BritishJournal ofHaernatology, 1979,41, 169-176.
Identification of a Pure Splenic Form of Chronic Lymphocytic Leukaemia G. DIGHIERO, D. CHARRON, P. DEBRE, M. LE PORRIER, G. VAUGIER, J. Y. FOLLEZOU, L. DEGOS,CL.JACQUILLAT A N D J.-L. BINET Groupe Hdpital Pitie-Salpetrih, Paris (Received
20
March 1978; accepted for publication 5 July 1978)
SUMMARY. We have recently proposed a new staging system for chronic lymphocytic leukaemia (CLL) in which patients with isolated splenomegaly arc classified into a distinct stage (stage 11). Twenty-three such patients (from two institutions) have been studied without recorded death in a follow-up of 1 8 months to 3 0 years. This favourable prognosis justifies separation of these ‘pure splenic forms’ (SCLL) which must be distinguished from what Galton has termed prolymphocytic leukaemia (PL). This distinction can be made on the basis of three criteria: (i) Clinically, SCLL has a slow uneventful course and neither anaemia and/or thrombocytopenia; (ii) cytologically PL can be distinguished from other forms of CLL though atypical forms of CLL may be confused with the former; and (iii) the study of surface membrane immunoglobulins (SmIg) showed that while lymphocytes from most patients with both PL and SCLL bore uniform SmIg, suggesting a monoclonal B-cell proliferation, there was a major quantitative difference in that whereas PL lymphocytes had a number of antigenic sites close to that of normal lymphocytes (mean: 82 ooo sites per cell), SCLL lymphocytes had a drastically reduced number of sites. It is our opinion that this is an important criterion for the differential diagnosis between PL and SCLL. Splenomegaly is a common feature of chronic lymphocytic leukaemia (CLL). As a rule, this splenomegaly is accompanied by lymph node enlargement. However, in some cases it may appear alone without clinically perceptible lymph node enlargement (Touraine, 1922). Patients with isolated splenomegaly appear to be rare in series reported in the literature: I I of 188 in the review by Hansen (1973), 1 5 of 394 at the Haematology Clinic of the St Louis Hospital in Paris (Maral et al, 1976), eight of 129 in our personal experience (Binet etal, 1977). After dividing our patients into five stages on the basis of clinical and haematological criteria (Binet et al, 1977),it came to our attention that all the patients in group 11, defined by clinically isolated splenomegaly, were still alive. Moreover, these patients had a slow and uneventful course, even when not treated. It appeared to us that separation of this pure splenic form of CLL (SCLL) was justified in view of the good prognosis with which it was associated. Following the work of Bouroncle (1958), haematologists have learnt to recognize the Correspondence: Dr G. Dighiero, Departement d’HCmatologie, Groupe Hbpital Pitie-Salpetricre, 147 Boulevard de I’Hbpital, 75634 Paris Cedex 1 3 , France. 0007-1o~8/~g/ozoo-o16~$o2.oo 01979 Blackwell Scientific Publications
169
170
G. Digkiero et a f
cytological features of the hairy cell, characteristic of hairy cell leukaemia and no longer include them among cases of CLL. More recently, Galton et at (1974) have identified prolymphocytic Ieukaemia (PL) on the basis of its course and clinical and cytological features. We have previously demonstrated the usefulness of quantitative analysis of surface membrane immunoglobulin (SmIg) in these forms (Dighiero et al, 1976; Follezou et al, 1977). The good prognosis associated with SCLL has not been highlighted to date and it is not impossible that the two above-mentioned disorders have long been confused with SCLL. In this work we have combined our cases of SCLl with those of the Haematology Clinic of the St Louis Hospital. The clinical, cytological and immunological features of these cases have been compared to those of six cases of PL studied by us. MATERIALS AND METHODS Patients Twenty-nine patients with peripheral and bone marrow lymphocytosis, accompanied by splenomegaly, were included in the study. The patients were divided into two groups: (i) Six patients fell into a category of ‘prolymphocytic leukaemia’ as described by Galton et a1 (1974).
(ii) The 23 remaining patients fulfilled the classical criteria for CLL. In all patients, initial diagnosis was made on the basis of peripheral lymphocytosis of greater than 1 5 x 109/l combined with bone marrow lymphocyte infiltration greater than 40%. In six cases the diagnosis of CLL was made on the basis of a lymphocytosis of greater than 4 x 109/l persisting for longer than 6 months and accompanied by a medullary infiltration of greater than 40% on bone marrow and lymphocytic infiltration on histological biopsy. All these patients had isolated splenic involvement and were classified as stage I1 in an anatomico-clinical staging system, similar to that of Rai et al(1975),which we have recently proposed. Eight patients were from the Haematology Department of Pitii SalpCtrikre,the other I 5 being from the Haematology Clinic of the St Louis Hospital. For these last 1 5 patients, belonging to a series of 394 cases of CLL, only a retrospective study was performed. Immunocytological Methods These studies could only be performed in the 14 patients (eight CLL and six PL) from our own series. The methods used have been previously described in detail (Ternynck et al, 1974) and are briefly summarized. Preparation of anti-kappa and anti-lambda chain antibodies. Anti-kappa and anti-lambda antibodies were isolated from goat antisera using the appropriate light chain immunoabsorbent. They were rendered monospecific for kappa and lambda chains by absorption with an immunoabsorbent of the opposite type, prepared by coupling Bence-Jones proteins to glutaraldehyde-treated polyacrylamide beads. Peroxidase-labelled antibodies. Purified antibodies were coupled to peroxidase with glutaraldehyde using a two-step procedure (Ternynck et al, 1974). Purification of peripheral blood lymphocytes. Lymphocytes were purified from heparinized
Pure Splenic Forms of CLL
171
peripheral blood by sedimentation for I h a t 37°C or by centrifugation on a ficoll-triosil gradient. Cell preparations contained an average of 95% lymphoid cells and cell death determined by trypan-blue exclusion never exceeded 10%. Reaction ofcells with enzyme-labelled antibody. Samples of 5 x 10’ cells were incubated either at 4°C or 37°C with 0.5 ml of peroxidase-labelled anti-kappa or anti-lambda antibodies for 2 h. A concentration of 0.125 mg/ml of labelled antibody was used. Light microscopy. Aliquots of cells treated at 37°C were washed four times and then examined under the light microscope after staining with a solution of 3-amino-9-ethylcarbazole and hydrogen peroxide for 10min. One drop of the cell suspension was mounted under a cover slip and the percentage of labelled versus unlabelled cells was obtained after counting 500 cells. Measurement ofperoxiduse activity. Peroxidase activity was determined by adding 2.9 ml of the 0-dianizidine reagent to 0.1 ml aliquots containing 5 x 106 cells. The reaction was stopped after I 5 min with a drop of 6 N HC1. Tubes were then centrifuged at 5000 g for 5 min and the absorbancy of the supernatant was read at 400 nm. Taking into account the number of cells and the activity found, the quantity of peroxidase fixed per 106 cells was calculated by referring to a standard curve of absorbancy versus reaction time for known quantities of peroxidase. Since the mean ratio of peroxidase coupled to antibodies is one to one, the number of molecules of antibody found per I O cells ~ could be calculated, and the mean number of antigenic sites on the surface of each positive lymphocyte could be determined. RESULTS The clinical, haematological and follow-up features of patients with SCLL and PL are summarized in Table I. The picture was uniform for the 23 SCLL patients: the majority of patients (14123) were over 60 years old; the spleen though clinically enlarged in all, remained above the umbilicus in 12 cases. Anaemia and thrombocytopenia were always absent and lymphocytosis was moderate, exceeding 75 x 109/l in only one patient. The cytological picture did not differ from that usually seen in CLL. No deaths were recorded among these patients with a follow-up as long as 30 years. All patients have now been followed up for at least 18 months and 14 have a follow-up of more than 5 years. The course of these patients was uneventful, even without treatment. The clinical picture of the six patients with PL (Table I) contrasted with that of the former. Marked splenic enlargement below the umbilicus was seen in all cases. Peripheral lymphocytosis exceeded 75 x 109/l in four patients and even 500 x ro9/l in two of these patients, and anaemia and thrombocytopenia were present in all cases. Lastly, and most important, the prognosis was extremely poor, four patients having a survival of less than 6 months. The survivors now have a follow-up of 6 months. O n smears, the cells from these patients did not present a uniform appearance. Small apparently normal lymphocytes, blastlike cells and particularly prolymphocytes were encountered in variable proportions. Immunocytologicalfeatures are shown in Table 11. In five of the six patients with PL, peripheral lymphocytes bore easily detectable SmIg. Over 90% of the cells were tagged either by peroxidase-labelled anti-lc (three patients) or anti-1 antibodies, thus suggesting a monoclonal B-cell proliferation. Quantitation of cell surface antigenic sites revealed a mean value of 82 ooo sites per cell.
53 73 62
62
58 66
M F F
M
M F
2
4
6
5
3
I
(yr)
Age
Sex
14
9
23
9 0123
Superficial nodes
0
Microlymphadenopathy Microlymphadenopathy
0
0
0
Lymph nodes
14
> 60
Age (Y r)
< 60
Patient
F
Sex
M
~
No. of patients
7
I2
I0
14
I0
I2
I1
>6cm
WBC
25
30 4s
8.8 8.7
24.9 580
IS 30
50
Platelets ( x 109/r)
8.2
7 6 7. s
Hb (g/dl)
13.3
24
700 7s 75
( x 109/1)
m=
11-18
4-120
m = 22.13
Hb @/do
Peripheral lymphocytes ( x 10911)
Prolymphocytic leukaemia
Spleen (cm)
I2
t6cm
Spleen
Treatment
I 1/23
Treatment
6 months
6 months 4 monthst
Splenectomy, steroids CVP
6 monthst 7 monthst 8 monthst
Survival
I0
5-10 yr
Follow-up
CVP
9
< 5 yr
Chlorambucil, MOPP Chlorambucil, vincristine Splenic irradiation
m = 188
120-256
(x1o9/1)
Platelets
TABLE I. Pure splenic forms of chronic lymphocytic leukaemia (stage 11)
4
>lo
yr
3
EL
Y
""s -_ rn
N
4
Pure Splenic Forms of CLL
I73
A monoclonal B-cell appearance was also found in seven patients with SCLL (five IC and two
A). However, the mean number of antigenic sites found on the surface of these lymphocytes was 1 2 500 which is similar to that found on lymphocytes from patients with CLL in general and is appreciably lower than the number of antigenic sites found on control lymphocytes. One case of SCLL was found to correspond to a proliferation of 'null' cells since less than 5% of cells were positive for SmIg, 3 % of cells formed B-rosettes and 6% formed E-rosettes. One patient with PL had less than 5 % SmIg positive lymphocytes, but 80% of cells formed B-rosettes and 93 % formed E-rosettes. TABLE 11. Surface membrane immunoglobulins (quantitativestudy) Patient
Anti-x
Anti-2 Sites per cell
Sites per cell % Positive cells
x
103
% Positive cells
Prolymphocytic leukaemia B-cell type (five patients) I
-
99 98
97-5 65
99 -
-
-
93
6
-
2
3 4 5
Pure splenic forms of CLL B-cell type (seven patients) I
2
3 4 5 6 7
CLL
68 -
10.5
70 86 68 63
8.2
36/46
Mean 74% Normal subjects
Mean 19.4
-
88
-
13.5
77 94 -
Mean 9.5 (range 2-25)
Mean 70%
5.5 9.7
Mean 93.7 (range 86-185)
10146
Mean 9.5
x
103
75 113
-
11.5
29
-
Mean 10.2 (range 2.5-29) Mean 106 (range 32.5-190)
DISCUSSION For a long time CLL was considered a disease of unpredictable course. Recently, Rai et al(1975) and Binet et al(1977)suggested a new five-stage anatomico-clinical classification system which has proved to be of prognostic significance (Philipps et al, 1977). The actuarial survival curves (Fig I) of 129 patients followed-up in our department for periods ranging from 6 to 168 months have revealed: (i) an overall median survival exceeding I 14 months; (ii) no difference between the survival curves of stage o (peripheral and bone marrow involvement only), stage I (stage o plus lymphadenopathy) and stage I1 (SCLL) patients; (iii) a significant difference (P ~ 0 . 0 1between ) the survival curves of stages 0 , I, I1 combined and stage 111 (stage o+lympha-
G. Dighiero et a1
I74
denopathy +splenomegaly) as well as stage IV (haemoglobin less than 10g/dl and/or platelets greater than 1 0 0 ~ 1 0 ~ / l(iv) ) ; the median survival of stage 111 patients was 70 months as compared to 23 months for stage IV patients. Unlike Rai et aZ(1975), we have classified SCLL in a distinct stage (stage 11). No deaths have been recorded among the eight patients in our series. Fifteen additional stage I1 patients have been reported here and their course does not differ from previous reports since no deaths have been recorded with a follow-up ranging from 18 month to 3 0 years. This warrants classification of these patients into a distinct stage associated with a hitherto unrecognized good prognosis. The failure to recognize the good prognosis of these patients is probably due to the fact that they were often confused with other predominantly splenic haematological disorders
I
2
3
4
5
6
7
8
9
10
Yeors
FIG I . Actuarial survival surves of 129 CLL patients according to disease stage at time of initial diagnosis. 0 , Stage o (48); x , stage I (33); -, stage I1 (8); O, stage 111 (28); 0,stage IV (12); A, total (129). The numbers in parentheses refer to the number of patients in each stage.
such as hairy cell leukaemia and PL. The latter was described by Galton et al(1974) and can be distinguished from CLL on the basis of clinical and cytological criteria. A third criterion appears to be useful for this distinction, namely the quantitation of SmIg. Clinically, PL is characterized by considerable splenic enlargement, the presence of anaemia and often thrombocytopenia and a usually rapidly fatal course. The cytological appearance of cells in PL is also characteristic; a high percentage of lymphoid cells with large vesicular nucleoli, relatively well condensed nuclear chromatin and large cell size. The cells that Galton called ‘prolymphocytes’ are associated with a variable percentage of mature small lymphocytes and blast-like cells. This picture differs from that usually encountered in CLL although typical lymphocytes may be seen in more slowly progressive forms, creating confusion with PL. Conversely, in certain cases of authentic PL, differential cytological diagnosis from atypical forms of CLL may be difficult. Thus, although the morphological appearance of the cells in PL differs from that of cells in CLL and acute lymphoblastic leukaemia, we feel that cytological criteria alone are not sufficient to establish a definite diagnosis. In this respect, the results of the immunocytochemical techniques presented in this work appear to provide an additional criterion for cell
Pure Splenic Forms of C L L
I75
identification. Indeed, compared to typical cases of CLL and SCLL, there was marked increase in the number of antigenic sites on the surface lymphoid cells from five of the six patients with PL. Although in four out of five cases quantitation was performed a very short time after initial diagnosis, and even though the results did not vary during the course of the disease (four patients were assayed at least twice), available data was insufficient to determine whether increased levels were present from the time of the onset of the disorder. This increase in SmIg is a good criterion only for the B type ofPL, and T-cell forms have been reported (Catovsky et al, 1973; Brouet et al, 1975). The only patient with PL in whom no SmIg was detected had both B-cell markers (80% EAC-rosettes) and T-cell markers (94% E-rosettes). A similar observation of a lymphoproliferative disorder with both T and B cell markers was reported by Shevach et a1 (1973). The simultaneous presence of both B- and T-cell markers on the same lymphocyte may reflect a disturbance in differentiation or it may mean that the proliferation alone arose from the small percentage of normal cells which bear both B- and T-cell markers (Shevach et al, 1973). Unlike Brouet et al (1975) who found that out of eight patients with T-cell CLL seven fulfilled the criterion for SCLL, we found that seven of our eight SCLL cases were of B-cell origin. Furthermore, of the four T-cell forms observed in a series of 129 CLL patients (Binet et al, 1977), all were stage I11 (lymphadenopathy + splenomegaly) (Dighiero et al, 1976).The only SCLL patient whose lymphocytes exhibited no SmIg, had a ‘null’ cell form (McLaughlin et al, 1973; Hamblin & Hough, 1977). This study has shown the usefulness of clinical cytological and immunological identification of SCLL which is associated with a favourable prognostic outlook. This individuality justifies a particular place in the anatomico-clinical classification of CLL and in the classification proposed by Rai et al (1975); they should be separated from forms in which adenopathy and splenomegaly coexist. REFERENCES BINET,J.L., LEPORRIER, M., DIGHIERO, G., CHARRON, D., D’ATHIS, PH., VAUGIER, G., MERLE BERAL, H., J.C., RAPHAEL, M., NIZET, M.G. & FOLLENATALI, ZOU, J.Y. (1977) A clinical staging system for CLL: prognostic significance. Cancer, 40, 855-864. BOURONCLE, B.A., WISEMAN, B.K. 81 DOAN,C.A. (1958) Leukemic reticuloendotheliosis. Blood, 13, 60-30. BROUET,J.-C., FLANDRIN,G., SASPORTES, M., PREUD’HOMME, J.L. & SELIGMANN, M. (1975)Chronic lymphocytic leukaemia of T-cell origin. Immunological and clinical evaluation in eleven patients. Lancet, ii, 890-893. CATOVSKY, D . , GALETI-0,J., OKOS,A , , GALTON, D.A.G.,WILTSHAW, E. & STATHOPOULOS, G. (1973) Prolymphocytic leukaemia of B and T cell type. Lancet, ii, 232-234. DIGHIERO, G., FOLLEZOU, J.Y ., ROISIN,J.P., TERNYNCK, T . & BINET, J.-L. (1976) Comparison of normal and chronic lymphocytic leukaemia lymphocytes surface Ig determinants using peroxidase-
labeled antibodies. 11. Quantitation of light chain determinants in atypical lymphocytic leukemia. Blood, 48, 559-566. D., D’ATHIS, DIGHIERO, G., VAUGIER, G., CHARRON, J.-L. (1977) Variation in lymphocyte PH. & BINET, counts four hours after administration of hydrocortisone in patients with chronic lymphocytic leukemia. Blood, 49, 719-728. FOLLEZOU, J.Y., DIGXIERO, G., ROISIN,J.P., TERNYNCK, T. & BINET,J.-L. (1977) Contribution of quantitative immunocytology to the study of lymphoid hemopathies. Biomedicine, 26, 330-336. GALTON, D.A.G., GOLDMAN, J.M., WILTSHAW, E., CATOVSKY, D., HENRY,K . & GOLDENBERG, G.J. (1974) Prolymphocytic leukaemia. Britishlournal of Haematology, 27, 7-23, HAMBLIN, T . & HOUGH, D. (1977) Chronic lymphatic leukaemia: correlation of immunofluorescent characteristics and clinical features. BritishJournal of Haematology, 36, 359-365. HANSEN, M.M. (1973) Chronic lymphocytic leukae-
176
G. Dighiero et a1
mia: clinical studies based on 189 cases followed for RAI, K.R., SAWITSKY, A., CRONKITE, E.P., CHANNA, a long time. Scandinavian Journal of Haematology , A.D., LEVY,R.N. & PASTERNACK, B.S. (1975)Clinical staging ofchronic lymphocytic leukemia. Blood, suppl. 18. JAYASWAL, U., ROATH,S., HYDE,R.D., CHISHOLM, 46, 219-234. D.M. & SMITH, J.L. (1977) Blood lymphocyte surSHEVACH, E.M.,JAFFE, E.S. & GREEN, I. (1973)Recepface lymphoproliferative disorders. BritishJournal of tors for complement and immunoglobulin on human and animal lymphoid cells. Transplantation Haematology, 37. 207-21 5. Reviews, 16,3-28. MARAL, J., AUCLERC,C., DENICOL, J.J., WEIL,M. & JACQUILLAT, C. (1976)LLC: Etude ritrospective de TERNYNCK, T., DIGHIERO, G., FOLLEZOU, J.Y. & 394 observations. IIPme Congrh Frangais d’HCmaBIN^, J.-L. (1974)Comparison of normal and CLL tologie, Le Touqiet, 24-27 Mai. lymphocyte surface Ig determinants using peroxiMCLAUGHLIN, G., WETHERLY-MEIN, G., PITCHER, C. dase-labeled antibodies. I. Detection and quanti& HOBBS, J.R. (1973) Non-immunoglobulin-beartation of light chain determinants. Blood, 43. ing ‘B’ lymphocytes in chronic lymphatic leukae789-79 5 . mia? BritirhJournal of Haematology, 25, 7-14, TOURAINE, A. (1922) Variktks cliniques de la leuckmie lymphoide chronique. Formes spleniques pures, PHILIPS,E.A., KEMPIN,S., PASSE,S., MIKB, V. & CLARKSON, B. (1977) Prognostic factors in chronic lymphocytkrniques et splknocytkmiques. Journal Mgdicine Francaise, 10, 427-43 I . lymphocytic leukaemia and their implications for therapy. Clinics in Haematology, 6, 203-222.
Identification of a Pure Splenic Form of Chronic Lymphocytic Leukaemia G. DIGHIERO, D. CHARRON, P. DEBRE, M. LE PORRIER, G. VAUGIER, J. Y. FOLLEZOU, L. DEGOS,CL.JACQUILLAT A N D J.-L. BINET Groupe Hdpital Pitie-Salpetrih, Paris (Received
20
March 1978; accepted for publication 5 July 1978)
SUMMARY. We have recently proposed a new staging system for chronic lymphocytic leukaemia (CLL) in which patients with isolated splenomegaly arc classified into a distinct stage (stage 11). Twenty-three such patients (from two institutions) have been studied without recorded death in a follow-up of 1 8 months to 3 0 years. This favourable prognosis justifies separation of these ‘pure splenic forms’ (SCLL) which must be distinguished from what Galton has termed prolymphocytic leukaemia (PL). This distinction can be made on the basis of three criteria: (i) Clinically, SCLL has a slow uneventful course and neither anaemia and/or thrombocytopenia; (ii) cytologically PL can be distinguished from other forms of CLL though atypical forms of CLL may be confused with the former; and (iii) the study of surface membrane immunoglobulins (SmIg) showed that while lymphocytes from most patients with both PL and SCLL bore uniform SmIg, suggesting a monoclonal B-cell proliferation, there was a major quantitative difference in that whereas PL lymphocytes had a number of antigenic sites close to that of normal lymphocytes (mean: 82 ooo sites per cell), SCLL lymphocytes had a drastically reduced number of sites. It is our opinion that this is an important criterion for the differential diagnosis between PL and SCLL. Splenomegaly is a common feature of chronic lymphocytic leukaemia (CLL). As a rule, this splenomegaly is accompanied by lymph node enlargement. However, in some cases it may appear alone without clinically perceptible lymph node enlargement (Touraine, 1922). Patients with isolated splenomegaly appear to be rare in series reported in the literature: I I of 188 in the review by Hansen (1973), 1 5 of 394 at the Haematology Clinic of the St Louis Hospital in Paris (Maral et al, 1976), eight of 129 in our personal experience (Binet etal, 1977). After dividing our patients into five stages on the basis of clinical and haematological criteria (Binet et al, 1977),it came to our attention that all the patients in group 11, defined by clinically isolated splenomegaly, were still alive. Moreover, these patients had a slow and uneventful course, even when not treated. It appeared to us that separation of this pure splenic form of CLL (SCLL) was justified in view of the good prognosis with which it was associated. Following the work of Bouroncle (1958), haematologists have learnt to recognize the Correspondence: Dr G. Dighiero, Departement d’HCmatologie, Groupe Hbpital Pitie-Salpetricre, 147 Boulevard de I’Hbpital, 75634 Paris Cedex 1 3 , France. 0007-1o~8/~g/ozoo-o16~$o2.oo 01979 Blackwell Scientific Publications
169
170
G. Digkiero et a f
cytological features of the hairy cell, characteristic of hairy cell leukaemia and no longer include them among cases of CLL. More recently, Galton et at (1974) have identified prolymphocytic Ieukaemia (PL) on the basis of its course and clinical and cytological features. We have previously demonstrated the usefulness of quantitative analysis of surface membrane immunoglobulin (SmIg) in these forms (Dighiero et al, 1976; Follezou et al, 1977). The good prognosis associated with SCLL has not been highlighted to date and it is not impossible that the two above-mentioned disorders have long been confused with SCLL. In this work we have combined our cases of SCLl with those of the Haematology Clinic of the St Louis Hospital. The clinical, cytological and immunological features of these cases have been compared to those of six cases of PL studied by us. MATERIALS AND METHODS Patients Twenty-nine patients with peripheral and bone marrow lymphocytosis, accompanied by splenomegaly, were included in the study. The patients were divided into two groups: (i) Six patients fell into a category of ‘prolymphocytic leukaemia’ as described by Galton et a1 (1974).
(ii) The 23 remaining patients fulfilled the classical criteria for CLL. In all patients, initial diagnosis was made on the basis of peripheral lymphocytosis of greater than 1 5 x 109/l combined with bone marrow lymphocyte infiltration greater than 40%. In six cases the diagnosis of CLL was made on the basis of a lymphocytosis of greater than 4 x 109/l persisting for longer than 6 months and accompanied by a medullary infiltration of greater than 40% on bone marrow and lymphocytic infiltration on histological biopsy. All these patients had isolated splenic involvement and were classified as stage I1 in an anatomico-clinical staging system, similar to that of Rai et al(1975),which we have recently proposed. Eight patients were from the Haematology Department of Pitii SalpCtrikre,the other I 5 being from the Haematology Clinic of the St Louis Hospital. For these last 1 5 patients, belonging to a series of 394 cases of CLL, only a retrospective study was performed. Immunocytological Methods These studies could only be performed in the 14 patients (eight CLL and six PL) from our own series. The methods used have been previously described in detail (Ternynck et al, 1974) and are briefly summarized. Preparation of anti-kappa and anti-lambda chain antibodies. Anti-kappa and anti-lambda antibodies were isolated from goat antisera using the appropriate light chain immunoabsorbent. They were rendered monospecific for kappa and lambda chains by absorption with an immunoabsorbent of the opposite type, prepared by coupling Bence-Jones proteins to glutaraldehyde-treated polyacrylamide beads. Peroxidase-labelled antibodies. Purified antibodies were coupled to peroxidase with glutaraldehyde using a two-step procedure (Ternynck et al, 1974). Purification of peripheral blood lymphocytes. Lymphocytes were purified from heparinized
Pure Splenic Forms of CLL
171
peripheral blood by sedimentation for I h a t 37°C or by centrifugation on a ficoll-triosil gradient. Cell preparations contained an average of 95% lymphoid cells and cell death determined by trypan-blue exclusion never exceeded 10%. Reaction ofcells with enzyme-labelled antibody. Samples of 5 x 10’ cells were incubated either at 4°C or 37°C with 0.5 ml of peroxidase-labelled anti-kappa or anti-lambda antibodies for 2 h. A concentration of 0.125 mg/ml of labelled antibody was used. Light microscopy. Aliquots of cells treated at 37°C were washed four times and then examined under the light microscope after staining with a solution of 3-amino-9-ethylcarbazole and hydrogen peroxide for 10min. One drop of the cell suspension was mounted under a cover slip and the percentage of labelled versus unlabelled cells was obtained after counting 500 cells. Measurement ofperoxiduse activity. Peroxidase activity was determined by adding 2.9 ml of the 0-dianizidine reagent to 0.1 ml aliquots containing 5 x 106 cells. The reaction was stopped after I 5 min with a drop of 6 N HC1. Tubes were then centrifuged at 5000 g for 5 min and the absorbancy of the supernatant was read at 400 nm. Taking into account the number of cells and the activity found, the quantity of peroxidase fixed per 106 cells was calculated by referring to a standard curve of absorbancy versus reaction time for known quantities of peroxidase. Since the mean ratio of peroxidase coupled to antibodies is one to one, the number of molecules of antibody found per I O cells ~ could be calculated, and the mean number of antigenic sites on the surface of each positive lymphocyte could be determined. RESULTS The clinical, haematological and follow-up features of patients with SCLL and PL are summarized in Table I. The picture was uniform for the 23 SCLL patients: the majority of patients (14123) were over 60 years old; the spleen though clinically enlarged in all, remained above the umbilicus in 12 cases. Anaemia and thrombocytopenia were always absent and lymphocytosis was moderate, exceeding 75 x 109/l in only one patient. The cytological picture did not differ from that usually seen in CLL. No deaths were recorded among these patients with a follow-up as long as 30 years. All patients have now been followed up for at least 18 months and 14 have a follow-up of more than 5 years. The course of these patients was uneventful, even without treatment. The clinical picture of the six patients with PL (Table I) contrasted with that of the former. Marked splenic enlargement below the umbilicus was seen in all cases. Peripheral lymphocytosis exceeded 75 x 109/l in four patients and even 500 x ro9/l in two of these patients, and anaemia and thrombocytopenia were present in all cases. Lastly, and most important, the prognosis was extremely poor, four patients having a survival of less than 6 months. The survivors now have a follow-up of 6 months. O n smears, the cells from these patients did not present a uniform appearance. Small apparently normal lymphocytes, blastlike cells and particularly prolymphocytes were encountered in variable proportions. Immunocytologicalfeatures are shown in Table 11. In five of the six patients with PL, peripheral lymphocytes bore easily detectable SmIg. Over 90% of the cells were tagged either by peroxidase-labelled anti-lc (three patients) or anti-1 antibodies, thus suggesting a monoclonal B-cell proliferation. Quantitation of cell surface antigenic sites revealed a mean value of 82 ooo sites per cell.
53 73 62
62
58 66
M F F
M
M F
2
4
6
5
3
I
(yr)
Age
Sex
14
9
23
9 0123
Superficial nodes
0
Microlymphadenopathy Microlymphadenopathy
0
0
0
Lymph nodes
14
> 60
Age (Y r)
< 60
Patient
F
Sex
M
~
No. of patients
7
I2
I0
14
I0
I2
I1
>6cm
WBC
25
30 4s
8.8 8.7
24.9 580
IS 30
50
Platelets ( x 109/r)
8.2
7 6 7. s
Hb (g/dl)
13.3
24
700 7s 75
( x 109/1)
m=
11-18
4-120
m = 22.13
Hb @/do
Peripheral lymphocytes ( x 10911)
Prolymphocytic leukaemia
Spleen (cm)
I2
t6cm
Spleen
Treatment
I 1/23
Treatment
6 months
6 months 4 monthst
Splenectomy, steroids CVP
6 monthst 7 monthst 8 monthst
Survival
I0
5-10 yr
Follow-up
CVP
9
< 5 yr
Chlorambucil, MOPP Chlorambucil, vincristine Splenic irradiation
m = 188
120-256
(x1o9/1)
Platelets
TABLE I. Pure splenic forms of chronic lymphocytic leukaemia (stage 11)
4
>lo
yr
3
EL
Y
""s -_ rn
N
4
Pure Splenic Forms of CLL
I73
A monoclonal B-cell appearance was also found in seven patients with SCLL (five IC and two
A). However, the mean number of antigenic sites found on the surface of these lymphocytes was 1 2 500 which is similar to that found on lymphocytes from patients with CLL in general and is appreciably lower than the number of antigenic sites found on control lymphocytes. One case of SCLL was found to correspond to a proliferation of 'null' cells since less than 5% of cells were positive for SmIg, 3 % of cells formed B-rosettes and 6% formed E-rosettes. One patient with PL had less than 5 % SmIg positive lymphocytes, but 80% of cells formed B-rosettes and 93 % formed E-rosettes. TABLE 11. Surface membrane immunoglobulins (quantitativestudy) Patient
Anti-x
Anti-2 Sites per cell
Sites per cell % Positive cells
x
103
% Positive cells
Prolymphocytic leukaemia B-cell type (five patients) I
-
99 98
97-5 65
99 -
-
-
93
6
-
2
3 4 5
Pure splenic forms of CLL B-cell type (seven patients) I
2
3 4 5 6 7
CLL
68 -
10.5
70 86 68 63
8.2
36/46
Mean 74% Normal subjects
Mean 19.4
-
88
-
13.5
77 94 -
Mean 9.5 (range 2-25)
Mean 70%
5.5 9.7
Mean 93.7 (range 86-185)
10146
Mean 9.5
x
103
75 113
-
11.5
29
-
Mean 10.2 (range 2.5-29) Mean 106 (range 32.5-190)
DISCUSSION For a long time CLL was considered a disease of unpredictable course. Recently, Rai et al(1975) and Binet et al(1977)suggested a new five-stage anatomico-clinical classification system which has proved to be of prognostic significance (Philipps et al, 1977). The actuarial survival curves (Fig I) of 129 patients followed-up in our department for periods ranging from 6 to 168 months have revealed: (i) an overall median survival exceeding I 14 months; (ii) no difference between the survival curves of stage o (peripheral and bone marrow involvement only), stage I (stage o plus lymphadenopathy) and stage I1 (SCLL) patients; (iii) a significant difference (P ~ 0 . 0 1between ) the survival curves of stages 0 , I, I1 combined and stage 111 (stage o+lympha-
G. Dighiero et a1
I74
denopathy +splenomegaly) as well as stage IV (haemoglobin less than 10g/dl and/or platelets greater than 1 0 0 ~ 1 0 ~ / l(iv) ) ; the median survival of stage 111 patients was 70 months as compared to 23 months for stage IV patients. Unlike Rai et aZ(1975), we have classified SCLL in a distinct stage (stage 11). No deaths have been recorded among the eight patients in our series. Fifteen additional stage I1 patients have been reported here and their course does not differ from previous reports since no deaths have been recorded with a follow-up ranging from 18 month to 3 0 years. This warrants classification of these patients into a distinct stage associated with a hitherto unrecognized good prognosis. The failure to recognize the good prognosis of these patients is probably due to the fact that they were often confused with other predominantly splenic haematological disorders
I
2
3
4
5
6
7
8
9
10
Yeors
FIG I . Actuarial survival surves of 129 CLL patients according to disease stage at time of initial diagnosis. 0 , Stage o (48); x , stage I (33); -, stage I1 (8); O, stage 111 (28); 0,stage IV (12); A, total (129). The numbers in parentheses refer to the number of patients in each stage.
such as hairy cell leukaemia and PL. The latter was described by Galton et al(1974) and can be distinguished from CLL on the basis of clinical and cytological criteria. A third criterion appears to be useful for this distinction, namely the quantitation of SmIg. Clinically, PL is characterized by considerable splenic enlargement, the presence of anaemia and often thrombocytopenia and a usually rapidly fatal course. The cytological appearance of cells in PL is also characteristic; a high percentage of lymphoid cells with large vesicular nucleoli, relatively well condensed nuclear chromatin and large cell size. The cells that Galton called ‘prolymphocytes’ are associated with a variable percentage of mature small lymphocytes and blast-like cells. This picture differs from that usually encountered in CLL although typical lymphocytes may be seen in more slowly progressive forms, creating confusion with PL. Conversely, in certain cases of authentic PL, differential cytological diagnosis from atypical forms of CLL may be difficult. Thus, although the morphological appearance of the cells in PL differs from that of cells in CLL and acute lymphoblastic leukaemia, we feel that cytological criteria alone are not sufficient to establish a definite diagnosis. In this respect, the results of the immunocytochemical techniques presented in this work appear to provide an additional criterion for cell
Pure Splenic Forms of C L L
I75
identification. Indeed, compared to typical cases of CLL and SCLL, there was marked increase in the number of antigenic sites on the surface lymphoid cells from five of the six patients with PL. Although in four out of five cases quantitation was performed a very short time after initial diagnosis, and even though the results did not vary during the course of the disease (four patients were assayed at least twice), available data was insufficient to determine whether increased levels were present from the time of the onset of the disorder. This increase in SmIg is a good criterion only for the B type ofPL, and T-cell forms have been reported (Catovsky et al, 1973; Brouet et al, 1975). The only patient with PL in whom no SmIg was detected had both B-cell markers (80% EAC-rosettes) and T-cell markers (94% E-rosettes). A similar observation of a lymphoproliferative disorder with both T and B cell markers was reported by Shevach et a1 (1973). The simultaneous presence of both B- and T-cell markers on the same lymphocyte may reflect a disturbance in differentiation or it may mean that the proliferation alone arose from the small percentage of normal cells which bear both B- and T-cell markers (Shevach et al, 1973). Unlike Brouet et al (1975) who found that out of eight patients with T-cell CLL seven fulfilled the criterion for SCLL, we found that seven of our eight SCLL cases were of B-cell origin. Furthermore, of the four T-cell forms observed in a series of 129 CLL patients (Binet et al, 1977), all were stage I11 (lymphadenopathy + splenomegaly) (Dighiero et al, 1976).The only SCLL patient whose lymphocytes exhibited no SmIg, had a ‘null’ cell form (McLaughlin et al, 1973; Hamblin & Hough, 1977). This study has shown the usefulness of clinical cytological and immunological identification of SCLL which is associated with a favourable prognostic outlook. This individuality justifies a particular place in the anatomico-clinical classification of CLL and in the classification proposed by Rai et al (1975); they should be separated from forms in which adenopathy and splenomegaly coexist. REFERENCES BINET,J.L., LEPORRIER, M., DIGHIERO, G., CHARRON, D., D’ATHIS, PH., VAUGIER, G., MERLE BERAL, H., J.C., RAPHAEL, M., NIZET, M.G. & FOLLENATALI, ZOU, J.Y. (1977) A clinical staging system for CLL: prognostic significance. Cancer, 40, 855-864. BOURONCLE, B.A., WISEMAN, B.K. 81 DOAN,C.A. (1958) Leukemic reticuloendotheliosis. Blood, 13, 60-30. BROUET,J.-C., FLANDRIN,G., SASPORTES, M., PREUD’HOMME, J.L. & SELIGMANN, M. (1975)Chronic lymphocytic leukaemia of T-cell origin. Immunological and clinical evaluation in eleven patients. Lancet, ii, 890-893. CATOVSKY, D . , GALETI-0,J., OKOS,A , , GALTON, D.A.G.,WILTSHAW, E. & STATHOPOULOS, G. (1973) Prolymphocytic leukaemia of B and T cell type. Lancet, ii, 232-234. DIGHIERO, G., FOLLEZOU, J.Y ., ROISIN,J.P., TERNYNCK, T . & BINET, J.-L. (1976) Comparison of normal and chronic lymphocytic leukaemia lymphocytes surface Ig determinants using peroxidase-
labeled antibodies. 11. Quantitation of light chain determinants in atypical lymphocytic leukemia. Blood, 48, 559-566. D., D’ATHIS, DIGHIERO, G., VAUGIER, G., CHARRON, J.-L. (1977) Variation in lymphocyte PH. & BINET, counts four hours after administration of hydrocortisone in patients with chronic lymphocytic leukemia. Blood, 49, 719-728. FOLLEZOU, J.Y., DIGXIERO, G., ROISIN,J.P., TERNYNCK, T. & BINET,J.-L. (1977) Contribution of quantitative immunocytology to the study of lymphoid hemopathies. Biomedicine, 26, 330-336. GALTON, D.A.G., GOLDMAN, J.M., WILTSHAW, E., CATOVSKY, D., HENRY,K . & GOLDENBERG, G.J. (1974) Prolymphocytic leukaemia. Britishlournal of Haematology, 27, 7-23, HAMBLIN, T . & HOUGH, D. (1977) Chronic lymphatic leukaemia: correlation of immunofluorescent characteristics and clinical features. BritishJournal of Haematology, 36, 359-365. HANSEN, M.M. (1973) Chronic lymphocytic leukae-
176
G. Dighiero et a1
mia: clinical studies based on 189 cases followed for RAI, K.R., SAWITSKY, A., CRONKITE, E.P., CHANNA, a long time. Scandinavian Journal of Haematology , A.D., LEVY,R.N. & PASTERNACK, B.S. (1975)Clinical staging ofchronic lymphocytic leukemia. Blood, suppl. 18. JAYASWAL, U., ROATH,S., HYDE,R.D., CHISHOLM, 46, 219-234. D.M. & SMITH, J.L. (1977) Blood lymphocyte surSHEVACH, E.M.,JAFFE, E.S. & GREEN, I. (1973)Recepface lymphoproliferative disorders. BritishJournal of tors for complement and immunoglobulin on human and animal lymphoid cells. Transplantation Haematology, 37. 207-21 5. Reviews, 16,3-28. MARAL, J., AUCLERC,C., DENICOL, J.J., WEIL,M. & JACQUILLAT, C. (1976)LLC: Etude ritrospective de TERNYNCK, T., DIGHIERO, G., FOLLEZOU, J.Y. & 394 observations. IIPme Congrh Frangais d’HCmaBIN^, J.-L. (1974)Comparison of normal and CLL tologie, Le Touqiet, 24-27 Mai. lymphocyte surface Ig determinants using peroxiMCLAUGHLIN, G., WETHERLY-MEIN, G., PITCHER, C. dase-labeled antibodies. I. Detection and quanti& HOBBS, J.R. (1973) Non-immunoglobulin-beartation of light chain determinants. Blood, 43. ing ‘B’ lymphocytes in chronic lymphatic leukae789-79 5 . mia? BritirhJournal of Haematology, 25, 7-14, TOURAINE, A. (1922) Variktks cliniques de la leuckmie lymphoide chronique. Formes spleniques pures, PHILIPS,E.A., KEMPIN,S., PASSE,S., MIKB, V. & CLARKSON, B. (1977) Prognostic factors in chronic lymphocytkrniques et splknocytkmiques. Journal Mgdicine Francaise, 10, 427-43 I . lymphocytic leukaemia and their implications for therapy. Clinics in Haematology, 6, 203-222.
