This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van Rhenen, A. Articles by Schuurhuis, G. J. Search for Related Content PubMed PubMed Citation Articles by van Rhenen, A. Articles by Schuurhuis, G. J.

High Stem Cell Frequency in Acute Myeloid Leukemia at Diagnosis Predicts High Minimal Residual Disease and Poor Survival

Authors' Affiliations: ¹ Department of Hematology, VU University Medical Center, Amsterdam, the Netherlands and ² Department of Hematology, Erasmus Medical Center, Rotterdam, the Netherlands

Requests for reprints: Gerrit Jan Schuurhuis, Department of Hematology, VU University Medical Center, brug 240, De Boelelaan 1117, 1081 HV Amsterdam, the Netherlands. Phone: 31-20-444-2604; Fax: 31-20-444-2601; E-mail: gj.schuurhuis{at}vumc.nl.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Purpose: In CD34-positive acute myeloid leukemia (AML), the leukemia-initiating event originates from the CD34⁺CD38^– stem cell compartment. Survival of these cells after chemotherapy may lead to minimal residual disease (MRD) and subsequently to relapse. Therefore, the prognostic impact of stem cell frequency in CD34-positive AML was investigated.

Experimental Design: First, the leukemogenic potential of unpurified CD34⁺CD38^– cells, present among other cells, was investigated in vivo using nonobese diabetic/severe combined immunodeficient mice transplantation experiments. Second, we analyzed whether the CD34⁺CD38^– compartment at diagnosis correlates with MRD frequency after chemotherapy and clinical outcome in 92 AML patients.

Results: In vivo data showed that engraftment of AML blasts in nonobese diabetic/severe combined immunodeficient mice directly correlated with stem cell frequency of the graft. In patients, a high percentage of CD34⁺CD38^– stem cells at diagnosis significantly correlated with a high MRD frequency, especially after the third course of chemotherapy. Also, it directly correlated with poor survival. In contrast, total CD34⁺ percentage showed no such correlations.

Conclusions: Both in vivo data, as well as the correlation studies, show that AML stem cell frequency at diagnosis offers a new prognostic factor. From our data, it is tempting to hypothesize that a large CD34⁺CD38^– population at diagnosis reflects a higher percentage of chemotherapy-resistant cells that will lead to the outgrowth of MRD, thereby affecting clinical outcome. Ultimately, future therapies should be directed toward malignant stem cells.

Relapse of AML is thought to originate from the outgrowth of persisting leukemic cells, called minimal residual disease (MRD; refs. 2–5). MRD can be detected in the majority of patients in (morphologic) CR using an immunophenotypical approach. To this end, leukemia-associated phenotypes are established at diagnosis, which can subsequently be used to identify MRD cells. The frequency of MRD cells persisting after chemotherapy has been found to have strong prognostic impact (2–5). It is thought that MRD cells resemble blasts at diagnosis (2) that survived chemotherapy and, therefore, likely originate from the stem cell compartment.

In CD34-positive AML, the CD34⁺ leukemic stem cell is characterized by the absence of CD38 (6, 7), irrespective of French-American-British type (FAB), although there is debate on FAB M3 (8). It has been hypothesized that the CD34⁺CD38^– compartment contains the hematopoietic cells in which the leukemia-initiating event has occurred that results in AML (9). CD34 expression, including both primitive (CD38^–) and more differentiated (CD38⁺) cells, has been proposed as a prognostic factor, although conflicting results have been described (reviewed in ref. 10).

These observations led us to hypothesize that a high frequency of CD34⁺CD38^– cells at diagnosis should result in a high MRD frequency after chemotherapy. Thereby, as MRD frequency itself has been shown to have strong prognostic impact, stem cell frequency at diagnosis should have direct prognostic impact.

To elucidate this, the frequency of CD34⁺CD38^– cells in AML patients at diagnosis was studied in 92 patients and correlated with MRD frequency after chemotherapy during CR.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Patient samples at diagnosis. Bone marrow samples from patients with AML were obtained at the VU University Medical Center after informed consent at diagnosis and after chemotherapeutic treatment. In 16 cases at diagnosis, bone marrow samples were not available and peripheral blood was used. Diagnosis of patients was based on morphology using the FAB classification, immunophenotyping, and cytogenetics (11). Patient characteristics are shown in Table 1. Cells were analyzed fresh (n = 56) or after storage in liquid nitrogen (n = 36). In fresh samples, for immunophenotypical analysis, RBCs were lysed using Pharm lyse (BD Biosciences, San Jose, CA) and washed twice with PBS with 0.1% bovine serum albumin added. Frozen samples were prepared using a Ficoll gradient (1.077 g/mL; Amersham Biosciences, Freiburg, Germany) and a 10-minute RBC lysis on ice using 10 mL lysis buffer [155 mmol/L NH₄Cl, 10 mmol/L KHCO₃, 0.1 mmol/L Na₂EDTA (pH 7.4)]. Cells were then frozen in RPMI (Life Technologies, Paisley, United Kingdom) with 20% heat-inactivated fetal bovine serum (Greiner, Alphen a/d Rijn, the Netherlands) and 10% DMSO (Riedel-de Haen, Seelze, Germany) in isopropyl alcohol–filled containers and subsequently stored in liquid nitrogen. When needed for analysis, cells were thawed and suspended in RPMI with 40% fetal bovine serum at 37°C. Cells were washed and enabled to recover for 45 minutes in RPMI with 40% fetal bovine serum at 37°C. Cells were washed and suspended in PBS with 0.1% bovine serum albumin. The different treatment of samples, i.e., lysis alone for fresh samples or Ficoll isolation combined with lysis and freeze-thawing, likely affects only mature cells and not the blast compartment. Procedures were validated for both the use of fresh and frozen material. Therefore, six samples, two with high stem cell frequency, one median stem cell frequency, and three with low stem cell frequency, were analyzed fresh and after cryopreservation. No significant differences were found between fresh and frozen samples using Wilcoxon statistics (P = 0.3 for CD38^–/CD34⁺). Also, both the median values for stem cell frequency and the range were quite similar in fresh (n = 56) and frozen (n = 36) samples, for CD38^–/CD34⁺: 3.2% (0.1-82.5%) and 3.7% (0.1-95.1%), respectively. Therefore, results obtained by both procedures were combined.

The HOVON 29 study was followed by the HOVON 42 study. The first cycle of both studies consisted of idarubicin and cytarabine. In the HOVON 42, patients were randomized to standard-dose or high-dose cytarabine. The second course consisted of amsacrin and cytarabine; again, for the latter, patients were randomized to different doses. Patients >60 years were treated according to HOVON 32 and 43 protocols. Induction cycles I and II of HOVON 32 consisted of cytarabine and granulocyte colony-stimulating factor with or without fludarabin. HOVON 43 started with an induction cycle containing cytarabine and low- or high-dose daunorubicin followed by a second-induction cycle with cytarabine. Patients under 60 years of age received either allogeneic (n = 15) or autologous (n = 7) stem cell transplantation or a course of intensive chemotherapy (n = 20) as postremission treatment.

Overall survival was calculated from date of diagnosis until date of death. Relapse-free survival was calculated from date of CR until date of death or relapse, with relapse defined as event. Finally, disease-free survival was calculated from date of CR until date of first event (i.e., death or relapse).

Cytogenetics and Flt3 internal tandem duplications. Patients were classified in three risk groups for cytogenetic abnormalities according to Grimwade et al. (12). Favorable risk was defined by the presence of t(15;17) or t(8;21) or inv(16), and poor risk was defined by the presence of either five unrelated abnormalities, or monosomy 5 or 7, or abnormalities of the long arm of chromosome 5, or inv(3) (q21:q26). Patients who did not meet the criteria for poor or favorable risk were classified as intermediate risk.

FLT3 mutant allelic expression was examined by reverse transcription-PCR with R5 and R6 primers previously described by Nakao et al. (13). This primer set covers the transmembrane and juxtamembrane domains of the FLT3 gene. PCR products were visualized by gel electrophoresis.

Transplantation of acute myeloid leukemia blasts in nonobese diabetic/severe combined immunodeficient mice. Engraftment of diagnosis AML blasts in nonobese diabetic/severe combined immunodeficient (NOD/SCID) mice was studied, in relation to the CD34⁺CD38^– and CD34⁺ frequency of the transplant. Samples were cryopreserved, after Ficoll isolation and T-cell depletion using S-adenosyl-L-ethionine–treated sheep RBC. These procedures resulted in fractions with very high percentages of blasts. Samples were obtained from patients with M₀ (n = 2), M₁ (n = 2), M₂ (n = 5), M₄ (n = 5), and M₅ (n = 7). Before transplantation, CD34 and CD38 expression were determined using flow cytometry. After total body irradiation of the mice with 3.5 Gy -irradiation, 3 x 10⁷ AML blasts, containing the total purified blast population, were injected in the lateral tail vein (14). The mice were killed after 30 days in accordance with the institutional animal research regulations. Chimerism was determined using flow cytometric detection of human CD45 expression combined with CD34, CD33, and CD38 expression. Criteria were defined for both chimerism and CD34 percentage. Chimerism should be >1% because the presence of AML blasts in the bone marrow of mice with chimerism <1% can be explained by simple sustenance of cells that homed to the bone marrow (15). Moreover, CD34 percentage had to be >1%, because in samples with a CD34 percentage <1%, the CD34⁺ cells are, in general, of normal origin (16). These criteria resulted in the exclusion of three samples because of lack of sufficient chimerism.

CD34⁺CD38^– frequency assessment using fluorescence-activated cell sorting analysis. Cells, prepared in PBS containing 0.1% bovine serum albumin, were incubated with monoclonal antibodies (mAb) for 15 minutes at room temperature, washed once in PBS containing 0.1% bovine serum albumin, and analyzed on the flow cytometer. mAb combinations are shown in Table 2. Combinations consisted of FITC-, peridinyl chlorophyllin–, or allophycocyanin-labeled mAbs. Anti-CD34 FITC, anti-CD34 peridinyl chlorophyllin, anti-CD45 peridinyl chlorophyllin, anti-CD45 allophycocyanin, anti-CD38 allophycocyanin, and Via-Probe (7-amino-actinomycin D) were all from BD Biosciences; anti-CD34 FITC was in part of the samples from Immunotech (Marseilles, France); and Annexin V FITC was from Nexins Research (Kattendijke, the Netherlands). Although different clones, both anti-CD34 FITC were IgG1 and class III. It is known that mAbs specific for class III CD34 epitopes detect similar numbers of CD34⁺ cells in large arrays of specimens and, therefore, results could be combined (17). When frozen-thawed cells were used, Annexin V FITC was included in the majority of cases to gate out dead/apoptotic cells in the tube, which further contained anti-CD34 peridinyl chlorophyllin and anti-CD45 allophycocyanin. In a second tube containing Annexin V FITC, anti-CD34 peridinyl chlorophyllin, and anti-CD38 allophycocyanin, these apoptotic/dead cells were then gated out before stem cell frequency assessment. In the remaining part, a sensitive approach was applied using Syto16 (Molecular Probes, Eugene, OR) together with 7-amino-actinomycin D (18) in one tube, which enabled to identify apoptotic/dead cells based on low forward and sideward scatter properties. In a second tube containing anti-CD34 FITC, anti-CD38 allophycocyanin, and anti-CD45 peridinyl chlorophyllin, these apoptotic/dead cells were then gated out before stem cell frequency assessment. Data acquisition was done using a FACScalibur (BD Biosciences) equipped with an argon and red diode laser and analysis was done using Cellquest software (BD Biosciences).

View larger version (47K): [in this window] [in a new window]   Fig. 1. Gating strategy in diagnosis AML to identify CD34+CD38– cells. A, after labeling of diagnosis AML cells with the appropriate antibody combinations, the CD34+CD38– cells were identified by a Boolean gating strategy. Gating on blasts characterized by CD45dim/low side scatter (SSC). B, gating of the blasts within the gate defined by forward scatter (FSC) and side scatter to identify a population that is roughly homogeneous for scatter properties. C, cells from the forward scatter/side scatter plot defined in (B) are shown in (C) in a plot defined by CD34 and CD38 expression. The CD38– population was defined using PBS as a negative control. D, the CD34+CD38– population defined in (C) is back gated in a forward scatter/side scatter plot to identify a CD34+CD38– population with homogeneous scatter properties. The frequency of the thus determined CD34+CD38– population was used in this study.

Minimal residual disease detection. At diagnosis, leukemia-associated phenotypes were established according to the methods described by us before (2, 21). In follow-up samples, 1 x 10⁶ bone marrow cells were incubated with mAb combinations that defined the leukemia-associated phenotype (2, 21). MRD was investigated in a situation of full bone marrow recovery after first and second course of chemotherapy and within 6 months after the third course of chemotherapy as previously described (2, 21). RBCs were lysed using fluorescence-activated cell sorting lysing solution (BD Biosciences); cells were washed twice in PBS with 0.1% bovine serum albumin, labeled with appropriate mAb combinations, washed again, and measured by flow cytometry. PBS was used as a negative control because isotype controls offered the same results in MRD analysis (2).

Statistical analyses. Statistical analysis of the data was done using the SPSS 9.0 software package (SPSS, Chicago, IL).

Cox regression analysis was done for initial screening of the correlations between both CD34⁺ and CD34⁺CD38^– expression and patient prognosis and for determining the prognostic significance of WBC count and age. Kaplan-Meier analysis was used to study the prognostic impact of Flt3-ITD, cytogenetics, and CR rate and to study the effect of cutoff values for CD38^–/CD34⁺ and CD38^–/CD45dim on both MRD frequency and clinical outcome in our patient group. The cutoff value for CD38^–/CD45dim or CD38^–/CD34⁺ was chosen at a percentage that resulted in the largest difference in survival between the two groups defined by that cutoff. The log-rank test was used to validate equality of the survival distributions.

To obtain cutoff values, part of the samples was used as a training set (n = 60) and the most significant cutoff value was chosen. The thus obtained cutoff values were subsequently used as cutoff values for the whole set (n = 92) of samples.

Multivariate analysis, using Cox regression analysis, was used for the correlation of stem cell frequencies with survival parameters to test the independence of known prognostic risk factors. Missing value analysis was done for Flt3ITD and validated using Little's test for Missing Completely at Random. Using this approach, multivariate analysis was possible on the total group.

The Mann-Whitney U test for nonparametric samples was used to correlate CD38^–/CD34⁺ and CD38^–/CD45dim with MRD frequency after different cycles of chemotherapy. Spearman correlation analysis was used for the correlation between the frequency of CD34⁺ or CD38^–/CD34⁺ and chimerism determined in NOD/SCID mice.

P < 0.05 was considered significant.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Nature and biological properties of CD34⁺CD38^– cells. To establish the stem cell character of CD34⁺CD38^– cells, the thus defined CD34⁺CD38^– population was purified using fluorescence-activated cell sorting and examined in long-term culture assays (22). The CD34⁺CD38^– showed long-term clonogenic capacity in contrast to the CD34⁺CD38⁺ population (data not shown). The gating strategy that was used is shown in Fig. 1 and described in Materials and Methods.

To establish the nature (leukemic or normal) of the CD34⁺CD38^– stem cell compartment, CD123 expression (known to be present on leukemic stem cells but not on normal stem cells; ref. 20) was established in 36 consecutive cases. In 35 of 36 cases, all stem cells were positive, according to the criteria described under Materials and Methods, confirming the leukemic nature of the CD34⁺CD38^– cells. In the remaining CD123-negative sample, the leukemic nature of the CD34⁺CD38^– cells was confirmed using functional, P-glycoprotein activity (16), and phenotypical characteristics (i.e., the presence of a leukemia-associated phenotype on the CD34⁺CD38^– cells). P-glycoprotein activity of the CD34⁺CD38^– cells was 1.3, which is below the reference range of 1.7 to 3.7. The leukemia-associated phenotype of this patient (aberrant expression of CD56, a well-known aberrant phenotype; refs. 2–5) was also present on the CD34⁺CD38^– cells. To conclude, both functional and phenotypical characteristics confirmed the malignant nature of the CD34⁺CD38^– cells in this patient. Control bone marrow CD34⁺CD38^– cells (n = 5) were all CD123 negative.

We investigated engraftment of 21 AML samples. Stem cell frequencies were determined before transplantation, using flow cytometry. A fixed number of cells, representing the total blast population, were injected. Chimerism in NOD/SCID mice turned out to be strongly dependent on the stem cell frequency in the graft, with P < 0.001 for CD38^–/CD34⁺ (n = 21, R = 0.701; Fig. 2A) and P = 0.007 for CD38^–/CD45dim (n = 21, R = 0.57; data not shown), while engraftment was completely independent of total CD34 frequency (Fig. 2B). This indicates that the leukemia-initiating cell is present within the CD34⁺CD38^– compartment and, in this study, present among other cells, gives rise to leukemia in a dose-dependent way.

View larger version (12K): [in this window] [in a new window]   Fig. 2. Correlation between stem cell frequency of human AML blasts at diagnosis and chimerism in NOD/SCID mice. Equal numbers of AML blasts from 21 diagnosis samples were injected into NOD/SCID mice as described in Materials and Methods. A, engraftment (depicted as the percentage of chimerism, which is a measure for the engraftment of human AML cells in a NOD/SCID mouse) is dependent on the stem cell frequency (CD38–/CD34+) in the AML bone marrow sample at diagnosis. B, engraftment is not related to the absolute number of injected CD34+ cells.

First, stem cell frequencies, defined either as CD38^–/CD34⁺ or CD38^–/CD45dim (for definitions, see Materials and Methods), were determined in a training set of samples of 60 newly diagnosed patients. A significant correlation was found between CD38^–/CD34⁺ and all survival parameters (overall survival overall survival, relapse-free survival, and disease-free survival) and between CD38^–/45dim and relapse-free survival and disease-free survival (Cox regression analysis). The most significant cutoff values, determined to identify two groups with the largest difference in survival, were 7.5% for CD38^–/CD34⁺ and 3.5% for CD38^–/CD45dim for all survival parameters (data not shown).

Next, the same correlations were examined for the complete group of 92 patients. CD38^–/CD34⁺ correlated best with all survival parameters, overall survival (n = 92, P = 0.02), disease-free survival (n = 60, P = 0.06), and relapse-free survival (n = 60, P = 0.005) using Cox regression analysis. Figure 3A shows the results for CD38^–/CD34⁺ and relapse-free survival for these 60 patients who fulfilled criteria for relapse-free survival in the total group of 92 patients. Using the cutoff values defined in the training set, results were best for CD38^–/CD45dim. Figure 3B shows results for CD38^–/CD45dim and relapse-free survival in a Kaplan-Meier plot, using the most significant cutoff value of 3.5% determined in the training set. Due to the particular shape of the Cox regression curves, like that shown in Fig. 3A, cutoff points are not based on an objective median value. However, it was nevertheless checked whether the most significant cutoff value was not the only significant P value. It is clear that only cutoff values to the right of the optimal cutoff value should be tested for this. Therefore, other cutoff values, to the right of 3.5% (for CD38^–/CD45dim), were tested for significance and this was found up to 30%. To conclude, high stem cell frequency was found to correlate with poor survival.

View larger version (14K): [in this window] [in a new window]   Fig. 3. Correlation between stem cell frequency and relapse-free survival (RFS). A, relation between the stem cell frequency CD38–/CD34+ and relapse-free survival (in months) for all 60 patients of the total group of 92 patients, who fulfilled criteria for relapse-free survival. Cox regression analysis showed a significant correlation (P = 0.005). B, a Kaplan-Meier plot of the same patients but now CD38–/CD45dim with relapse-free survival. The cutoff used is 3.5%. The patients with a high stem cell frequency at diagnosis (>3.5%) had a median relapse-free survival of 5.6 months (n = 15). This is in strong contrast to the patients with a low stem cell frequency who had a median relapse-free survival of 16.0 months (n = 45). This difference was significant using log-rank statistics (P = 0.02).

Finally, CR rate was analyzed in the groups defined by low and high stem cell frequency. Only a trend toward high stem cell frequency in the non-CR group was apparent (for CD38^–/CD34⁺, P = 0.20, n = 84). CR rate showed no correlation with total CD34 expression (P = 0.14, n = 84).

Risk factors, such as WBC count at diagnosis, age, CR rate, and Flt3-ITD status were found to correlate with overall survival, showing that our patient cohort is representative for AML patients (not shown). Cytogenetics showed no prognostic significance in our study, probably due to small sample size in the poor and favorable group (Table 1).

In a multivariate analysis, results for relapse-free survival with CD38^–/CD34⁺ (P = 0.004) were independent of WBC, age, and Flt3ITD (P > 0.3 for all variables). Results for disease-free survival with CD38^–/CD34⁺ (P = 0.05) were also independent of WBC, age, and Flt3ITD (P > 0.2 for all variables).

Because CD123 expression is high on CD34⁺CD38^– cells in 35 of 36 cases, correlations with survival were almost similar (not shown).

To conclude, a high stem cell frequency at diagnosis correlates with poor survival, independent of known prognostic risk factors.

Correlation of CD34⁺CD38^– frequency with minimal residual disease frequency. Minimal residual disease (MRD) frequency, as detected after different cycles of chemotherapy, is a strong independent prognostic risk factor for survival, in particular after the third cycle of chemotherapy (2). In the present study, this could be confirmed using the same cutoff values as reported in the study we did before (data not shown; ref. 2).

Therefore, the prognostic impact on survival of stem cell frequency at diagnosis might be mediated by its direct correlation with MRD frequency after chemotherapy. This analysis was done using the same cutoff values defined in our correlation with survival parameters (3.5% for CD38^–/CD34⁺ and 7.5% for CD38^–/CD45dim). Indeed, stem cell frequency at diagnosis was found to correlate with MRD frequency after the third cycle of chemotherapy in CR, with borderline significance for CD38^–/CD34⁺ (P = 0.05; Fig. 4A) and significance for CD38^–/CD45dim (P = 0.03; Fig. 4B). No correlations were found between stem cell frequency at diagnosis and MRD frequency after the first course of chemotherapy. After the second cycle, a trend toward a higher MRD frequency in the group defined by a high stem cell frequency at diagnosis was observed. No correlations were found between total CD34 expression and MRD frequency (P > 0.3 for all three cycles).

View larger version (13K): [in this window] [in a new window]   Fig. 4. Correlations between stem cell frequency at diagnosis and MRD frequency after third cycle of chemotherapy. A, patients were grouped according to the CD38–/CD34+ stem cell frequency of <7.5% and 7.5%, for which corresponding log-transformed MRD frequency after third cycle of chemotherapy is depicted (n = 22, P = 0.05). B, patients were grouped according to the CD38–/CD45dim stem cell frequency of <3.5% and 3.5%, for which corresponding log-transformed MRD frequency after third cycle of chemotherapy is depicted as well (n = 22, P = 0.03). The box-and-whisker plots show in the box the median and the 25th and 75th percentile, while the whiskers show the 2.5th and 97.5th percentile.

	Discussion

Top Abstract Materials and Methods Results Discussion References

Previously, using purified subpopulations, it has been established that SCID leukemia-initiating cells in CD34-positive AML are contained within the primitive CD34⁺CD38^– fraction (6). Although the issue was not specifically addressed, Monaco et al. (24) found, when using purified CD34⁺ blasts, the highest engraftment in NOD/SCID mice transplanted with samples from two patients with the highest stem cell frequency. In accordance with this, we found that the stem cell frequency in unpurified AML samples strongly correlated with engraftment in the NOD/SCID mouse experiments.

There is evidence that leukemic stem cells, compared with more mature blasts, are difficult to eradicate in patients using conventional chemotherapy, due to their chemotherapy-resistant character (25). This is probably due to the fact that most of these cells are quiescent (in G₀ phase), which has been shown for normal CD34⁺CD38^– cells, when compared with normal CD34⁺CD38⁺ cells (26), and for AML (27). It thus seemed obvious to hypothesize that, with a relatively large number of intrinsically resistant CD34⁺CD38^– cells at diagnosis, the stem cell frequency in remission should be relatively high. Due to the very low frequency of CD34⁺CD38^– cells in remission bone marrow, together with their similarities with normal CD34⁺CD38^– stem cells, the leukemic stem cells are difficult to study in follow-up samples. However, the progeny of these stem cells, called MRD, often have different degrees of differentiation and with immunophenotypical make-up quite comparable with blasts at diagnosis. MRD can be studied more easily, using leukemia-associated phenotype technology: Immunophenotypical aberrances (leukemia-associated phenotype) detected at diagnosis are used to establish MRD frequency, which, in turn, has been shown to have strong prognostic impact (2–5, 28). It turned out that there was no correlation between stem cell frequency at diagnosis with MRD frequency after the first course of therapy. This is probably due to the least predictive value of MRD frequency for survival after first cycle. However, correlations became evident after subsequent courses: A higher stem cell frequency at diagnosis significantly correlated with a higher MRD frequency. The observation that MRD frequency after the second and third course of chemotherapy correlates most strongly with clinical outcome (2) gives further support for a direct causal relationship between stem cell frequency at diagnosis and MRD frequency.

Our observations are thus compatible with a model in which a higher stem cell frequency at diagnosis has no major impact on the CR rate and MRD frequency after one course of chemotherapy. After all, major leukemic cell reduction, and not reduction of the low-frequency stem cell fraction, likely determines CR rate. With additional courses of chemotherapy, the most resistant fractions of the CD34⁺CD38^– compartment will be selected. These resistant CD34⁺CD38^– cells will grow out, resulting in the spectrum of populations, referred to in all studies as MRD. The frequencies of these MRD cells reflect the resistant CD34⁺CD38^– frequencies. Depending on these frequencies, relapse will occur. The hypothesized direct relationship of CD34⁺CD38^– frequency at diagnosis with in vivo outgrowth in patients has been confirmed in NOD/SCID mice experiments, in which CD34⁺CD38^– frequency correlated with leukemogenic potential.

In AML other prognostic factors have been proposed, such as apoptosis and multidrug resistance (29–34). Multidrug resistance at diagnosis as well as apoptosis-related protein profile at diagnosis have been proven to be strong independent prognostic risk factors. If P-glycoprotein activity and Bcl-2 expression are both high in patients with high stem cell frequency, this could provide a molecular basis for our findings, which needs to be elucidated in the future. When defining new prognostic factors for future use in diagnostics, simple assays that can be done in routine settings are required. Assessment of the presence and size of a CD34⁺CD38^– population at diagnosis is easy to incorporate, if not already, in current diagnostic immunophenotypical protocols.

In many studies, the total CD34⁺ fraction is still considered to be the progenitor/stem cell fraction of interest and although the number of CD34⁺ cells has indeed been shown to be a prognostic risk factor in some studies (35, 36); others could not confirm that finding (37–39). In our patient cohort, a complete lack of correlation was found between CD34⁺ frequency at diagnosis on the one hand and CR rate, MRD frequency, and clinical outcome on the other. Similarly, in the NOD/SCID mouse studies, the total number of CD34⁺ cells had no predictive value for leukemogenic outgrowth.

A particular FAB class of leukemia in which the stem cell frequency has been addressed is FAB M₀: These have a higher stem cell frequency and poor prognosis compared with the other FAB types (40, 41). However, we were not able to confirm these data in our study, probably due to only nine patients with AML M₀.

In the future, it would be of great interest to study the frequency of CD34⁺CD38^– cells during follow-up. MRD detection is thus far mainly based on the presence of blasts with primitive markers (CD117 and/or CD133 and/or CD34) combined with aberrant marker expression. In CD34-positive AML, however, such blasts represent, in most cases, a more mature population compared with the CD34⁺CD38^– stem cells. In at least part of the patients, the aberrances are also found on CD34⁺CD38^– cells at diagnosis, but not on normal bone marrow stem cells.³ In principle, this would allow to correlate stem cell frequency at diagnosis with stem cell frequency in CR after chemotherapy. Next, it would be of value to study if the prognostic impact of the MRD stem cell frequency might improve the already strong impact of total MRD frequency on outcome. Apart from leukemia-associated phenotype expression, the leukemia stem cell–selective expression of CD123 (20) and of the recently described C-type lectin–like molecule 1 (42) should be very helpful in this respect.

In CD34– AML, the compartmentalization of stem cells is still unknown. As 20% of AML samples in our clinic are completely CD34 negative, more effort is needed to identify stem cells in CD34-negative AML.

Eradication of malignant stem cells can probably be reached in part of the patients, using conventional chemotherapy, e.g., via the application of repetitive cycles of consolidation therapy (43). However, given the prognostic significance of CD34⁺CD38^– cells in AML and the apparent difficulties in eradicating malignant stem cells from the patient, it will be necessary to develop therapies that specifically target these malignant stem cells after reaching CR. The possibility to detect and characterize stem cells will help to reach this goal. There is emerging evidence that such therapies are needed to offer curative possibilities for patients with AML.

	Footnotes

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

³ A. van Rhenen, A. Kelder, N. Feller, A.B. Bakker, B. Moshver, A.H. Westra, G.J. Ossenkoppele, and G. Schuurhuis, unpublished data.

Received 3/ 2/05; revised 6/10/05; accepted 6/30/05.

	References

Top Abstract Materials and Methods Results Discussion References

 

1. Lowenberg B, Griffin JD, Tallman MS. Acute myeloid leukemia and acute promyelocytic leukemia. Hematology (Am Soc Hematol Educ Program) 2003;82–101.
2. Feller N, van der Pol MA, van Stijn A, et al. MRD parameters using immunophenotypic detection methods are highly reliable in predicting survival in acute myeloid leukaemia. Leukemia 2004;18:1380–90.[CrossRef][Medline]
3. San Miguel JF, Vidriales MB, Lopez-Berges C, et al. Early immunophenotypical evaluation of minimal residual disease in acute myeloid leukemia identifies different patient risk groups and may contribute to postinduction treatment stratification. Blood 2001;98:1746–51.[Abstract/Free Full Text]
4. Venditti A, Buccisano F, Del Poeta G, et al. Level of minimal residual disease after consolidation therapy predicts outcome in acute myeloid leukemia. Blood 2000;96:3948–52.[Abstract/Free Full Text]
5. Kern W, Voskova D, Schoch C, et al. Determination of relapse risk based on assessment of minimal residual disease during complete remission by multiparameter flow cytometry in unselected patients with acute myeloid leukemia. Blood 2004;104:3078–85.[Abstract/Free Full Text]
6. Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 1997;3:730–7.[CrossRef][Medline]
7. Ailles LE, Gerhard B, Kawagoe H, Hogge DE. Growth characteristics of acute myelogenous leukemia progenitors that initiate malignant hematopoiesis in nonobese diabetic/severe combined immunodeficient mice. Blood 1999;94:1761–72.[Abstract/Free Full Text]
8. Grimwade D, Enver T. Acute promyelocytic leukemia: where does it stem from? Leukemia 2004;18:375–84.[CrossRef][Medline]
9. Warner JK, Wang JC, Hope KJ, Jin L, Dick JE. Concepts of human leukemic development. Oncogene 2004;23:7164–77.[CrossRef][Medline]
10. Kanda Y, Hamaki T, Yamamoto R, et al. The clinical significance of CD34 expression in response to therapy of patients with acute myeloid leukemia: an overview of 2483 patients from 22 studies. Cancer 2000;88:2529–33.[CrossRef][Medline]
11. Lowenberg B, Downing JR, Burnett A. Acute myeloid leukemia. N Engl J Med 1999;341:1051–62.[Free Full Text]
12. Grimwade D, Walker H, Harrison G, et al. The predictive value of hierarchical cytogenetic classification in older adults with acute myeloid leukemia (AML): analysis of 1065 patients entered into the United Kingdom Medical Research Council AML11 trial. Blood 2001;98:1312–20.[Abstract/Free Full Text]
13. Nakao M, Yokota S, Iwai T, et al. Internal tandem duplication of the flt3 gene found in acute myeloid leukemia. Leukemia 1996;10:1911–8.[Medline]
14. Rombouts WJ, Martens AC, Ploemacher RE. Identification of variables determining the engraftment potential of human acute myeloid leukemia in the immunodeficient NOD/SCID human chimera model. Leukemia 2000;14:889–97.[CrossRef][Medline]
15. Verstegen MM, Wognum AW, Wagemaker G. Thrombopoietin is a major limiting factor for selective outgrowth of human umbilical cord blood cells in non-obese diabetic/severe combined immunodeficient recipient mice. Br J Haematol 2003;122:837–46.[CrossRef][Medline]
16. van der Pol MA, Feller N, Roseboom M, et al. Assessment of the normal or leukemic nature of CD34⁺ cells in acute myeloid leukemia with low percentages of CD34 cells. Haematologica 2003;88:983–93.[Abstract/Free Full Text]
17. Gratama JW, Orfao A, Barnett D, et al. Flow cytometric enumeration of CD34⁺ hematopoietic stem and progenitor cells. European Working Group on Clinical Cell Analysis. Cytometry 1998;34:128–42.[CrossRef][Medline]
18. van der Pol MA, Broxterman HJ, Westra G, Ossenkoppele GJ, Schuurhuis GJ. Novel multiparameter flow cytometry assay using Syto16 for the simultaneous detection of early apoptosis and apoptosis-corrected P-glycoprotein function in clinical samples. Cytometry 2003;55B:14–21.[Medline]
19. Lacombe F, Durrieu F, Briais A, et al. Flow cytometry CD45 gating for immunophenotyping of acute myeloid leukemia. Leukemia 1997;11:1878–86.[CrossRef][Medline]
20. Jordan CT, Upchurch D, Szilvassy SJ, et al. The interleukin-3 receptor chain is a unique marker for human acute myelogenous leukemia stem cells. Leukemia 2000;14:1777–84.[CrossRef][Medline]
21. van der Pol MA, Pater JM, Feller N, et al. Functional characterization of minimal residual disease for P-glycoprotein and multidrug resistance protein activity in acute myeloid leukemia. Leukemia 2001;15:1554–63.[CrossRef][Medline]
22. Sutherland H, Blair A, Vercauteren S, Zapf R. Detection and clinical significance of human acute myeloid leukaemia progenitors capable of long-term proliferation in vitro. Br J Haematol 2001;114:296–306.[CrossRef][Medline]
23. Keyhani A, Huh YO, Jendiroba D, et al. Increased CD38 expression is associated with favorable prognosis in adult acute leukemia. Leuk Res 2000;24:153–9.[CrossRef][Medline]
24. Monaco G, Konopleva M, Munsell M, et al. Engraftment of acute myeloid leukemia in NOD/SCID mice is independent of CXCR4 and predicts poor patient survival. Stem Cells 2004;22:188–201.[Abstract/Free Full Text]
25. Costello RT, Mallet F, Gaugler B, et al. Human acute myeloid leukemia CD34^+/. Cancer Res 2000;60:4403–11.[Abstract/Free Full Text]
26. Terstappen LW, Safford M, Unterhalt M, et al. Flow cytometric characterization of acute myeloid leukemia: IV. Comparison to the differentiation pathway of normal hematopoietic progenitor cells. Leukemia 1992;6:993–1000.[Medline]
27. Guan Y, Gerhard B, Hogge DE. Detection, isolation, and stimulation of quiescent primitive leukemic progenitor cells from patients with acute myeloid leukemia (AML). Blood 2003;101:3142–9.[Abstract/Free Full Text]
28. San Miguel JF, Martinez A, Macedo A, et al. Immunophenotyping investigation of minimal residual disease is a useful approach for predicting relapse in acute myeloid leukemia patients. Blood 1997;90:2465–70.[Abstract/Free Full Text]
29. van der Pol MA, Broxterman HJ, Pater JM, et al. Function of the ABC transporters, P-glycoprotein, multidrug resistance protein and breast cancer resistance protein, in minimal residual disease in acute myeloid leukemia. Haematologica 2003;88:134–47.[Abstract/Free Full Text]
30. Ong YL, McMullin MF, Bailie KE, et al. High bax expression is a good prognostic indicator in acute myeloid leukaemia. Br J Haematol 2000;111:182–9.[CrossRef][Medline]
31. Estrov Z, Thall PF, Talpaz M, et al. Caspase 2 and caspase 3 protein levels as predictors of survival in acute myelogenous leukemia. Blood 1998;92:3090–7.[Abstract/Free Full Text]
32. Legrand O, Simonin G, Beauchamp-Nicoud A, Zittoun R, Marie JP. Simultaneous activity of MRP1 and Pgp is correlated with in vitro resistance to daunorubicin and with in vivo resistance in adult acute myeloid leukemia. Blood 1999;94:1046–56.[Abstract/Free Full Text]
33. van der Kolk DM, de Vries EG, van Putten WJ, et al. P-glycoprotein and multidrug resistance protein activities in relation to treatment outcome in acute myeloid leukemia. Clin Cancer Res 2000;6:3205–14.[Abstract/Free Full Text]
34. van Stijn A, Feller N, Kok A, et al. Minimal residual disease in acute myeloid leukemia is predicted by an apoptosis-resistant protein profile at diagnosis. Clin Cancer Res 2005;11:2540–6.[Abstract/Free Full Text]
35. Geller RB, Zahurak M, Hurwitz CA, et al. Prognostic importance of immunophenotyping in adults with acute myelocytic leukaemia: the significance of the stem-cell glycoprotein CD34 (My10). Br J Haematol 1990;76:340–7.[Medline]
36. Guinot M, Sanz GF, Sempere A, et al. Prognostic value of CD34 expression in de novo acute myeloblastic leukaemia. Br J Haematol 1991;79:533–4.[Medline]
37. Selleri C, Notaro R, Catalano L, et al. Prognostic irrelevance of CD34 in acute myeloid leukaemia. Br J Haematol 1992;82:479–81.[Medline]
38. Bradstock K, Matthews J, Benson E, Page F, Bishop J. Prognostic value of immunophenotyping in acute myeloid leukemia. Australian Leukaemia Study Group. Blood 1994;84:1220–5.[Abstract/Free Full Text]
39. Kyoda K, Nakamura S, Hattori N, et al. Lack of prognostic significance of CD34 expression in adult AML when FAB M0 and M3 are excluded. Am J Hematol 1998;57:265–6.[CrossRef][Medline]
40. Haferlach T, Schoch C, Loffler H, et al. Morphologic dysplasia in de novo acute myeloid leukemia (AML) is related to unfavorable cytogenetics but has no independent prognostic relevance under the conditions of intensive induction therapy: results of a multiparameter analysis from the German AML Cooperative Group studies. J Clin Oncol 2003;21:256–5.[Abstract/Free Full Text]
41. Costello R, Mallet F, Chambost H, et al. The immunophenotype of minimally differentiated acute myeloid leukemia (AML-M0): reduced immunogenicity and high frequency of CD34⁺/CD38^– leukemic progenitors. Leukemia 1999;13:1513–8.[CrossRef][Medline]
42. Bakker AB, van den OS, Bakker AQ, et al. C-type lectin-like molecule-1: a novel myeloid cell surface marker associated with acute myeloid leukemia. Cancer Res 2004;64:8443–50.[Abstract/Free Full Text]
43. Mayer RJ, Davis RB, Schiffer CA, et al. Intensive postremission chemotherapy in adults with acute myeloid leukemia. Cancer and Leukemia Group B. N Engl J Med 1994;331:896–903.[Abstract/Free Full Text]

This article has been cited by other articles:

				M. L. Guzman, X. Li, C. A. Corbett, R. M. Rossi, T. Bushnell, J. L. Liesveld, J. Hebert, F. Young, and C. T. Jordan Rapid and selective death of leukemia stem and progenitor cells induced by the compound 4-benzyl, 2-methyl, 1,2,4-thiadiazolidine, 3,5 dione (TDZD-8) Blood, December 15, 2007; 110(13): 4436 - 4444. [Abstract] [Full Text] [PDF]

				M. L. Guzman, R. M. Rossi, S. Neelakantan, X. Li, C. A. Corbett, D. C. Hassane, M. W. Becker, J. M. Bennett, E. Sullivan, J. L. Lachowicz, et al. An orally bioavailable parthenolide analog selectively eradicates acute myelogenous leukemia stem and progenitor cells Blood, December 15, 2007; 110(13): 4427 - 4435. [Abstract] [Full Text] [PDF]

				L. M. Hansford, A. E. McKee, L. Zhang, R. E. George, J. T. Gerstle, P. S. Thorner, K. M. Smith, A. T. Look, H. Yeger, F. D. Miller, et al. Neuroblastoma Cells Isolated from Bone Marrow Metastases Contain a Naturally Enriched Tumor-Initiating Cell Cancer Res., December 1, 2007; 67(23): 11234 - 11243. [Abstract] [Full Text] [PDF]

				C. Tang, B. T. Ang, and S. Pervaiz Cancer stem cell: target for anti-cancer therapy FASEB J, December 1, 2007; 21(14): 3777 - 3785. [Abstract] [Full Text] [PDF]

				M. Grovdal, R. Khan, A. Aggerholm, P. Antunovic, J. Astermark, P. Bernell, L.-M. Engstrom, L. Kjeldsen, O. Linder, L. Nilsson, et al. Negative Effect of DNA Hypermethylation on the Outcome of Intensive Chemotherapy in Older Patients with High-Risk Myelodysplastic Syndromes and Acute Myeloid Leukemia following Myelodysplastic Syndrome Clin. Cancer Res., December 1, 2007; 13(23): 7107 - 7112. [Abstract] [Full Text] [PDF]

				A. van Rhenen, G. A. M. S. van Dongen, A. Kelder, E. J. Rombouts, N. Feller, B. Moshaver, M. S.-v. Walsum, S. Zweegman, G. J. Ossenkoppele, and G. Jan Schuurhuis The novel AML stem cell associated antigen CLL-1 aids in discrimination between normal and leukemic stem cells Blood, October 1, 2007; 110(7): 2659 - 2666. [Abstract] [Full Text] [PDF]

				G. J. L. Kaspers Win-WNT pathway in ALL Blood, April 15, 2007; 109(8): 3135 - 3136. [Full Text] [PDF]

				C. T. Jordan, M. L. Guzman, and M. Noble Cancer stem cells. N. Engl. J. Med., September 21, 2006; 355(12): 1253 - 1261. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by van Rhenen, A. Articles by Schuurhuis, G. J. Search for Related Content PubMed PubMed Citation Articles by van Rhenen, A. Articles by Schuurhuis, G. J.