high-sensitive immunophenotyping and dna ploidy studies for the investigation of minimal residual...

High-sensitive immunophenotyping and DNA ploidy studies

for the investigation of minimal residual disease

in multiple myeloma

JULI A ALMEIDA,1,7 ALBERTO ORFAO,1,7 MAURICIO OCQUETE AU,2 GE MA MATEO,3,7 MERCEDES CORRAL,3,7

Ma DELORES CABALLERO,3,7 JOA N BLADE,4 Ma JESUS MORO,5 JO SE HE R NA N DE Z6

AN D JESUS F. SA N MIGUEL3,7

1Departamento de Medicina, Universidad de Salamanca, Spain, 2Facultad de Medicina, Ponti®cia Universidad CatoÂlica,

Chile, 3Servicio de HematologõÂa, Hospital Universitario de Salamanca, Spain, 4Hospital ClõÂnico de Barcelona, Spain,5Hospital Virgen Blanca de LeoÂn, Spain, 6Hospital General Segovia, Spain, and 7Centro de InvestigacioÂn del Cancer,

University of Salamanca, Spain

Received 16 April 1999; accepted for publication 14 July 1999

Summary. Sensitive techniques for monitoring minimalresidual disease (MRD) in multiple myeloma (MM) areneeded to evaluate the effectiveness of new intensive treat-ment strategies. The aim of the present study was to explorethe applicability and sensitivity of ¯ow cytometry immuno-phenotyping and DNA ploidy studies for the investigation ofresidual myelomatous plasma cells (PC) in MM patients.Bone marrow (BM) samples from 61 untreated MM patientswere immunophenotypically analysed with a panel of21 monoclonal antibodies, using a high-sensitive methodbased on a two-step acquisition procedure through a SSC/CD38��-CD138� `live-gate'. Overall, in 87% of MM cases,PC displayed an aberrant phenotype at diagnosis. The mostimportant aberrant criteria were: antigen over-expression ofCD56 (62%), CD28 (16%) and CD33 (6%) and asynchronousexpression of CD117 (28%), sIg (21%) and CD20 (10%). DNAaneuploidy was found in 62% of cases. The simultaneous use

of these two techniques allowed the detection of aberrant/aneuploid PC in 95% of the cases. Based on dilutionalexperiments, the detection limit of both techniques rangedfrom 10ÿ4 to 10ÿ5. In 29 stem cells harvests and 19 BMsamples obtained 3 months after autologous transplantation,we have investigated the presence of residual myelomatousPC; they were detected in 44% of the stem cell collections andin 61% of the BM samples obtained after transplant. Thepercentage of pathological PC did not signi®cantly changeduring the days of harvest. In summary, the present studyshows that the combined use of immunophenotyping andDNA ploidy studies is a suitable approach for MRD investiga-tion in MM patients based on their applicability (95% of cases)and sensitivity (up to 10ÿ5).

Keywords: multiple myeloma, minimal residual disease,immunophenotype, DNA content, ¯ow cytometry.

High-dose chemotherapy followed by allogeneic and parti-cularly autologous stem cell support is increasingly used forthe treatment of multiple myeloma (MM) patients. This hasled to complete remission (CR) rates between 25% and 75%and a 3-year probability of event-free survival between 40%and 60% (Barlogie et al, 1986, 1995, 1997; Fermand et al,1992; Cunningham et al, 1994; Vesole et al, 1994, 1996;Tricot et al, 1996; Alegre et al, 1998). The increase in CRrate is an important step forward in the treatment of MM,since a widely accepted concept in haematology maintains

that the achievement of CR is a primary requirement for thecure of patients. From the morphological point of view, thebone marrow (BM) is considered to be in CR when there are<5% malignant plasma cells (PC). However, we have learnedfrom acute leukaemias that at the detection limit of mor-phology a BM sample in CR may still harbour as many as1010 malignant cells, this being termed minimal residualdisease (MRD). Moreover, in contrast to the acute leukae-mias, whose blast cells are morphologically different fromnormal cells, in MM morphology is usually unable to distin-guish normal from myelomatous PC, which would reduce,even more, the sensitivity of morphology for the evaluationof remission in MM patients. Accordingly, at present, moresensitive techniques are needed to evaluate the effectiveness

British Journal of Haematology, 1999, 107, 121±131

121q 1999 Blackwell Science Ltd

Correspondence: Professor JesuÂ s F. San Miguel, Servicio deHematologõÂa, Hospital Universitario de Salamanca, Paseo San Vicente

58-182, 37007-Salamanca, Spain. e-mail: [email protected].

of new treatment strategies in MM, to lay the foundationfor the design of patient-adapted treatments, to assess thetumoural contamination of stem cell harvests and to deter-mine the ef®cacy of ex-vivo purging protocols. Most MRDstudies reported in MM have been based on molecularbiology techniques (Bakkus et al, 1996; Voena et al, 1997;Moos et al, 1995), either employing a PCR-based strategyusing clone-speci®c sequences derived from the Ig generearrangement or using an IgH ®ngerprinting technique.These reports show that although some MM cases achievemolecular remission following allografting or double auto-grafting (Bird et al, 1993; Bjorkstrand et al, 1995), in mostcases MRD persists (Corradini et al, 1995; Schiller et al,1995). Moreover, in spite of their high sensitivity, moleculartechniques usually provide only qualitative or semi-quanti-tative information and are time-consuming.

Alternative approaches to MRD detection include: cyto-genetics/in situ hybridization, immunophenotyping and theanalysis of the cell DNA contents by ¯ow cytometry (SanMiguel & van Dongen, 1997). Immunophenotyping hasbeen used frequently for the investigation of MRD in acuteleukaemias and it has proved to be of great utility forpredicting relapses in these patients (Campana & Pui, 1995;San Miguel et al, 1997; Coustan-Smith et al, 1998). It hasbeen shown that blast cells from acute leukaemias frequentlydisplay aberrant/uncommon phenotypic features (`leukaemicphenotypes') that enable their distinction from normal cells(Campana & Pui, 1995; Coustan-Smith et al, 1998). Althoughour group (Ocqueteau et al, 1998), as well as Harada et al(1993) and Pope et al (1997), have shown that the phenotypiccharacteristics of myelomatous PC differ from those of normalPC, to the best of our knowledge the real incidence of aberrantphenotypes which would enable us to monitor MRD in MMpatients still needs to be determined. Moreover, the sensitivityof this strategy has not been explored in MM and this is anadditional prerequisite before it is translated into MRDclinical studies. The analysis of DNA content by ¯ow cyto-metry may be an additional approach for MRD detection,since the presence of DNA aneuploidy constitutes a speci®ctumour marker (Orfao et al, 1995) which would enable theidenti®cation of residual clonal cells. The overall incidence ofDNA aneuploidy in MM is arround 60% (San Miguel et al,1996), indicating that this approach could be applicable to alarge proportion of MM patients, providing it is sensitiveenough to detect low numbers of aneuploid plasma cells inBM samples.

The aim of the present study was to explore the appli-cability and sensitivity of two different approaches ±immunophenotyping and DNA ploidy analysis (both assessedby ¯ow cytometry) ± for the investigation of residualmyelomatous PC in MM patients who are in morphologicalCR. For that purpose we have developed a sensitive assaywhich enables the speci®c discrimination between myelo-matous and normal PC based on the existence of eitherantigenic aberrancies or DNA aneuploidy. Through thecombined use of both methods, >90% of MM patients couldbe monitored for the detection of residual disease, with adetection limit of at least 10ÿ4 (one neoplastic cell among10 000 normal cells).

MATERIAL AND METHODS

Patients. A total of 61 patients with non-treated MMdiagnosed according to the criteria of the Chronic Leukemia-Myeloma Task Force (1973) were included in this study. Themean age of MM patients was 61 6 12 years, 37 were malesand 24 females. In all cases BM samples were obtained afterinformed consent according to the University Hospital ofSalamanca's Ethics Committee.

Follow-up studies after autologous stem cell transplantwere performed in 19 MM patients. All these patients showed<5% PC by morphology at the 3-month post-transplantevaluation. In 12 cases both immunophenotypic and cellDNA content studies were performed, whereas in the remain-ing seven cases only immunophenotyping (n�6) or DNAploidy studies (n�1) were conducted. All cases wereconditioned with high-dose melphalan alone (200 mg/m2).One patient was transplanted in complete remission (CR;disappearance of monoclonal component in both serum andurine, and <5% of bone marrow plasma cells as assessed byconventional morphology), 17 in partial response (PR; >50%reduction in serum and/or urinary measurable monoclonalprotein) and one with non-responding but stable disease.

Peripheral blood stem cell (PBSC) collection. A total of 29apheresis samples from 10/19 transplanted patients wereanalysed by ¯ow cytometry. In 20 apheresis products bothimmunophenotyping and cell DNA content studies wereperformed, whereas in the remaining nine apheresis onlyimmunological investigations were made. Mobilization ofPBSC was performed by administering granulocyte-colonystimulating factor (G-CSF) at a dose of 5 mg/kg/d given over5 consecutive days (four cases) or together with cyclophos-phamide, 4 g/m2 (six cases).

Immunophenotypic studies. BM samples obtained by aspira-tion were collected in EDTA anticoagulant and immediatelydiluted 1/1 (vol/vol) in phosphate-buffered saline (PBS).Whole BM samples (approximately 2 ´ 106 cells in 100 ml/test)were stained using direct immuno¯uorescence and simulta-neous triple-labellings with the combination of monoclonalantibodies (MoAb) shown in Table I. Brie¯y, BM sampleswere incubated (15 min at room temperature in the dark), inthe presence of 5±20 ml of each of the MoAb, accordingto the recommendations of the manufacturers. After lysingthe non-nucleated red cells with the FACS lysing solution(Becton Dickinson), cells were centrifuged (5 min at 540 g)and the cell pellet washed with 4 ml of PBS and resuspendedin 0´5 ml of PBS until analysed (San Miguel et al, 1997).Since plasma cells usually display a higher level of auto-¯uorescence with respect to other nucleated BM cells, in allcases an FL1/FL2 isotype-matched negative control plusCD38-PE/Cy5 was used, in order to speci®cally evaluate thePC's auto¯uorescence level. Data acquisition was performedon a FACSort ¯ow cytometer (Becton Dickinson). In orderto increase the sensitivity and precision of the analysis, inthose MM patients who displayed a low proportion of bonemarrow PC at diagnosis, as well as in all normal BM samples,myelomatous BM samples following chemotherapy and PBSCcollections, acquisition was performed in two consecutivesteps. First, a total of 15 000 events/tube were acquired,

q 1999 Blackwell Science Ltd, British Journal of Haematology 107: 121±131

122 Julia Almeida et al

123Immunophenotype and DNA Ploidy Studies for MRD in MM


corresponding to the total of nucleated BM cells. In thesecond step, acquisition through a `live-gate' drawn on SSC/CD38 strong� cells (where PC are located) was performed.In this latter step between 3 ´ 105 and 2 ´ 106 events weremeasured (Fig 1). A minimum of 15 myelomatous PC weremeasured in each experiment. For data analysis, the Paint-A-Gate PRO software (Becton Dickinson) was used. Celldebris were excluded from the analysis based on their typicallight-scatter pattern, according to well-established methods(San Miguel et al, 1997). PC were identi®ed according totheir positivity for CD138, their strong reactivity for CD38and their typical light-scatter distribution.

In BM samples selected from 14 MM patients, serialdilutional experiments of myelomatous PC with normal BMcells from healthy subjects (1:10; 1:100; 1:1000; 1:10 000;1:100 000; 1:1 000 000) were performed in order to estab-lish the in vitro sensitivity of the immunophenotypic tech-nique for the detection of myelomatous PC when present inBM samples at very low frequencies. The immunophenotypicstudies were performed as described above (triple-stainingsand two-step acquisition).

DNA measurements. A simultaneous staining for DNA(with propidium iodide) and PC surface antigens (CD38 andCD138) was performed, according to a previously describedtechnique (Orfao et al, 1994). Data acquisition was also per-formed in two consecutive steps, as described above for theimmunophenotypic studies. DNA analysis was made usingthe Paint-A-Gate PRO and MODFIT software programs

Table I. Combinations of monoclonal anti-

bodies used for the phenotypic characteriza-

tion of plasma cells.

FITC PE PE/Cy5

FL11 FL21 CD382

CD381 CD561 CD192

CD201 CD191 CD382

CD1383 CD1174 CD382

CD451 CD331 CD382

CD1383 CD281 CD382

CD101 HLA-DR1 CD382

CD403 CD341 CD382

CD91 CD801 CD382

CD1383 CD131 CD382

CD49e1 CD221 CD382

FMC74 CD231 CD382

kappa1 lambda1 CD382

FITC: ¯uorescein isothiocyanate. PE: phyco-erythrin. PE/Cy5: PE/cyanin 5 ¯uorochrome

tamdem.1 Becton Dickinson, San JoseÂ, U.S.A.2 CALTAG Laboratories, San Francisco,

U.S.A.3 Serotec, Oxford, U.K.4 Immunotech, Marseille, France.

Fig 1. Two-step acquisition procedure in a BM

sample from a normal donor. Firstly, 15 000events from the total BM celullarity are acquired

(A). In the second step, acquisition through a

SSC/CD38�� `live gate' is performed (B), in order

to increase the number of PC for analysis.

(Verity Software House Inc., Topsham, Maine, U.S.A.). Adjust-ments to the electronics of the instrument were performedaccording to previously reported methods (Orfao et al, 1994).The percentage of PC (CD38strong�/CD138� cell fraction)was calculated after excluding cell doublets and debris. TheDNA index was obtained by dividing the modal ¯uorescencechannel of the G0/G1 peak of the myelomatous PC by themodal ¯uorescence channel of the residual G0/G1 normalcells present in the sample.

In order to investigate the level of sensitivity of theDNA ploidy technique in detecting residual aneuploid PC,serial dilutional experiments of myelomatous PC in normalBM cells (1:10; 1:100; 1:1000; 1:10 000; 1:100 000;1:1 000 000) were performed in six aneuploid MM cases.These studies were performed following the procedurespreviously described in this section.

Statistical analysis. The statistical signi®cance of the differ-ences observed between groups of patients for continuousvariables was assessed using nonparametric tests (Mann-Whitney U and Wilcoxon tests for non-paired and pairedvariables, respectively). For qualitative parameters, the stati-stical signi®cance of the differences found between groupswas evaluated using the chi-square test (SPSS 6.1.2 Inc.,Chicago, Ill., U.S.A.). P values <0´05 were considered to bestatistically signi®cant.

RESULTS

Incidence of aberrant phenotypes in MMA prerequisite for the immunophenotypic investigation ofresidual disease in MM is to assess whether or not myelo-matous PC display unique phenotypic features which wouldallow their distinction from normal PC. Therefore a pre-requisite for this study is the comprehensive analysis of theantigenic pro®le of normal PC. As we have recently shown(Ocqueteau et al, 1998), normal PC are identi®ed by theirexpression of CD138, their strong positivity for CD38, andtheir distribution according to light scatter characteristics. Inaddition, in all cases CD19 is expressed in the majority ofbone marrow PC; in contrast, normal PC are considered to beCD56 negative, although in all cases a minor proportion ofPC (<15%) display dim expression for this antigen. AlthoughCD13 and CD33 are expressed in PC from around 60% and20% of cases, respectively, the expression intensity is con-stantly low. Reactivity for CD28 is usually negative, althougharound 40% of BM PC from one healthy donor showed dimexpression. PC from normal individuals are constantly nega-tive for the remaining markers analysed (CD10, CD20, CD22,CD23, CD34, CD80, CD117, FMC7 and sIg).

The phenotypic characteristics of myelomatous PC fromthe 61 MM cases analysed were compared with those of PCfrom the healthy subjects described above, in order to estab-lish the incidence of aberrant phenotypes. In 62% of MMpatients PC showed over-expression of the CD56 antigen,usually in the absence of reactivity for CD19. Antigen over-expression was established on the basis of the existence ofabnormally high levels of ¯uorescence for that speci®cmarker, taking as reference its expression on normal PC. Thesame type of phenotypic aberrancy was found for CD28 and

CD33 that were over-expressed in 16% and 6´5% of MMcases, respectively (Fig 2). In addition, three markers (CD20,sIg and CD117) which were constantly negative in normalPC, were positive in 10%, 21% and 28% of MM BM samples.Additionally, in a high proportion of MM patients PC dis-played an abnormally lower ¯uorescence intensity of theCD38 antigen (62% of cases) and a higher FSC/SSC values(70% of cases) as compared to normal PC; however, in ourexperience these latter criteria can be only used when bothnormal and myelomatous PC coexist in the sample, sinceotherwise it would be dif®cult to be sure of the discriminationbetween both PC populations (Fig 2). According to thesedata, the overall incidence of aberrant phenotypes in MMpatients at diagnosis was 87% (53/61 cases) (Table II).Moreover, in 46% of MM cases two or more phenotypicaberrancies coexisted (two in 35%, three in 10% and fourin 1´6%).

DNA aneuploidy in MMFrom the 61 MM cases included in this study, 62% (n�38)showed the presence of DNA aneuploid PC. The DNA indexwas higher than 1 in 35 cases (mean DNA index of 1´19 60´09; median 1´18; range 1´05±1´44), whereas the remain-ing three aneuploid cases showed the existence of two distinctPC populations, with different cell DNA contents (biclonal).In one of these cases a diploid and a hyperdiploid PC popu-lation (DNA index 1´98) were found; in the second patienttwo hyperdiploid populations of PC were observed (DNAindices 1´13 and 2´18) and in the third case a hypodiploid(DNA index 0´92) and a hyperdiploid population of PC (DNAindex 1´20) coexisted. These results were concordant withprevious reports from our group in larger series of patientsusing propidium iodide and CD38/CD138 for the speci®cassessment of the cell DNA contents of myelomatous PC(San Miguel et al, 1995).

Taking the number of cases with phenotypic aberrancies(53/61 cases), together with the number of aneuploid cases



Table II. Incidence and type of the different

phenotypic aberrancies and DNA aneuploidy in

MM patients.

Aberrant characteristic % of cases

Phenotypic aberrancies* 87%

Antigen over-expression* 74%CD56 62%

CD28 16%

CD33 6%

Asynchronous antigen expression* 49%

CD117 28%

sIg 21%CD20 10%

DNA aneuploidy 62%

Total 95%

* Note that more than one aberrancy was presentin a large proportion of cases (46%).



Fig

2.

Ph

eno

typ

icab

erra

nci

esin

mye

lom

ato

us

PC

.A

ber

ran

ta

nti

gen

exp

ress

ion

of

CD

56

,C

D3

3,

CD

28

,C

D1

17

,C

D2

0a

nd

CD

38

inm

yelo

ma

tou

sP

C(r

edd

ots

)a

sco

mp

are

dto

no

rma

lP

C(g

reen

do

ts).

(38/61 cases), it was found that in a total of 58 cases of the61 MM patients analysed (95%), BM PC displayed a pheno-typic aberrancy and/or an abnormal cell DNA content whichcould be used as tumour-associated markers for the investi-gation of residual myelomatous PC later on during follow-up(Table II).

Investigation of the in vitro sensitivity levelfor detection of MRDThe sensitivity level achieved with the immunopheno-typic approach (triple-staining and two-step acquisition)was > 10ÿ4 (detection of at least one aberrant PC among10 000 normal cells) in all cases (n�14), the detection limitreaching up to 10ÿ5 in four cases (Fig 3). It should be notedthat the level of sensitivity was independent of the type ofaberrancy displayed by the myelomatous PC (Table III).Regarding DNA ploidy analysis, in all six experiments aneu-ploid PC could be detected even when present at frequenciesof 10ÿ4 (one aneuploid PC among 10 000 diploid bonemarrow cells) (Table IV, Fig 4).

Investigation of MRD in leukapheresis productsand BM samples after transplantation

MRD investigation in leukapheresis samples from MMpatients. The persistence of myelomatous PC was investigatedby immunophenotyping in 29 leukapheresis samples obtainedfrom 10 patients who showed an aberrant phenotypic markerat diagnosis and <5% PC in BM samples obtained prior to theleukapheresis collection. Six patients had no detectablepathological PC (< one aberrant cell in 104 normal cells)either on the initial or sequential harvested samples (n�16apheresis samples). In the remaining four patients, myelo-matous PC were detected and these were present in all stemcell leukapheresis products (n�13 apheresis samples).Interestingly, normal residual PC were found in all leuka-pheresis samples analysed, and usually predominated overmyelomatous PC: from the total number of PC, 81 6 27%displayed a normal phenotype and only 19 6 20% displayedaberrant immunological features.

In 20 apheresis samples (corresponding to seven patients)we have simultaneously performed immunophenotypingand DNA ploidy evaluations. In four aphereses (two patients)discrepant results were found (positive by immunopheno-typing and negative by DNA ploidy studies), which was prob-ably relative to the lower number of cells acquired in theDNA analysis (105) as compared to the immunophenotypicinvestigation (>106), reducing by 1 log the theoreticalsensitivity in these two particular cases. In the remaining 16aphereses the results obtained with both techniques wereconcordant.

BM samples obtained after autologous PBSC transplantation.BM samples from 18 patients obtained 3 months after auto-logous transplantation (all showing <5% BM PC onmorphological examination) were studied by immuno-phenotypic technique. In 14/18 cases pathological PCdisplayed aberrant phenotypes that could be used as a



Table IV. Sensitivity level of DNA ploidy studies based on dilutionalexperiments of DNA aneuploid myelomatous PC in normal BM cells.

Case DNA index An-PC D-PC SL

1 1´33 0´016% 0´24% 10ÿ4

2 1´15 0´09% 1´41% 10ÿ4

3 1´27 0´026% 0´49% 10ÿ4

4 1´18 0´03% 0´35% 10ÿ4

5 1´12 0´08% 0´18% 10ÿ4

6 1´30 0´027% 0´57% 10ÿ4

An-PC: aneuploid PC; D-PC: diploid PC; SL: sensitivity level.

Table III. Sensitivity level of immunophenotyping based on dilutional experiments of

myelomatous PC in normal BM cells.

Phenotypic aberrancy Case A-PC N-PC SL

CD38�� or ��/CD56��/CD19ÿ 1 0´05% 0´09% 10ÿ4

2 0´0027% 0´08% 10ÿ5

3 0´05% 0´31% 10ÿ4

4 0´003% 0´23% 10ÿ5

5 0´035% 0´08% 10ÿ4

6 0´06% 0´18% 10ÿ4

7 0´007% 0´7% 10ÿ5

8 0´011% 0´23% 10ÿ4

9 0´012% 0´49% 10ÿ4

CD38��/CD56��/CD19ÿ/CD117� 10 0´017% 0´3% 10ÿ4

CD38��/CD56��/CD19ÿ/CD117� 11 0´01% 0´45% 10ÿ4

CD38��/CD56��/CD19ÿ/CD28�� 12 0´002% 0´49% 10ÿ5

CD38��/CD56ÿ/CD19ÿ/CD28�� 13 0´082% 0´12% 10ÿ4

14 0´01% 0´5% 10ÿ4

A-PC: aberrant PC; N-PC: normal PC; SL: sensitivity level.



hallmark for malignancy ± such as over-expression of CD56and/or CD28 or reactivity for CD117 or sIg (Table V) ±whereas in four patients myelomatous PC were followed juston the basis of higher light scatter values and lower CD38expression as compared to normal PC. In 11/18 patientsanalysed (61%) residual myelomatous PC could be clearlydetected by immunophenotype, with a mean value of patho-logical PC of 0´28 6 0´41% (range 0´03±1´34%; median0´13%). In the negative cases the limit of sensitivity forMRD was 10ÿ5. Interestingly, at diagnosis, from the totalnumber of PC identi®ed, 98 6 7% were considered to bemyelomatous based on the presence of aberrant phenotypesand only 2 6 1´7% corresponded to normal residual PC.Before transplant, the percentage of myelomatous PC fromthe total BM PC was still high (72 6 38%), and 3 months

after transplantation the number had dropped to 28 6 32%(P�0´0002).

Using DNA aneuploidy as a marker for MRD, we analysed13 patients after autologous transplantation (Table V). In54% of these patients (7/13) residual myelomatous PC couldbe detected (0´22 6 0´18%; range 0´06±0´18%; median0´10%). In all patients diploid PC could also be identi®ed.Similarly, as we have seen with immunophenotyping, theproportion of aneuploid cells with respect to total PCprogressively decreased with treatment: diagnostic sample:94 6 11% before transplant: 74 6 26%; after transplant:17 6 27% (P�0´001).

As shown in Table V, once immunophenotyping and DNAploidy studies were simultaneously explored in the samesamples (n�12), similar results were obtained with the

Table V. MRD levels in BM samples from MM patients obtained 3 months after autologous PBSC transplantation.

3 months post-transplantDiagnosis

(% myelomatous % myelomatous % myelomatous

Patient Follow-up criteria for MRD PC/total BM PC*) PC/total BM PC* PC²

1 CD56��/19ÿ/28�� 100 0 0DNA aneuploidy (DI 1´1) 100 0 0

2 CD56��/19ÿ/117�/28�� 100 0 0

DNA aneuploidy (DI 1´12) 100 0 0

3 CD56��/19ÿ 100 0 0

DNA aneuploidy (DI 1´15±1´29) 100 0 0

4 CD56��/19ÿ 100 0 0


5 CD56��/19ÿ 100 0 0


6 CD56ÿ/19ÿ 100 0 0

7 CD56ÿ/19ÿ 100 0 0

8 DNA aneuploidy (DI 1´1) 86 0 0

9 CD56��/19ÿ 98 19 0´19

DNA aneuploidy (DI 1´3) 94 25 0´4

10 CD56��/19ÿ/28�� 100 27 0´03

11 CD56��/19ÿ 100 25 0´2

12 CD56��/19ÿ 71 11 0´043


13 CD56��/19ÿ 99 56 0´13


14 CD56��/19ÿ 97 14 0´042


15 CD56��/19ÿ/l� 100 53 0´16

16 CD56ÿ/19ÿ/28�� 100 68 0´82

17 CD56��/19ÿ 100 20 0´12


18 CD56ÿ/19ÿ 100 98 0´49DNA aneuploidy (DI 1´25) 100 99 0´50

19 CD56ÿ/19ÿ 100 91 1´34


* Results expressed as % of myelomatous PC from the total number of PC identi®ed (myelomatous plus normal PC).

² Proportion of myelomatous PC identi®ed in the total BM cell population.



two methods as regards both the level of residual myelo-matous PC and the proportion of abnormal PC from the totalpopulation of BM PC.

DISCUSSION

Techniques for detection of MRD are increasingly used inacute leukaemias in order to develop patient-adapted con-solidation treatment strategies (Campana & Pui, 1995; SanMiguel et al, 1997; Coustan-Smith et al, 1998). Moreover,these techniques should contribute not only to the treatmentdecision-making process but also to provide answers toimportant questions: (i) what treatment is more ef®cient inreducing tumour load? (ii) what is the best source for auto-logous stem cells in terms of contamination by malignantcells? (iii) what is the degree of depletion of malignant cells inpositive and negative selection procedures? MRD studies inmultiple myeloma were inconceivable some years ago, andperhaps today are still rather premature. Nevertheless, thehigh rate of CR obtained in recent years with intensivechemotherapy has changed our attitude to MM patients(Barlogie et al, 1997; Alegre et al, 1998). The methods to beused to detect residual disease should be speci®c, sensitive,and rapid enough for clinical applicability (San Miguel & vanDongen, 1997). Most MRD studies which have been per-formed so far in MM are based on PCR assays, which, thoughhighly sensitive, are time-consuming and, as shown byVescio et al (1997), ®nally applicable to less than half of thepatients due to either polyclonal ampli®cation, lack of Iggene ampli®cation or insuf®cient BM RNA for analysis. Thepresent study shows that both immunophenotyping andthe analysis of cell DNA contents may represent suitableapproaches for the detection of residual disease in MM.Regarding immunophenotyping, we have found that in mostMM patients (87%) myelomatous PC display phenotypicaberrancies that enable their distinction from normal PC.Moreover, these aberrancies can be easily identi®ed by usingonly four triple combinations of MoAb (CD38/CD56/CD19;CD20/CD28/CD38; CD117/CD33/CD38; kappa/lambda/CD38) which makes this approach cost-effective. In addition,two other parameters (relatively low reactivity for CD38 andhigh FSC/SSC values) could be used to detect myelomatousPC, but in our experience they can only be used as MRDcriteria if the myelomatous PC coexist with normal PC in thesame sample, in order to contrast the low CD38 expressionand high FSC/SSC with a reference cell population (normalPC). The highly sensitive method described here, which isbased mainly on a two-step acquisition procedure with aspeci®c `live gate' for PC, shows that it is possible to reach adetection limit of 10ÿ4±10ÿ5 (one aberrant PC among10 000 or 100 000 normal cells), just 1 log below that

reported by PCR, independent of the type of aberrancy.Similar results have been obtained upon using DNA aneu-ploidy as a tool for the identi®cation of clonal PC, providedthat, in order to increase the sensitivity of the method, adouble-staining technique is used to speci®cally analyse theDNA contents of PC. This second strategy can be applied toaround two-thirds of all MM patients. Moreover, once com-bined with immunophenotyping it would allow the investi-gation of residual myelomatous PC in up to 95% of all MMpatients. In a similar way to the immunophenotypic method,the detection limit of DNA aneuploidy studies for the investi-gation of residual clonal PC was 10ÿ4 (one aneuploidPC among 10 000 diploid cells) as assessed by dilutionalexperiments.

Although the aim of this study was to set up and stan-dardize two different methods for MRD detection in MM andto explore both their applicability (proportion of patients whocan bene®t) and the in vitro sensitivity of the method, wehave also explored the potential utility of these techniques foridentifying residual tumour cells in vivo in the stem cellharvests and following autologous stem cell transplantation.Our results show that most patients (61%) still harbourresidual disease 3 months after the transplant, but at a verylow level (0´28 6 0´14%). In addition, the proportion ofmyelomatous in the total amount of PC signi®cantly changedfollowing high-dose chemotherapy: at diagnosis a mean of98% of total PC were considered to be aberrant, whereasfollowing transplantation, normal PC clearly predominatedover myelomatous PC, with only 28% of total PC pheno-typically aberrant. Interestingly, this second pro®le is similarto that observed in monoclonal gammopathies of undeter-mined signi®cance (MGUS) (Ocqueteau et al, 1998) andcould represent a truth plateau phase. Whether or not thisregeneration of the normal PC population will result in asigni®cant prolongation of disease-free survival, whereas afurther increase in the proportion of myelomatous PC willherald an impending clinical relapse, will only be clari®edthrough a prospective analysis of a large series of patientsincluded in a uniform transplant clinical trial.

Forty per cent of the patients analysed showed detectablemyelomatous PC at a sensitivity level of ´10ÿ4 in the PBSCautograft. In the remaining patients the apheresis productshad no detectable pathological PC at the detection limitreached with this approach. These ®ndings are in line withthose reported by Pope et al (1997), who have demonstratedthat the majority of PC from PBSC harvests from MMpatients are polyclonal. Since all six cases in which patho-logical PC were not detected in the harvest showed residualdisease following high-dose chemotherapy, this would con-stitute additional evidence that relapse in MM is due mainlyto the lack of ef®cacy of the myeloablative regimens to

Fig 3. Upper. Representative dilutional experiment showing the in vitro sensitivity level of immunophenotyping for the detection of residualmyelomatous PC displaying over-expression of CD56. In this case aberrant PC could be detected at a level of 10ÿ4±10ÿ5 (one myelomatous PC in

up to 104±105 normal BM cells).

Fig 4. Lower. Representative dilutional experiment showing the in vitro sensitivity level for MRD detection of DNA ploidy estudies. Serial dilutional

experiments of aneuploid PC displaying a DNA index of 1´30 in increasingly higher numbers of normal bone marrow cells were performed.

Aneuploid PC could be detected at a level of up to 2 ´ 10ÿ4 (two aneuploid PC in 104 normal BM cells).

erradicate the malignant clone and is not directly related tograft contamination. Additional information provided in thepresent study is that the ratio between pathological andnormal PC did not signi®cantly change during the courseof apheresis, at least with the schedule we used in whichapheresis procedures usually were not prolonged for >4 d.

In summary, in the present study we have shown thehigh applicability and sensitivity of two different approaches± immunophenotyping and DNA cell contents analysis ±suitable for the detection of residual disease in MM. Follow-up studies are necessary in order to explore whether theseapproaches are useful for predicting impending relapsesprior to clinical manifestations and could therefore be usedto make early treatment interventions (i.e lymphocyteinfusions at immunological relapse before clinical relapse)or to adapt treatment in individual patients.

ACKNOWLEDGMENTS

This work has been partially supported by a grant fromthe FundacioÂn RamoÂn Areces (Spain), from the Junta deCastilla y LeoÂn (CL SA26/96) (Valladolid, Spain), and fromthe DireccioÂn General de EnsenÄ anza Superior (DGES) (PM97-0161), Ministerio de EducacioÂn y Cultura (Madrid, Spain).M. Ocqueteau was supported by a grant from the Ponti®ciaUniversidad CatoÂlica from Chile (1980989).

REFERENCES

Alegre, A., DõÂaz-Mediavilla, J., San Miguel, J., MartõÂnez, R., GarcõÂa-

LaranÄ a, J., Sureda, A., Lahuerta, J.J., Morales, D., BladeÂ, J.,

Caballero, D., De la Rubia, J., Escudero, A., DõÂez-MarõÂn, J.L.,HernaÂndez-Navarro, F., RifoÂn, J., Odriozola, J., Brunet, S., De la

Serna, J., Besalbuch, J., Vidal, M.J., Solano, C., LeoÂn, A., SaÂnchez,

J.J., MartõÂnez-Chamorro, C. & FernaÂndez-RanÄ ada, J.M. (1998)

Autologous peripheral blood stem cell transplantation for multiplemyeloma: a report of 259 cases from the Spanish Registry. Bone

Marrow Transplantation, 21, 133±140.

Bakkus, M.H., Juge-Morineau, N. & van der Werff ten Boch, J.E.

(1996) Detection of minimal residual disease in multiple myelomaand acute leukemia. Medical Oncology, 13, 121±131.

Barlogie, B., Anderson, K.C., Berenson, J., Crowley, J., Cunningham,

D., Gertz, M., Henon, P., Horowitz, M., Jagannath, S., Powles, R.,Reece, D., Reiffers, J., Salmon, S., Tricot, G. & Vesole, D. (1995)

Transplants for multiple myeloma. Bone Marrow Transplantation,

16, (Suppl. 1), 234±239.

Barlogie, B., Hall, R., Zander, A., Dicke, K. & Alexanian, R. (1986)High-dose melphalan with autologous bone marrow transplanta-

tion for multiple myeloma. Blood, 67, 1298±1301.

Barlogie, B., Jagannath, S., Vesole, D.H., Nancke, S., Cheson, B.,

Mattox, S., Bracy, D., Salmon, S., Jacobson, J., Crowley, J. & Tricot,G. (1997) Superiority of tandem autologous transplantation over

standard therapy for previously untreated multiple myeloma.

Blood, 89, 789±793.Bird, J.M., Russell, N.H. & Samson, D. (1993) Minimal residual

disease after bone marrow transplantation for multiple myeloma:

evidence for cure in long-term survivors. Bone Marrow Transplan-

tation, 12, 651±654.Bjorkstrand, B., Ljungman, P., Bird, J.M., Samson, D. & Gharton, G.

(1995) Double high-dose chemoradiotherapy with autologous

stem cell transplantation can induce molecular remissions in

multiple myeloma. Bone Marrow Transplantation, 15, 367±371.

Campana, D. & Pui, C.H. (1995) Detection of minimal residual

disease in acute leukemia: methodologic advances and clinical

signi®cance. Blood, 85, 1416±1434.

Chronic Leukemia-Myeloma Task Force, National Cancer Institute(1973) Proposed guidelines for protocol studies. II. Plasma cell

myeloma. Cancer Chemotherapy Reports, 4, 145±158.

Corradini, P., Voena, C., Astol®, M., Ladetto, M., Tarella, C., Boccadoro,

M. & Pileri, A. (1995) High-dose sequential chemoradiotherapy inmultiple myeloma: residual tumor cells are detectable in bone

marrow and peripheral blood cell harvests and after autografting.

Blood, 85, 1596±1602.Coustan-Smith, E., Behm, F.G., Sanchez, J., Boyett, J.M., Hancock,

M.L., Raimondi, S.C., Rubnitz, J.E., Rivera, G.K., Sandlund, J.T.,

Pui, C.-H. & Campana, D. (1998) Immunological detection of

minimal residual disease in children with acute lymphoblasticleukaemia. Lancet, 351, 550±554.

Cunningham, D., Paz-Ares, L., Milan, S., Powles, R., Nicolson, M.,

Hickish, T., Selby, P., Treleavan, J., Viner, C. & Malpas, J. (1994)

High-dose melphalan and autologous bone marrow transplanta-tion as consolidation in previously untreated myeloma. Journal of

Clinical Oncology, 12, 759±763.

Fermand, J.P., Chreve, S., Levy, Y., Miclea, J.M., Tsapis, A., Gerota, J.,Benbunan, M. & Brovet, J.C. (1992) The role of autologous blood

stem cells support of high-dose therapy for multiple myeloma.

Hematological/Oncological Clinics of North America, 6, 541±562.

Harada, H., Kawano, M.M., Huang, N., Harada, Y., Iwato, K., Tanabe,O., Tanaka, H., Sakai, A., Asaoku, H. & Kuramoto, A. (1993)

Phenotypic difference of normal plasma cells from mature

myeloma cells. Blood, 81, 2658±2663.

Moos, M., Schulz, R., Cremer, F., Sucker, D., Schmohl, D., Dohner, H.,Goldschmidt, H., Haas, R. & Hunstein, W. (1995) Detection of

minimal residual disease by polimerase chain reaction in B-cell

malignancies. Stem Cells, 13, (Suppl. 3), 42±51.

Ocqueteau, M., Orfao, A., Almeida, J., BladeÂ, J., GonzaÂlez, M., GarcõÂa-Sanz, R., LoÂpez-Berges, M.C., Moro, M.J., HernaÂndez, J., Escribano,

L., Caballero, M.D., Rozman, M. & San Miguel, J.F. (1998)

Immunophenotypic characterization of plasma cells from mono-clonal gammopathy of undetermined signi®cance (MGUS) patients:

implications for the differential diagnosis between MGUS and

multiple myeloma. American Journal of Pathology, 152, 1655±

1665.Orfao, A., Ciudad, J., Gonzalez, M., Lopez, A., Del Mar Abad, M., Paz-

Bouza, J.I., Cruz, J.J., Gomez Alonso, A. & San Miguel, J.F. (1995)

Flow cytometry in the diagnosis of cancer. Scandinavian Journal of

Clinical Laboratory Investigations, 55, (Suppl. 221), 145±152.Orfao, A., GarcõÂa-Sanz, R., LoÂpez-Berges, M.C., Vidriales, B.,

GonzaÂlez, M., Caballero, M.D. & San Miguel, J.F. (1994) A new

method for the analysis of plasma cell DNA content in multiplemyeloma samples using a CD38/propidium iodide double staining

technique. Cytometry, 17, 332±339.

Pope, B., Brown, R., Gibson, J. & Joshua, D. (1997) Plasma cells in

peripheral blood stem cell harvests from patients with multiplemyeloma are predominantly polyclonal. Bone Marrow Transplanta-

tion, 20, 205±210.

San Miguel, J.F., GarcõÂa-Sanz, R., GonzaÂlez, M. & Orfao, A. (1995)

Immunophenotype and DNA cell content in multiple myeloma.Bailliere's Clinical Haematology, 8, 735±759.

San Miguel, J.F., GarcõÂa-Sanz, R., GonzaÂlez, M. & Orfao, A. (1996)

DNA cell content studies in multiple myeloma: a review. Leukemia

and Lymphoma, 23, 33±41.San Miguel, J.F., MartõÂnez, A., Macedo, A., Vidriales, M.B., Lopez-

Berges, M.C., Gonzalez, M., Caballero, D., Garcia-Marcos, M.A.,

Ramos, F., Fernandez-Calvo, J., Calmuntia, M.J., Diaz-Mediavilla, J.& Orfao, A. (1997) Immunophenotyping investigation of minimal





residual disease is a useful approach for predicting relapse in acute

myeloid leukemia patients. Blood, 90, 2465±2470.

San Miguel, J.F. & van Dongen, J.J.M. (1997) Methods for detection of

minimal residual disease. Acute Leukemias. VI. Prognostic Factorsand Treatment Strategies (ed. by T. BuÈ chner, W. Hiddeman, B.

WoÈrmann, G. Schellong, J. Ritter and U. Creutzig), pp. 307±312.

Springer, Berlin.

Schiller, G.J., Vescio, R.A., Freytes, C., Spitzer, G., Sahebi, F., Lee, M.,Hua, C., Cao, J., Lee, J.C., Hong, C.H., Bearman, S., Lill, M.,

Berenson, R. & Berenson, J. (1995) Transplantation of CD34�

peripheral blood progenitor cells after high-dose chemotherapy forpatients with advanced multiple myeloma. Blood, 86, 390±397.

Tricot, G., Jagannath, S., Vesole, D.H., Bracy, D., Desikan, K.R., Siegel,

D. & Barlogie, B. (1996) Hematopoietic stem cell transplants for

multiple myeloma. Leukemia and Lymphoma, 22, 25±36.Vescio, R., Stewart, A., Ballester, O., Noga, S., Rugo, H., Freytes, C.,

Stadtmauer, E., Kogut, N., Tarantolo, S., Stiff, P., PinÄ ero, L., Dalton,

W., Ho, A., Schiller, G., Benyunes, M., Jacobs, C., Dipersio, J.,

Anderson, K. & Berenson, J. (1997) Myeloma cell tumor reduc-

tion in PBPC autografts following CD34 selection: the results of a

phase III trial using the CEPRATE device. (Abstract). Blood, 90,421a.

Vesole, D.H., Barlogie, B., Jagannath, S., Cheson, B., Tricot, G.,

Alexanian, R. & Crowley, J. (1994) High-dose therapy for refractory

multiple myeloma: improved prognosis with better supportive careand double transplants. Blood, 84, 950±956.

Vesole, D.H., Tricot, G., Jagannath, S., Desikan, K.R., Siegel, D.,

Bracy, D., Miller, L., Cheson, B., Crowley, J. & Barlogie, B. (1996)Autotransplants in multiple myeloma: what have we learnt? Blood,

88, 838±847.

Voena, C., Ladetto, M., Astol®, M., Provan, D., Gribben, J.G.,

Boccadoro, M., Pileri, A. & Corradini, P. (1997) A novel nested-PCR strategy for the detection of rearranged immunoglobulin

heavy-chain genes in B-cell tumors. Leukemia, 11, 1793±1798.

high-sensitive immunophenotyping and dna ploidy studies for the investigation of minimal residual...

Documents