Leukemia Research 25 (2001) 103 – 107 www.elsevier.com/locate/leukres
Primary plasma cell leukaemia: a report of 18 cases Re´gis Costello a,b,c,d,*, Daniele Sainty a, Reda Bouabdallah a,c, Jean-Pierre Fermand c, Alain Delmer c, Marine Divine´ c, Jean-Pierre Marolleau c, Jean-Albert Gastaut a, Daniel Olive b, Philippe Rousselot a,d, Pascal Chaı¨bi a,d a
b
De´partement d’He´matologie, Institut Paoli-Calmettes, 232 bd de Sainte Marguerite, 13009 Marseille, France Unite´ d’Immunologie des Tumeurs, Institut Paoli-Calmettes, 232 bd Sainte Marguerite, 13009 Marseille, France c Club de Re´flexion en He´matologie, Assistance Publique des Hoˆpitaux de Paris, Paris, France d Ser6ice Clinique des Maladies du Sang, Hoˆpital Saint-Louis, 1 a6 Claude Vellefaux, 75010 Paris, France Received 30 January 2000; accepted 22 July 2000
Abstract Primary plasma cell leukaemia (P-PCL) is a variant of multiple myeloma (MM) first diagnosed in the leukemic phase, with \2000/mm3 circulating plasma cells (PCs) and plasmacytosis \ 20% of the white cell count. We investigated the clinical characteristics, therapy, immunophenotype and prognosis factors of 18 patients. Common features at diagnosis were asthenia (seven patients), renal insufficiency (ten patients), bone pain (seven patients), splenomegaly or hepatomegaly (five patients). Hypercalcemia was present at diagnosis in seven patients and was the most potent poor prognosis factor (PB 0.05). Most patients (16 out of 18) were treated with an anthracyclin containing regiment; complete remission was attained in one patient and partial remission in 11 patients while six patients had no response. The median survival time from diagnosis was 7 months (2 –12, 95% confidence interval), but response to treatment had favorable predictive value (PB 0.05). The PCs were usually positive for mature B-cell markers (PCA-1, CD38). They expressed integrins which may increase their binding to endothelial cells and thus participate in PCL physiopathology by favoring plasmocyte extramedullary spread. © 2001 Elsevier Science Ltd. All rights reserved. Keywords: Primary plasma cell leukaemia; Multiple myeloma
1. Introduction Plasma cell leukaemia (PCL) is a very rare acute leukaemia considered as a variant of multiple myeloma (MM) and defined by more than 2000/mm3 circulating plasma cells (PCs) and plasmacytosis accounting for \ 20% of the differential white cell count [1 – 3]. Primary PCL (P-PCL) is defined as a malignant PC proliferation first diagnosed in the leukemic phase, while secondary PCL corresponds to the leukemic transformation of a previously diagnosed MM. Both the primary and secondary forms of PCL share a poor prognosis. Most series [1,2,4,5] reported a median surAbbre6iations: PCL, plasma cell leukaemia; MM, multiple myeloma; CR, complete response; PR, partial response; UPN, unique patient number. * Corresponding author. Tel.: + 33-4-91223640; fax: + 33-491223610. E-mail address:
[email protected] (R. Costello).
vival not exceeding 1 year, but the use of more intensive chemotherapy lead up to a 20-month median survival [6]. Moreover, some patients [6–10] experience ‘longterm’ survival (i.e. \ 3 years). In order to evaluate PCL prognosis, treatment and physiopathology, we report here on 18 patients with P-PCL diagnosed from 1983 to 1997, with flow cytometry analysis of multiple cell membrane molecules.
2. Materials and methods
2.1. Cytological and flow cytometry study Written informed consent was obtained from all patients before study. Peripheral blood and marrow smears were stained with standard May–GrunwaldGiemsa. Cell separation was done using standard procedures of Ficoll–Hypaque gradient centrifugation and
0145-2126/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 1 4 5 - 2 1 2 6 ( 0 0 ) 0 0 1 0 2 - 8
104
R. Costello et al. / Leukemia Research 25 (2001) 103–107
cytospins were performed to control the results of this separation. Mononuclear cell immunophenotype was performed by an immunofluorescence-activated cell sorter with different panels of monoclonal antibodies. Monoclonal antibodies were purchased from different manufacturers. Anti-DR, CD8, CD10, CD18, CD19, CD5, CD7, CD38, CD1a, CD25, CD11a, CD34, CD56, and CD19 were from Immunotech (Marseille, France). Anti-DP, DQ and CD4 were from Becton – Dickinson (Erembodegen, Belgium). CD20, CD21and CD23 were from Coulter (Mydrecht, The Netherlands). Anti-Igk, Igl, CD11b, and CD11c were from Dako (Glostrup, Denmark). The PC samples were considered as positive for a marker when more than 20% of leukemic cells were marked after subtraction of the isotype-matched immunoglobulin control used in all experiments.
2.2. Chemotherapy regimens The chemotherapy regimen used were the following. VAD: 4-day continuous infusions of vincristine (0.4 mg daily), doxorubicin (9 mg/m2) daily and oral dexamethasone (40 mg daily) from day 1 to day 4, day 9 to day 12 and day 17 to day 22. C2H2OP: vincristine 1,4 mg/m2 on day 1, adriblastine 90 mg/m2 on day 1, cyclophosphamide 1500 mg/m2 on day 1, and prednisone 80 mg/m2 for 4 days. Cy/Mpred.: cyclophosphamide 1000 mg/m2 on day 1, methylprednisolone 400 mg daily for 4 days. VMCP: vincristine 1 mg/m2 on day 1, melphalan 6 mg/m2 orally for 4 days, cyclophosphamide 125 mg/m2 orally for 4 days, and prednisone 60 mg/m2 for 4 days. MP: melphalan 8 mg/m2 orally for 4 days, with prednisone 60 mg/ m2 orally for 4 days, HD-M: high-dose melphalan 140 mg/m2 on day 1. DEX: dexamethasone 40 mg/day orally for 4 days.
2.3. Response and sur6i6al A complete response (CR) to chemotherapy was defined as a disappearance of circulating PCs and of the monoclonal protein, detected by immunofixation, both in plasma and urine. A decrease of more than 50% in serum or 75% in urine of the monoclonal protein, associated with the disappearance of circulating PCs, was defined as a partial response (PR). Survival was calculated from primary treatment (not delayed for more than three days, from diagnosis) by the Kaplan– Meier method. 3. Results
3.1. Clinical presentation and biological 6alues The patients studied here had primary PCL and they were not treated for MM prior to the diagnosis of PCL. There were seven women and 11 men, with a median
age of 51.5 years (range 30–75). The main presenting symptoms were asthenia (seven patients) and bone pain (seven patients). Only five patients had splenomegaly and/or hepatomegaly, and none of our cases showed a clinically relevant lymphadenopathy as presenting symptom. Bone lesions were observed in 11 patients. Pleural effusions due to plasmocyte infiltrates were present in two patients, meningeal myelomatosis in one patient and pulmonary nodules (that also corresponded to plasmocytes) in one patient. The laboratory data are listed in Table 1. Median hemoglobin (Hb) was 9 g/dl (range, 4.5–15.5); 15 out of the 18 patients had Hb value B 12 g/dl. The median leukocyte count was 24× 109 /l (range 4–62); eight patients had leukocytes \20× 109/l. Median relative value of blood plasmocytes was 46% (range 11–77), corresponding to a median count of 7.5× 109/l (range 2.1–42). Median platelet count was 62× 109/l (range 9–342); 11 patients had platelet counts B 100×109/l. Examination of bone marrow aspirates revealed a median PC value of 76% (range 24–100). Median creatinine was 130 mM/l (range 74–1177); nine patients had creatinine level \ 120 mM/l. Median calcemia was 2.5 mmol/l (range 2.2–4.3); nine patients had calcemia \ 2.5 mM/l.. Median serum monoclonal component level was 2.5 g/l (range 0–100) and median light chain excretion was 0.2 g/day (range 0–15). Data on LDH were available in 13 patients; the median LDH level was 407 IU/ml (range 171–2970, nB 220); nine patients had LDH \ 220 IU/ml (data not shown). Data on b2-microglobulin were available in 14 patients; the median b2-microglobulin was 6 (range 2.9–17, nB2.2 mg/l): all the patients tested had value \ 2.2 mg/l (data not shown). The monoclonal protein was k IgG in three cases, k IgA in four cases, k IgM in one case, l IgG in two cases, k light chains in four cases, l light chains in one case, with three non-secretory cases.
3.2. Flow cytometry study For technical considerations this study was only performed in nine patients (Unique patient numbers [UPN] 1,2,7,8,9,11,16,19 and 22). No patient was positive for the T-cell markers CD3, CD4, CD8 or CD25. All patients examined were positive for the PC-associated antigen CD38 or PCA1. Concerning B-cell-associated antigens, no patient was positive for CD10, CD21, CD24 or CD22. The CD19 and CD20 early B-cell antigens were present in three out of eight and two out of seven tested cases. The different chains of the b2-integrin family were expressed in two out of eight (CD11b and CD11c) and six out of eight (CD11a) of the tested patients. The adhesion molecule CD56 is usually expressed by MM plasmocytes, but was detected in only three out of the eight cases tested, in accordance with previous observations in PCL [11].
R. Costello et al. / Leukemia Research 25 (2001) 103–107
3.3. Treatment and e6olution The most common first line therapy was the VAD regimen (eight patients), followed by C2H2OP (three patients), VMCP (two patients), DEX (two patients). The CR was obtained in one patient, PR in 11 patients, while six patients were non responders (NR) (Table 2). Using the Kaplan–Meier analysis method, the overall median survival from primary PCL therapy was seven months (two–12, 95% confidence interval [CI], Fig. 1,
105
panel A). The causes of death were sepsis (five cases), uncontrolled plasmocyte proliferation with end-stage multivisceral involvement (seven patients), hepatic veno-occlusive disease following allogenic BMT (one patient), and acute renal failure (one case). We analyzed prognosis factors using log-rank analysis of Kaplan–Meier survival curves. The presence of hypercalcemia was a bad prognosis factor (P B0.05) with median survival of 12 months (1–23, 95% CI) when absent versus 6 months (3–9, 95% CI) when present
Table 1 Biological values at presentationa UPN
Hb g/dl
WBC 109/l
Blood PC%
PLT 109/l
BM PC%
Cr. mmol/l
Ca mmol/l
MP g/l
MP type
Ur. LC g/day
1 2 3 4 5 6 7 8 9 11 12 13 14 15 16 19 21 22
6.6 4.5 10.0 6.7 10.0 10.7 9.1 9.7 7.5 12.2 12.1 15.5 9.0 7.7 7 8 9.2 10.5
16.0 25.0 62.0 52.5 30.0 11.7 1.40 3.9 20.0 20.0 8.2 26.4 9.6 9.8 13.3 22 25 60.6
40 77 67 76 28 65 20 30 30 60 23 68 62 42 53 40 50 70
16 25 78 17 220 331 46 270 9 194 20 106 84 294 17 118 19 94
70 80 76 88 80 57 81 40 NDc 45 24 70 80 ND 100 ND 97 70
1177 Nb 411 600 350 87 96 130 102 110 106 140 160 79 114 481 150 95
2.4 2.4 4.3 3.0 2.4 2.39 2.33 2.4 3.5 2.74 3.65 3.7 2.45 2.16 3.0 3.03 2.57 2.45
0 52 0 29 B3 58 0 0 0 0 40 0.8 100.0 90.0 0 0 82 B2
L, k IgM, k L, k IgG, l IgA, k IgG, l L, k L, k L, l NS IgA, k IgA, k IgG, k IgG, k NS NS IgG, k IgA, k
6.0 15.0 0 13.0 12.5 6.0 0.32 12.0 2.5 0 0 0.5 ND 0.11 0 0 ND 0
a WBC, white blood cell; PC, plasma cells; PLT, platelets; BM, bone marrow; Ca, calcium; MP, monoclonal protein; Cr, creatinine; LC, light chain. b N: normal range; l, lambda; k, kappa. c ND: not done.
Table 2 Treatment and outcome UPN
First line treatment
Response
Survival (months)
Cause of death
1 2 3 4 5 6 7 8 9 11 12 13 14 15 16 19 21 22
VAD C2H2OP/Allo-BMT VAD C2H2OP VAD VAD VAD Cy/MPred. C2H2OP MP VMCP VAD DEX VMCP VAD DEX HD-M+ABMT VAD
PR CR PR PR PR PR PR PR PR PR PR NR NR PR NR NR NR NR
12 27 4 9 \6 12 \24 \6 2 14 12 3 1 6 2 1 7 6
Sepsis Hepatic veno-occlusive disease Sepsis Tumour progression // Tumour progression // // Septic shock Tumour progression Septic shock Tumour progression Acute renal failure Tumour progression Pneumopathy Tumour progression Pneumopathy Tumour progression
106
R. Costello et al. / Leukemia Research 25 (2001) 103–107
(Fig. 1, panel B). The presence of response to therapy resulted in a better prognosis (PB 0.05) with a median survival of 2 months (0–4.5, 95% CI) for non-responder versus 12 months (9–15, 95% CI) for responders (Fig. 1, panel C).
4. Discussion
Fig. 1. Kaplan – Meier analysis of the survival of 18 patients: overall survival (panel A), survival according to the presence or absence of hypercalcemia (panel B: \2.6 mM/l, dashed line; 5 2.6 mM/l, full line), survival according to the response to therapy (panel C: NR, dashed line; CR or PR, full line). n: number of patients. .
The most frequent presentation features of PCL were asthenia, severe anemia and thrombocytopenia, while clinical examination evidenced in some cases hepatomegaly and splenomegaly. This mimics, to some extent, the presentation of acute leukaemia and differs from classical MM, suggesting that PCL not only relies on the hematological definition of circulating PCs but also on clinical characteristics. In accordance with the report of Noel and Kyle [2], the presence of isolated free light chains was more frequent in PCL than in MM. Nonetheless, in contrast with the same report [2], most of our primary PCL with isolated light chains had the k isotype instead of the l isotype. This discrepancy may reflect the small size of the populations studied. Flow cytometry analysis was performed in order to characterize better the circulating myeloma cells in P-PCL. The phenotypic markers of PCs and of B cells in advanced stages of development (PCA-1, CD38, cytoplasmic Ig) were detected in most cases tested, while the pre-B cell antigens CD10 and CD21 were absent. Nonetheless, some patients expressed the early B-cell antigens CD19 and CD20. These data extend the observations made in classical MM that showed co-expression of B-cell and PC markers [12,13]. This implies that immunophenotype immaturity is not a specific characteristic of PCL that may distinguish this entity from ‘classical’ MM. At least one member of the a chain of b-2 integrins LFA-1, Mac-1 or p150,95 (respectively CD11 a, b and c) or the common b-chain (CD18) was detected on the PC surface. These integrins represent the leukocyte adhesion molecule involved in cell–cell adhesion, which may contribute to PC outgrowth by mediating their adhesion to lymphokine-producing cells of the bone marrow micro-environment and favor extramedullary spread by increased binding to endothelial cells. Most patients received anthracyclin-containing regimens which resulted in two-third cases to a response. Death related to sepsis was of major concern, in relation with both PCL immunosuppression and neutropenia induced by intensive and/or cumulative therapy. The patient treated by allogenic bone marrow transplantation was the only one who attained CR, but he died from graft–versus–host disease. The remaining patients died from complications related to disease progression leading to massive multi-organ tumor in-
R. Costello et al. / Leukemia Research 25 (2001) 103–107
volvement. The two patients treated with dexamethasone alone rapidly died, in line with a recent report which suggests a better survival for patients treated with polychemotherapy [5]. In our study, the response to treatment predicted a better outcome. This argues for a front-line therapy able to promptly reverse the complications of PCL, and for the early application of some form of intensive chemotherapy with the goal of achieving at least a partial remission. Programs such as VAD with early consideration of myeloablative therapy supported by autologous stem cells should receive high priority [14].
[6] [7] [8] [9]
[10]
[11]
References [1] Kyle RA, Maldonado JE, Bayrd ED. Plasma cell leukaemia: report on 17 cases. Arch Intern Med 1974;133:813. [2] Noel P, Kyle RA. Plasma cell leukaemia: an evaluation of response to therapy. Am J Med 1987;83:1062. [3] Costello R, Chaı¨bi P, Rousselot P. La leuce´mie a` plasmocytes, une forme particulie`re du mye´lome multiple. He´matologie 1996;2(4):296. [4] Bernasconi C, Castelli G, Pagnucco G, Brusamolino E. Plasma cell leukaemia: a report on 15 patients. Eur J Haematol Suppl 1989;51:76. [5] Garcia-Sanz R, Orfao A, Gonzalez M, Tabernero MD, Blade´ J, Moro MJ, Fernadez-Calvo MA, Sanz MA, Perez-Simon JA, Rasillo A, San Miguel JF. Primary plasma cell leukaemia:
.
[12]
[13] [14]
107
clinical, immunophenotypic, DNA ploidy, and cytogenetic characteristics. Blood 1999;93(3):1032. Dimopoulos MA, Palumbo A, Delasalle KB, Alexanian R. Primary plasma cell leukaemia. Br J Haematol 1994;88:754. McElwain TJ, Powles RL. High-dose intravenous melphalan for plasma-cell leukaemia and myeloma. Lancet 1983;2:822. Montecucco C, Riccardi A, Merlini G. Complete remission in plasma cell leukaemia. Br J Haematol 1986;62:525. Yeh KH, Lin MT, Tang JL, Yang CH, Tsay W, Chen YC. Long-term disease-free survival after autologous bone marrow transplantation in a primary plasma cell leukaemia: detection of minimal residual disease in the transplant marrow by third-complementarity-determining region-specific probes. Br J Haematol 1995;89:914. Bareford D, Pamphilon DH, Barnard DL. Plasma cell leukaemia relapsing in the dermis: correspondence. Acta Haematol 1984;71:359. Pellat-Deceunynck C, Barille´ S, Jego G, Puthier D, Robillard N, Pineau D, Rapp M-J, Harousseau J-L, Amiot M, Bataille R. The absence of CD56 (NCAM) on malignant plasma cells is a hallmark of plasma cell leukaemia and of a special subset of multiple myeloma. Leukaemia 1998;12(12):1977. Shimazaki C, Gotoh H, Ashihara E, Oku N, Inaba T, Murakami S, Itoh K, Ura Y, Nakagawa M, Fujita N. Immunophenotype and DNA content of myeloma cells in primary plasma cell leukaemia. Am J Hematol 1992;39:159. Linden MD, Fishleder AJ, Tubbs RR, Park H. Immunophenotypic spectrum of plasma cell leukaemia. Cancer 1989;63:859. Hovenga S, DeWolf JT, Klip H, Vellenga E. Consolidation therapy with autologous stem cell transplantation in plasma cell leukaemia after VAD, high-dose cyclophosphamide and EDAP courses: a report of three cases and a review of the literature. Bone Marrow Transplantation 1997;20(10):901.
