Document zzVpzQvX4RLDorQxyMYwbYq2z

... Associated Chronic Lymphocytic Leukemia and Multiple Myeloma: Origin From a Single Clone By J.P. Fermand, J.M. James, P. Herait, and J.C. Brouet We investigatedthe clonalrelationshipof malignant cells in a patient affected with both chronic lymphocytic leukemia (CLL) and multiple myeloma (MM).CLL cells and malignant plasma cells synthesized igG,K and lgAK molecules, respectively; these monoclonal Ig shared idiotypic determinants, providing evidence that a single clonal disease occurred in this patient. Furthermore, when leukemic CLL cells were driven to differentiatein vitro to immunoblastsand pla )ma cells, a switch from IgG to IgA occurred in a significant percentage of cells that were double producers. Tt 888 data suggest that, in some circumstances, CLL leukemic B cells may reach a more mature state, leading to the occurrence of clinical MM. a 1985 by Grune & Stratton, Inc. THE OCCURRENCE of both chronic lymphocytic leukemia (CLL) and multiple myeloma (MM) in one patient is a rare and noteworthy association of two B cell neoplasias. Immunologic investigations in such cases point so far to a biclonal disease process, although only few cases could be adequately studied.' We report here a case of a patient with shared idiotypes between the IgGK molecules synthesized by the CLL cells and the IgAK molecules secreted by the plasma cell proliferation. Moreover, in vitro studies showed that the CLL cells could be driven to differentiate to immunoblastic or plasmacytoid cells with a switch from IgG to IgA. These data provide strong evidence for a clonal disease and suggest that in some cases CLL leukemic B ctlls may be triggered in vivo or in vitro to evolve to a more mature but still malignant state. MATERIALS AND METHODS Case record. The diagnosis of CLL was made in 1969 in a 61-year-old woman with no significant past medical history. Physical examinationwas normal; WBCs were 60.109/mL with 90Csmall lymphocytes. During the next ten years, WBCs were stable at approximately 20.109/mLunder intermittent chlorambucil therapy. In August 1979, a fracture of the patient's left humerus occurred after minor trauma; x-ray studies revealed a single lytic lesion. NO Mhcr investigationswere done until two years later, when bone pains Wrred. WBCs were unchanged at this time; bone marrow biopsy h w e d invasion by small lymphocytesand few plasma cells. Needle YPration of the lytic humeral lesion showed a diffuse plasma cell Infiltration.Serum electrophoresis disclosed a monoclonal spike (20 8/L) identified as lgAK by immunoelectrophoresis.Proteinuria was O.15 g/L with traces of monoclonal K chains. During the next two yam. bone lytic lesionsdeveloped in the patient's skulland vertebral hie, with diffuse osteopenia despite chemotherapy with cyclohphamide. cell cultures. Peripheral blood lymphocytes (PBLs) were isoby centrifugation on a Ficoll-Hypaque density gradient, culat a concentration of 1.106/mL with or without the following phytohemagglutinin (PHA), 5 pg/mL; concanavalin A (%A). 5 rg/mL; and Staphylococcus aureus (SAC) (Cowan rrain 1, 1.10d3vol/vol). In other experiments, the patient's PBLs CO-cultured with allogeneic T cells (E rosetting cells from a -1 donor) with or without addition of mitogens. Finally, the Pknt's PBLs were activated for three days by SAC before the rddi~iOonf conditionedT cell medium (at a 20% final concentration) Was obtained after a six-day stimulation of purified T cells with * and phorbol myristate acetate (PMA), 10 ng/mL. Cultures out for six to I2 days; at different times, a sample of examined for surface and cytoplasmic Ig. AMi-idiolypic antiserum to patient 3 monoclonal IgA. The on a Sepharose 6 B column. The ascending limb of the first peak was concentrated by vaci ium dialysis; it contained mainly the monoclonal 1gAK and a few cont tmhating proteins as assessed by double immunodiffusion using i: ntisera to IgA, IgG, IgM, and a polyvalent rabbit antiserum to hul nan serum. An anti-idiotypic serum was prepared in rabbits by repei ted footpad or subcutaneous (SC) immunizations with 350 pg of the purified IgA in Freund's adjuvant. Immune serum was absorbed with polyclonal IgG, IgM, and IgA. In double diffusion experiments, the absorbed serum reacted only with the patient's IgAn. By indi 'ect immunofluorescence,this serum was negative on pokeweed mita ten (PWM)-stimulated PBLs, which contained a fair percentage of I tA, IgG, and 1gM plasma cells. Immunopuorescenr procedures. Surface and cytoplasmic Ig were characterized by direct immunofluorescence using rhodan ine or fluorescein-conjugated Fab'2 fragments of rabbit IgG mono! pecific for c. -y,& K and X Ig chains as describedelsewhere' Determi nation of IgG subclass of surface and cytoplasmic IgG synthesized by the patient's leukemic cells was performed by using monoclc nal antibodies to the various IgG subclasses (a kind gift of Dr J.D. Capra) and a second layer of goat antibodies to mouse Ig. Doul lelabeling for surface and cytoplasmic Ig were performed by f rst staining living cells with rhodamine reagents; after three washii gs, cells were smeared on slides, fixed in 95% ethanol, and thereal ter were restained for cytoplasmic Ig with fluoresceinated antibod es. Reactivity of the anti-idiotypic serum was studied by indir :ct immunofluorescence, using as a second layer Fab2 fragments of goat IgG antibodies to rabbit Ig that had no cross-reactivity with human lg by direct immunofluorescence. RESULTS Membrane markers of leukemic CLL cells and malign6 nt plasma cells. Nearly 100%of blood lymphocytes featur Ed a dim surface staining with anti-y and -K sera; no positive cells were seen with anti-p, -a,-6, or -A antibodies. Intracyl oplasmic staining of CLL cells showed numerous crystalli ie inclusions brightly reactive with anti-y and -K antibodit :s. These IgG molecules belonged to the y , subclass. Besides - From the Laboratory of Immunochemistry and Immunopatho, D gy (INSERM U108,Research Insiiiute on Blood Diseases of 1 he University of Paris VII, and the Laboratory of Oncology ard Immunohematology of CNRS). Hbpital Saint-Louis. and Servr re d'H6matologie. Hbiel-Dieu, Paris. Submitted Nov 29, 1984;accepted Jan 26. 1985. Address reprint requests to Dr J.C. Brouet. Laboratory o/ Immunochemistry and immunopathology, INSERM lJ108.Ha)li- tal Saint-Louis. Place du Dr Fournier. 75475 Paris. Cedex 10, France. B 1985 by Grune & Stratton. lnc. 0006-497 I/85/6602-ooO7$03.00/0/0 f I--- 2s 1 292 FERMAND ET y small lymphocytes, the bone marrow was infiltrated with plasma cells containing IgAK molecules. No plasma cells (or IgA-positive cells) were detected among PBLs. Induction of digerentiation of leukemic CLL cells. In order to induce differentiation of leukemic lymphocytes, the patient's PBLs were cultured in the presence of different mitogens and of allogeneic T cells. The Ig phenotype of the stimulated cells was determined after various times of culture. Table l indicates results of representative experiments. No modification was observed in unstimulated cultures. In the presence of mitogens and/or allogeneic T cells or T cell conditioned medium, two main changes were observed. At day 8, a few blast cells bearing surface a chains were seen; double-labeling experiments for surface and cytoplasmic Ig showed that some of these surface a-positive cells exhibited simultaneously characteristic crystalline y inclusions. At day 12 of culture, cytoplasmic immunofluorescence revealed that 3% to 10% of the cells contained a chains (Table I); some had the morphological aspect of plasma cells. Some of these cells also contained y heavy chains, as shown by simultaneous staining with rhodamine anti-a! and fluorescein anti-y antibodies. In the latter cells, the characteristic aspect of the crystalline inclusions was often modified because of a decrease in the number of crystals. Idiotypic studies. An anti-idiotypic rabbit serum was raised against the patient's serum monoclonal IgA. After suitable absorptions, its specificity was confirmed by lack of staining of polyclonal plasma cells. This antiserum stained by indirect fluorescence the cytoplasmic YK crystalline-like inclusions of leukemic CLL B cells. The staining was abolished by further absorption of the antiserum with patient's IgAK but not by unrelated I g h . DISCUSSION The clinical and hematological diagnosis of CLL is not questionable in patient L.H.E. The immunologic characterization of the leukemic C L L cells revealed a homogeneous proliferation of lymphocytes featuring surface and cytoplasmic IgGlK molecules. For ten years, the patient had a marked blood and bone marrow lymphocytosis when M M occurred: lytic bone lesions developed, with progressive infiltration of the bone marrow by malignant plasma cells that secreted 1gAK molecules. The assxiation in one patient oT CLL and M M is quite rare, sinc: less than 40 cases have been reported.' In most cases, as ir the present one, CLL ha, been diagnosed one to 15 years >efore MM. The clinical picture is usually a composite of typical features of both diseases; however, the incidence o ' extraosseous plasma cell tumors may be outlined. The clas i distribution of serum or urinary monoclonal components of M M supervening on CLL appears rather unusual, since Igl L cases predominate over IgG, light chains, IgM, IgD, or nc Insecreting myleomas. T,, account for this rare association, two main hypotheses are currently offered: the malignant B CLL clone might funher mature to give rise to a proliferatia n of plasma cells; alterna. tively. the two diseases could repre rent distinct clonal prolif. erations. Immunologic studies pe .formed SO far favor the latter hypothesis. In six of 12cases it~died,'.~C-'L L cells and plasma cells synthesized Ig mole1:ules with different light chain types, a widely accepted alt iough circumstantial evi. dence for distinct clonal origin. Moreover, in three c a s e in which surface Ig of CLL cells and cytoplasmic Ig of plasma cells had the same light chain type, they did not share idiotypic determinants.`" It should be pointed out, however, that the finding of two clones synthesizing idiotypically different molecules (or even Ig mol m l e s with different light chain types) does not definitely e iminate a clonal disease affecting a precursor cell. Indeed. somatic mutations may modify idiotopes on one cell line a i d , on the other hand, the occurrence of a common V, gene rearrangement with different V, gene rearrangements ir different subclones has been suspected in a recent study.' 1n contrast, in the present report, CLL cells and malignant plasma cells synthesized IgG and IgA molecules, respectivtly, that shared common idiotypic determinants. Indeed, a r abbit anti-idiotypic antiserum to the serum monoclonal IglLstained the cytoplasmic IgG crystalline-like inclusions char 3cteristic of the patient's CLL cells. Furthermore, we obtained in vitrl) evidence that CLL cells were the precursors of IgA plasma cells. To explore further the clonal relationship between the two diseases in our patient, we investigated whether le ukemic C L L cells could be induced to differentiate in vit .o.~-" According to the stimulus used, 3% to 10%aK-SyntheSiZing immunoblasts or plasma cells were observed in late cultures (at day 12) of Table 1. Percentage of Positive Cells for Slg or clG 'Y or a Heavy Chains at Days 0 , 8 , and 12 of Cult!ire PELSof L.H.E. Medium Alone Culture Conditions and Day of Culture + +SAC PHA Allogeneic PHA T Cdls 160%) SAC + T W l s Supernatant Added at Day 3 of Culture 0 8 12 8 12 8 12 8 12 sb Y 60 ND ND ND ND ND ND ND ND (Weak) a0 0000 3 03 0 clG Y 95 a0 95 95 9 0 90 50 50 92 90 0 0 0 3 0 7 0 10 slg, surface immunoglobulin;clg, intracytoplasmic immunoglobulin;PHA, phytohemagglutinin;SAC, S aureus; ND, not done. CLONAL ORIGIN OF ASSOCIATED CLL AND MM 2! I3 PBLs that initially contained over 95% monoclonal YKbearing cells. That these a cells did not originate from precursors or residual B cells could be ascertained by double-staining experiments that disclosed cells having both cytoplasmic a and y chains; the latter cells displayed the crystalline inclusions characteristic of the patient's CLL cells. Finally, earlier in culture (at day 8 ) , cells with surface (YK molecules and cytoplasmic Y K inclusions were observed. Altogether, these findings offer strong evidence that the leukemic CLL cells could be induced to differentiate in vitro with a switch from IgG to IgA synthesis. The in vitro switch we observed was triggered by T cells (or T cell factors) as it probably occurs in vivo for normal B cells. This finding raises the important issue of whether leukemic clones may be sometimes susceptible to regulation by physiologic T mediators or cognate interaction with T cells. To what extent such an event occurs in vivo in our patient is presently unknown. On the other hand, it should be emphasized that there are many exceptions to the conventional view that leukemic CLL cells are frozen a t a given step of maturation. Apart from the possible maturation and switch to plasma cells as was shown in this patient, CLL cells more often transform into large blastic cells (the so-called Richter's syndrome) having usually (but not always) the same clonal origin.'**'' The occurrence of in vitro heavy chain switch is remarkable and has been seldom noted in human lymphoid malig- nancies. Rudders" reported a patient with CLL having t\ro leukemic clones featured by idiotypically related YK and I KK molecules; in this patient, a small percentage of cells synth:- sized both molecules. In vitro experiments in two other CL L patients demonstrated that stimulated leukemic cells we -e able to mature in vitro, with a switch from IgM to IgG ~ y n t h e s i s . ~W. ' ~e have no data about the molecular everts implied in the switch phenomenon we observed. Howevcr, the occurrence of cells synthesizing both IgG and IgA and If IgA plasma cells suggests two possibilities that are nit mutually exclusive. First, the C y l and CaI genes could ,e cotranscribed and expressed by an RNA processing mech Inism as occurs in B lymphocytes coexpressing IgM and Igl>. A long transcript would be needed, since y, and oIgenes a -e separated by 32 kb. Such a model of a large multi-C, gel le transcript has experimental support.I6 Alternatively, yI (IH gene (and 5' constant genes) may be deleted, as occurs universally in plasmocytomas and also in murine B ce Is stimulated by polyclonal activators"; in this case, a Ion Jlived y chain mRNA could account for our finding of double producers. Whatever the mechanism involved, such a disea ie association offers a unique opportunity for the study of tl re heavy class switch in B cells. For instance, our data strong iy suggest the existence of a switch from IgG to IgA, where IS most studies of the differentiation of human circulating B cells are rather consistent with the occurrence of dire:t switches from p to each of the other isotypes." REFERENCES 1. Brouet JC, Fermand JP, Laurent G, Grange MJ. Chevalier A. Jaquillat C, Seligmann M: The simultaneousoccurrenceof chronic lymphocyte leukemia and multiple myeloma: A study of eleven patients. Br J Haematol 5955, 1985 2. Preud'homme JL, Labaume S: Immunofluorescent staining of human lymphocytes for the detection of surface immunoglobulins. Ann NY Acad Sci 254:254, 1975 3. Hoffman KD, Rudders RA: Multiple myeloma and chronic lymphocytic leukemia in a single individual. Immunologic studies and review of the literature. Arch Intern Med 137:232, 1977 4. Kough RM, Makary AZ: Chronic lymphocytic leukemia (CLL) terminating in multiple myeloma. Blood 52:532, 1978 5. kdersen-Biergaard J, Draminsky Petersen H, Thomsen M, Wiik A, Wolff-JensenJ: Chronic lymphocytic leukaemia with s u h u e n t development of multiple myeloma. Scand J Haematol 21:256, 1978 6. Zalcherg JR, Cornel1 FN, Ireton JC, McGrath KM, McLach- lan R. WoodruffRK, Wiley JS: Chronic lymphatic developing in a mtient with multiple myeloma. Cancer 50594. 1982 7 . Jeha MT, Hamblin TJ, Smith JL: Coincident chronic lympho- W l C leukemia and osteosclerotic multiple myeloma. Blood S7:617. 1981 8. Skhr J, Cleary ML, Thielemans K, Gralow J, Warnke R, Levy R: Biclonal B-cell lymphoma. N Engl J Med 3 I I :20, 1984 9. Guglielmi P, Preud'homme JL, Ciorbaru-Barot R, Seligmann M:Mibgen-induced maturation of chronic lymphocytic leukemia B lyrnpkYta.J Clin lmmunol 2:186, 1982 . lo- T6tterman TH, Nilsson K, Sundstrom C: Phorbol esterlndUdwifferentiation of chronic lymphocytic leukaemic cells. Nature 288:176, 1980 1 1 . Fu SM, Chiorazzi N, Kunkel HG: Differentiation capaci .y and other properties of the leukemic cells of chronic lymphocyl ic leukemia. lmmunol Rev 48:23, 1979 12. Harousseau JL, Flandrin G, Tricot G, Brouet JC, Seligmai in M, Bernard J: Malignant lymphoma superveningin chronic lymph Jcytic leukemia and related disorders. Richter's syndrome: A study >f 25 cases. Cancer 15: 1302, 1981 13. van Dongen JJM, Hooijkaas H, Grosveld G, Hagemeijer , i , Michiels JJ, Abels J, van der Kwast H, Prins MEF, de Klein A: Ca ;e 31-1983: Richter's syndrome. N Engl J Med 309:1582, 191 3 (letter) 14. Rudders RA, Ross R: Partial characterization of the sh ft from IgG to IgA synthesis in the clonal differentiation of hum m leukemic bone marrow-derived lymphocytes. J Exp Med 14254 9, 1975 15. Saiki 0,KishimotoT, Kuritani T, Muraguchi A, Yamamu pa Y: In vitro induction of IgM secretion and switching to IgG production in human B leukemic cells with the help of T cells. J lmmunol 124:2609, I980 16. Shimizu A, Honjo T: lmmunoglobulin class switching. CII1 36:801, 1984 17. Radbruch A, Sablitzky F: Deletion of Cu genes in mouse B lymphocytes upon stimulation with LPS. EMBO J 2:1929. 1983 18. Kuritani T, Cooper MD: Human B cell differentiation. I. Analysis of immunoglobulin heavy chain switching using mon )clonal anti-immunoglobulin M, G, and A antibodies and pokewo :d mitogen-induced plasma cell differentiation. J Exp Med 15533 3, 1982