British lournal of Haematology. 1992. 82, 4 6 4 9
Comparison of plasma cell infiltration in bone marrow biopsies and aspirates in patients with multiple myeloma W. E. TERPSTRA, H. M. I,OKHORST,*F. B L O M J O U S0. , ~J . A . T H . MEUWISSENA N D A. W. DEKKER* Department of Internal Medicine. St Antonius Hospital Nieuwegein, Koekoekslaan 1lP.0. Box 2500, 3 4 3 0 EM Nieuwegein, *Department of Haematology. P.O. Box 85500, 3508 G.A. Utrecht, and -)Department of Pathology. St Antonius Hospital Nieuwegein
Recfived 1 4 January 1992: accepted for publication 1 3 May 1992
Summary. In 54 patients with multiple myeloma plasma cell infiltration was compared in bone marrow biopsies and aspirates. In 48% of cases plasma cell infiltration was comparable, in 48% infiltration in the aspirate was lower than in the biopsy. In only two cases more plasma cells were found in the aspirate. Eleven patients (20%) had less than 20% plasma cells in the aspirate and more than 50% in the biopsy. Underestimation of plasma cell load especially seems to occur in patients with a focal growth pattern of multiple myeloma or when strong fibrosis is present. 69% of patients
with stage 111, according to Durie & Salmon (1975), and 76% of patients with a high f12-microglobulin had more than 50% plasma cells in the biopsy, indicating that these parameters, which are based on tumour load, are influenced by other factors as well. The bone marrow biopsy is of superior value for direct estimation of the tumour load in multiple myeloma compared to bone marrow aspirates. A prospective study is needed to determine its prognostic significance.
Multiple myeloma is characterized by a n increase in monoclonal plasma cells in the bone marrow. The percentage of plasma cells in a bone marrow aspirate requested for establishing the diagnosis of multiple myeloma, however. varies: the National Cancer Institute requires more than 5% plasma cells (Committee of the Chronic Leukemia-Myeloma Task Force, 1973),but other groups require 10% (SouthWest Oncology Group. 1 9 i 6 ) or even more than 40% (WesternCancer Study Group. 197 5 ). 1 0-30% plasma cells is considered a minor criterion for the diagnosis of multiple myeloma and more than 30% a major criterion (Durie. 1986). The degree of plasma cell infiltration has been investigated as an independent prognostic factor by only a few authors. One author found a significant relationship between clinical stage, percentage of plasma cells in the bone marrow aspirate and survival (Vercelli et al. 1980). Cavo et a l ( 1 9 8 3 showed that survival within clinical stage I (Durie & Salmon. 1975)was significantly shorter when more than 50%plasma cells in the bone marrow smear were present. The abovenamed criteria for diagnosis and prognosis of multiple myeloma were determined in bone marrow aspirates. A histological classification and staging of multiple mye-
loma based on plastic embedded bone marrow sections was proposed by Bartl and coworkers (Bartl, 1988; Bartl et uJ, 1982. 1987, 1989). Important differences in survival were registered. The prognostic value of this system has not yet been prospectively investigated and no comparison was made with bone marrow smears. Very few studies (Carter et aJ, 1987; Paule et aJ, 1988; Thiele et ul. 1988: Buss et al, 1987: Pileri et al. 1989) compared the value of bone marrow sections to smears. Considerable discordance in plasma cell infiltration was found in a high number of cases. The prognostic consequence of this difference was not analysed in these studies. The aim of our study was to investigate plasma cell infiltration in aspirates compared to the plasma cell infiltration, proliferation pattern and degree of fibrosis in plasticembedded bone marrow biopsies. The impact on clinical stage and Br microglobulin was analysed. MATERlALS A N D METHODS
Patimts chnracteristics. Sixty-one consecutive patients with multiple myeloma diagnosed between 1986 and 1990 were evaluated at the St Antonius Hospital, Nieuwegein, and the University Hospital Utrecht. Seven patients were excluded because aspirates (four patients) or biopsies (three patients) were not representative (see evaluation of smears and
C'orrespondence: L)r A . U'. lkkker. Uniifersity Hospital Utrecht. Ikpartrnent of Hacrnatology. GO3.h-t;. P.O. Box 8 5 5 0 0 . 3508 (;A I.trecht. The Netherlands.
46
Bone Marrow Biopsies i n Multiple Myeloma Table I. Characteristics of patients. No. of patients Median age (yr) (range) Male/female IgA
54 66 (47-87) 33/21 13
IgG
34
Light chain Stage I* Stage I1 Stage 111 BL-microglobulin > 4 mg/l
7 15 5 34 20/44t
* Clinical stage according to Durie & Salmon (1975).
t Pz-microglobulin was determined in 44 patients. biopsies). No patient was treated before the aspirate and biopsy were performed. Diagnosis of multiple myeloma was made according to the criteria of Durie (1986). Fifty-four patients were studied, 33 males and 2 1 females, the mean age was 66 years, range 47-87. Fifteen patients (28%)had stage I disease, five (10%)stage I1 and 34 (63%)stage I11 disease. Thirteen patients had IgA myeloma, 34 patients had IgG myeloma and seven patients had light chain disease. In 44 patients serum p2 microglobulin could be evaluated as prognostic factor (Table I). Marrow aspirates. Bone marrow was aspirated through a Jamshidi needle Uamshidi & Swaim, 1971). Smears were stained with May-Grunewald-Giemsa. Biopsy techniques. All biopsies were taken from the iliac crest with a Jamshidi needle. Histological preparation. All biopsies were fixed, dehydrated and embedded in plastic without decalcification (Burkhardt, 1972). Sections of 3 pm in thickness were cut and stained with Gallamina Blue Giemsa for cytological details, and Gomori stain was used for reticulin staining. Evaluation of bone marrow smears. Smears were examined using a 100 x oil immersion objective. The percentage of plasma cells were determined by counting 500 bone marrow cells. Smears were regarded as adequate for analysis when a sufficient number of megakaryocytes, erythroblasts, mononuclear cells and myeloid cells were present. Evaluation ofbone marrow sections. The proliferationpattern of plasma cells, plasma cell mass and degree of fibrosis were evaluated according to Bart1 et a1 (1989). The proliferation pattern was described: diffuse (non-nodular), nodular, packed or sarcomatous. Four histological stages were defined as: stage I < 5 vol %plasma cells, stage 11: 5-20 vol % plasma cells, stage 111: 20-50 vol % and stage IV: > 50 vol % plasma cells. Three patterns of fibrosis were distinguished: fine, patchy and coarse. RESULTS
Marrow aspirates In 28 patients (52%)the number of plasma cells was 5-20%, in 1 4 patients (26%) 20-50% and in 12 patients (22%)more than 50%.
47
Bone marrow histology No patient had stage I infiltration, stage I1 infiltration was present in 12 cases (22744, stage I11 in nine cases (17%)and stage IV in 33 cases (61%). The proliferation pattern was regarded as diffuse in 16 patients (30%), nodular in 27 patients (50%),packed in nine patients (17%) and sarcomatous in two patients (4%). Fine fibrosis was observed in 31 cases (57%), patchy fibrosis in 10 cases (19%) and coarse fibrosis in 1 3 cases (24%).The mean surface of bone marrow sections was 29 mm2,cortical bone excluded. Comparison of bone marrow smears and bone marrow sections (Fig 1) Overall plasma cell infiltration in bone marrow smears and in bone marrow biopsies showed a weak correlation (Spearman correlation coefficient = 0.52). In 26/54 cases (48%)plasma cell infiltration was comparable between smears and biopsies. However, in 26 cases (48%)plasma cell infiltration was lower in the aspirate than in the biopsy. In only two cases (4%)a higher plasma cell infiltration was found in the aspirate. 11/ 28 patients with less than 20% plasma cells in the aspirate had more than 50% plasma cells in the biopsy. Six of them had fine fibrosis and five coarse fibrosis in the biopsy. Coarse fibrosis was present in 1 3 out of the other 43 patients (30%). The infiltration pattern in the group of 11patients with less than 20% plasma cells in the smear and more than 50% plasma cells in the biopsy was interstitial in one patient, nodular in seven patients, and packed in three patients. From 16 patients with an interstitial proliferation pattern only one had less than 20% plasma cells in the aspirate and more than 50% in the biopsy.
100 %plasma cells aspirate
1
90-
0
8
8070
-
60-
0
B8
5040
0
30 20
-
0
II (5.20%)
0
88
00
Ill
IV
(20-50%)
(>SO%)
Infiltration in biopsy
Fig 1 . Comparison of plasma cell infiltration in bone marrow aspirate
and biopsy in patients with multiple myeloma.
48 25,
W. E. Terpstra et a1 Number o t patlents
Stage I1
50
50
c20 20-50 >50
%plasma cell inliltration biopsy
Fig 2. Plasma cell infiltration in bone marrow in patients with multiple myeloma in relation to the I>urie8- Salmon staging system.
Table 11. Bone marrow biopsy plasma cell infiltration related to Pj-microglobulin.
/?z-microglobulin (mg/l) Plasma cell infiltration in biopsy (%)
> 50 < 50
>4
Comparison of plasma cell infiltration in bone marrow biopsies and aspirates in patients with multiple myeloma W. E. TERPSTRA, H. M. I,OKHORST,*F. B L O M J O U S0. , ~J . A . T H . MEUWISSENA N D A. W. DEKKER* Department of Internal Medicine. St Antonius Hospital Nieuwegein, Koekoekslaan 1lP.0. Box 2500, 3 4 3 0 EM Nieuwegein, *Department of Haematology. P.O. Box 85500, 3508 G.A. Utrecht, and -)Department of Pathology. St Antonius Hospital Nieuwegein
Recfived 1 4 January 1992: accepted for publication 1 3 May 1992
Summary. In 54 patients with multiple myeloma plasma cell infiltration was compared in bone marrow biopsies and aspirates. In 48% of cases plasma cell infiltration was comparable, in 48% infiltration in the aspirate was lower than in the biopsy. In only two cases more plasma cells were found in the aspirate. Eleven patients (20%) had less than 20% plasma cells in the aspirate and more than 50% in the biopsy. Underestimation of plasma cell load especially seems to occur in patients with a focal growth pattern of multiple myeloma or when strong fibrosis is present. 69% of patients
with stage 111, according to Durie & Salmon (1975), and 76% of patients with a high f12-microglobulin had more than 50% plasma cells in the biopsy, indicating that these parameters, which are based on tumour load, are influenced by other factors as well. The bone marrow biopsy is of superior value for direct estimation of the tumour load in multiple myeloma compared to bone marrow aspirates. A prospective study is needed to determine its prognostic significance.
Multiple myeloma is characterized by a n increase in monoclonal plasma cells in the bone marrow. The percentage of plasma cells in a bone marrow aspirate requested for establishing the diagnosis of multiple myeloma, however. varies: the National Cancer Institute requires more than 5% plasma cells (Committee of the Chronic Leukemia-Myeloma Task Force, 1973),but other groups require 10% (SouthWest Oncology Group. 1 9 i 6 ) or even more than 40% (WesternCancer Study Group. 197 5 ). 1 0-30% plasma cells is considered a minor criterion for the diagnosis of multiple myeloma and more than 30% a major criterion (Durie. 1986). The degree of plasma cell infiltration has been investigated as an independent prognostic factor by only a few authors. One author found a significant relationship between clinical stage, percentage of plasma cells in the bone marrow aspirate and survival (Vercelli et al. 1980). Cavo et a l ( 1 9 8 3 showed that survival within clinical stage I (Durie & Salmon. 1975)was significantly shorter when more than 50%plasma cells in the bone marrow smear were present. The abovenamed criteria for diagnosis and prognosis of multiple myeloma were determined in bone marrow aspirates. A histological classification and staging of multiple mye-
loma based on plastic embedded bone marrow sections was proposed by Bartl and coworkers (Bartl, 1988; Bartl et uJ, 1982. 1987, 1989). Important differences in survival were registered. The prognostic value of this system has not yet been prospectively investigated and no comparison was made with bone marrow smears. Very few studies (Carter et aJ, 1987; Paule et aJ, 1988; Thiele et ul. 1988: Buss et al, 1987: Pileri et al. 1989) compared the value of bone marrow sections to smears. Considerable discordance in plasma cell infiltration was found in a high number of cases. The prognostic consequence of this difference was not analysed in these studies. The aim of our study was to investigate plasma cell infiltration in aspirates compared to the plasma cell infiltration, proliferation pattern and degree of fibrosis in plasticembedded bone marrow biopsies. The impact on clinical stage and Br microglobulin was analysed. MATERlALS A N D METHODS
Patimts chnracteristics. Sixty-one consecutive patients with multiple myeloma diagnosed between 1986 and 1990 were evaluated at the St Antonius Hospital, Nieuwegein, and the University Hospital Utrecht. Seven patients were excluded because aspirates (four patients) or biopsies (three patients) were not representative (see evaluation of smears and
C'orrespondence: L)r A . U'. lkkker. Uniifersity Hospital Utrecht. Ikpartrnent of Hacrnatology. GO3.h-t;. P.O. Box 8 5 5 0 0 . 3508 (;A I.trecht. The Netherlands.
46
Bone Marrow Biopsies i n Multiple Myeloma Table I. Characteristics of patients. No. of patients Median age (yr) (range) Male/female IgA
54 66 (47-87) 33/21 13
IgG
34
Light chain Stage I* Stage I1 Stage 111 BL-microglobulin > 4 mg/l
7 15 5 34 20/44t
* Clinical stage according to Durie & Salmon (1975).
t Pz-microglobulin was determined in 44 patients. biopsies). No patient was treated before the aspirate and biopsy were performed. Diagnosis of multiple myeloma was made according to the criteria of Durie (1986). Fifty-four patients were studied, 33 males and 2 1 females, the mean age was 66 years, range 47-87. Fifteen patients (28%)had stage I disease, five (10%)stage I1 and 34 (63%)stage I11 disease. Thirteen patients had IgA myeloma, 34 patients had IgG myeloma and seven patients had light chain disease. In 44 patients serum p2 microglobulin could be evaluated as prognostic factor (Table I). Marrow aspirates. Bone marrow was aspirated through a Jamshidi needle Uamshidi & Swaim, 1971). Smears were stained with May-Grunewald-Giemsa. Biopsy techniques. All biopsies were taken from the iliac crest with a Jamshidi needle. Histological preparation. All biopsies were fixed, dehydrated and embedded in plastic without decalcification (Burkhardt, 1972). Sections of 3 pm in thickness were cut and stained with Gallamina Blue Giemsa for cytological details, and Gomori stain was used for reticulin staining. Evaluation of bone marrow smears. Smears were examined using a 100 x oil immersion objective. The percentage of plasma cells were determined by counting 500 bone marrow cells. Smears were regarded as adequate for analysis when a sufficient number of megakaryocytes, erythroblasts, mononuclear cells and myeloid cells were present. Evaluation ofbone marrow sections. The proliferationpattern of plasma cells, plasma cell mass and degree of fibrosis were evaluated according to Bart1 et a1 (1989). The proliferation pattern was described: diffuse (non-nodular), nodular, packed or sarcomatous. Four histological stages were defined as: stage I < 5 vol %plasma cells, stage 11: 5-20 vol % plasma cells, stage 111: 20-50 vol % and stage IV: > 50 vol % plasma cells. Three patterns of fibrosis were distinguished: fine, patchy and coarse. RESULTS
Marrow aspirates In 28 patients (52%)the number of plasma cells was 5-20%, in 1 4 patients (26%) 20-50% and in 12 patients (22%)more than 50%.
47
Bone marrow histology No patient had stage I infiltration, stage I1 infiltration was present in 12 cases (22744, stage I11 in nine cases (17%)and stage IV in 33 cases (61%). The proliferation pattern was regarded as diffuse in 16 patients (30%), nodular in 27 patients (50%),packed in nine patients (17%) and sarcomatous in two patients (4%). Fine fibrosis was observed in 31 cases (57%), patchy fibrosis in 10 cases (19%) and coarse fibrosis in 1 3 cases (24%).The mean surface of bone marrow sections was 29 mm2,cortical bone excluded. Comparison of bone marrow smears and bone marrow sections (Fig 1) Overall plasma cell infiltration in bone marrow smears and in bone marrow biopsies showed a weak correlation (Spearman correlation coefficient = 0.52). In 26/54 cases (48%)plasma cell infiltration was comparable between smears and biopsies. However, in 26 cases (48%)plasma cell infiltration was lower in the aspirate than in the biopsy. In only two cases (4%)a higher plasma cell infiltration was found in the aspirate. 11/ 28 patients with less than 20% plasma cells in the aspirate had more than 50% plasma cells in the biopsy. Six of them had fine fibrosis and five coarse fibrosis in the biopsy. Coarse fibrosis was present in 1 3 out of the other 43 patients (30%). The infiltration pattern in the group of 11patients with less than 20% plasma cells in the smear and more than 50% plasma cells in the biopsy was interstitial in one patient, nodular in seven patients, and packed in three patients. From 16 patients with an interstitial proliferation pattern only one had less than 20% plasma cells in the aspirate and more than 50% in the biopsy.
100 %plasma cells aspirate
1
90-
0
8
8070
-
60-
0
B8
5040
0
30 20
-
0
II (5.20%)
0
88
00
Ill
IV
(20-50%)
(>SO%)
Infiltration in biopsy
Fig 1 . Comparison of plasma cell infiltration in bone marrow aspirate
and biopsy in patients with multiple myeloma.
48 25,
W. E. Terpstra et a1 Number o t patlents
Stage I1
50
50
c20 20-50 >50
%plasma cell inliltration biopsy
Fig 2. Plasma cell infiltration in bone marrow in patients with multiple myeloma in relation to the I>urie8- Salmon staging system.
Table 11. Bone marrow biopsy plasma cell infiltration related to Pj-microglobulin.
/?z-microglobulin (mg/l) Plasma cell infiltration in biopsy (%)
> 50 < 50
>4
