- Email: [email protected]
Molecular Monitoring of Minimal Residual Disease in the Peripheral Blood of Patients with Multiple Myeloma
Biol Blood Marrow Transplant 19 (2013) 1109e1115
Molecular Monitoring of Minimal Residual Disease in the Peripheral Blood of Patients with Multiple Myeloma Mark Korthals 1, 2, Nina Sehnke 1, Ralf Kronenwett 1, y, Thomas Schroeder 1, Tobias Strapatsas 1, Guido Kobbe 1, Rainer Haas 1, Roland Fenk 1, * 1 2
ASBMT
American Society for Blood and Marrow Transplantation
Department of Hematology, Oncology and Clinical Immunology, Heinrich-Heine-University, Duesseldorf, Germany Department of Functional Genomics and Medical Toponomics, University Magdeburg, Magdeburg, Germany
Article history: Received 16 July 2012 Accepted 26 April 2013 Key Words: Minimal residual disease IgH-PCR Multiple myeloma
a b s t r a c t The prognostic relevance of minimal residual disease (MRD) in patients with multiple myeloma is still an open question. In bone marrow, the level of residual myeloma cells is associated with treatment outcome, but the role of clonotypic cells in the peripheral blood (PB) for the prognosis of patients is not identified yet. In this study, we retrospectively analyzed MRD by quantitative real-time IgH-PCR (IgH-qPCR) in the PB of 42 patients undergoing high-dose therapy followed by autologous PB stem cell transplantation as first-line therapy for multiple myeloma. The MRD level of PB samples was in median 40-fold lower than in bone marrow samples, collected on the same day, with a wide intra- and interindividual variation (range, .4- to 4628-fold). The presence or absence of detectable MRD levels in PB did not correlate with the serological remission status. Still, patients with negative PCR results in PB 3 months after high-dose therapy and PB stem cell transplantation had lower International Staging System stage (P ¼ .01), lower levels of b2-microglobulin (P ¼ .02), higher hemoglobin levels (P ¼ .01), and a prolonged event-free (median, 15 versus 4 months; P ¼ .004) and overall (median, 52 versus 17 months; P ¼ .03) survival. Importantly, by sequential monitoring of clonotypic cells in PB, in 19 of 29 patients (66%) with progressive disease, an increase of the 2IgH/b-actin ratio of at least 1 log step could be detected in median 4 months (range, .8 to 13 months) before the relapse was diagnosed on the basis of the European Group for Blood and Marrow Transplantation criteria. These patients with a molecular relapse in PB before a serological relapse had a significantly shorter overall survival than other patients (median, 17 months versus median not reached, P ¼ .02). In conclusion, IgH-qPCR is a sensitive technique for the detection of clonotypic cells in PB, which precede clinical relapse. Future studies are needed to evaluate whether these circulating tumor cells play a role in promoting disease recurrence. Ó 2013 American Society for Blood and Marrow Transplantation.
INTRODUCTION Multiple myeloma (MM) is a B cell malignancy characterized by a monoclonal expansion of plasma cells in bone marrow (BM) and the secretion of paraprotein in the serum and/or urine of patients. High-dose chemotherapy (HDT) followed by autologous peripheral blood stem cell transplantation (PBSCT) has improved the clinical outcome, but eventually all patients relapse [1,2]. The quantitative measurement of minimal residual disease (MRD) on a molecular level is a highly sensitive and specific method to detect tumor cells of clonal origin [3,4] that can help to estimate the risk of relapse. Evidence shows that the achievement of a serological complete remission (CR) is associated with a prolonged survival of patients with MM [5]. Treatment protocols using chemotherapy in combination with novel agents such as bortezomib [6] or lenalidomide [7,8] are continuously improving the depth of response. Thus, the assessment of MRD becomes more and more important, and the level of MRD may help to guide therapeutic decisions in patients with MM in the future [9], as it is already clinical practice in patients with acute lymphoblastic leukemia or chronic myeloid leukemia [10,11].
Financial disclosure: See Acknowledgments on page 1114. * Correspondence and reprint requests: PH Dr. Roland Fenk, Dept. of Hematology, Oncology and Clinical Immunology, Heinrich-HeineUniversity, Moorenstr. 5, 40225 Duesseldorf, Germany. E-mail address: [email protected] (R. Fenk). y Current address: PH Dr. Ralf Kronenwett: Sividon Diagnostics, Cologne, Germany.
Residual tumor cells in BM have a prognostic value in newly diagnosed patients with MM treated with HDT followed by autologous PBSCT, as has been shown by flow cytometry [12-14] and quantitative real-time IgH-PCR (IgH-qPCR) [14-17]. In the same line are results from Corradini et al. [18], who used qualitative IgH-PCR in BM samples of 48 patients undergoing allogeneic PBSCT. In this setting, negative MRD results were also associated with prolonged overall survival (OS) and progression-free survival. Whereas the prognostic impact of MRD in the BM is well documented, experimental studies of MRD in PB of patients with MM are rare, and patient numbers of reported results are very low [19-23]. As with BM, fluorescent activated cell sorter (FACS) analysis or IgH-PCR are the methods of choice for the detection of very low numbers of myeloma cells in the PB of patients. To acquire more data on this topic, we used IgH-qPCR [3,16] and sequentially measured the amount of clone-specific cells in the PB of 42 patients with MM during first-line therapy with HDT and autologous PBSCT. METHODS Patients and Samples Patients were treated at our hospital between November 2000 and April 2008. All patients presented with symptomatic MM and received the same first-line therapy consisting of HDT followed by autologous PBSCT. The induction treatment of all patients consisted of 2 to 4 cycles of idarubicine (20 mg/m2 i.v., days 1 to 4) and dexamethasone (40 mg oral, days 1 to 4, 9 to 12, and 17 to 21) in monthly intervals. The patients then received pegfilgrastim or granulocyte colony-stimulating factor in combination with cyclophosphamide (2 g/m2, days 1 þ 2) for PB stem cell mobilization. Thereafter, a single HDT was administered at a dose of 200 mg/m2 melphalan followed by autologous PBSCT.
1083-8791/$ e see front matter Ó 2013 American Society for Blood and Marrow Transplantation. http://dx.doi.org/10.1016/j.bbmt.2013.04.025
1110
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
Table 1 Patient Characteristics MRD before Transplantation
Age, median yrs (range) Sex (male:female), % Ig subtype (IgG:others), % Stage SþD (IþII:III), % Stage ISS (1:2:3), % Cytogenetics (abnormal:normal), % CRP, median mg/dL (range) LDH, median U/L (range) Albumin, median g/dL (range) b2-microglobulin, median mg/dL (range) Hemoglobin, median g/dL (range) Platelet count, median 103 (range) Maintenance therapy (IFN:Thal:RIC), %
MRD after Transplantation *
All Patients (n ¼ 42)
Negative (n ¼ 12)
Positive (n ¼ 9)
P
Negative (n ¼ 24)
Positive (n ¼ 8)
P*
54 (31-66) 67:33 74:26 12:88 43:38:19 27:73 .5 (.3-9.8) 160 (79-346) 3.9 (2.8-5.1) 2.8 (1.2-15.3) 11.1 (6.9-17.0) 261 (108-475) 48:36:17
54 (45-62) 75:25 58:42 8:92 25:67:8 42:58 .6 (.3-9.8) 149 (98-285) 3.6 (2.9-5.1) 3.0 (1.5-8.5) 11.0 (6.9-13.9) 239 (108-351) 50:42:8
55 (44-65) 67:33 78:22 11:89 56:33:11 22:78 .7 (.3-1.8) 171 (94-346) 4.2 (3.2-4.9) 2.4 (1.6-8.0) 11.4 (9.3-17.0) 320 (115-475) 33:56:11
1.0 1.0 .6 1.0 .3 .6 .9 .5 .08 .7 .1 .3 .7
54 (31-65) 67:33 67:33 13:88 58:33:8 22:78 .6 (.3-9.8) 149 (79-335) 4.0 (3.1-4.9) 2.3 (1.2-8.5) 11.7 (7.6-17) 259 (108-360) 42:46:13
55 (42-62) 63:38 75:25 0:100 0:63:38 25:75 .8 (.3-2.4) 227 (107-346) 3.7 (2.9-4.9) 3.9 (2.1-10.2) 9.2 (6.9-13.6) 245 (115-475) 63:25:13
.8 1.0 1.0 .6 .01 1.0 .8 .08 .3 .02 .01 .8 .6
CRP indicates C-reactive protein; ISS, international staging system; LDH, lactate dehydrogenase; RIC, reduced-intensity conditioning allogeneic transplantation; SþD, Salmon and Durie; Thal, thalidomide. * Correlations were calculated by Fisher’s exact test for two categories, by chi-square test for three categories or the 2-tailed Mann-Whitney U test for continuous variables.
Sequencing of the Patient-Specific IgH Sequence and Real-Time Quantitative IgH-PCR IgH sequences were derived from total RNA preparations of BM samples at the time of diagnosis, and a patient-specific IgH-qPCR was performed on the LightCycler system (Roche, Mannheim, Germany) using Taqman technology as described [25]. The assays were performed as previously published [16,25]. To achieve optimal specificity, an allelle-specific oligonucleotide Taqman probe complementary to the complementarity determining region 3 and an allelle-specific oligonucleotide sense primer and an allelle-specific oligonucleotide antisense primer were designed. IgH-PCR was performed according to previously published guidelines [26]. Instead of serial dilutions of DNA from the diagnostic samples with known plasma cell infiltration, a serial 10-fold dilution of 103 to 105 plasmid IgH copy numbers was used as an external, exogenous standard. As a reference for relative quantification, a b-actin PCR was run on the same instrument with the same DNA samples. Each patient-specific PCR reaction required a minimal sensitivity of 104 (105 in 87% of cases) and a linear amplification within the quantitative range with r .98. To confirm analytical specificity, no background amplification of each IgH-PCR was allowed using a pool of DNA from healthy donors. Negativity was ensured by PCR without specific amplification in 5 replicates, and the proportion of clonotypic cells among mononuclear cells was expressed as the 2IgH/b-actin ratio in percent. Statistical Analysis OS and event-free survival(EFS) were measured from the time of first diagnosis and the time of HDT or from 3 months after HDT, respectively, to the date of disease progression or death and analyzed by Kaplan-Meier plots
and the log rank test. Differences between nominal patient characteristics were analyzed using Fisher’s exact test for 2 categories and the 2-tailed chisquare test for 3 categories. Differences between continuous variables were tested by two-tailed Mann-Whitney U test. Uni- and multivariate analyses were performed using Cox regression analysis. For statistical analysis and graphical representations, SPSS 15.0 (SPSS Inc., Chicago, IL), Microsoft Office Excel 2007 (Microsoft Corp., Redmond, WA) and Graphpad Prism (GraphPad Software, San Diego, CA) were used.
RESULTS Samples and MRD Measurements We measured the level of clonotypic cells by IgH-qPCR in 533 PB samples collected during the course of disease in 42 patients undergoing HDT and autologous PBSCT as first-line treatment for MM. Samples were available at the time of diagnosis in 12 patients, approximately 1 week (median, 8 days; range, 3 to 30 days) before HDT in 21 patients and approximately 3 months (median, 3.2 months; range, 2.0 to 4.4 months) after HDT in 32 patients (Figure 1). At time of relapse, samples of 22 patients were available. At time of diagnosis, clonotypic cells were detectable in 67% of PB samples (n ¼ 8) with a median MRD level of .001% 2IgH/b-actin ratio (median of positive samples, .02% 2IgH/bactin ratio; range, .001% to 2.5%). During the course of first-line P = .03 P = .2
P = .3
P = .002
100
2 IgH / β-actin ratio in PB (%)
Three months after HDT, all patients underwent maintenance therapy with IFN-a (n ¼ 20) or thalidomide (n ¼ 15). IFN-a was given at a dose between 1.0 and 4.5 106 U subcutaneously twice per week. The daily dose of thalidomide varied between 100 and 400 mg. Maintenance therapy with IFN-a or thalidomide was continued until disease progression or discontinued in case of toxicity. In case of an HLA-identical sibling donor, an allogeneic transplantation was performed (n ¼ 7) after reduced-intensity conditioning with fludarabine 30 mg/m2 on days e3 to e1 and 2 Gy total body irradiation on day 0. Remission status and disease progression were defined according to the European Group for Blood and Marrow Transplantation (EBMT) criteria [24], with the only exception that a CR was not confirmed by a bone marrow biopsy. All patients with a suitable MRD marker were included in the study regardless of a negative or positive immunofixation or the remission status after HDT. We treated 121 patients with first-line HDT and autologous PBSCT. BM samples were obtained from 71 patients with MM of IgG/A subtype at the time of diagnosis. For 57 of them (80%), we could establish a patient-specific IgH-PCR. We included patients with the following criteria: (1) at least 1 PB sample for MRD detection obtained after HDT and (2) in case of clinical relapse, 1 sample at least 12 months before disease progression. A total of 42 patients entered the analysis. Baseline characteristics of these patients are listed in Table 1. All patients gave written informed consent for treatment and molecular analysis. The study was performed according to the guidelines of the ethical committee of the University of Duesseldorf.
10 1 0.1 0.01 0.001 0
First diagnosis
Before
After
Relapse
HDT + PBSCT
Figure 1. MRD in PB at a single time point. MRD levels in PB were analyzed in samples obtained either at the time of first diagnosis (n ¼ 12), approximately 1 week before (n ¼ 21) or 3 months after (n ¼ 32) HDT and PBSCT, or at the time of relapse (n ¼ 22).
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
Table 2 Correlation between the MRD Level and the Remission Status before or 3 Months after HDT and PBSCT
p < 0.0001 100
Remission Status before Transplantation
10
nCR
PR
MRSD
0 (0%) 2 (22%)
9 (75%) 3 (33%)
3 (25%) 4 (44%)
Remission Status 3 Mo after Transplantation
MRD in PB after transplantationy Negative Positive
nCR
PR
MRSD
7 (29%) 2 (25%)
16 (67%) 5 (63%)
1 (4%) 1 (13%)
% 2IgH / β-actin ratio
MRD in PB before transplantation* Negative Positive
1111
1
0.1
0.01
0.001
n=6
0
MRSD indicates minimal response/stable disease; nCR, near complete remission; PR, partial remission. * P ¼ .09. y P ¼ .7. Significance was calculated by chi-square test.
therapies, the number of circulating clonotypic cells was stepwise reduced in comparison with the MRD levels at first diagnosis. This was shown by a lower detection rate of 43% in samples obtained at the time before HDT (P ¼ .3, median of positive samples, .02% 2IgH/b-actin ratio; range, .001% to .9%) and only 25% in samples obtained 3 months after HDT (P ¼ .03, median MRD level in positive samples, .004% 2IgH/b-actin ratio; range, .001% to .9%). At the time of relapse, the number of tumor cells circulating in the PB was elevated and detectable again in 73% of patients (P ¼ .002, median level in positive samples, .04% 2IgH/b-actin ratio; range, .001% to 6.4%). In 6 of 22 patients the IgH-qPCR was negative at the time of relapse. Surprisingly, we did not find a significant correlation between remission rates, as defined by the EBMT criteria, and MRD levels in PB, either before or 3 months after HDT and PBSCT (Table 2). Interestingly, 4 patients with a CR had positive PCR results in their PB samples, whereas 4 patients who did not achieve a remission had negative PCR results at the same time. Comparison between MRD Level in BM and PB At first, we looked for the intraindividual varieties between 67 paired samples from BM and PB that were taken on the same day. These samples were obtained from 36 patients at different time points during the course of disease (Figure 2). Overall, these PB samples had a median MRD level of 0% 2IgH/b-actin ratio (range, 0 to 6%) that was significantly lower than the median MRD level of .3% 2IgH/b-actin ratio in the corresponding BM samples (range, 0 to 100%, P < .0001). By direct comparison of paired positive PB and BM samples (excluding pairs with negative PB samples), the MRD level in BM was in median 40-fold higher than the MRD level in the corresponding PB sample with ratios ranging from .4 to 4628. Paired PB and BM samples obtained before or 3 months after HDT and PBSCT were available from 20 and 24 patients, respectively. Although all PCR-negative BM samples were also paired with PCR-negative PB samples, PCR-positive BM samples (median, 2.3% 2IgH/b-actin ratio; range, .001% to 100%) were accompanied by PCR-positive results in 37% of corresponding PB samples (median, .02% 2IgH/b-actin ratio; range, .001% to 6%) (Table 3). Prognostic Value of MRD in PB at a Single Time Point We correlated MRD levels at 3 different time points, namely at the time of first diagnosis, 1 week before HDT and
PB
BM
Figure 2. Comparison of MRD levels of 67 paired samples from PB and BM of 36 patients. PB and BM samples were taken on the same day and represent different time points during the course of disease. PCR results of paired samples are indicated by lines.
autologous PBSCT, and 3 months after transplantation, with the clinical outcome of patients. Looking at MRD levels at the time of first diagnosis and treatment outcome, we found that patients with negative PCR results (n ¼ 4) had a significantly prolonged EFS (P ¼ .006) of median 66 months in comparison with patients with positive PCR results at this time point (n ¼ 8) and a median EFS of 26 months (Figure 3A). The OS of these patients (median, 66 versus 53 months) was not significantly different (P ¼ .3), probably due to the low number of cases (Figure 3B). On the other hand, the MRD levels in PB samples obtained at the time before HDT had no prognostic relevance in terms of EFS and OS (Figure 3C,D). Interestingly, a negative PCR result in PB 3 months after HDT was associated with a better treatment outcome. Patients with a molecular remission in PB after HDT and autologous PBSCT (n ¼ 24) had a significantly prolonged EFS of 15 months (P ¼ .004) in comparison with patients with positive PCR results at this time point (n ¼ 8) and a median EFS of 4 months (Figure 3E). Consistently, patients without detectable MRD levels in PB after transplantation also had a prolonged OS of 52 months (P ¼ .03) in comparison with MRD-positive patients who had an OS of only 17 months (Figure 3F). In line with this finding, MRD negativity in PB after HDT was associated with other well-established prognostic factors (Table 1), such as low International Staging Table 3 Correlation between the MRD Levels in BM and PB before or 3 Months after HDT and PBSCT MRD in PB before Transplantation
MRD in BM before transplantation* Negative Positive
Negative
Positive
3 (100%) 9 (53%)
0 (0%) 8 (47%)
MRD in PB after Transplantation
MRD in BM after transplantationy Negative Positive * y
Negative
Positive
3 (100%) 14 (67%)
0 (0%) 7 (33%)
P ¼ .2. P ¼ .5. Significance was calculated by Fisher’s exact test.
1112
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
MRD at diagnosis
MRD at diagnosis
MRD negative (n = 4)
MRD negative (n = 4)
MRD positive (n = 8)
MRD positive (n = 8) 66 vs 53 months, p = 0.3
66 vs 26 months, p = 0.006
100
100
OS Proportion of patients (%)
Proportion of patients (%)
EFS 80 60 40 20
80 60 40 20 0
0 0
12
24
36
48
60
72
0
84
12
EFS after diagnosis (months)
24
36
84
MRD negative (n = 12)
100
80
EFS 60 40 20
Proportion of patients (%)
MRD positive (n = 9) 13 vs 19 months, p = 0.6
Proportion of patients (%)
72
MRD before HDT
MRD negative (n = 12)
MRD positive (n = 9) 60 vs 65 months, p = 0.5
80
OS 60 40 20 0
0 0
12
24
36
48
60
72
0
84
12
24
36
60
72
84
MRD after HDT
MRD after HDT
MRD negative (n = 24)
MRD negative (n = 24)
100
48
OS after HDT (months)
EFS after HDT (months)
100
MRD positive (n = 8) 15 vs 4 months, p = 0.004
Proportion of patients (%)
Proportion of patients (%)
60
OS after diagnosis (months)
MRD before HDT
100
48
80
EFS 60 40 20
MRD positive (n = 8) 52 vs 17 months, p = 0.03
80
OS 60 40 20 0
0 0
12
24
36
48
60
72
84
EFS from 3 months after HDT (months)
0
12
24
36
48
60
72
84
OS from 3 months after HDT (months)
Figure 3. Correlation of the MRD levels in PB at single time points with the clinical outcome of patients. The EFS of patients with negative or positive PCR results in PB either at the time of first diagnosis (A), 1 week before (C), or 3 months after (E) HDT and autologous PBSCT was analyzed by Kaplan-Meier plots and the log rank test. Similarly, the OS of the patients with negative or positive PCR results at the time of first diagnosis (B), before (D), or after (F) HDT and autologous PBSCT was analyzed. FD indicates first diagnosis; pre-MRD, MRD level before HDT and autologous PBSCT; post-MRD, MRD level after HDT and autologous PBSCT.
System stage (P ¼ .01), low b2-microglobulin (median, 2.3 mg/dL versus 3.9 mg/dL; P ¼ .02), and high levels of hemoglobin (median, 11.7 g/dL versus 9.2 g/dL; P ¼ .01). Prognostic Value of Sequential MRD Monitoring in PB After a median follow-up of survivors of 45 months (range, 13 to 80 months) after HDT, 29 of 42 patients (69%) had suffered from relapse, whereas 13 patients (31%) were still in ongoing remission. Sequential monitoring of clonotypic cells in PB showed that in 19 of 29 patients (66%) with progressive disease, a loss of molecular remission or an increase of the 2IgH/b-actin ratio of more than 1 log step could be observed (Figure 4). In these patients, the rise in circulating clone-specific cells occurred in median 4 months (range, .8 to 13 months) before the relapse was diagnosed by
conventional diagnostic tests on the basis of the EBMT criteria. In 10 patients negative results in PB could be documented until the time of relapse. Looking at the 13 patients in ongoing remission, the duration of remission was in median 45 months (range, 18 to 78 months). The MRD results were continuously negative in 8 of these patients, whereas in 5 patients single positive results could not be confirmed in following samples. Next, we investigated whether circulating tumor cells detected before serological relapse were associated with a poor prognosis. Therefore, we looked at all patients stratified according to whether or not they showed a molecular relapse before serological relapse. The median OS from the time of first clinical relapse was 17 months and not reached, respectively (P ¼ .02, Figure 5).
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
1113
Figure 4. Sequential monitoring of MRD in PB. MRD levels in consecutive PB samples are shown for 29 patients with relapse after HDT and autologous PBSCT as a function of time after HDT. One log steps of the 2IgH/b-actin ratio are depicted by circles with different color fills. In 19 patients (66%) an increase of the 2IgH/b-actin ratio of more than 1 log step (arrows) could be detected in PB before clinical disease progression was observed.
DISCUSSION To date, monitoring of MRD by IgH-PCR or flow cytometry is not a standard diagnostic procedure in patients with MM [9], but there is growing evidence that the measurement of the level of residual tumor cells in the BM may help to identify patients at different risk of relapse [12,16,18,25]. Although MRD assessment in BM is a valuable tool to evaluate treatment outcome, the role of MRD in the PB of patients with MM is currently not clear. In this study, we present data of 42 patients with MM undergoing HDT and autologous PBSCT who were monitored for MRD in the PB by IgH-qPCR at different time points during the course of disease. The absence of detectable MRD levels in PB at the time of first diagnosis as well as 3 months after HDT was associated with a prolonged EFS and OS. Moreover, by sequentially measuring residual clonotypic cells in samples collected on a monthly basis, relapse could be identified in median 4 months earlier than with conventional diagnostic methods. Further, the appearance of clonotypic cells circulating in PB before the diagnosis of a clinical relapse was also associated with a shorter OS. When addressing the role of MRD in PB, the first essential question is whether there is a difference between MRD in BM and PB. Our comparison of both compartments showed a great difference in the level of clonotypic cells with a variation of the MRD ratio between BM and PB from less than 1 up to approximately 5000-fold without any pattern
according to sampling time point or remission status. Therefore, the MRD level in PB is not only a mere reflection of the MRD level in BM; as a result, in patients with MM it is worth looking at MRD in both compartments. The second question is whether MRD levels in PB at a single time point are of prognostic relevance. Using flow cytometry, it has been shown that circulating myeloma cells in the PB at a given time point are associated with a poor outcome of patients undergoing HDT and autologous PBSCT. This is true for the time of first diagnosis [27], the time before [28], and 3 months after [29] HDT and autologous transplantation. Using IgH-PCR to detect MRD in PB, the published patient numbers range from 6 to 20 [19-23]. In our study we report results on molecular MRD analysis in 42 patients using IgH-qPCR. In line with the published data acquired by FACS analysis [27-29], our data confirm the poor prognosis of circulating myeloma cells at the time of first diagnosis and also confirm the prognostic benefit of a response on a molecular level in PB after HDT and autologous PBSCT in terms of a prolonged EFS or OS. The absence or presence of detectable myeloma cells in PB at this time point also correlated with known prognostic factors such as International Staging System stage and serum levels of lactate dehydrogenase and b2-microglobulin, although there was no correlation with the remission status as defined by the EBMT criteria. Regarding the fact that the number of clonotypic cells in PB do not correlate with the number in BM, it is tempting to
1114
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
Molecular relapse = serological relapse (n = 10) Molecular prior to serological relapse (n = 19) n.r. vs 17 months, P = .02
Proportion of patients (%)
100
monitoring of MRD can help to detect disease progression earlier than conventional diagnostic methods. Even though the clinical utility is limited, evaluation of MRD in PB by IgHqPCR may help us to better understand the pathophysiology of disease recurrence during therapy with investigational agents in the future.
OS 80 60 40 20 0 0
12
24
36
48
60
72
84
OS after serological relapse (months) Figure 5. Comparison of the survival of patients with a molecular relapse before clinical relapse. The OS was calculated as time from clinical relapse and analyzed by Kaplan-Meier plots and the log rank test.
ask if these 2 measures reflect qualitatively different aspects of MRD. Because of relatively low patient numbers with paired PB and BM samples, we were not able to finally clarify if MRD levels in PB and BM are of prognostic relevance in a complementary manner. The third question addresses the impact of MRD not only at a single time point but as a sequential monitoring of dynamic changes of the number of residual tumor cells during times of clinical remission. In this field, FACS data are rare, whereas IgH-PCR studies [23,30,31] with small patient numbers (n ¼ 1 to 6) showed that an increase in tumor load could be detected in PB before disease progression was proven by conventional means. We confirm these data in our group of 42 patients, where a molecular relapse could be observed in 66% of 29 patients with progressive disease in median 4 months before the clinical relapse was diagnosed. Moreover, we show that patients with a molecular relapse preceding clinical relapse have a very short survival after disease progression. This interesting finding hints at a role of these clonotypic cells in spreading the disease, which should be further investigated. It will also be of interest to see whether new agents may be effective in overcoming the bad prognosis associated with these circulating tumor cells. On the other hand, the clinical relevance of this finding is relatively small, because most physicians will not treat patients with MM due to a molecular relapse, in the same way they do not treat patients with relapse from CR unless the clinical relapse is associated with increasing disease activity or end-organ damage [32]. In combination with the fact that IgH-PCR can only be applied to about 75% of patients due to technical reasons [14,16,25], we do not see a great potential in the future of MRD detection in the PB as a routine diagnostic tool for all patients with MM. However, IgH-qPCR may be a very suitable tool to evaluate therapy response to investigational drugs, whose mechanism of action often are not completely understood, and to define the origin of drugresistant tumor cells. In conclusion, molecular remission in PB is a prognostic parameter for a prolonged EFS and OS, and monthly
ACKNOWLEDGMENTS Financial disclosure: Supported by Deutsche Krebshilfe (Bonn, Germany), Leukämie Liga e.V. (Duesseldorf, Germany), and Novartis. Conflict of interest statement: G.K. was a consultant to Amgen and Celgene; received lecture fees from Amgen, Celgene, and Janssen-Cilag; and received research funds from Amgen and Celgene. R.F. was a consultant to Celgene, received lecture fees from Celgene and Janssen-Cilag, and received research funds from Novartis and Celgene. Authorship statement: R.F. designed research, collected data, performed statistical analysis, interpreted data, and drafted the manuscript. M.K. performed the experiments and statistical analysis, interpreted data, and drafted the manuscript. N.S. performed the experiments, collected data, and drafted the manuscript. R.K. designed research and contributed substantially to the development of the methodology. T.S. and T.S. were involved in patient recruitment and correlation of clinical data with experimental findings. G.K. was significantly involved in patient recruitment and correlation of clinical data with experimental findings. R.H. was significantly involved in the induction and conception of the study and interpreted data. All authors critically revised the manuscript. REFERENCES 1. Attal M, Harousseau JL, Stoppa AM, et al. A prospective, randomized trial of autologous bone marrow transplantation and chemotherapy in multiple myeloma. Intergroupe Francais du Myelome. N Engl J Med. 1996;335:91-97. 2. Haas R, Bruns I, Kobbe G, Fenk R. High-dose therapy and autologous peripheral blood stem cell transplantation in patients with multiple myeloma. Rec Res Cancer Res. 2011;183:207-238. 3. Fenk R, Haas R, Kronenwett R. Molecular monitoring of minimal residual disease in patients with multiple myeloma. Hematology. 2004; 9:17-33. 4. Szczepanski T, Orfao A, van der Velden, et al. Minimal residual disease in leukaemia patients. Lancet Oncol 2001 Jul;2:409-417. 5. Harousseau JL, Attal M, Avet-Loiseau H. The role of complete response in multiple myeloma. Blood. 2009;114:3139-3146. 6. Richardson PG, Sonneveld P, Schuster MW, et al. Bortezomib or highdose dexamethasone for relapsed multiple myeloma. N Engl J Med. 2005;352:2487-2498. 7. Dimopoulos M, Spencer A, Attal M, et al. Lenalidomide plus dexamethasone for relapsed or refractory multiple myeloma. N Engl J Med. 2007;357:2123-2132. 8. Weber DM, Chen C, Niesvizky R, et al. Lenalidomide plus dexamethasone for relapsed multiple myeloma in North America. N Engl J Med. 2007;357:2133-2142. 9. Hart AJ, Jagasia MH, Kim AS, et al. Minimal residual disease in myeloma: Are we there yet? Biol Blood Marrow Transplant. 2012;18: 1790-1799. 10. Bruggemann M, Schrauder A, Raff T, et al. Standardized MRD quantification in European ALL trials: Proceedings of the Second International Symposium on MRD assessment in Kiel, Germany, 18-20 September 2008. Leukemia. 2010;24:521-535. 11. Neumann F, Markett J, Fenk R, et al. Therapy adapted to molecular response in patients with chronic myelogenous leukaemia in first chronic phase: results of the Duesseldorf study. Hematol Oncol. 2008; 26:213-218. 12. Paiva B, Vidriales MB, Cervero J, et al. Multiparameter flow cytometric remission is the most relevant prognostic factor for multiple myeloma patients who undergo autologous stem cell transplantation. Blood. 2008;112:4017-4023. 13. Paiva B, Gutierrez NC, Rosinol L, et al. High-risk cytogenetics and persistent minimal residual disease by multiparameter flow cytometry
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
14.
15.
16.
17.
18.
19.
20.
21.
22.
predict unsustained complete response after autologous stem cell transplantation in multiple myeloma. Blood. 2012;119:687-691. Sarasquete ME, Garcia-Sanz R, Gonzalez D, et al. Minimal residual disease monitoring in multiple myeloma: a comparison between allelic-specific oligonucleotide real-time quantitative polymerase chain reaction and flow cytometry. Haematologica. 2005;90:1365-1372. Bakkus MH, Bouko Y, Samson D, et al. Post-transplantation tumour load in bone marrow, as assessed by quantitative ASO-PCR, is a prognostic parameter in multiple myeloma. Br J Haematol. 2004;126: 665-674. Korthals M, Sehnke N, Kronenwett R, et al. The level of minimal residual disease in the bone marrow of patients with multiple myeloma before high-dose therapy and autologous blood stem cell transplantation is an independent predictive parameter. Biol Blood Marrow Transplant. 2012;18:423-431. Putkonen M, Kairisto V, Juvonen V, et al. Depth of response assessed by quantitative ASO-PCR predicts the outcome after stem cell transplantation in multiple myeloma. Eur J Haematol. 2010;85:416-423. Corradini P, Cavo M, Lokhorst H, et al. Molecular remission after myeloablative allogeneic stem cell transplantation predicts a better relapse-free survival in patients with multiple myeloma. Blood. 2003; 102:1927-1929. Billadeau D, Prosper F, Verfaillie C, et al. Sequential analysis of bone marrow and peripheral blood after stem cell transplant for myeloma shows disparate tumor involvement. Leukemia. 1997;11:1565-1570. Cremer FW, Ehrbrecht E, Kiel K, et al. Evaluation of the kinetics of the bone marrow tumor load in the course of sequential high-dose therapy assessed by quantitative PCR as a predictive parameter in patients with multiple myeloma. Bone Marrow Transplant. 2000;26:851-858. Kiel K, Cremer FW, Rottenburger C, et al. Analysis of circulating tumor cells in patients with multiple myeloma during the course of high-dose therapy with peripheral blood stem cell transplantation. Bone Marrow Transplant. 1999;23:1019-1027. Lipinski E, Cremer FW, Ho AD, et al. Molecular monitoring of the tumor load predicts progressive disease in patients with multiple myeloma
23.
24.
25.
26.
27.
28.
29.
30. 31.
32.
1115
after high-dose therapy with autologous peripheral blood stem cell transplantation. Bone Marrow Transplant. 2001;28:957-962. Raab MS, Cremer FW, Breitkreutz IN, et al. Molecular monitoring of tumour load kinetics predicts disease progression after nonmyeloablative allogeneic stem cell transplantation in multiple myeloma. Ann Oncol. 2005;16:611-617. Blade J, San Miguel JF, Fontanillas M, et al. Survival of multiple myeloma patients who are potential candidates for early high-dose therapy intensification/ autotransplantation and who were conventionally treated. J Clin Oncol. 1996;14:2167-2173. Fenk R, Ak M, Kobbe G, et al. Levels of minimal residual disease detected by quantitative molecular monitoring herald relapse in patients with multiple myeloma. Haematologica. 2004;89:557-566. van der Velden VH, Cazzaniga G, Schrauder A, et al. Analysis of minimal residual disease by Ig/TCR gene rearrangements: guidelines for interpretation of real-time quantitative PCR data. Leukemia. 2007;21: 604-611. Nowakowski GS, Witzig TE, Dingli D, et al. Circulating plasma cells detected by flow cytometry as a predictor of survival in 302 patients with newly diagnosed multiple myeloma. Blood. 2005;106:2276-2279. Dingli D, Nowakowski GS, Dispenzieri A, et al. Flow cytometric detection of circulating myeloma cells before transplantation in patients with multiple myeloma: A simple risk stratification system. Blood. 2006;107:3384-3388. Liu H, Yuan C, Heinerich J, et al. Flow cytometric minimal residual disease monitoring in patients with multiple myeloma undergoing autologous stem cell transplantation: A retrospective study. Leuk Lymph. 2008;49:306-314. Cooke F, Bakkus M, Thielemans K, et al. Use of quantitative ASO-PCR to predict relapse in multiple myeloma. Br J Haematol. 1999;105:317-319. Rasmussen T, Hansson L, Osterborg A, et al. Idiotype vaccination in multiple myeloma induced a reduction of circulating clonal tumor B cells. Blood. 2003;101:4607-4610. Durie BG, Salmon SE. Cellular kinetics staging, and immunoglobulin synthesis in multiple myeloma. Annu Rev Med. 1975;26:283-288.
Molecular Monitoring of Minimal Residual Disease in the Peripheral Blood of Patients with Multiple Myeloma Mark Korthals 1, 2, Nina Sehnke 1, Ralf Kronenwett 1, y, Thomas Schroeder 1, Tobias Strapatsas 1, Guido Kobbe 1, Rainer Haas 1, Roland Fenk 1, * 1 2
ASBMT
American Society for Blood and Marrow Transplantation
Department of Hematology, Oncology and Clinical Immunology, Heinrich-Heine-University, Duesseldorf, Germany Department of Functional Genomics and Medical Toponomics, University Magdeburg, Magdeburg, Germany
Article history: Received 16 July 2012 Accepted 26 April 2013 Key Words: Minimal residual disease IgH-PCR Multiple myeloma
a b s t r a c t The prognostic relevance of minimal residual disease (MRD) in patients with multiple myeloma is still an open question. In bone marrow, the level of residual myeloma cells is associated with treatment outcome, but the role of clonotypic cells in the peripheral blood (PB) for the prognosis of patients is not identified yet. In this study, we retrospectively analyzed MRD by quantitative real-time IgH-PCR (IgH-qPCR) in the PB of 42 patients undergoing high-dose therapy followed by autologous PB stem cell transplantation as first-line therapy for multiple myeloma. The MRD level of PB samples was in median 40-fold lower than in bone marrow samples, collected on the same day, with a wide intra- and interindividual variation (range, .4- to 4628-fold). The presence or absence of detectable MRD levels in PB did not correlate with the serological remission status. Still, patients with negative PCR results in PB 3 months after high-dose therapy and PB stem cell transplantation had lower International Staging System stage (P ¼ .01), lower levels of b2-microglobulin (P ¼ .02), higher hemoglobin levels (P ¼ .01), and a prolonged event-free (median, 15 versus 4 months; P ¼ .004) and overall (median, 52 versus 17 months; P ¼ .03) survival. Importantly, by sequential monitoring of clonotypic cells in PB, in 19 of 29 patients (66%) with progressive disease, an increase of the 2IgH/b-actin ratio of at least 1 log step could be detected in median 4 months (range, .8 to 13 months) before the relapse was diagnosed on the basis of the European Group for Blood and Marrow Transplantation criteria. These patients with a molecular relapse in PB before a serological relapse had a significantly shorter overall survival than other patients (median, 17 months versus median not reached, P ¼ .02). In conclusion, IgH-qPCR is a sensitive technique for the detection of clonotypic cells in PB, which precede clinical relapse. Future studies are needed to evaluate whether these circulating tumor cells play a role in promoting disease recurrence. Ó 2013 American Society for Blood and Marrow Transplantation.
INTRODUCTION Multiple myeloma (MM) is a B cell malignancy characterized by a monoclonal expansion of plasma cells in bone marrow (BM) and the secretion of paraprotein in the serum and/or urine of patients. High-dose chemotherapy (HDT) followed by autologous peripheral blood stem cell transplantation (PBSCT) has improved the clinical outcome, but eventually all patients relapse [1,2]. The quantitative measurement of minimal residual disease (MRD) on a molecular level is a highly sensitive and specific method to detect tumor cells of clonal origin [3,4] that can help to estimate the risk of relapse. Evidence shows that the achievement of a serological complete remission (CR) is associated with a prolonged survival of patients with MM [5]. Treatment protocols using chemotherapy in combination with novel agents such as bortezomib [6] or lenalidomide [7,8] are continuously improving the depth of response. Thus, the assessment of MRD becomes more and more important, and the level of MRD may help to guide therapeutic decisions in patients with MM in the future [9], as it is already clinical practice in patients with acute lymphoblastic leukemia or chronic myeloid leukemia [10,11].
Financial disclosure: See Acknowledgments on page 1114. * Correspondence and reprint requests: PH Dr. Roland Fenk, Dept. of Hematology, Oncology and Clinical Immunology, Heinrich-HeineUniversity, Moorenstr. 5, 40225 Duesseldorf, Germany. E-mail address: [email protected] (R. Fenk). y Current address: PH Dr. Ralf Kronenwett: Sividon Diagnostics, Cologne, Germany.
Residual tumor cells in BM have a prognostic value in newly diagnosed patients with MM treated with HDT followed by autologous PBSCT, as has been shown by flow cytometry [12-14] and quantitative real-time IgH-PCR (IgH-qPCR) [14-17]. In the same line are results from Corradini et al. [18], who used qualitative IgH-PCR in BM samples of 48 patients undergoing allogeneic PBSCT. In this setting, negative MRD results were also associated with prolonged overall survival (OS) and progression-free survival. Whereas the prognostic impact of MRD in the BM is well documented, experimental studies of MRD in PB of patients with MM are rare, and patient numbers of reported results are very low [19-23]. As with BM, fluorescent activated cell sorter (FACS) analysis or IgH-PCR are the methods of choice for the detection of very low numbers of myeloma cells in the PB of patients. To acquire more data on this topic, we used IgH-qPCR [3,16] and sequentially measured the amount of clone-specific cells in the PB of 42 patients with MM during first-line therapy with HDT and autologous PBSCT. METHODS Patients and Samples Patients were treated at our hospital between November 2000 and April 2008. All patients presented with symptomatic MM and received the same first-line therapy consisting of HDT followed by autologous PBSCT. The induction treatment of all patients consisted of 2 to 4 cycles of idarubicine (20 mg/m2 i.v., days 1 to 4) and dexamethasone (40 mg oral, days 1 to 4, 9 to 12, and 17 to 21) in monthly intervals. The patients then received pegfilgrastim or granulocyte colony-stimulating factor in combination with cyclophosphamide (2 g/m2, days 1 þ 2) for PB stem cell mobilization. Thereafter, a single HDT was administered at a dose of 200 mg/m2 melphalan followed by autologous PBSCT.
1083-8791/$ e see front matter Ó 2013 American Society for Blood and Marrow Transplantation. http://dx.doi.org/10.1016/j.bbmt.2013.04.025
1110
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
Table 1 Patient Characteristics MRD before Transplantation
Age, median yrs (range) Sex (male:female), % Ig subtype (IgG:others), % Stage SþD (IþII:III), % Stage ISS (1:2:3), % Cytogenetics (abnormal:normal), % CRP, median mg/dL (range) LDH, median U/L (range) Albumin, median g/dL (range) b2-microglobulin, median mg/dL (range) Hemoglobin, median g/dL (range) Platelet count, median 103 (range) Maintenance therapy (IFN:Thal:RIC), %
MRD after Transplantation *
All Patients (n ¼ 42)
Negative (n ¼ 12)
Positive (n ¼ 9)
P
Negative (n ¼ 24)
Positive (n ¼ 8)
P*
54 (31-66) 67:33 74:26 12:88 43:38:19 27:73 .5 (.3-9.8) 160 (79-346) 3.9 (2.8-5.1) 2.8 (1.2-15.3) 11.1 (6.9-17.0) 261 (108-475) 48:36:17
54 (45-62) 75:25 58:42 8:92 25:67:8 42:58 .6 (.3-9.8) 149 (98-285) 3.6 (2.9-5.1) 3.0 (1.5-8.5) 11.0 (6.9-13.9) 239 (108-351) 50:42:8
55 (44-65) 67:33 78:22 11:89 56:33:11 22:78 .7 (.3-1.8) 171 (94-346) 4.2 (3.2-4.9) 2.4 (1.6-8.0) 11.4 (9.3-17.0) 320 (115-475) 33:56:11
1.0 1.0 .6 1.0 .3 .6 .9 .5 .08 .7 .1 .3 .7
54 (31-65) 67:33 67:33 13:88 58:33:8 22:78 .6 (.3-9.8) 149 (79-335) 4.0 (3.1-4.9) 2.3 (1.2-8.5) 11.7 (7.6-17) 259 (108-360) 42:46:13
55 (42-62) 63:38 75:25 0:100 0:63:38 25:75 .8 (.3-2.4) 227 (107-346) 3.7 (2.9-4.9) 3.9 (2.1-10.2) 9.2 (6.9-13.6) 245 (115-475) 63:25:13
.8 1.0 1.0 .6 .01 1.0 .8 .08 .3 .02 .01 .8 .6
CRP indicates C-reactive protein; ISS, international staging system; LDH, lactate dehydrogenase; RIC, reduced-intensity conditioning allogeneic transplantation; SþD, Salmon and Durie; Thal, thalidomide. * Correlations were calculated by Fisher’s exact test for two categories, by chi-square test for three categories or the 2-tailed Mann-Whitney U test for continuous variables.
Sequencing of the Patient-Specific IgH Sequence and Real-Time Quantitative IgH-PCR IgH sequences were derived from total RNA preparations of BM samples at the time of diagnosis, and a patient-specific IgH-qPCR was performed on the LightCycler system (Roche, Mannheim, Germany) using Taqman technology as described [25]. The assays were performed as previously published [16,25]. To achieve optimal specificity, an allelle-specific oligonucleotide Taqman probe complementary to the complementarity determining region 3 and an allelle-specific oligonucleotide sense primer and an allelle-specific oligonucleotide antisense primer were designed. IgH-PCR was performed according to previously published guidelines [26]. Instead of serial dilutions of DNA from the diagnostic samples with known plasma cell infiltration, a serial 10-fold dilution of 103 to 105 plasmid IgH copy numbers was used as an external, exogenous standard. As a reference for relative quantification, a b-actin PCR was run on the same instrument with the same DNA samples. Each patient-specific PCR reaction required a minimal sensitivity of 104 (105 in 87% of cases) and a linear amplification within the quantitative range with r .98. To confirm analytical specificity, no background amplification of each IgH-PCR was allowed using a pool of DNA from healthy donors. Negativity was ensured by PCR without specific amplification in 5 replicates, and the proportion of clonotypic cells among mononuclear cells was expressed as the 2IgH/b-actin ratio in percent. Statistical Analysis OS and event-free survival(EFS) were measured from the time of first diagnosis and the time of HDT or from 3 months after HDT, respectively, to the date of disease progression or death and analyzed by Kaplan-Meier plots
and the log rank test. Differences between nominal patient characteristics were analyzed using Fisher’s exact test for 2 categories and the 2-tailed chisquare test for 3 categories. Differences between continuous variables were tested by two-tailed Mann-Whitney U test. Uni- and multivariate analyses were performed using Cox regression analysis. For statistical analysis and graphical representations, SPSS 15.0 (SPSS Inc., Chicago, IL), Microsoft Office Excel 2007 (Microsoft Corp., Redmond, WA) and Graphpad Prism (GraphPad Software, San Diego, CA) were used.
RESULTS Samples and MRD Measurements We measured the level of clonotypic cells by IgH-qPCR in 533 PB samples collected during the course of disease in 42 patients undergoing HDT and autologous PBSCT as first-line treatment for MM. Samples were available at the time of diagnosis in 12 patients, approximately 1 week (median, 8 days; range, 3 to 30 days) before HDT in 21 patients and approximately 3 months (median, 3.2 months; range, 2.0 to 4.4 months) after HDT in 32 patients (Figure 1). At time of relapse, samples of 22 patients were available. At time of diagnosis, clonotypic cells were detectable in 67% of PB samples (n ¼ 8) with a median MRD level of .001% 2IgH/b-actin ratio (median of positive samples, .02% 2IgH/bactin ratio; range, .001% to 2.5%). During the course of first-line P = .03 P = .2
P = .3
P = .002
100
2 IgH / β-actin ratio in PB (%)
Three months after HDT, all patients underwent maintenance therapy with IFN-a (n ¼ 20) or thalidomide (n ¼ 15). IFN-a was given at a dose between 1.0 and 4.5 106 U subcutaneously twice per week. The daily dose of thalidomide varied between 100 and 400 mg. Maintenance therapy with IFN-a or thalidomide was continued until disease progression or discontinued in case of toxicity. In case of an HLA-identical sibling donor, an allogeneic transplantation was performed (n ¼ 7) after reduced-intensity conditioning with fludarabine 30 mg/m2 on days e3 to e1 and 2 Gy total body irradiation on day 0. Remission status and disease progression were defined according to the European Group for Blood and Marrow Transplantation (EBMT) criteria [24], with the only exception that a CR was not confirmed by a bone marrow biopsy. All patients with a suitable MRD marker were included in the study regardless of a negative or positive immunofixation or the remission status after HDT. We treated 121 patients with first-line HDT and autologous PBSCT. BM samples were obtained from 71 patients with MM of IgG/A subtype at the time of diagnosis. For 57 of them (80%), we could establish a patient-specific IgH-PCR. We included patients with the following criteria: (1) at least 1 PB sample for MRD detection obtained after HDT and (2) in case of clinical relapse, 1 sample at least 12 months before disease progression. A total of 42 patients entered the analysis. Baseline characteristics of these patients are listed in Table 1. All patients gave written informed consent for treatment and molecular analysis. The study was performed according to the guidelines of the ethical committee of the University of Duesseldorf.
10 1 0.1 0.01 0.001 0
First diagnosis
Before
After
Relapse
HDT + PBSCT
Figure 1. MRD in PB at a single time point. MRD levels in PB were analyzed in samples obtained either at the time of first diagnosis (n ¼ 12), approximately 1 week before (n ¼ 21) or 3 months after (n ¼ 32) HDT and PBSCT, or at the time of relapse (n ¼ 22).
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
Table 2 Correlation between the MRD Level and the Remission Status before or 3 Months after HDT and PBSCT
p < 0.0001 100
Remission Status before Transplantation
10
nCR
PR
MRSD
0 (0%) 2 (22%)
9 (75%) 3 (33%)
3 (25%) 4 (44%)
Remission Status 3 Mo after Transplantation
MRD in PB after transplantationy Negative Positive
nCR
PR
MRSD
7 (29%) 2 (25%)
16 (67%) 5 (63%)
1 (4%) 1 (13%)
% 2IgH / β-actin ratio
MRD in PB before transplantation* Negative Positive
1111
1
0.1
0.01
0.001
n=6
0
MRSD indicates minimal response/stable disease; nCR, near complete remission; PR, partial remission. * P ¼ .09. y P ¼ .7. Significance was calculated by chi-square test.
therapies, the number of circulating clonotypic cells was stepwise reduced in comparison with the MRD levels at first diagnosis. This was shown by a lower detection rate of 43% in samples obtained at the time before HDT (P ¼ .3, median of positive samples, .02% 2IgH/b-actin ratio; range, .001% to .9%) and only 25% in samples obtained 3 months after HDT (P ¼ .03, median MRD level in positive samples, .004% 2IgH/b-actin ratio; range, .001% to .9%). At the time of relapse, the number of tumor cells circulating in the PB was elevated and detectable again in 73% of patients (P ¼ .002, median level in positive samples, .04% 2IgH/b-actin ratio; range, .001% to 6.4%). In 6 of 22 patients the IgH-qPCR was negative at the time of relapse. Surprisingly, we did not find a significant correlation between remission rates, as defined by the EBMT criteria, and MRD levels in PB, either before or 3 months after HDT and PBSCT (Table 2). Interestingly, 4 patients with a CR had positive PCR results in their PB samples, whereas 4 patients who did not achieve a remission had negative PCR results at the same time. Comparison between MRD Level in BM and PB At first, we looked for the intraindividual varieties between 67 paired samples from BM and PB that were taken on the same day. These samples were obtained from 36 patients at different time points during the course of disease (Figure 2). Overall, these PB samples had a median MRD level of 0% 2IgH/b-actin ratio (range, 0 to 6%) that was significantly lower than the median MRD level of .3% 2IgH/b-actin ratio in the corresponding BM samples (range, 0 to 100%, P < .0001). By direct comparison of paired positive PB and BM samples (excluding pairs with negative PB samples), the MRD level in BM was in median 40-fold higher than the MRD level in the corresponding PB sample with ratios ranging from .4 to 4628. Paired PB and BM samples obtained before or 3 months after HDT and PBSCT were available from 20 and 24 patients, respectively. Although all PCR-negative BM samples were also paired with PCR-negative PB samples, PCR-positive BM samples (median, 2.3% 2IgH/b-actin ratio; range, .001% to 100%) were accompanied by PCR-positive results in 37% of corresponding PB samples (median, .02% 2IgH/b-actin ratio; range, .001% to 6%) (Table 3). Prognostic Value of MRD in PB at a Single Time Point We correlated MRD levels at 3 different time points, namely at the time of first diagnosis, 1 week before HDT and
PB
BM
Figure 2. Comparison of MRD levels of 67 paired samples from PB and BM of 36 patients. PB and BM samples were taken on the same day and represent different time points during the course of disease. PCR results of paired samples are indicated by lines.
autologous PBSCT, and 3 months after transplantation, with the clinical outcome of patients. Looking at MRD levels at the time of first diagnosis and treatment outcome, we found that patients with negative PCR results (n ¼ 4) had a significantly prolonged EFS (P ¼ .006) of median 66 months in comparison with patients with positive PCR results at this time point (n ¼ 8) and a median EFS of 26 months (Figure 3A). The OS of these patients (median, 66 versus 53 months) was not significantly different (P ¼ .3), probably due to the low number of cases (Figure 3B). On the other hand, the MRD levels in PB samples obtained at the time before HDT had no prognostic relevance in terms of EFS and OS (Figure 3C,D). Interestingly, a negative PCR result in PB 3 months after HDT was associated with a better treatment outcome. Patients with a molecular remission in PB after HDT and autologous PBSCT (n ¼ 24) had a significantly prolonged EFS of 15 months (P ¼ .004) in comparison with patients with positive PCR results at this time point (n ¼ 8) and a median EFS of 4 months (Figure 3E). Consistently, patients without detectable MRD levels in PB after transplantation also had a prolonged OS of 52 months (P ¼ .03) in comparison with MRD-positive patients who had an OS of only 17 months (Figure 3F). In line with this finding, MRD negativity in PB after HDT was associated with other well-established prognostic factors (Table 1), such as low International Staging Table 3 Correlation between the MRD Levels in BM and PB before or 3 Months after HDT and PBSCT MRD in PB before Transplantation
MRD in BM before transplantation* Negative Positive
Negative
Positive
3 (100%) 9 (53%)
0 (0%) 8 (47%)
MRD in PB after Transplantation
MRD in BM after transplantationy Negative Positive * y
Negative
Positive
3 (100%) 14 (67%)
0 (0%) 7 (33%)
P ¼ .2. P ¼ .5. Significance was calculated by Fisher’s exact test.
1112
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
MRD at diagnosis
MRD at diagnosis
MRD negative (n = 4)
MRD negative (n = 4)
MRD positive (n = 8)
MRD positive (n = 8) 66 vs 53 months, p = 0.3
66 vs 26 months, p = 0.006
100
100
OS Proportion of patients (%)
Proportion of patients (%)
EFS 80 60 40 20
80 60 40 20 0
0 0
12
24
36
48
60
72
0
84
12
EFS after diagnosis (months)
24
36
84
MRD negative (n = 12)
100
80
EFS 60 40 20
Proportion of patients (%)
MRD positive (n = 9) 13 vs 19 months, p = 0.6
Proportion of patients (%)
72
MRD before HDT
MRD negative (n = 12)
MRD positive (n = 9) 60 vs 65 months, p = 0.5
80
OS 60 40 20 0
0 0
12
24
36
48
60
72
0
84
12
24
36
60
72
84
MRD after HDT
MRD after HDT
MRD negative (n = 24)
MRD negative (n = 24)
100
48
OS after HDT (months)
EFS after HDT (months)
100
MRD positive (n = 8) 15 vs 4 months, p = 0.004
Proportion of patients (%)
Proportion of patients (%)
60
OS after diagnosis (months)
MRD before HDT
100
48
80
EFS 60 40 20
MRD positive (n = 8) 52 vs 17 months, p = 0.03
80
OS 60 40 20 0
0 0
12
24
36
48
60
72
84
EFS from 3 months after HDT (months)
0
12
24
36
48
60
72
84
OS from 3 months after HDT (months)
Figure 3. Correlation of the MRD levels in PB at single time points with the clinical outcome of patients. The EFS of patients with negative or positive PCR results in PB either at the time of first diagnosis (A), 1 week before (C), or 3 months after (E) HDT and autologous PBSCT was analyzed by Kaplan-Meier plots and the log rank test. Similarly, the OS of the patients with negative or positive PCR results at the time of first diagnosis (B), before (D), or after (F) HDT and autologous PBSCT was analyzed. FD indicates first diagnosis; pre-MRD, MRD level before HDT and autologous PBSCT; post-MRD, MRD level after HDT and autologous PBSCT.
System stage (P ¼ .01), low b2-microglobulin (median, 2.3 mg/dL versus 3.9 mg/dL; P ¼ .02), and high levels of hemoglobin (median, 11.7 g/dL versus 9.2 g/dL; P ¼ .01). Prognostic Value of Sequential MRD Monitoring in PB After a median follow-up of survivors of 45 months (range, 13 to 80 months) after HDT, 29 of 42 patients (69%) had suffered from relapse, whereas 13 patients (31%) were still in ongoing remission. Sequential monitoring of clonotypic cells in PB showed that in 19 of 29 patients (66%) with progressive disease, a loss of molecular remission or an increase of the 2IgH/b-actin ratio of more than 1 log step could be observed (Figure 4). In these patients, the rise in circulating clone-specific cells occurred in median 4 months (range, .8 to 13 months) before the relapse was diagnosed by
conventional diagnostic tests on the basis of the EBMT criteria. In 10 patients negative results in PB could be documented until the time of relapse. Looking at the 13 patients in ongoing remission, the duration of remission was in median 45 months (range, 18 to 78 months). The MRD results were continuously negative in 8 of these patients, whereas in 5 patients single positive results could not be confirmed in following samples. Next, we investigated whether circulating tumor cells detected before serological relapse were associated with a poor prognosis. Therefore, we looked at all patients stratified according to whether or not they showed a molecular relapse before serological relapse. The median OS from the time of first clinical relapse was 17 months and not reached, respectively (P ¼ .02, Figure 5).
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
1113
Figure 4. Sequential monitoring of MRD in PB. MRD levels in consecutive PB samples are shown for 29 patients with relapse after HDT and autologous PBSCT as a function of time after HDT. One log steps of the 2IgH/b-actin ratio are depicted by circles with different color fills. In 19 patients (66%) an increase of the 2IgH/b-actin ratio of more than 1 log step (arrows) could be detected in PB before clinical disease progression was observed.
DISCUSSION To date, monitoring of MRD by IgH-PCR or flow cytometry is not a standard diagnostic procedure in patients with MM [9], but there is growing evidence that the measurement of the level of residual tumor cells in the BM may help to identify patients at different risk of relapse [12,16,18,25]. Although MRD assessment in BM is a valuable tool to evaluate treatment outcome, the role of MRD in the PB of patients with MM is currently not clear. In this study, we present data of 42 patients with MM undergoing HDT and autologous PBSCT who were monitored for MRD in the PB by IgH-qPCR at different time points during the course of disease. The absence of detectable MRD levels in PB at the time of first diagnosis as well as 3 months after HDT was associated with a prolonged EFS and OS. Moreover, by sequentially measuring residual clonotypic cells in samples collected on a monthly basis, relapse could be identified in median 4 months earlier than with conventional diagnostic methods. Further, the appearance of clonotypic cells circulating in PB before the diagnosis of a clinical relapse was also associated with a shorter OS. When addressing the role of MRD in PB, the first essential question is whether there is a difference between MRD in BM and PB. Our comparison of both compartments showed a great difference in the level of clonotypic cells with a variation of the MRD ratio between BM and PB from less than 1 up to approximately 5000-fold without any pattern
according to sampling time point or remission status. Therefore, the MRD level in PB is not only a mere reflection of the MRD level in BM; as a result, in patients with MM it is worth looking at MRD in both compartments. The second question is whether MRD levels in PB at a single time point are of prognostic relevance. Using flow cytometry, it has been shown that circulating myeloma cells in the PB at a given time point are associated with a poor outcome of patients undergoing HDT and autologous PBSCT. This is true for the time of first diagnosis [27], the time before [28], and 3 months after [29] HDT and autologous transplantation. Using IgH-PCR to detect MRD in PB, the published patient numbers range from 6 to 20 [19-23]. In our study we report results on molecular MRD analysis in 42 patients using IgH-qPCR. In line with the published data acquired by FACS analysis [27-29], our data confirm the poor prognosis of circulating myeloma cells at the time of first diagnosis and also confirm the prognostic benefit of a response on a molecular level in PB after HDT and autologous PBSCT in terms of a prolonged EFS or OS. The absence or presence of detectable myeloma cells in PB at this time point also correlated with known prognostic factors such as International Staging System stage and serum levels of lactate dehydrogenase and b2-microglobulin, although there was no correlation with the remission status as defined by the EBMT criteria. Regarding the fact that the number of clonotypic cells in PB do not correlate with the number in BM, it is tempting to
1114
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
Molecular relapse = serological relapse (n = 10) Molecular prior to serological relapse (n = 19) n.r. vs 17 months, P = .02
Proportion of patients (%)
100
monitoring of MRD can help to detect disease progression earlier than conventional diagnostic methods. Even though the clinical utility is limited, evaluation of MRD in PB by IgHqPCR may help us to better understand the pathophysiology of disease recurrence during therapy with investigational agents in the future.
OS 80 60 40 20 0 0
12
24
36
48
60
72
84
OS after serological relapse (months) Figure 5. Comparison of the survival of patients with a molecular relapse before clinical relapse. The OS was calculated as time from clinical relapse and analyzed by Kaplan-Meier plots and the log rank test.
ask if these 2 measures reflect qualitatively different aspects of MRD. Because of relatively low patient numbers with paired PB and BM samples, we were not able to finally clarify if MRD levels in PB and BM are of prognostic relevance in a complementary manner. The third question addresses the impact of MRD not only at a single time point but as a sequential monitoring of dynamic changes of the number of residual tumor cells during times of clinical remission. In this field, FACS data are rare, whereas IgH-PCR studies [23,30,31] with small patient numbers (n ¼ 1 to 6) showed that an increase in tumor load could be detected in PB before disease progression was proven by conventional means. We confirm these data in our group of 42 patients, where a molecular relapse could be observed in 66% of 29 patients with progressive disease in median 4 months before the clinical relapse was diagnosed. Moreover, we show that patients with a molecular relapse preceding clinical relapse have a very short survival after disease progression. This interesting finding hints at a role of these clonotypic cells in spreading the disease, which should be further investigated. It will also be of interest to see whether new agents may be effective in overcoming the bad prognosis associated with these circulating tumor cells. On the other hand, the clinical relevance of this finding is relatively small, because most physicians will not treat patients with MM due to a molecular relapse, in the same way they do not treat patients with relapse from CR unless the clinical relapse is associated with increasing disease activity or end-organ damage [32]. In combination with the fact that IgH-PCR can only be applied to about 75% of patients due to technical reasons [14,16,25], we do not see a great potential in the future of MRD detection in the PB as a routine diagnostic tool for all patients with MM. However, IgH-qPCR may be a very suitable tool to evaluate therapy response to investigational drugs, whose mechanism of action often are not completely understood, and to define the origin of drugresistant tumor cells. In conclusion, molecular remission in PB is a prognostic parameter for a prolonged EFS and OS, and monthly
ACKNOWLEDGMENTS Financial disclosure: Supported by Deutsche Krebshilfe (Bonn, Germany), Leukämie Liga e.V. (Duesseldorf, Germany), and Novartis. Conflict of interest statement: G.K. was a consultant to Amgen and Celgene; received lecture fees from Amgen, Celgene, and Janssen-Cilag; and received research funds from Amgen and Celgene. R.F. was a consultant to Celgene, received lecture fees from Celgene and Janssen-Cilag, and received research funds from Novartis and Celgene. Authorship statement: R.F. designed research, collected data, performed statistical analysis, interpreted data, and drafted the manuscript. M.K. performed the experiments and statistical analysis, interpreted data, and drafted the manuscript. N.S. performed the experiments, collected data, and drafted the manuscript. R.K. designed research and contributed substantially to the development of the methodology. T.S. and T.S. were involved in patient recruitment and correlation of clinical data with experimental findings. G.K. was significantly involved in patient recruitment and correlation of clinical data with experimental findings. R.H. was significantly involved in the induction and conception of the study and interpreted data. All authors critically revised the manuscript. REFERENCES 1. Attal M, Harousseau JL, Stoppa AM, et al. A prospective, randomized trial of autologous bone marrow transplantation and chemotherapy in multiple myeloma. Intergroupe Francais du Myelome. N Engl J Med. 1996;335:91-97. 2. Haas R, Bruns I, Kobbe G, Fenk R. High-dose therapy and autologous peripheral blood stem cell transplantation in patients with multiple myeloma. Rec Res Cancer Res. 2011;183:207-238. 3. Fenk R, Haas R, Kronenwett R. Molecular monitoring of minimal residual disease in patients with multiple myeloma. Hematology. 2004; 9:17-33. 4. Szczepanski T, Orfao A, van der Velden, et al. Minimal residual disease in leukaemia patients. Lancet Oncol 2001 Jul;2:409-417. 5. Harousseau JL, Attal M, Avet-Loiseau H. The role of complete response in multiple myeloma. Blood. 2009;114:3139-3146. 6. Richardson PG, Sonneveld P, Schuster MW, et al. Bortezomib or highdose dexamethasone for relapsed multiple myeloma. N Engl J Med. 2005;352:2487-2498. 7. Dimopoulos M, Spencer A, Attal M, et al. Lenalidomide plus dexamethasone for relapsed or refractory multiple myeloma. N Engl J Med. 2007;357:2123-2132. 8. Weber DM, Chen C, Niesvizky R, et al. Lenalidomide plus dexamethasone for relapsed multiple myeloma in North America. N Engl J Med. 2007;357:2133-2142. 9. Hart AJ, Jagasia MH, Kim AS, et al. Minimal residual disease in myeloma: Are we there yet? Biol Blood Marrow Transplant. 2012;18: 1790-1799. 10. Bruggemann M, Schrauder A, Raff T, et al. Standardized MRD quantification in European ALL trials: Proceedings of the Second International Symposium on MRD assessment in Kiel, Germany, 18-20 September 2008. Leukemia. 2010;24:521-535. 11. Neumann F, Markett J, Fenk R, et al. Therapy adapted to molecular response in patients with chronic myelogenous leukaemia in first chronic phase: results of the Duesseldorf study. Hematol Oncol. 2008; 26:213-218. 12. Paiva B, Vidriales MB, Cervero J, et al. Multiparameter flow cytometric remission is the most relevant prognostic factor for multiple myeloma patients who undergo autologous stem cell transplantation. Blood. 2008;112:4017-4023. 13. Paiva B, Gutierrez NC, Rosinol L, et al. High-risk cytogenetics and persistent minimal residual disease by multiparameter flow cytometry
M. Korthals et al. / Biol Blood Marrow Transplant 19 (2013) 1109e1115
14.
15.
16.
17.
18.
19.
20.
21.
22.
predict unsustained complete response after autologous stem cell transplantation in multiple myeloma. Blood. 2012;119:687-691. Sarasquete ME, Garcia-Sanz R, Gonzalez D, et al. Minimal residual disease monitoring in multiple myeloma: a comparison between allelic-specific oligonucleotide real-time quantitative polymerase chain reaction and flow cytometry. Haematologica. 2005;90:1365-1372. Bakkus MH, Bouko Y, Samson D, et al. Post-transplantation tumour load in bone marrow, as assessed by quantitative ASO-PCR, is a prognostic parameter in multiple myeloma. Br J Haematol. 2004;126: 665-674. Korthals M, Sehnke N, Kronenwett R, et al. The level of minimal residual disease in the bone marrow of patients with multiple myeloma before high-dose therapy and autologous blood stem cell transplantation is an independent predictive parameter. Biol Blood Marrow Transplant. 2012;18:423-431. Putkonen M, Kairisto V, Juvonen V, et al. Depth of response assessed by quantitative ASO-PCR predicts the outcome after stem cell transplantation in multiple myeloma. Eur J Haematol. 2010;85:416-423. Corradini P, Cavo M, Lokhorst H, et al. Molecular remission after myeloablative allogeneic stem cell transplantation predicts a better relapse-free survival in patients with multiple myeloma. Blood. 2003; 102:1927-1929. Billadeau D, Prosper F, Verfaillie C, et al. Sequential analysis of bone marrow and peripheral blood after stem cell transplant for myeloma shows disparate tumor involvement. Leukemia. 1997;11:1565-1570. Cremer FW, Ehrbrecht E, Kiel K, et al. Evaluation of the kinetics of the bone marrow tumor load in the course of sequential high-dose therapy assessed by quantitative PCR as a predictive parameter in patients with multiple myeloma. Bone Marrow Transplant. 2000;26:851-858. Kiel K, Cremer FW, Rottenburger C, et al. Analysis of circulating tumor cells in patients with multiple myeloma during the course of high-dose therapy with peripheral blood stem cell transplantation. Bone Marrow Transplant. 1999;23:1019-1027. Lipinski E, Cremer FW, Ho AD, et al. Molecular monitoring of the tumor load predicts progressive disease in patients with multiple myeloma
23.
24.
25.
26.
27.
28.
29.
30. 31.
32.
1115
after high-dose therapy with autologous peripheral blood stem cell transplantation. Bone Marrow Transplant. 2001;28:957-962. Raab MS, Cremer FW, Breitkreutz IN, et al. Molecular monitoring of tumour load kinetics predicts disease progression after nonmyeloablative allogeneic stem cell transplantation in multiple myeloma. Ann Oncol. 2005;16:611-617. Blade J, San Miguel JF, Fontanillas M, et al. Survival of multiple myeloma patients who are potential candidates for early high-dose therapy intensification/ autotransplantation and who were conventionally treated. J Clin Oncol. 1996;14:2167-2173. Fenk R, Ak M, Kobbe G, et al. Levels of minimal residual disease detected by quantitative molecular monitoring herald relapse in patients with multiple myeloma. Haematologica. 2004;89:557-566. van der Velden VH, Cazzaniga G, Schrauder A, et al. Analysis of minimal residual disease by Ig/TCR gene rearrangements: guidelines for interpretation of real-time quantitative PCR data. Leukemia. 2007;21: 604-611. Nowakowski GS, Witzig TE, Dingli D, et al. Circulating plasma cells detected by flow cytometry as a predictor of survival in 302 patients with newly diagnosed multiple myeloma. Blood. 2005;106:2276-2279. Dingli D, Nowakowski GS, Dispenzieri A, et al. Flow cytometric detection of circulating myeloma cells before transplantation in patients with multiple myeloma: A simple risk stratification system. Blood. 2006;107:3384-3388. Liu H, Yuan C, Heinerich J, et al. Flow cytometric minimal residual disease monitoring in patients with multiple myeloma undergoing autologous stem cell transplantation: A retrospective study. Leuk Lymph. 2008;49:306-314. Cooke F, Bakkus M, Thielemans K, et al. Use of quantitative ASO-PCR to predict relapse in multiple myeloma. Br J Haematol. 1999;105:317-319. Rasmussen T, Hansson L, Osterborg A, et al. Idiotype vaccination in multiple myeloma induced a reduction of circulating clonal tumor B cells. Blood. 2003;101:4607-4610. Durie BG, Salmon SE. Cellular kinetics staging, and immunoglobulin synthesis in multiple myeloma. Annu Rev Med. 1975;26:283-288.