- Email: [email protected]
T-cell prolymphocytic leukaemia: Does the expression of CD8+ phenotype justify the identification of a new subtype?
Annals of Oncology 10: 649-653, 1999. © 1999 Kluwer Academic Publishers Printed in the Netherlands.
Commentary T-cell prolymphocytic leukaemia: Does the expression of CD8+ phenotype justify the identification of a new subtype? Description of two cases and review of the literature
1
Service of Pathologic Anatomy and Hematopathology. Institute of Haematology and Clinical Oncology L & A. Seragnoli'. Bologna University. Bologna; ~ Chair of Haematology; Institute of Pathologic Anatomy, Umberto I Hospital, Ancona University, Ancona: 4 Immunohaematology and Transfusion Centre, St. Maria Hospital, Terni; 5 Institute of Haemutologv, Perugia University, Perugia, Italy
Summary T-cell chronic lymphocytic leukaemia (T-CLL) has recently been reclassified under the heading of T-cell prolymphocytic leukaemia (T-PLL) because of its unfavourable clinical course, independently of the morphologic features. This rare neoplasm usually shows C D 4 + / C D 8 - phenotype. Herein we report on two cases of T-PLL with CD8 expression that correspond to a possible variant of the disease first proposed by Hui et al. in 1987. These cases presented with malignant cells showing immunophenotypic features that can be easily identified and distinguished from other peripheral T-cell leukemias. However,
Introduction
T-cell chronic lymphocytic leukaemia (T-CLL) accounts for less than 5% of all chronic lymphoproliferative disorders, most of which stem from peripheral B cells [1]. The classification of mature T-cell leukaemias has evolved considerably in recent years. In the late 1980s, the French-American-British (FAB) Cooperative Group distinguished five subgroups, namely T-cell lymphocytosis and T-CLL, T-cell prolymphocytic leukaemia (T-PLL), HTLV-1+ adult T-cell leukaemia/lymphoma (ATLL), and Sezary syndrome [2]. Later on, when a new entity designated 'the large granular lymphocyte leukaemia' (LGL) was identified [3-6], the existence of T-CLL became a matter of debate [7-10]. The World Health Organisation (WHO) recently commissioned a panel of experts to draft a new classification of the neoplasms of the haemo-lymphopoietic system [11]. During a meeting held at the Airlie House (Virginia, USA) in November 1997, a categorisation of peripheral T/NKcell neoplasms was proposed, which was largely based on the REAL Classification [12]. In the WHO proposal, T-CLL was reclassified under the heading T-PLL, because of its aggressive clinical behaviour, independently of the morphologic features. Furthermore, it was included among peripheral T/NK-cell neoplasms, predominantly
the total number of cases studied is inadequate for defining a discrete clinico-pathologic entity with characteristic clinical features and cytogenetical abnormalities. An international collaboration in which tissue from similar cases is referred to a central pathologist for immunophenotyping and cytogenetical study, and clinical data are centrally compiled, may assist in defining this rare malady as a discrete clinico-pathologic entity.
Key words: cytogenetics, peripheral T-cell lymphoma, phenotype, T-cell chronic lymphocytic leukaemia, T-cell prolymphocytic leukaemia
leukaemic/disseminated, along with T-cell LGL, NKcell leukaemia, and ATLL. T-PLL represents about 30% of T-cell leukaemias with mature phenotype [13] and more often shows CD4+/CD8- phenotype, with only 10%-15% of the cases being CD4-/CD8+ [1, 7, 13]. In this report, we describe two cases of T-PLL with CD8 phenotype and clinicopathologic characteristics similar to those previously observed in three patients by Hui et al. [14]. Case reports Patient 1
In October 1996, the patient - a 43-year-old man entered Ancona University Hospital because of hepatosplenomegaly and thrombocytopenia (85.0 x 109/l). His white blood cell (WBC) count was 6.0 x 109/l with 62% lymphocytes. No lymphadenopathies were observed. Results of laboratory tests for HIV and HTLV-1 were negative. As a bone marrow aspirate produced a dry-tap, a Jamshidi needle biopsy was performed, which on purely morphological grounds was regarded as suggestive of hairy cell leukaemia. However, on the basis of the results of biopsy immunophenotyping and peripheral
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
S. Ascani,1 P. Leoni,2 G. Fraternali Orcioni,1 I. Bearzi,3 M. Piccioli,1 M. Materazzi,4 P. L. Zinzani,1 F. Gherlinzoni,1 B. Falini5 & S. A. Pileri1
650
Patient 2
Figure I Cytological details of leukaemic cells in a peripheral blood smear: please note the irregularity of nuclear profiles, the obvious central nucleolus, and the moderate amount of cytoplasm (case 2, May-Griinwald-Giemsa, xlOOO).
In December 1996, a 46-year-old man, affected by diabetes mellitus of the insulin-independent type, was admitted to Terni University Hospital because of severe hyperglycaemia and marked asthaenia. Laboratory findings on admission revealed anaemia (Hb 9 g/dl) and thrombocytopaenia (41.0 x 109/l). The WBC count was 5.5 x 109/l with 58% lymphocytes. Mild hepatosplenomegaly was detected unaccompanied by any lymphadenopathies. As a bone marrow aspirate produced a drytap, a Jamshidi needle biopsy was performed, which on purely morphological grounds was regarded as suggestive of hairy cell leukaemia. However, on the basis of the results of biopsy immunophenotyping and peripheral blood examination - at light microscopy and flow cytometric analysis (see below) - a diagnosis of CD8+ T-PLL was made. In February 1997, the patient was treated with four courses of chemotherapy according to the ProMACE schedule [16], with reduction of the leukocytosis to a level of 8.0-12.0 x 109/l. In May 1997, following progression of the disease (WBC count > 20.0 x 109/l), the therapy was changed and the patient treated with pentostatin (4 mg/m2 i.v., for a total of two administrations two weeks apart). In August 1997, the onset of opportunistic infections was recorded: herpes zoster of the lips and oral cavity, followed by staphylococcal sepsis that resolved with antibiotic therapy. In September 1997, nine months after the diagnosis, the patient died of sepsis. Permission for post-mortem examination was denied. Histology, phenotype and cytogenetics On blood smears treated according to routine procedures [17], the leukaemia cells (about 12 urn in diameter) showed nuclei, with a distinct centrally located nucleolus and moderately condensed chromatin (Figure 1). They often appeared notched along half of the nuclear border or exhibited nuclear lobations or indentations. The cytoplasm was moderate in amount, pale blue, and agranular. No positiviry for tartrate-resistant acid phosphatase was seen. At ultrastructural analysis (Figure 2), the neoplastic cells revealed an irregular nuclear profile with hetero-
Figure 2. Electronmicrograph of leukaemic cells showing moderately irregular shaped nuclei, with the heterochromatin distributed in the periphery and a central prominent nucleolus (case 1, magnification: x3000. uranyl acetate and lead citrate staining).
chromatin distributed in the periphery and a central obvious nucleolus. The cytoplasm was moderate to relatively abundant, with rather poorly developed organelles and a variable number of protrusions. At microscopic examination of pathology samples, which had been processed according to previously reported protocols [18]. the bone-marrow biopsies from both patients turned out to be hypercellular with a
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
blood examination - at light microscopy and FACScan analysis (see below) - a diagnosis of CD8+ T-PLL was made. In June 1997, a chemotherapy with low doses of prednisone and methotrexate (5.0 and 17.5 mg/week respectively, for three months) was started, but the patient showed progression of his disease, with significant increase of both lymphocytosis (164.0 x 109/l) and splenomegaly. Therefore, in October 1997 he underwent splenectomy with open liver biopsy, followed by two cycles of 2-chloro-deoxy-adenosin (2-CDA) (0.9 mg/kg/day s.c, for seven days). At the time of this writing, the patient is receiving immunotherapy with human CD52 monoclonal antibody CAMPATH-1H [15], because of resistant disease (WBC count 100.0 x 109/l).
651 -
•E-
4
J
"C* '.
••
i'
t
m
si
\
t -*
r
;
A
it
•>
Table 2. Immunohistochemistry in paraffin sections. Results
Target molecule
Antibody
Source
CD20 CD79a CD72 CDla
L26 JCBI17 DBA.44 -
CD3
S-100b protein TIA-1
Polyclonal DF-T1 UCHL-1 OPD4 114B T199 Leu 7 (3F1 Polyclonal -
Granzyme B
Grb-7
Dako. Denmark Prof. D. Y Mason Prof. G. Delsol Immunotech. Marseille, France + Dako, Denmark + Dako, Denmark + Dako, Denmark Dako, Denmark + Prof. D. Y. Mason Dako. Denmark Becton Dickinson + Dako. Denmark Dako. Denmark Coulter Immunology Hialeah. USA Kamiya Biomedical Company, Seattle WA, USA Dako, Denmark
* * *
1
•'
j
;•»
*
*
*
*
'• • /
»«
"S I«
t
>
t
'*•
***
,;
«.' **' ' > •r* 1
* *
#
•
*
"*
*
«4
" *4
/A
•.
•
. *
'
*
• ' J
* '
4"' -
••l
^
_
' *
CD4 CD8
'
.
CD43 CD45R0
"
CD56 CD57
a/pTCR
Table 1. Results offlowcytometric analysis Antibody
Source
Casel
Case 2
SK9 S52 SIC7 SK3 L17F12 4H9 SKI D731 1QP222FR 4G7 L27 HB5 CH92 11F2
Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Immuno Quality Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson
CD1 = 1 6%, CD2 = 97 7% CD3 = 90%, CD4 = 2%, CD5 = 90%, CD7 = 95%, CD8 = 90% CD16 = 6%, CD56 = 3% C D 1 9 = 1% CD20 = 1% CD21 = 1%, CD8 + VP3+ = 95%, TCR-yS = neg
CD1 = 0 8% CD2 = 93 7% CD3 = 93 l% CD4 = 23 8% CD5 = 93.5%, CD7 = 96%, CD8 = 88%, CD16 = 4%o CD56 = 2% CD19 = 0 3%, CD20 = 0 6% CD21 = 1.2% CD8+ VP3+ = 93%, TCR-y5 = neg
TdT
diffuse lymphoid infiltrate, sparing some foci of normal haematopoietic cells (Figure 3). The cellular density was rather loose (Figure 3). At higher magnification, the neoplastic elements were small- to medium-sized, with irregular nuclear profiles (notched, lobated, at times kidney-shaped), and had somewhat condensed chromatin, distinct nucleoli and a moderate amount of cytoplasm, pale at Giemsa staining. Gomori silver impregnation showed a moderate diffuse increase in reticulin fibres. The spleen (whose weight was 2.250 g) revealed a diffuse infiltration of both the red and white pulp by the neoplastic cells. The liver biopsy was characterised by involvement of the portal tracts and sinusoids, which appeared dilated. The results of flow cytometry on the peripheral blood [19] from both cases are summarised in Table 1. In particular, at single and double labelling, leukaemia cells displayed positivity for V(33, CD2, CD3, CD5, CD7, and CD8. In case 2, a small minority of the neoplastic cells co-expressed CD4 and CD8 (see Table 1). Immunophenotyping in routine sections performed according to previous reports (Table 2) [20-22] showed that the neoplastic cells carried the following molecules: pFl, CD3, CD43, CD45R0, and CD8. All of the corresponding specific antibodies produced a distinct membrane-bound positivity (Figures 4 and 5).
Abbreviations: CD - cluster of differentiation; TCR - T-cell receptor; TIA - T-cell intracellular antigen, TdT - terminal deoxynucleotidyl transferase
-•• w%« %if
^
'*. * *\*
"#*i A •'
•>;^
>
v
- •'«%.i"} 1 * .**.• z -"'- *r»i'J V*;fti "» «*" ,**4, *«'<; '\*"*jm »i * > ' "*
Figure 4 Positivity of the neoplastic cells at the determination of CD3 in paraffin sections (case 1, EnVision technique, Gill's haematoxylin counterstain, x200).
Figure 5. Positivity of the neoplastic cells at the determination of CD8 in paraffin sections (case 1. APAAP technique, Gill's haematoxylin counterstain, x400).
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
Figure 3 The bone marrow is diffusely infiltrated by a i•ather loose lymphoid population, sparing some groups of normal haematopoietic cells (case 2, Giemsa. x400).
652 Cytogenetic studies [23] performed in case 1 demonstrated the following karyotype: 45 XY; add (7p), add (8p), -9, -13, -14, +Mar, +Ring. Discussion
Acknowledgements We thank Prof. L. Moretta for his kind suggestions and Mrs. C. Ercolessi and Miss F. Sandri for their skilled technical assistance. This paper was supported by grants from AIRC (Milan) and MURST (Rome).
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
In 1987, Hui et al. [14] described a new variant of T-CLL showing CD8+ phenotype, a lack of cytoplasmic azurophilic granules and NK. cell antigens, irregularly shaped nuclei, and aggressive clinical course characterised by a progressive increase in the number of circulating leukaemic cells, poor response to therapy, and death within 20 months. In particular, according to Hui et al. [14] this variant morphologically resembled 'the CD4+ CLL' (i.e., the present T-PLL), and needs to be distinguished for therapeutical and prognostic reasons from 'the usual CD8+ CLL with azurophilic granules' (i.e.,T-cell LGL), T-acute lymphoblastic leukaemia and ATLL. Since then, the examples of T-CLL/T-PLL with CD4-/CD8+ phenotype have not been regarded by most authors [2,13,24] as a specific subtype of the process, although the disease appears to have a more aggressive natural history than that of CD4+ T-PLL [7,13, 25]. The two examples of CD4-/CD8+ T-PLL in this report correspond to the original description of Hui et al. [14]. Their correspondence was confirmed by their direct comparison with case 3 of Hui's series, which was initially observed by one of us (SAP). The present cases provide further clues that may be useful in establishing the correct diagnosis and classification of this rare disorder. Firstly, the interpretation of the bone-marrow biopsies appeared problematic at conventional light microscopy: in fact, the neoplastic infiltrate was rather reminiscent of hairy cell leukaemia (HCL) since it showed indented nuclear profiles and a moderate amount of weakly stained cytoplasm, which produced a loose cellular density [12, 26]. However, the application of immunohistochemistry to paraffin sections - which nowadays is quite easy and reliable [27] - allowed a straightforward distinction between the two diseases on the basis of their different phenotypes (CD3+, CD20-, CD79a-, DBA.44- for T-PLL vs. CD3-, CD20+, CD79a+, DBA.44+ for HCL). Secondly, cytogenetic studies in one of our patients revealed a complex karyotype, which did not include abnormalities on chromosome 14 at bands qll and q32, or trisomy 8q. The latter aberrations are regarded as characteristic of T-PLL, occurring in about 70% of cases [7, 13, 28, 29]. However, we think that their absence does not preclude a diagnosis of T-PLL. In fact, in our cases this diagnosis was based on a constellation of morphologic, ultrastructural, phenotypic, virologic and clinical findings which were all in keeping with a definition of T-PLL, but not of other leukaemic peripheral T-cell tumours [7]. In particular, the positivity for CD7 and the negativity for HTLV-1 ruled out the possibility of ATLL, the lack of azurophilic granules and NK markers
together with the aggressive course excluded T-cell LGL, and the absence of erythroderma or lymphadenopathies precluded a diagnosis of Sezary syndrome [7]. Thirdly, the fact that the tumor cells in both cases expressed antigen receptors from the V(33 family is interesting. In fact, the fraction of normal T-cells from this family is small (about 10%) [30] and seems involved in autoimmune phenomena [31]. It might be argued that in our cases a possibly common antigen or superantigen stimulation evoked a CD8+ T-lymphocyte expansion, that represented the substrate from which the tumor originated. Fourthly, our cases did not stain for S-100P protein, thereby differing from the examples of so-called 'S-100(3positive T-cell leukaemia' described by others [32, 33] and also expressing CD4-/ CD8+ phenotype. Besides the staining for S-100, S-lOOp positive T-cell leukaemia displays additional features differentiating it from the present series and from T-PLL in general. In fact, the neoplastic cells are medium to large in size, with round to oval nuclei, can show erythrophagocytosis, carry CD16 and CD11 molecules, and tend to spread more diffusely (to the liver, mesenteric nodes, bone marrow, kidneys, lungs and heart) [29]. S-100P-positive T-cell leukaemia seems to stem from S-100p-positive T-lymphocytes in the normal human peripheral blood, but further studies are needed to better define this entity. More in general, our cases draw attention to the problem of CD8+ T-PLL, which - as stated above - is regarded as more aggressive than the CD4+ variety [13, 25]. Most reports [2,13, 24] make no distinction between CD4+/CD8- and CD4-/CD8+ cases, with the only recognised variants being morphological (small-cell or cerebriform) [29]. In light of the non-specific chromosomal aberrations observed in one of our cases, we scanned the literature in search of possible relationships between the karyotype and CD4-/CD8+ phenotype in T-PLL, and found a single case, quoted by Brito-Babapulle et al. [28], which carried neither aberrations on chromosome 14, nor trisomy 8q (i.e., the aberrations regarded as characteristic of T-PLL) [7, 13, 28, 29]. This suggests a need to plan further studies in order to assess whether the rare cases of CD4-/CD8+ T-PLL do indeed show definite bio-pathological characteristics (including chromosomal abnormalities) that could explain their poor response to therapy and more unfavourable clinical course, and justify their specific mention in the classification scheme.
653 References
Received 3 November 1998; accepted 9 February 1999.
Correspondence to: Prof. S. A. Pileri Servizio di Anatomia Patologica ed Ematopatologia Istituto di Ematologia e Oncologia Medica 'L. e A. Seragnoli' Universita di Bologna. Policlinico S. Orsola Via Massarenti 9 40138 Bologna Italy E-mail: pilerifealmadns.unibo.it
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
1. Catovsky D, Matutes E. Leukemias of mature T cells. In Knowles DM (ed): Neoplastic Hematopathology. Baltimore: Williams and Wilkins 1992: 1267-79. 2. Bennett JM. Catovsky D. Daniel M-T et al. and the FrenchAmerican-British (FAB) Cooperative Group. Proposal for the classification of chronic (mature) B and T lymphoid leukaemias. J Clin Pathol 1989, 42: 567-84. 3. Reynolds CW, Foon KA. T-gamma-lymhoproliferative disorders in man and experimental animals: A review of the clinical, cellular, and functional characteristics. Blood 1984: 64: 1146-58. 4. Pandolfi F. Pezzuto A, De Rossi G et al. Characterization of two patients with rymphomas of large granular lymphocytes. Cancer 1984: 53- 445-52. 5. LoughranTP, Draves K.E. Starkebaum G et al. Induction of NK activity in large granular leukemia: Activation with anti-CD3 monoclonal antibody and interleukin 2 Blood 1987; 69:72-8. 6. Loughran TP, Starkebaum G, Aprile JA. Rearrangement and expression of T-cell receptor genes in large garnular lymphocyte leukemia. Blood 1988; 71: 822-4. 7. Matutes E. Catovsky D. Mature T-cell leukemias and leukemia/ lymphoma syndromes. Review of our experience in 175 cases. Leuk Lymph 1991; 4: 81-91. 8. Foon KA, Gale RP. Is there a T-cell form of chronic lymphocytic leukemia? Leukemia 1992; 6: 867-8. 9. Matutes E, Catovsky D. CLL should be used only for the disease with B-cell phenotype. Leukemia 1993; 7: 917-8. 10. Wong KF. Chan JKC, Sin VC. T-cell form of chronic lymphocytic leukaemia: A reaffirmation of its existence Br J Haematol 1996; 93:157-9 11 Pileri SA, Milani M, Fraternali-Orcioni G, Sabattini E. From the REAL Classification to the upcoming WHO scheme: A step forward to a universal categorizing of lymphoma entities'' Ann Oncol 1998; 9: 607-12. 12. Harris NL, Jaffe ES, Stein H et al. A Revised European-American Classification of lymphoid neoplasms. A proposal from the International Lymphoma Study Group Blood 1994; 84: 1361-92. 13. Matutes E, Brito-Babapulle V, Swansbury J et al. Clinical and laboratory features of 78 cases of T-prolymphocytic leukemia Blood 1991; 78: 3269-74. 14. Hui PK. Feller AC, Pileri S et al. New aggressive variant of suppressor/cytotoxicT-CLL. Am J Clin Pathol 1987; 87: 55-9 15. Pawson RJC, Dyer MJ. Barge R et al. Successful treatment of T-cell prolymphocytic leukemia with human CDw52 antibody, CAMPATH-IH. J Clin Oncol 1997; 7: 2667-72. 16. Longo DL, De Vita VT Jr, Duffey PL et al. Superiority of Pro-MACE-CytaBOM over Pro-MACE MOPP in the treatment of advanced diffuse aggressive lymphoma: Results of a prospective randomized trial. J Clin Oncol 1991; 9: 25-38. 17. Hayhoe FG, Quaglino D. Haematological Cytochemistry (3rd ed). Edinburgh: Churchill Livingstone 1994; 199-252. 18. Pileri S. Linfomi maligni non-Hodgkin: Dalla tecnica alia diagnostica. Bologna: Societa Editnce Esculapio 1985; 8-18. 19. Raspadori D. Lauria F, Ventura MA et al. Incidence and prognostic relevance of CD34 expression in acute myeloblastic leukemia: Analysis of 141 cases. Leuk Res 1997; 7: 603-7.
20. Cordell JC. Falini B. Erber WNet al. Immunoenzymatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J Histochem Cytochem 1984: 32: 219-29. 21. Sabattini E. Bisgaard K. Ascani S et al. Envision Plus: A new immunohistochemical method of choice for diagnostics and research. Critical comparison with the APAAP. ChemMate. CSA. LABC. and LABC techniques. J Clin Pathol 1998: 51: 506-11. 22. Pileri SA. Roncador G. Ceccarelli C et al. Antigen retrieval techniques in immunohistochemistry: Comparison of different methods. J Pathol 1997: 183: 116-23. 23. Sandberg AA. The Chromosomes in Human Cancer and Leukemia. New York: Elsevier 1980. 24. Matutes E. Garcia Talavera J. O'Brien M. Catovsky D The morphologic spectrum of T-prolymphocytic leukaemia. Br J Haematol 1986:64: 111-24. 25. Hoyer JD. Ross CW, Li C-Yet al. True T-cell chronic lymphocytic leukemia' A morphologic and immunophenotypic study of 25 cases Blood 1995; 86: 1163-9. 26. Polliack A. Hairy cell leukemia and allied chronic lymphoid leukemias: Current knowledge and new therapeutic options. Leuk Lymph 1997; 26: 41-51. 27. Pileri SA, Roncador G, Ceccarelli C et al. Immunohistochemistry of bone-marrow biopsy. Leuk Lymph 1997; 26: 69-75. 28. Brito-Babapulle V, Pomfet M, Matutes E, Catovsky D. Cytogenetic studies on prolymphocytic leukemia. II. T cell prolymphocytic leukemia. Blood 1987: 70: 926-31. 29. Brito-Babapulle V, Maljaie SH, Matutes E et al Relationship of T leukaemias with cerebnform nuclei to T-prolymphocytic leukemia: A cytogenetic analysis with in situ hybridization. Br J Haematol 1997; 96: 724-32. 30. Poggi A, Demarest JF. Costa Pet al. Expression of a wideT cell receptor V beta repertoire in human T lymphocytes derived in vitro from embryonic liver cell precursors. Eur J Immunol 1994; 9: 2258-61. 31. Arenz M, Meyer zum Biischenfelde K-H, Lohr HF. Limited Tcell receptor V|3-chain repertoire of liver-infiltrating T cells in autoimmune hepatitis. J Hepatol 1998: 28: 70-7. 32. Takahashi K, Ohtsuki Y, Sonobe H et al. S-IOOb positive T cell leukemia. Blood 1988; 71: 1299-303. 33 Hanson CA, Bockenstedt PL, Schmtzer B et al. SlOO-positive, T-cell chronic lymphoproliferative disease: An aggressive disorder of an uncommon T-cell subset. Blood 1991: 78. 1803-13.
Commentary T-cell prolymphocytic leukaemia: Does the expression of CD8+ phenotype justify the identification of a new subtype? Description of two cases and review of the literature
1
Service of Pathologic Anatomy and Hematopathology. Institute of Haematology and Clinical Oncology L & A. Seragnoli'. Bologna University. Bologna; ~ Chair of Haematology; Institute of Pathologic Anatomy, Umberto I Hospital, Ancona University, Ancona: 4 Immunohaematology and Transfusion Centre, St. Maria Hospital, Terni; 5 Institute of Haemutologv, Perugia University, Perugia, Italy
Summary T-cell chronic lymphocytic leukaemia (T-CLL) has recently been reclassified under the heading of T-cell prolymphocytic leukaemia (T-PLL) because of its unfavourable clinical course, independently of the morphologic features. This rare neoplasm usually shows C D 4 + / C D 8 - phenotype. Herein we report on two cases of T-PLL with CD8 expression that correspond to a possible variant of the disease first proposed by Hui et al. in 1987. These cases presented with malignant cells showing immunophenotypic features that can be easily identified and distinguished from other peripheral T-cell leukemias. However,
Introduction
T-cell chronic lymphocytic leukaemia (T-CLL) accounts for less than 5% of all chronic lymphoproliferative disorders, most of which stem from peripheral B cells [1]. The classification of mature T-cell leukaemias has evolved considerably in recent years. In the late 1980s, the French-American-British (FAB) Cooperative Group distinguished five subgroups, namely T-cell lymphocytosis and T-CLL, T-cell prolymphocytic leukaemia (T-PLL), HTLV-1+ adult T-cell leukaemia/lymphoma (ATLL), and Sezary syndrome [2]. Later on, when a new entity designated 'the large granular lymphocyte leukaemia' (LGL) was identified [3-6], the existence of T-CLL became a matter of debate [7-10]. The World Health Organisation (WHO) recently commissioned a panel of experts to draft a new classification of the neoplasms of the haemo-lymphopoietic system [11]. During a meeting held at the Airlie House (Virginia, USA) in November 1997, a categorisation of peripheral T/NKcell neoplasms was proposed, which was largely based on the REAL Classification [12]. In the WHO proposal, T-CLL was reclassified under the heading T-PLL, because of its aggressive clinical behaviour, independently of the morphologic features. Furthermore, it was included among peripheral T/NK-cell neoplasms, predominantly
the total number of cases studied is inadequate for defining a discrete clinico-pathologic entity with characteristic clinical features and cytogenetical abnormalities. An international collaboration in which tissue from similar cases is referred to a central pathologist for immunophenotyping and cytogenetical study, and clinical data are centrally compiled, may assist in defining this rare malady as a discrete clinico-pathologic entity.
Key words: cytogenetics, peripheral T-cell lymphoma, phenotype, T-cell chronic lymphocytic leukaemia, T-cell prolymphocytic leukaemia
leukaemic/disseminated, along with T-cell LGL, NKcell leukaemia, and ATLL. T-PLL represents about 30% of T-cell leukaemias with mature phenotype [13] and more often shows CD4+/CD8- phenotype, with only 10%-15% of the cases being CD4-/CD8+ [1, 7, 13]. In this report, we describe two cases of T-PLL with CD8 phenotype and clinicopathologic characteristics similar to those previously observed in three patients by Hui et al. [14]. Case reports Patient 1
In October 1996, the patient - a 43-year-old man entered Ancona University Hospital because of hepatosplenomegaly and thrombocytopenia (85.0 x 109/l). His white blood cell (WBC) count was 6.0 x 109/l with 62% lymphocytes. No lymphadenopathies were observed. Results of laboratory tests for HIV and HTLV-1 were negative. As a bone marrow aspirate produced a dry-tap, a Jamshidi needle biopsy was performed, which on purely morphological grounds was regarded as suggestive of hairy cell leukaemia. However, on the basis of the results of biopsy immunophenotyping and peripheral
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
S. Ascani,1 P. Leoni,2 G. Fraternali Orcioni,1 I. Bearzi,3 M. Piccioli,1 M. Materazzi,4 P. L. Zinzani,1 F. Gherlinzoni,1 B. Falini5 & S. A. Pileri1
650
Patient 2
Figure I Cytological details of leukaemic cells in a peripheral blood smear: please note the irregularity of nuclear profiles, the obvious central nucleolus, and the moderate amount of cytoplasm (case 2, May-Griinwald-Giemsa, xlOOO).
In December 1996, a 46-year-old man, affected by diabetes mellitus of the insulin-independent type, was admitted to Terni University Hospital because of severe hyperglycaemia and marked asthaenia. Laboratory findings on admission revealed anaemia (Hb 9 g/dl) and thrombocytopaenia (41.0 x 109/l). The WBC count was 5.5 x 109/l with 58% lymphocytes. Mild hepatosplenomegaly was detected unaccompanied by any lymphadenopathies. As a bone marrow aspirate produced a drytap, a Jamshidi needle biopsy was performed, which on purely morphological grounds was regarded as suggestive of hairy cell leukaemia. However, on the basis of the results of biopsy immunophenotyping and peripheral blood examination - at light microscopy and flow cytometric analysis (see below) - a diagnosis of CD8+ T-PLL was made. In February 1997, the patient was treated with four courses of chemotherapy according to the ProMACE schedule [16], with reduction of the leukocytosis to a level of 8.0-12.0 x 109/l. In May 1997, following progression of the disease (WBC count > 20.0 x 109/l), the therapy was changed and the patient treated with pentostatin (4 mg/m2 i.v., for a total of two administrations two weeks apart). In August 1997, the onset of opportunistic infections was recorded: herpes zoster of the lips and oral cavity, followed by staphylococcal sepsis that resolved with antibiotic therapy. In September 1997, nine months after the diagnosis, the patient died of sepsis. Permission for post-mortem examination was denied. Histology, phenotype and cytogenetics On blood smears treated according to routine procedures [17], the leukaemia cells (about 12 urn in diameter) showed nuclei, with a distinct centrally located nucleolus and moderately condensed chromatin (Figure 1). They often appeared notched along half of the nuclear border or exhibited nuclear lobations or indentations. The cytoplasm was moderate in amount, pale blue, and agranular. No positiviry for tartrate-resistant acid phosphatase was seen. At ultrastructural analysis (Figure 2), the neoplastic cells revealed an irregular nuclear profile with hetero-
Figure 2. Electronmicrograph of leukaemic cells showing moderately irregular shaped nuclei, with the heterochromatin distributed in the periphery and a central prominent nucleolus (case 1, magnification: x3000. uranyl acetate and lead citrate staining).
chromatin distributed in the periphery and a central obvious nucleolus. The cytoplasm was moderate to relatively abundant, with rather poorly developed organelles and a variable number of protrusions. At microscopic examination of pathology samples, which had been processed according to previously reported protocols [18]. the bone-marrow biopsies from both patients turned out to be hypercellular with a
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
blood examination - at light microscopy and FACScan analysis (see below) - a diagnosis of CD8+ T-PLL was made. In June 1997, a chemotherapy with low doses of prednisone and methotrexate (5.0 and 17.5 mg/week respectively, for three months) was started, but the patient showed progression of his disease, with significant increase of both lymphocytosis (164.0 x 109/l) and splenomegaly. Therefore, in October 1997 he underwent splenectomy with open liver biopsy, followed by two cycles of 2-chloro-deoxy-adenosin (2-CDA) (0.9 mg/kg/day s.c, for seven days). At the time of this writing, the patient is receiving immunotherapy with human CD52 monoclonal antibody CAMPATH-1H [15], because of resistant disease (WBC count 100.0 x 109/l).
651 -
•E-
4
J
"C* '.
••
i'
t
m
si
\
t -*
r
;
A
it
•>
Table 2. Immunohistochemistry in paraffin sections. Results
Target molecule
Antibody
Source
CD20 CD79a CD72 CDla
L26 JCBI17 DBA.44 -
CD3
S-100b protein TIA-1
Polyclonal DF-T1 UCHL-1 OPD4 114B T199 Leu 7 (3F1 Polyclonal -
Granzyme B
Grb-7
Dako. Denmark Prof. D. Y Mason Prof. G. Delsol Immunotech. Marseille, France + Dako, Denmark + Dako, Denmark + Dako, Denmark Dako, Denmark + Prof. D. Y. Mason Dako. Denmark Becton Dickinson + Dako. Denmark Dako. Denmark Coulter Immunology Hialeah. USA Kamiya Biomedical Company, Seattle WA, USA Dako, Denmark
* * *
1
•'
j
;•»
*
*
*
*
'• • /
»«
"S I«
t
>
t
'*•
***
,;
«.' **' ' > •r* 1
* *
#
•
*
"*
*
«4
" *4
/A
•.
•
. *
'
*
• ' J
* '
4"' -
••l
^
_
' *
CD4 CD8
'
.
CD43 CD45R0
"
CD56 CD57
a/pTCR
Table 1. Results offlowcytometric analysis Antibody
Source
Casel
Case 2
SK9 S52 SIC7 SK3 L17F12 4H9 SKI D731 1QP222FR 4G7 L27 HB5 CH92 11F2
Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Immuno Quality Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson Becton Dickinson
CD1 = 1 6%, CD2 = 97 7% CD3 = 90%, CD4 = 2%, CD5 = 90%, CD7 = 95%, CD8 = 90% CD16 = 6%, CD56 = 3% C D 1 9 = 1% CD20 = 1% CD21 = 1%, CD8 + VP3+ = 95%, TCR-yS = neg
CD1 = 0 8% CD2 = 93 7% CD3 = 93 l% CD4 = 23 8% CD5 = 93.5%, CD7 = 96%, CD8 = 88%, CD16 = 4%o CD56 = 2% CD19 = 0 3%, CD20 = 0 6% CD21 = 1.2% CD8+ VP3+ = 93%, TCR-y5 = neg
TdT
diffuse lymphoid infiltrate, sparing some foci of normal haematopoietic cells (Figure 3). The cellular density was rather loose (Figure 3). At higher magnification, the neoplastic elements were small- to medium-sized, with irregular nuclear profiles (notched, lobated, at times kidney-shaped), and had somewhat condensed chromatin, distinct nucleoli and a moderate amount of cytoplasm, pale at Giemsa staining. Gomori silver impregnation showed a moderate diffuse increase in reticulin fibres. The spleen (whose weight was 2.250 g) revealed a diffuse infiltration of both the red and white pulp by the neoplastic cells. The liver biopsy was characterised by involvement of the portal tracts and sinusoids, which appeared dilated. The results of flow cytometry on the peripheral blood [19] from both cases are summarised in Table 1. In particular, at single and double labelling, leukaemia cells displayed positivity for V(33, CD2, CD3, CD5, CD7, and CD8. In case 2, a small minority of the neoplastic cells co-expressed CD4 and CD8 (see Table 1). Immunophenotyping in routine sections performed according to previous reports (Table 2) [20-22] showed that the neoplastic cells carried the following molecules: pFl, CD3, CD43, CD45R0, and CD8. All of the corresponding specific antibodies produced a distinct membrane-bound positivity (Figures 4 and 5).
Abbreviations: CD - cluster of differentiation; TCR - T-cell receptor; TIA - T-cell intracellular antigen, TdT - terminal deoxynucleotidyl transferase
-•• w%« %if
^
'*. * *\*
"#*i A •'
•>;^
>
v
- •'«%.i"} 1 * .**.• z -"'- *r»i'J V*;fti "» «*" ,**4, *«'<; '\*"*jm »i * > ' "*
Figure 4 Positivity of the neoplastic cells at the determination of CD3 in paraffin sections (case 1, EnVision technique, Gill's haematoxylin counterstain, x200).
Figure 5. Positivity of the neoplastic cells at the determination of CD8 in paraffin sections (case 1. APAAP technique, Gill's haematoxylin counterstain, x400).
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
Figure 3 The bone marrow is diffusely infiltrated by a i•ather loose lymphoid population, sparing some groups of normal haematopoietic cells (case 2, Giemsa. x400).
652 Cytogenetic studies [23] performed in case 1 demonstrated the following karyotype: 45 XY; add (7p), add (8p), -9, -13, -14, +Mar, +Ring. Discussion
Acknowledgements We thank Prof. L. Moretta for his kind suggestions and Mrs. C. Ercolessi and Miss F. Sandri for their skilled technical assistance. This paper was supported by grants from AIRC (Milan) and MURST (Rome).
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
In 1987, Hui et al. [14] described a new variant of T-CLL showing CD8+ phenotype, a lack of cytoplasmic azurophilic granules and NK. cell antigens, irregularly shaped nuclei, and aggressive clinical course characterised by a progressive increase in the number of circulating leukaemic cells, poor response to therapy, and death within 20 months. In particular, according to Hui et al. [14] this variant morphologically resembled 'the CD4+ CLL' (i.e., the present T-PLL), and needs to be distinguished for therapeutical and prognostic reasons from 'the usual CD8+ CLL with azurophilic granules' (i.e.,T-cell LGL), T-acute lymphoblastic leukaemia and ATLL. Since then, the examples of T-CLL/T-PLL with CD4-/CD8+ phenotype have not been regarded by most authors [2,13,24] as a specific subtype of the process, although the disease appears to have a more aggressive natural history than that of CD4+ T-PLL [7,13, 25]. The two examples of CD4-/CD8+ T-PLL in this report correspond to the original description of Hui et al. [14]. Their correspondence was confirmed by their direct comparison with case 3 of Hui's series, which was initially observed by one of us (SAP). The present cases provide further clues that may be useful in establishing the correct diagnosis and classification of this rare disorder. Firstly, the interpretation of the bone-marrow biopsies appeared problematic at conventional light microscopy: in fact, the neoplastic infiltrate was rather reminiscent of hairy cell leukaemia (HCL) since it showed indented nuclear profiles and a moderate amount of weakly stained cytoplasm, which produced a loose cellular density [12, 26]. However, the application of immunohistochemistry to paraffin sections - which nowadays is quite easy and reliable [27] - allowed a straightforward distinction between the two diseases on the basis of their different phenotypes (CD3+, CD20-, CD79a-, DBA.44- for T-PLL vs. CD3-, CD20+, CD79a+, DBA.44+ for HCL). Secondly, cytogenetic studies in one of our patients revealed a complex karyotype, which did not include abnormalities on chromosome 14 at bands qll and q32, or trisomy 8q. The latter aberrations are regarded as characteristic of T-PLL, occurring in about 70% of cases [7, 13, 28, 29]. However, we think that their absence does not preclude a diagnosis of T-PLL. In fact, in our cases this diagnosis was based on a constellation of morphologic, ultrastructural, phenotypic, virologic and clinical findings which were all in keeping with a definition of T-PLL, but not of other leukaemic peripheral T-cell tumours [7]. In particular, the positivity for CD7 and the negativity for HTLV-1 ruled out the possibility of ATLL, the lack of azurophilic granules and NK markers
together with the aggressive course excluded T-cell LGL, and the absence of erythroderma or lymphadenopathies precluded a diagnosis of Sezary syndrome [7]. Thirdly, the fact that the tumor cells in both cases expressed antigen receptors from the V(33 family is interesting. In fact, the fraction of normal T-cells from this family is small (about 10%) [30] and seems involved in autoimmune phenomena [31]. It might be argued that in our cases a possibly common antigen or superantigen stimulation evoked a CD8+ T-lymphocyte expansion, that represented the substrate from which the tumor originated. Fourthly, our cases did not stain for S-100P protein, thereby differing from the examples of so-called 'S-100(3positive T-cell leukaemia' described by others [32, 33] and also expressing CD4-/ CD8+ phenotype. Besides the staining for S-100, S-lOOp positive T-cell leukaemia displays additional features differentiating it from the present series and from T-PLL in general. In fact, the neoplastic cells are medium to large in size, with round to oval nuclei, can show erythrophagocytosis, carry CD16 and CD11 molecules, and tend to spread more diffusely (to the liver, mesenteric nodes, bone marrow, kidneys, lungs and heart) [29]. S-100P-positive T-cell leukaemia seems to stem from S-100p-positive T-lymphocytes in the normal human peripheral blood, but further studies are needed to better define this entity. More in general, our cases draw attention to the problem of CD8+ T-PLL, which - as stated above - is regarded as more aggressive than the CD4+ variety [13, 25]. Most reports [2,13, 24] make no distinction between CD4+/CD8- and CD4-/CD8+ cases, with the only recognised variants being morphological (small-cell or cerebriform) [29]. In light of the non-specific chromosomal aberrations observed in one of our cases, we scanned the literature in search of possible relationships between the karyotype and CD4-/CD8+ phenotype in T-PLL, and found a single case, quoted by Brito-Babapulle et al. [28], which carried neither aberrations on chromosome 14, nor trisomy 8q (i.e., the aberrations regarded as characteristic of T-PLL) [7, 13, 28, 29]. This suggests a need to plan further studies in order to assess whether the rare cases of CD4-/CD8+ T-PLL do indeed show definite bio-pathological characteristics (including chromosomal abnormalities) that could explain their poor response to therapy and more unfavourable clinical course, and justify their specific mention in the classification scheme.
653 References
Received 3 November 1998; accepted 9 February 1999.
Correspondence to: Prof. S. A. Pileri Servizio di Anatomia Patologica ed Ematopatologia Istituto di Ematologia e Oncologia Medica 'L. e A. Seragnoli' Universita di Bologna. Policlinico S. Orsola Via Massarenti 9 40138 Bologna Italy E-mail: pilerifealmadns.unibo.it
Downloaded from https://academic.oup.com/annonc/article-abstract/10/6/649/155235 by guest on 28 July 2019
1. Catovsky D, Matutes E. Leukemias of mature T cells. In Knowles DM (ed): Neoplastic Hematopathology. Baltimore: Williams and Wilkins 1992: 1267-79. 2. Bennett JM. Catovsky D. Daniel M-T et al. and the FrenchAmerican-British (FAB) Cooperative Group. Proposal for the classification of chronic (mature) B and T lymphoid leukaemias. J Clin Pathol 1989, 42: 567-84. 3. Reynolds CW, Foon KA. T-gamma-lymhoproliferative disorders in man and experimental animals: A review of the clinical, cellular, and functional characteristics. Blood 1984: 64: 1146-58. 4. Pandolfi F. Pezzuto A, De Rossi G et al. Characterization of two patients with rymphomas of large granular lymphocytes. Cancer 1984: 53- 445-52. 5. LoughranTP, Draves K.E. Starkebaum G et al. Induction of NK activity in large granular leukemia: Activation with anti-CD3 monoclonal antibody and interleukin 2 Blood 1987; 69:72-8. 6. Loughran TP, Starkebaum G, Aprile JA. Rearrangement and expression of T-cell receptor genes in large garnular lymphocyte leukemia. Blood 1988; 71: 822-4. 7. Matutes E. Catovsky D. Mature T-cell leukemias and leukemia/ lymphoma syndromes. Review of our experience in 175 cases. Leuk Lymph 1991; 4: 81-91. 8. Foon KA, Gale RP. Is there a T-cell form of chronic lymphocytic leukemia? Leukemia 1992; 6: 867-8. 9. Matutes E, Catovsky D. CLL should be used only for the disease with B-cell phenotype. Leukemia 1993; 7: 917-8. 10. Wong KF. Chan JKC, Sin VC. T-cell form of chronic lymphocytic leukaemia: A reaffirmation of its existence Br J Haematol 1996; 93:157-9 11 Pileri SA, Milani M, Fraternali-Orcioni G, Sabattini E. From the REAL Classification to the upcoming WHO scheme: A step forward to a universal categorizing of lymphoma entities'' Ann Oncol 1998; 9: 607-12. 12. Harris NL, Jaffe ES, Stein H et al. A Revised European-American Classification of lymphoid neoplasms. A proposal from the International Lymphoma Study Group Blood 1994; 84: 1361-92. 13. Matutes E, Brito-Babapulle V, Swansbury J et al. Clinical and laboratory features of 78 cases of T-prolymphocytic leukemia Blood 1991; 78: 3269-74. 14. Hui PK. Feller AC, Pileri S et al. New aggressive variant of suppressor/cytotoxicT-CLL. Am J Clin Pathol 1987; 87: 55-9 15. Pawson RJC, Dyer MJ. Barge R et al. Successful treatment of T-cell prolymphocytic leukemia with human CDw52 antibody, CAMPATH-IH. J Clin Oncol 1997; 7: 2667-72. 16. Longo DL, De Vita VT Jr, Duffey PL et al. Superiority of Pro-MACE-CytaBOM over Pro-MACE MOPP in the treatment of advanced diffuse aggressive lymphoma: Results of a prospective randomized trial. J Clin Oncol 1991; 9: 25-38. 17. Hayhoe FG, Quaglino D. Haematological Cytochemistry (3rd ed). Edinburgh: Churchill Livingstone 1994; 199-252. 18. Pileri S. Linfomi maligni non-Hodgkin: Dalla tecnica alia diagnostica. Bologna: Societa Editnce Esculapio 1985; 8-18. 19. Raspadori D. Lauria F, Ventura MA et al. Incidence and prognostic relevance of CD34 expression in acute myeloblastic leukemia: Analysis of 141 cases. Leuk Res 1997; 7: 603-7.
20. Cordell JC. Falini B. Erber WNet al. Immunoenzymatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP complexes). J Histochem Cytochem 1984: 32: 219-29. 21. Sabattini E. Bisgaard K. Ascani S et al. Envision Plus: A new immunohistochemical method of choice for diagnostics and research. Critical comparison with the APAAP. ChemMate. CSA. LABC. and LABC techniques. J Clin Pathol 1998: 51: 506-11. 22. Pileri SA. Roncador G. Ceccarelli C et al. Antigen retrieval techniques in immunohistochemistry: Comparison of different methods. J Pathol 1997: 183: 116-23. 23. Sandberg AA. The Chromosomes in Human Cancer and Leukemia. New York: Elsevier 1980. 24. Matutes E. Garcia Talavera J. O'Brien M. Catovsky D The morphologic spectrum of T-prolymphocytic leukaemia. Br J Haematol 1986:64: 111-24. 25. Hoyer JD. Ross CW, Li C-Yet al. True T-cell chronic lymphocytic leukemia' A morphologic and immunophenotypic study of 25 cases Blood 1995; 86: 1163-9. 26. Polliack A. Hairy cell leukemia and allied chronic lymphoid leukemias: Current knowledge and new therapeutic options. Leuk Lymph 1997; 26: 41-51. 27. Pileri SA, Roncador G, Ceccarelli C et al. Immunohistochemistry of bone-marrow biopsy. Leuk Lymph 1997; 26: 69-75. 28. Brito-Babapulle V, Pomfet M, Matutes E, Catovsky D. Cytogenetic studies on prolymphocytic leukemia. II. T cell prolymphocytic leukemia. Blood 1987: 70: 926-31. 29. Brito-Babapulle V, Maljaie SH, Matutes E et al Relationship of T leukaemias with cerebnform nuclei to T-prolymphocytic leukemia: A cytogenetic analysis with in situ hybridization. Br J Haematol 1997; 96: 724-32. 30. Poggi A, Demarest JF. Costa Pet al. Expression of a wideT cell receptor V beta repertoire in human T lymphocytes derived in vitro from embryonic liver cell precursors. Eur J Immunol 1994; 9: 2258-61. 31. Arenz M, Meyer zum Biischenfelde K-H, Lohr HF. Limited Tcell receptor V|3-chain repertoire of liver-infiltrating T cells in autoimmune hepatitis. J Hepatol 1998: 28: 70-7. 32. Takahashi K, Ohtsuki Y, Sonobe H et al. S-IOOb positive T cell leukemia. Blood 1988; 71: 1299-303. 33 Hanson CA, Bockenstedt PL, Schmtzer B et al. SlOO-positive, T-cell chronic lymphoproliferative disease: An aggressive disorder of an uncommon T-cell subset. Blood 1991: 78. 1803-13.