three patients with T-cell lymphoma or leukemia. 4 Said and coworkers 5 reported TdT staining in formalin-fixed, paraffin-embedded lymph nodes with lymphoblastic lymphoma using DNAse pretreatment of the sections. This is the first report of anti-tdt staining in microwaveoven-processed, acid-decalcified, and paraffin-embedded bone marrow trephine biopsy specimens and particle sections obtained from a large number of patients with acute leukemia. The microwave oven technique applied in this study is considered a most effective way to enhance the immunoreactivity of antigenic epitopes in conventionally fixed histologic sections. 8,9 " In recent studies we have shown its ability to retrieve antigenic reactivity for p5 and Ki-67 antigens from acid decalcified bone marrow biopsy specimens. 0,6 In this study we compared the anti-tdt immunostaining of paraffin sections to the results obtained by the immunofluorescence staining. With both these techniques TdT showed the frequency of expression expected among the different subtypes of acute leukemia, that is, high level of expression in T-cell and "common" ALL and less frequent expression in acute myeloid leukemia, " 56 although the small number of cases of acute myeloid leukemia studied does not allow any conclusive statement in these patients. As expected from previous results, low level of expression was consistently found in nonneoplastic and normal bone marrow sections, the rare TdT-positive cells representing normal pre-b immature bone marrow precursors. ' 7 These cells were more numerous in the pediatric age group, as previously reported. 5 All cases that were TdT-positive by immunofluorescence were also TdT-positive by immunoperoxidase. In one case (50) that was TdT-negative by immunofluorescence, the enzyme was weakly positive by immunoperoxidase in a proportion of blasts. This discrepancy, although unexplained, may be due to degradation of the antigen in slides in which the methanol fixation was initiated before the cells were completely dried or other causes of loss of enzyme from the cells. The causes of loss of reactivity with nuclear antigens in slides and cytocentrifuge preparations stained by immunofluorescence have been recently discussed by Campana and colleagues 8 and Cattoretti and colleagues. 6 It is important to emphasize that, although the bone marrow biopsy specimens included in this study had a relatively short and uniform time in fixative, previous evidence suggests that the antigen retrieval technique here described can effectively unmask fixation-sensitive antigens in tissues fixed in formalin for more than week." Our results also show that B5 fixation can negatively influence the sensitivity of the immunoperoxidase detection of TdT, as previously reported. 5 This study indicates that it is now possible to evaluate TdT expression in leukemic cases using conventionally processed sections of bone marrow. This is especially important in cases in which aspirates are not available because of "dry taps" or hypocellular posttherapy marrow. In addition, this technique enables the pathologist to perform analysis for TdT in archival material. Acknowledgments. The authors thank Susan Cooper for assistance with photography. REFERENCES. Bollum FJ. Terminal deoxynucleotidyl transferase as a hematopoietic cell marker. Blood 979;54:0-5.. Janossy G, Hoffbrand AV, Greaves MF, et al. Terminal transferase enzyme assay and immunological membrane markers in the diagnosis of leukemia: A multiparameter analysis of 00 cases. Br J Haematol 980;44:-4.. Almasri NM, Iturraspe JA, Benson NA, Chen MG, Braylan RC. Flow cytometric analysis of terminal deoxynucleotidyl transferase. Am J Clin Pathol 99;95: Lanham GR, Melvin SL, Stass SA. Immunoperoxidase determination of terminal deoxynucleotidyl transferase in acute leukemia using PAP and ABC methods: Experience in 0 cases. Am J Clin Pathol 9;8: Erber WN, Mason DY. Immunoalkaline phosphatase labeling of terminal transferase in hematologic samples. Am J Clin Pathol 9;88:4-50. A.J.C.P. November 9 Downloaded from on 5 January 08

6 ORAZI ET AL. 645 Immunoperoxidase Labeling of TdT in Paraffin Sections 6. Wood GS, Warnke RA. The immunologic phenotyping of bone marrow biopsies and aspirates: Frozen section techniques. Blood 9;59: Racklin B, Bearman R, Sheibani K, WinbergC, Rappaport H. The demonstration of terminal deoxynucleotidyl transferase on frozen tissue sections and smears by the avidin-biotin complex (ABC) method. Leiik Res 98;7: Shi S-R, Key ME, Kalra KL. Antigen retrieval in formalin-fixed, paraffin-embedded tissues: An enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections. J Histochem Cytochem 99;9: Cattoretti G, Becker MHG, Key G, et al. Monoclonal antibodies against recombinant parts of the Ki-67 antigen (MIB and MIB ) detect proliferating cells in microwave-processed formalinfixed paraffin sections. J Paihol 99; 68: Orazi A, Cattoretti G, Heerema NA, et al. Frequent p5 overexpression in therapy related myelodysplastic syndromes and acute myeloid leukemias: An immunohistochemical study of bone marrow biopsies. Mod Pathol 9;6: Cattoretti G, Pileri S, Parravicini C, et al. Antigen unmasking on formalin-fixed, paraffin embedded tissue sections. J Pathol 9;7: Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of the acute leukemias (FAB co-operative group). Br J Haematol 976;: Jani P, Verbi W, Greaves MF, Bevan D, Bollum F. Terminal deoxynucleotidyl transferase in acute myeloid leukemia. Leuk Res 98;7: Kotylo PK, Baenzinger JC, Yoder MC, Engle WA, Bolinger CD. Rapid analysis of lymphocyte subsets in cord blood. Am J Clin Paihol 990;: Orazi A, Cattoretti G, Schiro' R, et al. rh-i L and GM-CSF administered in-vivo after high-dose cyclophosphamide cancer-chemotherapy: Effect on the hematopoiesis and microenviroment in the human bone marrow. Blood 99;79: Bradstock KF, Hoffbrand AV, Ganeshaguru K, et al. Terminal deoxynucleotidyl transferase expression in acute non-lymphoid leukemia: An analysis by immunofluorescence. Br J Haematol 98;47: Orazi A, Neiman RS, Cualing H, Heerema NA, John K. CD4 immunostaining of bone marrow biopsies is a reliable way to classify the phases of chronic myeloid leukemia. Am J Clin Pathol 9;0: Orazi A, Cattoretti G, Soligo D, Luksch R, Lambertenghi-Deliliers G. Therapy-related myelodysplastic syndromes: FAB classification, bone marrow histology, and immunohistology in the prognostic assessment. Leukemia 9;7: Andrade RE, Wick MR, Frizzera G, Gajl-Peczalska KJ. Immunophenotyping of hematopoietic malignancies in paraffin sections. Hum Pathol 988; 9: Kurec AS, Cruz VE, Barrett D, Mason DY, Davey FR. Immunophenotyping of acute leukemias using paraffin-embedded tissue sections. Am J Clin Pathol 990;: Taubenberger JK, Cole DE, Raffeld M, et al. Immunophenotypic analysis of acute lymphoblastic leukemia using routinely processed bone marrow specimens. Arch Pathol Lab Med 99;5:8-4.. Janossy G, Thomas JA, Bollum FJ, et al. The human thymic microenviroment: An immunohistologic study. J Immunol 980;5:0-.. Janossy G, Thomas JA, Pizzolo G, et al. Immuno-histological diagnosis of lymphoproliferative diseases by selected combinations of antisera and monoclonal antibodies. Br J Cancer 980; 4: Halverson CA, Falini B, Taylor CR, Parker JW. Detection of terminal transferase in paraffin sections with the immunoperoxidase technique. Am J Pathol 98; 05: Said JW, Shintaku IP, Pinkus GS. Immunohistochemical staining for terminal deoxynucleotidyl transferase (TDT): An enhanced method in routinely processed formalin-fixed tissue sections. Am J Clin Pathol 988;89: Cattoretti G, Orazi A, Gerdes J. Proliferating normal marrow cells do stain for Ki-67 antigen. Br J Haematol 9;: Janossy G, Bollum FJ, Bradstock KF, et al. Terminal transferasepositive human bone marrow cells exhibit the antigenic phenotype of common acute lymphoblastic leukemia. J Immunol 979;: Campana D, Coustan-Smith E, Janossy G. 'Cytoplasmic' expression of nuclear antigens. Leukemia 989;:9-4. Vol. 0-No. 5 Downloaded from on 5 January 08