2212

Telomerase Activity Is Prognostic in Pediatric Patients with Acute Myeloid Leukemia Comparison with Adult Acute Myeloid Leukemia

Srdan Verstovsek, M.D., Ph.D.1 Taghi Manshouri, M.S.2 Franklin O. Smith, M.D.3 Francis J. Giles, M.D.1 Jorge Cortes, M.D.1 Elihu Estey, M.D.1 Hagop Kantarjian, M.D.1 Michael Keating, M.D.1 Sima Jeha, M.D.4 Maher Albitar, M.D.2 1

Department of Leukemia, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.

2

Department of Hematopathology, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.

3

Division of Hematology-Oncology, Children’s Hospital Medical Center, Cincinnati, Ohio.

4

Department of Pediatrics, The University of Texas M. D. Anderson Cancer Center, Houston, Texas.

BACKGROUND. Significantly elevated telomerase activity (TA) has been found in samples from patients with almost all malignant hematologic diseases. The impact of elevated TA on the course of pediatric patients with acute myeloid leukemia (P-AML) is unknown. METHODS. Using a modified polymerase chain reaction-based, telomeric repeatamplification protocol assay, the authors measured TA in bone marrow samples from 40 patients with P-AML and, for comparison, in 65 adult patients with AML (A-AML), excluding patients with French–American–British M3 disease. The results were correlated with patient characteristics and survival. RESULTS. TA in patients with P-AML was significantly lower compared with TA in patients with A-AML (P ⫽ 0.005). Patients who had P-AML with low TA had a projected 5-year survival rate of 88%, whereas patients who had P-AML with high TA had a projected 5-year survival rate of 43% (P ⫽ 0.009). Conversely, patients who had A-AML with very high TA (upper quartile) had significantly longer survival compared with patients who had A-AML with lower TA (P ⫽ 0.03). There was no correlation between complete remission rate or disease free survival and TA in P-AML or A-AML. In the A-AML group, when patients were separated by cytogenetic findings (poor prognosis vs. others), it was found that TA was significantly lower in patients with a poor prognosis, but the prognostic value of TA was not independent of cytogenetic status. CONCLUSIONS. The current results suggest, that for patients with P-AML, bone marrow TA is a highly significant prognostic factor. Cancer 2003;97:2212–7. © 2003 American Cancer Society. DOI 10.1002/cncr.11313 KEYWORDS: telomerase activity, pediatric acute myeloid leukemia (AML), adult acute myeloid leukemia, prognosis.

T

Address for reprints: Maher Albitar, M.D., Department of Hematopathology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Boulevard, Box 72, Houston, TX 77030; Fax: (713) 794-1800; E-mail: [email protected] Received September 12, 2002; revision received December 30, 2002; accepted January 7, 2003. © 2003 American Cancer Society

he number of newly diagnosed pediatric patients with acute myeloid leukemia (P-AML) is relatively low: approximately 500 patients each year in the U.S. Because age is a major prognostic factor for survival in patients with AML, overall, patients with P-AML have a better prognosis compared with adult patients who have AML (AAML). In P-AML, as in A-AML, cytogenetic abnormalities are associated with specific clinical features and define prognostic groups.1,2 The distribution of particular cytogenetic abnormalities in P-AML differs from that in A-AML.2 The incidence of AML subtypes according to the French–American–British (FAB) classification system also varies with age.1,3 In addition, there are a number of constitutional disorders that predispose children to develop AML that probably have no role in the development of AML in adults.2 In some patients with

Telomerase Activity in Pediatric AML/Verstovsek et al.

P-AML, there is evidence that leukemia occurs prenatally, during fetal development.2 Thus, patients with P-AML and patients with A-AML may have markedly different pathophysiology; and the examination of relevant biologic markers, their expression, and their prognostic significance in these subgroups is needed. For example, we reported that cellular vascular endothelial growth factor levels were significantly lower in patients with P-AML than in patients with A-AML and had no prognostic significance in patients with PAML, whereas it did have prognostic significance in patients with A-AML.4 Telomerase is the only known ribonucleoprotein in human cells with reverse transcriptase activity.5 It contains an RNA component that provides a template for the synthesis of repeated telomeric sequences. These repeats, TTAGGG, are attached to the ends of existing telomeres to maintain telomere lengths. Telomerase activity (TA) is present in almost all types of malignancies, including hematologic malignancies, but essentially is absent in the great majority of normal somatic tissues.6 A number of researchers have examined the prognostic significance of telomerase overexpression in patients with A-AML.7–14 The level of TA and its impact on the prognosis of patients with P-AML has not been studied well.15,16

MATERIALS AND METHODS Patients Pretreatment TA was measured in bone marrow samples collected from 40 patients with P-AML who were enrolled on the Children’s Cancer Group Protocols 2941 and 2961 and from 65 patients with A-AML at the time of presentation to the University of Texas M. D. Anderson Cancer Center. Patients with acute promyelocytic leukemia (FAB classification, M3) were excluded from this study due to distinct features of the disease and its treatment. All patients were treated on front-line, AML-type chemotherapy clinical research protocols. Treatment results between patients with AML who are treated with different AML-type chemotherapies do not differ significantly; this is true both for patients with P-AML and patients with A-AML.17,18 Samples were stored at ⫺ 70 °C until analysis. All samples were obtained under protocols approved by the hospitals’ Internal Review Boards and with written informed consent from the patient. The characteristics of patients are shown in Table 1.

Protein Extraction and Quantification The entire bone marrow sample was analyzed. Protein extraction and quantification were performed as reported previously in detail.19 In brief, cell pellets were lysed for 30 minutes on ice in TENN buffer (50

2213

TABLE 1 Patient Characteristics Characteristic

P-AML (n ⴝ 40 patients)

Age (yrs) Median 7.6 Range 0.7–20.6 Leukocyte count (⫻ 103/␮L) Median 29.6 Range 0.0–436.0 Platelet count (⫻ 103/␮L) Median 42.0 Range 2–293 Hemoglobin (g/dL) Median NA Range NA French–American–British classification (no. of patients) M0–M2 15 M4–M5 16 M6–M7 4 Missing 5 Prognosis (karyotype) (no. of patients) Poor (⫺5, ⫺7, 11q23) 4 Other 8 Missing 28

A-AML (n ⴝ 65 patients)

60.5 20.0–87.0 15.6 0.4–196.5 41.5 4–597 7.9 2.9–15.0 43 15 7 0

21 44 0

P-AML: pediatric acute myeloid leukemia: A-AML: adult acute myeloid leukemia; NA: not available.

mmol/L Tris-HCl, pH 7.4; 5 mmol/L ethylenediamine tetraacetic acid; 0.5% Nonidet P-40, and 150 mmol/L NaCl supplemented with 1 mmol/L phenylmethylsulfonyl fluoride; and 2 ␮g/mL pepstatin). Frequent vortexing was performed during lysis, and samples were left on ice for an additional hour. Lysates were purified by microcentrifugation for 1 hour at 14,000 revolutions per minute. Protein concentrations were determined by the Bradford method, and 200 ␮g of each cell extract were run on a 9.5% sodium dodecyl sulfate polyacrylamide gel and stained with Coomassie blue R-250 to check protein profiles and to assure stability and proper quantification of protein. In addition, proteins were quantified using a solid-phase radioimmunoassay and actin antibodies.

Measurement of TA TA was determined by combining a modified polymerase chain reaction (PCR)-based telomeric repeatamplification protocol (TRAP) assay with PCR product detection/TA semiquantification in an ABI Prism 310 Genetic Analyzer (Perkin-Elmer Biosystems, Foster City, CA), as reported recently in detail.19 Each protein sample was diluted in lysis buffer to concentration of 1 ␮g/␮L. Two microliters of each sample were combined, for the total volume of 25 ␮L, with 0.2 ␮L (5 units/␮L) of AmpliTaq Gold polymerase (Perkin-

2214

CANCER May 1, 2003 / Volume 97 / Number 9

FIGURE 1. An example of an electropherogram of telomeric repeat amplification protocol-polymerase chain reaction products. Products of telomerase activity (TA) are detected as peaks at 6-base pair intervals. The intensity of each peak is measured and the sum is presented as units of TA.

Elmer), 2.5 ␮L of GeneAmp威 10 ⫻ PCR Gold buffer (Perkin-Elmer), 16 ␮L of sterile water, 3 ␮L of MgCl2, 0.15 ␮L (25 ␮M) of each primer (TS primer [5⬘-AATCCGTCGAGCAGAGTT-3⬘] labeled with the fluorescent dye FAM (6-carboxy-flurocin) and CX primer [5⬘CCCTTACCCTTACCCTTACCCTAA-3⬘]; Genosys, The Woodlands, TX), and 0.25 ␮L (10 mM) of each nucleotide (Perkin-Elmer). TRAP reactions began with elongation of forward TS primer for 30 minutes at 30 °C, followed by heat shock at 94 °C for 6 minutes. PCR was carried out for 33 cycles at 94 °C for 30 seconds, 55 °C for 30 seconds, and 72 °C for 90 seconds. The finishing step was at 72 °C, for 10 minutes. PCR products were diluted 250-fold in formaldehyde containing the ROX, a GeneScan internal lane size standard (PerkinElmer). Samples were then placed into the genetic analyzer. Software (GeneScan Analysis 2.1; PerkinElmer) automatically determines sizes and semiquantified DNA fragments. Each electropherogram shows fluorescence intensity as a function of fragment size, and tabular data provide precise sizing and semiquantitative information (peak areas). PCR products that are a consequence of TA are manifested as fluorescence intensity/peaks at 6-base pair intervals starting at 44 base pairs. TA is calculated as a sum of the areas under all such peaks in a sample (this is presented as units of TA in Fig. 1).

Statistical Analysis Associations among variables were assessed using Spearman rank-correlation analysis. The Kruskall– Wallis test was used to compare various groups of data. Survival was plotted using Kaplan–Meier plots and was compared by log-rank test. Survival was measured from the date the sample was obtained.

RESULTS TA varied among patient samples (Fig. 2). For samples from 40 patients with P-AML, the median and mean

FIGURE 2. Histograph showing telomerase activity distribution in samples from patients with adult acute myeloid leukemia (A-AML) and patients with pediatric acute myeloid leukemia (P-AML). No. of obs: number of observations.

TA values were 363 and 778, respectively. For samples from 65 patients with A-AML, median and mean TA values were 962 and 2168, respectively. Thus, compared with A-AML samples, a significantly lower TA was found in P-AML samples (P ⫽ 0.005) (Fig. 3). When we correlated TA levels with survival in patients with P-AML, using value of 340 (approximately the median) as a cut-off point, we found that patients who had high TA levels (⬎ 340) had significantly shorter survival compared with patients who had low TA levels (P ⫽ 0.009) (Fig. 4). Conversely, when we correlated TA levels with survival in patients with A-AML, we found no correlation with survival when approximately median TA values were used. Patients who had A-AML with high TA levels (cut-off point, 2700; separating the upper quartile) had significantly longer survival compared with patients who had A-AML with lower TA levels (⬍ 2700; P ⫽ 0.03) (Fig. 5). There was no correlation between complete remission (CR) rates and TA levels in patients with P-AML or patients with A-AML (data not shown). The disease-free survival

Telomerase Activity in Pediatric AML/Verstovsek et al.

FIGURE 3. Box plot showing telomerase activity (TA) in bone marrow samples obtained from patients with acute myeloid leukemia (AML). The mean (⫹ standard error [Std. Err.]) TA values in bone marrow samples obtained from 40 patients with pediatric AML (P-AML) and 65 patients with adult AML (A-AML) are compared.

2215

FIGURE 5. Survival of patients with adult acute myeloid leukemia (A-AML) based on telomerase activity (Telom Activity [TA]) in bone marrow. TA was measured in bone marrow samples from 65 patients with A-AML. Survival is presented in relation to TA. Pats: patients.

TABLE 2 Correlation between Telomerase Activity and Patient Characteristics P value Characteristic

P-AML

A-AML

Age Leukocyte count Platelet count Hemoglobin Poor prognosis karyotype

0.7 0.07a 0.4 NA NA

0.1 0.002b 0.4 0.8 0.0079

P-AML: pediatric acute myeloid leukemia: A-AML: adult acute myeloid leukemia; NA: not available. a Correlation coefficient (R) ⫽ ⫺0.28. b R ⫽ 0.43.

FIGURE 4.

Survival of patients with pediatric acute myeloid leukemia (P-AML) based on telomerase activity (Telom Activity [TA]) in bone marrow. TA was measured in bone marrow samples from 40 patients with P-AML. Survival is presented in relation to TA. Pats: patients.

(DFS) was not significantly different between the two groups according to TA (data not shown). P values for both A-AML groups according to TA were 0.5. P values for both P-AML groups according to TA were 0.15; the group with lower TA had somewhat better DFS, but it was not significantly better compared with the group with high TA. Correlations between TA levels and characteristics of patients with P-AML or with A-AML are shown on Table 2. TA levels in patients with AAML were correlated significantly with leukocyte count (P ⫽ 0.002; correlation coefficient [R] ⫽ 0.43); whereas, in patients with P-AML, there was no correlation with WBC (there was a trend toward an inverse correlation; P ⫽ 0.07; R ⫽ ⫺ 0.28). The patients studied had bone marrow samples with high percentages of

blasts. There was no significant difference in the blast percentage between patients with P-AML and patients with A-AML (P-AML: median, 78%; mean, 74%; AAML: median, 72%, mean, 70%; P ⫽ 0.22). There was no correlation between bone marrow blast percentages and TA levels in patients with P-AML (P ⫽ 0.28) or in patients with A-AML (P ⫽ 0.91). In patients with A-AML, when they were separated by cytogenetic findings (⫺ 5, ⫺ 7, and 11q23 [poor prognosis] vs. others), it was found that TA levels were significantly lower in patients who had a poor prognosis (Fig. 6), although the prognostic value of TA was not independent of cytogenetic status (Table 2). Cytogenetic findings were available for only 12 patients with P-AML (not done on other patients), an inadequate number for proper evaluation.

DISCUSSION AML is rare disease in pediatric patients. It is believed that P-AML differs from A-AML in its biology and in

2216

CANCER May 1, 2003 / Volume 97 / Number 9

FIGURE 6. Box plot showing telomerase activity (TA) in samples from patients with adult myeloid leukemia (A-AML) in relation to their karyotype. The mean (⫹ standard error [Std. Err.]) TA values in bone marrow samples obtained from 65 patients with A-AML are compared based on the cytogenetic findings (poor prognosis karyotype vs. others).

patient outcome.1,2 The study results presented here reflect this difference. Our results show that, in patients with A-AML, high TA levels were associated with the better prognosis karyotype and with better survival; there was no correlation between TA levels and CR rates or DFS. In contrast to patients with A-AML, we found that high TA levels were associated significantly with shorter survival in patients with P-AML. The finding that TA in patients who had P-AML was significantly lower compared with patients who had A-AML is quite surprising, because TA in normal individuals decreases with age, with very high TA levels in very young individuals.10,12,20 At least one study reported that TA in patients with P-AML overlapped with TA in a normal pediatric control group.12 Thus, the current data suggest that the telomerase system is switched on in patients with A-AML and that this event may play a role in the leukemogenesis of A-AML. TA is a subject to multiple levels of control and is regulated by different factors in different cellular contexts.21 Therefore, the cellular content and, in particular, associated factors may be different between patients with P-AML and patients with A-AML, explaining how telomerase may have different roles in the pathophysiology of P-AML and A-AML. This also may explain the finding that TA correlates significantly with WBC count in patients with A-AML but not in patients with P-AML. The mechanism behind the observed difference in TA in bone marrow samples between patients with P-AML and patients with A-AML remains to be determined and may represent a target for future drug therapy. This is the first report that assesses TA and its

prognostic significance in patients with P-AML. Previously, two studies addressed TA in pediatric patients with leukemia. Engelhardt et al. studied TA in 16 pediatric patients with acute leukemia, including 1 patient with AML and 15 patients with acute lymphoblastic leukemia (ALL); TA was up-regulated in patients’ bone marrow specimens compared with normal controls, TA was decreased after induction therapy, and TA was correlated with remission.15 Malaska et al. reported on TA in specimens from eight children (two patients with AML and six patients with ALL) during the course of therapy. TA levels at diagnosis were elevated compared to normal controls: Those authors also found a close correlation between TA changes and response to therapy.16 In our patients with P-AML, CR rates were not significantly different between groups according to TA level. The group with low TA had somewhat better DFS, but it was not significantly better compared with DFS for the group with high TA. Nevertheless, the tendency for better DFS after induction therapy translated subsequently to significantly better survival for the group with low TA, suggesting sensitivity of these patients to secondline and subsequent therapy. Although there is a lack of studies on TA in patients with P-AML, a number of investigators have investigated TA in patients with A-AML. TA is elevated in the great majority of A-AML samples compared with normal controls.8,10,13 TA levels decreased to normal in patients who achieved remission but were significantly higher at the time of recurrence or disease progression compared with TA levels at the time of diagnosis.8,10,12,13 Two studies reported a significant correlation between TA and cytogenetic findings. Xu et al. found that higher TA levels were associated with aberrant karyotype (compared with normal karyotype),13 whereas Ohyashiki et al. reported an opposite correlation: normal karyotype was associated with higher TA levels.10 Zhang et al. reported no significant correlation, however.14 Conflicting data also exist on the correlation between TA and the attainment of CR in patients with A-AML: Two studies reported a lack of correlation between TA levels and CR rates,13,14 and Seol et al. found higher CR rates in patients with A-AML who had high TA levels.11 Finally, Ohyashiki et al. found no correlation between TA levels and survival in 55 patients with A-AML.10 Thus, the role of telomerase in the propagation of A-AML has been suggested but is far from being established clearly. In conclusion, the current findings suggest that TA overexpression is a significant prognostic indicator for shortened survival in patients with P-AML but not in patients with A-AML. Thus, the data suggest that the

Telomerase Activity in Pediatric AML/Verstovsek et al.

potential therapeutic role of telomerase inhibitors in patients with P-AML should be investigated.

12.

REFERENCES 1.

Hall GW. Childhood myeloid leukaemias. Best Pract Res Clin Haematol. 2001;14:573–591. 2. Greaves M. Childhood leukaemia. BMJ. 2002;324:283–287. 3. Webb DK, Harrison G, Stevens RF, Gibson BG, Hann IM, Wheatley K. Relationships between age at diagnosis, clinical features, and outcome of therapy in children treated in the Medical Research Council AML 10 and 12 trials for acute myeloid leukemia. Blood. 2001;98:1714 –1720. 4. Jeha S, Smith FO, Estey E, et al. Comparison between pediatric acute myeloid leukemia (AML) and adult AML in VEGF and KDR (VEGF-R2) protein levels. Leuk Res. 2002;26:399 – 402. 5. Helder MN, Wisman GB, van der Zee GJ. Telomerase and telomeres: from basic biology to cancer treatment. Cancer Invest. 2002;20:82–101. 6. Meyerson M. Role of telomerase in normal and cancer cells. J Clin Oncol. 2000;18:2626 –2634. 7. Counter CM, Gupta J, Harley CB, Leber B, Bacchetti S. Telomerase activity in normal leukocytes and in hematologic malignancies. Blood. 1995;85:2315–2320. 8. Engelhardt M, Mackenzie K, Drullinsky P, Silver RT, Moore MA. Telomerase activity and telomere length in acute and chronic leukemia, pre- and post-ex vivo culture. Cancer Res. 2000;60:610 – 617. 9. Li B, Yang J, Andrews C, et al. Telomerase activity in preleukemia and acute myelogenous leukemia. Leuk Lymphoma. 2000;36:579 –587. 10. Ohyashiki JH, Ohyashiki K, Iwama H, Hayashi S, Toyama K, Shay JW. Clinical implications of telomerase activity levels in acute leukemia. Clin Cancer Res. 1997;3:619 – 625. 11. Seol JG, Kim ES, Park WH, Jung CW, Kim BK, Lee YY. Telomerase activity in acute myelogenous leukaemia: clini-

13.

14.

15.

16.

17. 18.

19.

20.

21.

2217

cal and biological implications. Br J Haematol. 1998;100: 156 –165. Tatematsu K, Nakayama J, Danbara M, et al. A novel quantitative “stretch PCR assay” that detects a dramatic increase in telomerase activity during the progression of myeloid leukemias. Oncogene. 1996;13:2265–2274. Xu D, Gruber A, Peterson C, Pisa P. Telomerase activity and the expression of telomerase components in acute myelogenous leukaemia. Br J Haematol. 1998;102:1367–1375. Zhang W, Piatyszek MA, Kobayashi T, et al. Telomerase activity in human acute myelogenous leukemia: inhibition of telomerase activity by differentiation-inducing agents. Clin Cancer Res. 1996;2:799 – 803. Engelhardt M, Ozkaynak MF, Drullinsky P, et al. Telomerase activity and telomere length in pediatric patients with malignancies undergoing chemotherapy. Leukemia. 1998;12: 13–24. Malaska J, Sklenickova M, Krejci K, et al. Telomerase activity and expression and telomere analysis in situ in the course of treatment of childhood leukemias. Blood Cells Mol Dis. 2000;26:534 –539. Loeb DM, Arceci RJ. What is the optimal therapy for childhood AML? Oncology (Huntington). 2002;16:1057–1066. Estey E, Thall P, Beran M, Kantarjian H, Pierce S, Keating M. Effect of diagnosis (refractory anemia with excess blasts, refractory anemia with excess blasts in transformation, or acute myeloid leukemia [AML]) on outcome of AML-type chemotherapy. Blood. 1997;90:2969 –2977. Verstovsek S, Manshouri T, Kantarjian H, et al. Highly reproducible detection and semi-quantification of telomerase activity. Biotechniques. 2001;30:930 –934. Iwama H, Ohyashiki K, Ohyashiki JH, et al. Telomeric length and telomerase activity vary with age in peripheral blood cells obtained from normal individuals. Hum Genet. 1998; 102:397– 402. Elenitoba-Johnson KS. Complex regulation of telomerase activity: implications for cancer therapy. Am J Pathol. 2001; 159:405– 410.