|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Additional Prognostic Value of Bone Marrow Histology in Patients Subclassified According to the International Prognostic Scoring System for Myelodysplastic Syndromes
From the Department of Hematology, Department of Morphology and Molecular Pathology, and Center for Human Genetics, University Hospitals, University of Leuven, Leuven, Belgium. Address reprint requests to Estelle Verburgh, MD, Department of Hematology, University Hospitals, University of Leuven, Herestraat 49, B-3000 Leuven, Belgium; email: estelle.verburgh{at}uz.kuleuven.ac.be.
Purpose: The most recent and powerful prognostic instrument established for myelodysplastic syndromes (MDS) is the International Prognostic Scoring System (IPSS), which is primarily based on medullary blast cell count, number of cytopenias, and cytogenetics. Although this prognostic system has substantial predictive power in MDS, further refinement is necessary, especially as far as lower-risk patients are concerned. Histologic parameters, which have long proved to be associated with outcome, are promising candidates to improve the prognostic accuracy of the IPSS. Therefore, we assessed the additional predictive power of the presence of abnormally localized immature precursors (ALIPs) and CD34 immunoreactivity in bone marrow (BM) biopsies of MDS patients. Patients and Methods: Cytogenetic, morphologic, and clinical data of 184 MDS patients, all from a single institution, were collected, with special emphasis on the determinants of the IPSS score. BM biopsies of 173 patients were analyzed for the presence of ALIP, and CD34 immunoreactivity was assessable in 119 patients. Forty-nine patients received intensive therapy. Results: The presence of ALIP and CD34 immunoreactivity significantly improved the prognostic value of the IPSS, with respect to overall as well as leukemia-free survival, in particular within the lower-risk categories. In contrast to the IPSS, both histologic parameters also were predictive of outcome within the group of intensively treated MDS patients. Conclusion: Our data confirm the importance of histopathologic evaluation in MDS and indicate that determining the presence of ALIP and an increase in CD34 immunostaining in addition to the IPSS score could lead to an improved prognostic subcategorization of MDS patients.
MYELODYSPLASTIC SYNDROMES (MDS) are clonal hematologic stem cell disorders with a variable propensity for leukemic transformation and a great heterogeneity of clinical outcome. The classification system proposed in 1982 by the French-American-British (FAB) group, which is based exclusively on morphologic features, distinguishes among five subgroups with significantly different prognoses.1 The classification of MDS into refractory anemia (RA), refractory anemia with ringed sideroblasts (RARS), RA with excess of blasts (RAEB), RAEB in transformation (RAEB-T), and chronic myelomonocytic leukemia (CMML) served as the standard in the evaluation of MDS for nearly two decades. Although the FAB classification predicts the risk of evolution toward acute myeloid leukemia (AML), the great disparity in outcome within the different subgroups makes it a poor model for predicting prognosis in individual patients.2 Several additional risk classification systems have since been developed and have been implemented in the clinical setting by various groups.3–5 In 1997, the International MDS Risk Analysis Workshop proposed an International Prognostic Scoring System (IPSS) that compared favorably with the previously most widely used MDS risk evaluation systems (FAB, Spanish, and Lille).3,5,6 Since then, the IPSS has been widely implemented to make individual patient-based therapeutic decisions. It is a powerful instrument in stratifying MDS patients into distinctive prognostic subgroups through its more refined cytogenetic categorizations, an improved subdivision of BM blast percentages, and the inclusion of cytopenias. However, bone marrow (BM) histopathologic findings were not considered for inclusion in this prognostic index, although the study of BM histology has long proved its prognostic and diagnostic utility in MDS.7–9 Indeed, the abnormal localization of immature precursors (ALIPs) has been identified as a histologic parameter with prognostic significance in MDS.8 Its detection proves to be difficult for the untrained eye, however, in particular when biopsies are of inferior quality or poorly processed. Therefore, we also performed CD34 immunostaining. An increase in CD34-positive immunostaining as well as the presence of CD34-positive aggregates on BM biopsy have been shown to be associated with an increased risk of leukemic transformation and poor overall survival (OS).10 In this study, we explored the possibility of further refining the IPSS risk evaluation system by adding histopathologic variables. We studied the BM biopsies of MDS patients for the presence of ALIPs and CD34-positive cells to determine whether these features constitute parameters with independent prognostic significance in MDS.
Patients We selected 184 MDS patients who were consecutively diagnosed and followed up at the university hospital of the University of Leuven, Belgium, between 1980 and 1997. All patients were included in a previous study on the application of the IPSS.11 Patients diagnosed before 1997 were chosen to ensure a sufficient length of follow-up. Similar selection criteria were used as in the IPSS workshop: therapy-related MDS cases were not included, CMML cases with white blood cell counts more than 12 x 103/µL at diagnosis were excluded. Forty-nine patients (28%) underwent intensive chemotherapy during the course of their disease; 14 of these patients were treated within the first month after diagnosis. All patients underwent routine BM investigation with cytogenetic analysis as well as BM biopsy at diagnosis. Repeat examinations were done in cases with insufficient material for histologic examination or failure of cytogenetic analysis. Because of lack of histologic material for review, 10 of the original 184 patients were excluded from this study. One other patient was excluded because the diagnosis of MDS type RA was changed to idiopathic thrombocytopenic purpura on review of follow-up data. Cytopenias were defined as a hemoglobin level of less than 10 g/dL, an absolute neutrophil count of less than 1,800/µL (according to an erratum published on the IPSS12), and a platelet count of less than 100,000/µL.
BM Morphology
Bone Marrow Histology
Immunohistochemistry for CD34 on BM Biopsy Successful CD34 immunostaining could be performed in 119 of 173 patients. Optimal immunohistochemistry could not be performed on the 54 patients diagnosed before 1986 when routine biopsies were plastic embedded. Paraffin-embedded sections of 3 µm were immunostained with the CD34 monoclonal antibody Anti-HPCA-1 (Beckton-Dickenson, Frankin Lakes, NJ). The immunocomplexes were detected by the peroxidase-labeled avidin-biotin complex method. Diaminobenzidine was used as chromogen. Positive CD34 staining showed cytoplasmic granular immunoreactivity in BM cells. Staining of endothelial cells was used as an internal control.
The CD34-positive cells were evaluated by visual assessment in 10 randomly selected high-power (x400) fields without knowledge of patient information. The normal CD34 expression of BM-nucleated cells on BM histology does not exceed 1%;10 therefore, MDS cases were categorized into CD34-negative (CD34-positive cells
Cytogenetic Analysis
Calculation of Prognostic Risk According to the IPSS
Statistical analysis.
The relationship between the clinical (age, sex, FAB and IPSS category, blast percentage, cytopenia, karyotype, and intensive therapy) and pathologic (ALIP and CD34 immunostaining) variables of interest on the one hand and OS and freedom-from-AML (FF-AML) on the other hand was assessed using the product limit method of Kaplan-Meier. The P values for these analyses are based on the log-rank test. OS and FF-AML were measured from the time of diagnosis to death and to time of progression to AML, respectively. Rates of progression to AML were compared by Pearson’s
Univariate Analyses Table 1  provides the distribution of the 173 patients in the IPSS risk categories with the relevant demographic, clinical, cytogenetic, and histologic variables, as well as estimated median survival, risk of AML, and time to 25% of patients evolving to AML. The median follow-up time was 2.3 years (range, 0.1 to 21 years). The median age was 64 years (range, 9 to 89 years). The effects of patient characteristics on OS, risk of AML progression, and FF-AML on univariate analysis are summarized in Table 2.
Overall survival. At last time of follow-up, 20 patients (12%) are alive. When these patients are subdivided according to the IPSS, 26% of low-risk patients (median survival, 62.5 months), 10% of intermediate-1 (INT-1)-risk patients (median survival, 36 months), 2% of intermediate-2(INT-2)-risk patients (median survival, 13 months), and 12% of high-risk patients (median survival, 13 months; Table 1  ) are alive. On univariate analysis, all demographic, clinical, cytogenetic, and histologic variables, except for therapeutic regimen, were found to correlate significantly with OS (Table 2). As shown in Fig 2, the IPSS as well as the FAB classification stratify MDS patients into distinctive prognostic subgroups.
Risk of progression to AML. There was leukemic evolution in 21% of low-risk patients (time to 25% AML, 87 months), in 34% of INT-1-risk patients (time to 25% AML, 35 months), in 61% of INT-2-risk patients (time to 25% AML, 5 months), and in 61% of high-risk patients (time to 25% AML, 4 months; Table 1 ). Risk of progression to AML was not influenced by age, sex, cytopenia, or karyotype. BM histologic findings (ALIP and CD34 immunoreactivity), FAB classification, and IPSS category (including its prognostic feature, BM blasts) were all found to be significantly associated with the risk of evolution to AML (Table 2).
FF-AML.
The time to 25% AML for all patients was 15 months. The time to leukemic evolution was associated with the presence of ALIPs, CD34 immunoreactivity, FAB classification, and IPSS category. Moreover, all the determinants of the IPSS (BM blasts, cytopenias, and karyotype) correlated with FF-AML. Leukemia-free survival times did not correlate with age or sex (Table 2
Multivariate Analyses
Age and sex. Age and sex remained independent prognostic factors for OS (P = .0097 and P = .0003, respectively), but not for risk of AML progression or FF-AML, after correction for IPSS. FAB category. The strong independent prognostic significance of FAB category for OS (P < .0001), risk of leukemic evolution (P = .0001), and FF-AML (P < .0001) is noteworthy.
Presence of ALIP on bone marrow histology.
The presence of ALIP on BM histology, as distributed according to IPSS category, is shown in Tables 1
CD34 immunohistochemical staining. Distribution according to IPSS category is shown in Tables 1 and 3. CD34 immunoreactivity greater than 1% was observed in 62 of 119 patients (52%), including four of 22 patients (18%) in the low-risk group, 23 of 54 patients (43%) in the INT-1-risk group, 22 of 29 patients (76%) in the INT-2-risk group, and 13 of 14 patients (93%) in the high-risk group. The presence of positive CD34 immunoreactivity (> 1%) on BM histology was a significant prognostic variable within the low-, INT-1-, and INT-2-risk groups, resulting in important differences in estimated median survival times as compared with patients considered CD34 negative (Fig 4A). In the high-risk group, 13 of 14 patients displayed positive CD34 immunoreactivity, and their estimated median survival was 8 months. The one high-risk patient with less than 1% CD34-positive BM cells is alive with no evidence of leukemic progression. Not only did CD34 immunohistochemistry remain an independent predictor of OS in a multivariate analysis, but it also was associated with leukemic evolution after taking IPSS score into account (Tables 3 and 5; Fig 4B). Predicted median survival times after correction for IPSS risk category are shown in Table 3.
Correlation of CD34 immunoreactivity with ALIPs. There was a close correlation between the presence of ALIPs and CD34 immunoreactivity, with ALIP-negative cases more often considered CD34 negative as well and ALIP-positive cases usually exhibiting CD34 immunoreactivity greater than 1% (P < .0001). This correlation was, however, not absolute, with five of 46 ALIP-negative cases (11%) being CD34 positive, and 16 of 73 ALIP-positive cases (22%) regarded as CD34 negative. Moreover, after stratification for ALIP, there was a clear trend toward prognostic significance for CD34 immunoreactivity: ALIP-negative/CD34-negative patients had an OS of 72 months, whereas ALIP-negative/CD34-positive patients had an OS of 42 months (P = .0443); and ALIP-positive/CD34-negative patients had an estimated OS of 28.5 months, whereas ALIP-positive/CD34-positive patients had an OS of 14 months (Fig 5 ).
Effect of intensive therapy. Forty-one percent of patients treated with intensive therapy were older than 60 years of age. Since 1991, patients older than 60 years of age have been considered eligible for intensive therapy in our institution. Only 17 of 109 patients (16%) in the low- and INT-1-risk groups were treated, as compared with 32 of 64 patients (50%) in the INT-2- and high-risk groups. In the latter IPSS risk categories, patients selected for intensive treatment (AML-type therapy, autologous and allogeneic BM transplantation) had a significantly better predicted OS than those who did not receive this type of therapy: 17 versus 15 months (P = .008) and 13 versus 7 months (P = .0118), respectively. Regarding the various demographic, clinical, and pathologic variables in the group of intensively treated patients, only ALIPs (P = .0095), CD34 immunoreactivity > 1% (P = .04), and proportion of BM blasts were found to be associated with OS (P = .0018).
The IPSS was shown in 1997 to be an improved method for predicting survival and AML evolution in MDS compared with the FAB, Spanish, and Lille risk classification systems.6 Its more refined prognostic subdivision has subsequently been confirmed in large, independent patient groups.11,16,17 Other investigators18–20 could not show it to be superior to existing prognostic systems. In their patient population subdivided according to the IPSS, Estey et al21 found significantly shorter survival times, which were particularly pronounced for the lower-risk categories. This finding, discordant with the indolent natural history predicted by a lower IPSS score, might be explained by a substantially different patient population or referral pattern. The IPSS is generated from refined cytogenetic subgroups, subdivisions of BM blast percentage, and number of cytopenias. Although blast percentage on BM cytology is a major predictive factor for prognosis, there is a risk of sampling error from dilution with peripheral blood, heterogeneous BM cellularity, or the presence of BM fibrosis.7 The karyotype is another powerful predictor of outcome in MDS3,16,17 that also has certain limitations in clinical practice. It is not always available at diagnosis because of unsuccessful karyotyping, nor is it readily available in all centers. The value of the IPSS remains restricted to patients with cytogenetic data. A further challenge is the remaining heterogeneity within each of the IPSS risk categories, in particular, the lower-risk groups.16,19,22,23 This heterogeneity could even increase when subdividing patients according to the new WHO proposal for MDS, which results in removing a substantial proportion of high-risk patients with a resultant shift to the lower-risk groups of the IPSS.22 Several characteristics have been examined for their potential independent prognostic value in MDS. Some groups have generated dysplastic indexes of BM cytology that could be shown to improve prognostic subcategorization in the lower risk categories of the IPSS.19,23 Molecular tools, such as ras and p53 mutations, or in situ hybridization, have yielded promising results.24–27 However, we found the incorporation of histopathologic parameters into the IPSS to have certain distinct advantages. Bone marrow histology is routinely performed at diagnosis of MDS and is readily available. It provides additional prognostic information not contained in other factors of the IPSS. Here we show its high predictive value in addition to the IPSS risk classification.
ALIPs are encountered in the majority of patients with MDS and are highly predictive of outcome for survival and leukemic evolution.2,7–9,13,28 Here we extend these findings by demonstrating that the presence of ALIP also discriminates significantly within the subgroups of the IPSS with respect to overall as well as leukemia-free survival, especially as far as the low-, INT-1-, and INT-2-risk groups are concerned. (Table 3
CD34 immunostaining is a useful adjunct to standard histologic examination because it reveals a proportion of immature hematopoietic precursors with a strong myeloid commitment. In normal BM, CD34 expression is less than 1%, and its increased expression (> 1%) is well correlated with adverse outcome.29–31 Similar to our findings in ALIP, CD34 immunoreactivity greater than 1% conveys prognostic significance for survival and leukemic evolution, after correction for the IPSS. (Table 3
Not unexpectedly, the presence of ALIPs and expression of the CD34 antigen correlated quite well, but this correlation was not absolute. Not all immature precursors, whether occurring separately or in clusters, were CD34 positive, and vice versa. Recognizing ALIPs remains a purely morphologic endeavor. There is no single immunostaining method available to unequivocally rule out pseudo-ALIPs or to prove the presence of ALIPs. For this purpose, Mangi and Mufti32 had to employ several markers for different lineages. Nevertheless, CD34 immunoreactivity is helpful in identifying some or all of the blast-like cells in an ALIP. Indeed, we observed that aggregates of CD34-positive cells correlated significantly with the intertrabecular localization of immature precursors as reported previously;30,33,34 in our experience, cases with CD34 aggregates invariably featured ALIPs as well. Both histopathologic parameters not only constitute meaningful predictors after correction for IPSS risk but to some extent also convey independent prognostic information. CD34 immunoreactivity could thus discriminate between two significantly different prognostic subgroups within the category of ALIP-negative patients, whereas a clear trend to significance was observed within the ALIP-positive subpopulation (Fig 5 Another observation of special interest with respect to ALIP and CD34 immunostaining relates to the group of patients selected for intensive therapy (AML-type therapy and allogeneic and autologous BM transplantation). Although originally excluded from the IPSS workshop patients, we chose to include these patients in the analysis because it is more representative of the current standard of care in our and other institutions. Whereas neither karyotype nor the IPSS score were predictive of outcome in these intensively treated patients, the presence of ALIP and CD34 immunoreactivity, in addition to the percentage of BM blasts, correlated with OS. Although the IPSS score and the karyotype usually correlate with outcome in intensively treated patients, this has not been invariably proved in recent studies.35,36 The population group from which the IPSS risk classification was derived received no intensive therapy, which could adversely affect its present-day applicability in MDS populations.23,36 Considering the promising results obtained in this group of patients, it would be attractive to further explore the value of ALIP and CD34 immunohistochemistry in relation to the IPSS in a larger group of recently treated patients. As the IPSS score was derived from an MDS population receiving supportive care only, this could serve as the standard against which to evaluate its role in predicting the prognosis in intensively treated patients. However promising the prospects of being able to further refine the prognostic subgroups in MDS are, some evident limitations remain. To some extent, the prognostic value of age and sex reflects characteristics of the population independent of the disease itself. Our data confirm that the higher mortality rate among elderly patients is not from AML progression; mortality from complications of BM failure or coexisting diseases thus plays a prominent role in less-advanced cases of MDS. This increasing comorbidity will always reflect on the accuracy of risk classification by MDS characteristics alone; therefore, age stratification as proposed by the IPSS is important. In summary, our results confirm previous studies that identified ALIP and CD34 immunostaining as significant prognostic variables in MDS. Here we show that these BM histologic parameters also can effectively discriminate patients at increased risk of death and leukemic transformation within the subgroups of the IPSS, especially as far as low-, INT-1-, and INT-2-risk groups are concerned. In contrast to the high-risk MDS patients for whom death or leukemic transformation is imminent, the need for better assessment of prognosis mainly relates to the lower-risk categories and to CMML, subgroups for which outcome remains highly variable at this time.16,19,23,37 A refined IPSS is especially indicated when selecting patients early on for curative therapies with a high risk of morbidity. The prognostic subcategorization of MDS patients according to the IPSS can be improved by the addition of histopathologic parameters, which may prove helpful in selecting patients for different treatment strategies. Determining the presence of ALIP and increased CD34 immunostaining in addition to the IPSS also might better define different risk groups as inclusion criteria for clinical trials in MDS.
Supported in part by the Flanders Fund for Scientific Research (F.W.O.) grant G.0112.98. R. Achten is a research fellow of the Flanders F.W.O.
1. Bennett JM, Catovsky D, Daniel MT, et al: Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 51:189–199, 1982[Medline] 2. Tricot G, Vlietinck R, Boogaerts MA, et al: Prognostic factors in the myelodysplastic syndromes: Importance of initial data on peripheral blood counts, bone marrow cytology, trephine biopsy and chromosomal analysis. Br J Haematol 60:19–32, 1985[Medline] 3. Morel P, Hebbar M, Lai JL, et al: Cytogenetic analysis has strong independent prognostic value in de novo myelodysplastic syndromes and can be incorporated in a new scoring system: A report on 408 cases. Leukemia 7:1315–1323, 1993[Medline] 4. Mufti GJ, Stevens JR, Oscier DG, et al: Myelodysplastic syndromes: A scoring system with prognostic significance. Br J Haematol 59:425–433, 1985[Medline]
5. Sanz GF, Sanz MA, Vallespi T, et al: Two regression models and a scoring system for predicting survival and planning treatment in myelodysplastic syndromes: A multivariate analysis of prognostic factors in 370 patients. Blood 74:395–408, 1989
6. Greenberg P, Cox C, LeBeau MM, et al: International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 89:2079–2088, 1997 7. Maschek H, Georgii A, Kaloutsi V, et al: Myelofibrosis in primary myelodysplastic syndromes: A retrospective study of 352 patients. Eur J Haematol 48:208–214, 1992[Medline] 8. Tricot G, De Wolf-Peeters C, Vlietinck R, et al: Bone marrow histology in myelodysplastic syndromes. II. Prognostic value of abnormal localization of immature precursors in MDS. Br J Haematol 58:217–225, 1984[Medline] 9. Werner M, Kaloutsi V, Walter K, et al: Immunohistochemical examination of routinely processed bone marrow biopsies. Pathol Res Pract 188:707–713, 1992[Medline] 10. Soligo DA, Oriani A, Annaloro C, et al: CD34 immunohistochemistry of bone marrow biopsies: Prognostic significance in primary myelodysplastic syndromes. Am J Hematol 46:9–17, 1994[Medline]
11. Maes B, Meeus P, Michaux L, et al: Application of the International Prognostic Scoring System for myelodysplastic syndromes. Ann Oncol 10:825–829, 1999
12. Erratum. Blood 91:1100, 1998 13. Verhoef G, De Wolf-Peeters C, Kerim S, et al: Update on the prognostic implication of morphology, histology, and karyotype in primary myelodysplastic syndromes. Hematol Pathol 5:163–175, 1991[Medline] 14. De Wolf-Peeters C, Stessens R, Desmet V, et al: The histological characterization of ALIP in the myelodysplastic syndromes. Pathol Res Pract 181:402–407, 1986[Medline] 15. Mitelman F: ISCN (1995). In Mitelman F (ed): An International System for Human Cytogenetic Nomenclature. Basel, Switzerland, Karger, 1995 16. Pfeilstocker M, Reisner R, Nosslinger T, et al: Cross-validation of prognostic scores in myelodysplastic syndromes on 386 patients from a single institution confirms importance of cytogenetics. Br J Haematol 106:455–463, 1999[CrossRef][Medline] 17. Sole F, Espinet B, Sanz GF, et al: Incidence, characterization and prognostic significance of chromosomal abnormalities in 640 patients with primary myelodysplastic syndromes. Grupo Cooperativo Espanol de Citogenetica Hematologica Br J Haematol 108:346–356, 2000[CrossRef][Medline] 18. Jaiyesimi IA, Friedline JA, Mattson JC, et al: Myelodysplastic syndrome at a large tertiary care community hospital: Analysis according to the international prognostic scoring system. Leuk Res 24:417–426, 2000[CrossRef][Medline] 19. Matsuda A, Jinnai I, Yagasaki F, et al: New system for assessing the prognosis of refractory anemia patients. Leukemia 13:1727–1734, 1999[CrossRef][Medline] 20. Sperr WR, Wimazal F, Kundi M, et al: Survival analysis and AML development in patients with de novo myelodysplastic syndromes: Comparison of six different prognostic scoring systems. Ann Hematol 80:272–277, 2001[CrossRef][Medline]
21. Estey E, Keating M, Pierce S, et al: Application of the International Scoring System for myelodysplasia to M. D. Anderson patients. Blood 90:2843–2846, 1997
22. Nosslinger T, Reisner R, Koller E, et al: Myelodysplastic syndromes, from French-American-British to World Health Organization: Comparison of classifications on 431 unselected patients from a single institution. Blood 98:2935–2941, 2001 23. Tassin F, Dewe W, Schaaf N, et al: A four-parameter index of marrow dysplasia has predictive value for survival in myelodysplastic syndromes. Leuk Lymphoma 36:485–496, 2000[Medline] 24. Constantinidou M, Chalevelakis G, Economopoulos T, et al: Codon 12 ras mutations in patients with myelodysplastic syndrome: Incidence and prognostic value. Ann Hematol 74:11–14, 1997[CrossRef][Medline] 25. Padua RA, Guinn BA, Al Sabah AI, et al: RAS, FMS and p53 mutations and poor clinical outcome in myelodysplasias: A 10-year follow-up. Leukemia 12:887–892, 1998[CrossRef][Medline]
26. Paquette RL, Landaw EM, Pierre RV, et al: N-ras mutations are associated with poor prognosis and increased risk of leukemia in myelodysplastic syndrome. Blood 82:590–599, 1993 27. Rigolin GM, Bigoni R, Milani R, et al: Clinical importance of interphase cytogenetics detecting occult chromosome lesions in myelodysplastic syndromes with normal karyotype. Leukemia 15:1841–1847, 2001[Medline] 28. Gonzalez Manso AI, Garcia MA, Barreiro E, et al: Cytohematologic and cytogenetic prognostic factors at diagnosis and in the evolution in 46 primary myelodysplastic syndromes. Cancer Genet Cytogenet 61:174–182, 1992[CrossRef][Medline]
29. Horny HP, Wehrmann M, Schlicker HU, et al: QBEND10 for the diagnosis of myelodysplastic syndromes in routinely processed bone marrow biopsy specimens. J Clin Pathol 48:291–294, 1995 30. Oriani A, Annaloro C, Soligo D, et al: Bone marrow histology and CD34 immunostaining in the prognostic evaluation of primary myelodysplastic syndromes. Br J Haematol 92:360–364, 1996[CrossRef][Medline] 31. Span LF, Dar SE, Shetty V, et al: Apparent expansion of CD34+ cells during the evolution of myelodysplastic syndromes to acute myeloid leukemia. Leukemia 12:1685–1695, 1998[CrossRef][Medline]
32. Mangi MH, Mufti GJ: Primary myelodysplastic syndromes: Diagnostic and prognostic significance of immunohistochemical assessment of bone marrow biopsies. Blood 79:198–205, 1992 33. Baur AS, Meuge-Moraw C, Schmidt PM, et al: CD34/QBEND10 immunostaining in bone marrow biopsies: An additional parameter for the diagnosis and classification of myelodysplastic syndromes. Eur J Haematol 64:71–79, 2000[CrossRef][Medline] 34. Lambertenghi DG, Annaloro C, Soligo D, et al: The diagnostic and prognostic value of bone marrow immunostaining in myelodysplastic syndromes. Leuk Lymphoma 28:231–239, 1998[Medline]
35. Castro-Malaspina H, Harris RE, Gajewski J, et al: Unrelated donor marrow transplantation for myelodysplastic syndromes: Outcome analysis in 510 transplants facilitated by the National Marrow Donor Program. Blood 99:1943–1951, 2002
36. de Witte T, Suciu S, Verhoef G, et al: Intensive chemotherapy followed by allogeneic or autologous stem cell transplantation for patients with myelodysplastic syndromes (MDSs) and acute myeloid leukemia following MDS. Blood 98:2326–2331, 2001
37. Onida F, Kantarjian HM, Smith TL, et al: Prognostic factors and scoring systems in chronic myelomonocytic leukemia: A retrospective analysis of 213 patients. Blood 99:840–849, 2002 Submitted April 26, 2002; accepted September 23, 2002.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
1%) and CD34-positive (CD34-positive cells > 1%, including formation of aggregates) cells. 
2. The Cox proportional hazards model was used to assess the joint effects of the covariates found to be associated with OS or FF-AML in the Kaplan-Meier analyses (the proportional hazards assumption appeared valid for all these variables). It was determined which of those covariates retained significant prognostic value after adjusting for the IPSS score. Variables comprised within the IPSS score were not considered separately. To obtain a more accurate view on the predictive value of each additional covariate of interest, however, it was decided to fit the Cox proportional hazards model for the IPSS score, stratifying on each of these variables. All statistical analyses were performed using the SAS system, version 8.1 (SAS system; SAS Institute Inc., NC). 
