| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Department of Clinical Chemistry and Transfusion Medicine, Institute of Laboratory Medicine and 2 Hematology and Coagulation Section, Department of Medicine, Sahlgrenska University Hospital, Göteborg, Sweden3 Department of Medicine, Uddevalla Hospital, Uddevalla, Sweden4 Department of Experimental Anaesthesiology, University Hospital, Freiburg, Germany
aaddress correspondence to this author at: Department of Clinical Chemistry and Transfusion Medicine, Sahlgrenska University Hospital, SE-413 45 Göteborg, Sweden; fax 46-31-828458, e-mail anne.ricksten{at}clinchem.gu.se
In the absence of pathognomonic markers, the diagnosis of the two chronic myeloproliferative disorders polycythemia vera (PV) and essential thrombocythemia (ET) has relied on a set of clinical and laboratory criteria (1)(2)(3)(4)(5). The cloning of the cell surface receptor polycythemia rubra vera-1 (PRV-1) has recently been described (6), and the consistent overexpression of PRV-1 mRNA observed in PV patients indicates that this might constitute a new diagnostic marker for the disease. In the initial cohort examined by Northern blot analysis, PRV-1 expression was increased in all PV patients examined as well as in some ET patients, but not in healthy controls (6). These results have also been verified and extended using a quantitative reverse transcription-PCR (RT-PCR) assay. All PV as well as 50% of ET patients displayed increased PRV-1 expression (7)(8). Patients with secondary erythrocytosis and healthy controls tested showed PRV-1 concentrations within the reference interval. Interestingly, the observed increase in PRV-1 mRNA expression does not lead to a corresponding increase in protein expression on the cell surface (9). Erythropoietin-independent colony growth and PRV-1 overexpression seem to go hand in hand in both PV and ET patients (7), raising the hope that RT-PCR for PRV-1 could replace the need for the technically demanding erythropoietin-independent colony assay, although a recent report suggests that the erythropoietin-independent colony growth assay is a more reliable method (10). The aim of the present work was to develop a quantitative RT-PCR method to measure PRV-1 transcripts in whole-blood leukocytes and to determine the potential usefulness of the method in the differential diagnosis of polycythemias and thrombocytosis. The assay was compared with a method using isolated granulocytes (7)(9) to assess whether granulocyte purification is necessary before RNA extraction. Granulocyte fractionation is cumbersome to standardize, and an assay that uses whole-blood leukocytes would simplify the analysis and perhaps make it more suitable for a routine laboratory setting.
Blood samples were collected from 78 patients with well-characterized myeloproliferative diseases, 25 with PV and 53 with ET (Table 1
). At the time of the study, 92% of the PV patients and 66% of the ET patients were receiving therapy with myelosuppressive agents in addition to phlebotomies (Table 1
). Duplicate samples were transferred to the two laboratories that participated in the study for analysis of PRV-1 expression by their respective methods. The samples arrived within 24 h from the time of sampling to both laboratories. All patients fulfilled the Polycythemia Vera Study Group criteria as well as the additional criteria proposed for their respective diagnoses (1)(2)(3)(4)(5). The study was approved by the Ethics Committee of Göteborg University, and informed written consent was obtained from all participants. Details for the laboratory procedures, assay performance, and establishment of reference intervals for the two methods are given in the Data Supplement that accompanies the online version of this Technical Brief athttp://www.clinchem.org/content/vol50/issue3/ as well as in the text and in the report by Klippel et al. (8).
|
Both RT-PCR assays used in this report are quantitative, which was not the case in some of the earlier reports (6)(11). The cycle threshold (CT) for PRV-1 was determined by real-time PCR and was divided by the CT value for the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) transcript. Consequently, a lower PRV-1:GAPDH CT ratio indicates higher PRV-1 mRNA expression. Both methods had approximately the same assay imprecision, as determined by their between-day CV (3%). A PRV-1:GAPDH CT ratio <0.93 indicated increased expression of PRV-1 in peripheral whole-blood leukocytes, and a CT ratio <1.16 indicated increased expression of PRV-1 in purified granulocytes.
The PRV-1:GAPDH CT ratios did not follow a gaussian distribution for the patients; therefore, the Kruskal–Wallis nonparametric method was used to determine the significance of the difference between the groups. All individual results can be found in Table 1 of the online Data Supplement. With either assay, the differences between PV and ET patients vs healthy controls were significant (P <0.0001), as was the difference between the PV and ET patients (P <0.0001). In 23 of the 25 PV patients, PRV-1 was overexpressed in whole-blood leukocytes. PRV-1 was overexpressed in the same number of patients when purified granulocytes were used for the assay. One patient was found to express normal amounts of PRV-1 by both methods. One additional PV patient had normal PRV-1 expression when measured with one or the other method, but the PV patient with normal PRV-1 expression in granulocytes but overexpression in leukocytes was one of two outliers in the method comparison (see below) and was excluded because we suspected that this sample was a mix up with the other outlier, an ET patient. This gave an assay sensitivity of 92% (23 of 25) for PV patients when we used whole-blood leukocytes and 96% (23 of 24) for PV patients when we used purified granulocytes. Thus, the assay sensitivity for PV does not seem to be severely decreased by omitting granulocyte purification from whole-blood leukocytes. The fact that two PV patients did not show increased PRV-1 expression in one or both assays suggests that treatment could have some effect on PRV-1 expression.
It has been shown in a recent study that two PV patients reverted from clonal to polyclonal hematopoiesis after interferon-
treatment (12), indicating that treatment could affect PRV-1 expression. This study also reported several PRV-1-negative PV patients (4 of 13). A treatment effect is also supported by the finding that interferon- might normalize PRV-1 expression (13). In contrast, the PV patient who was normal in both of the RT-PCR assays used in the present study has never been treated with interferon-, but with hydroxyurea instead. It is unclear whether this patient, who fulfilled all of the criteria for PV, has ever been PRV-1-positive. Both patients who were PRV-1-negative in the whole-blood leukocyte assay have been reanalyzed with concordant results. Furthermore, we found no significant difference in PRV-1 expression between treated (n = 35) and untreated (n = 18) ET patients (whole-blood leukocytes, P = 0.32; isolated granulocytes, P = 0.88, Kendall rank correlation) that would indicate any inhibitory or stimulatory effect on PRV-1 expression by myelosuppressive treatment. PRV-1 was overexpressed in 14 of 53 (25%) whole-blood leukocyte samples from ET patients. In contrast, 24 of the 52 ET patients (46%) displayed overexpression when purified granulocytes were analyzed. Ten patients who were PRV-1-negative in the whole-blood leukocyte assay showed PRV-1 overexpression in the granulocyte assay. No patient was found to be positive in the whole-blood leukocyte assay but negative in isolated granulocyte assay. The difference between the two methods in detecting PRV-1 overexpression in ET patients can to some extent be explained by the fact that nongranulocytes (e.g., lymphocytes) contribute GAPDH transcripts but no PRV-1 transcripts when whole blood is used for the RNA preparation (6). We also observed a significant correlation between PRV-1 expression in whole-blood leukocytes and lymphocyte count (P = 0.014). However, regression analysis demonstrated a poor fit (r2 = 0.10; data not shown), which indicates that additional factors other than lymphocyte number contribute to the difference between the two methods. When whole-blood leukocytes were used, a correlation between PRV-1 overexpression and increased risk for vascular complications was detected in ET patients (14) as well as an association between PRV-1 overexpression and low plasma erythropoietin concentrations (15). These findings indicate that the whole-blood leukocyte assay is a good predictor for patients with a more aggressive ET.
A comparison of PRV-1 expression in whole-blood leukocytes and purified granulocytes is shown in Fig. 1
. Deming regression analysis was used to compare the two methods. A regression residual plot between the two methods identified two possible outliers that were outside ±4 SD, and the samples were excluded from the comparison (Fig. 1A
). Deming analysis detected both a constant difference and a proportional difference. The presence of a constant difference between the two assays (-0.38; 95% confidence interval, -0.49 to -0.26) was expected because non-PRV-1-expressing cells were included in one of the methods. The proportional difference (slope, 1.17; 95% confidence interval, 1.07–1.27) might reflect the differences in sample preparation and analysis procedures (e.g., RNA preparation method and PCR primer choice) for the two methods. However, the numeric results for the samples analyzed in the two laboratories are clearly linear, and the regression analysis showed a high correlation coefficient (r = 0.94; 95% confidence interval, 0.90–0.96; Fig. 1B
).
|
In conclusion, our study shows that a quantitative RT-PCR assay of PRV-1 expression in whole-blood leukocytes is a potentially valuable diagnostic tool in the differential diagnosis of polycythemia, especially if the granulocyte purification procedure is unfeasible. The potential usefulness of this method in differential diagnosis and prognosis prediction of thrombocytosis was indicated in two previous studies (14)(15), but it needs further investigation. Furthermore, the finding of normal PRV-1 expression in PV patients as well as the possibility that treatment might affect PRV-1 expression needs further investigation. These questions can only be answered in prospective studies involving newly diagnosed patients whose PRV-1 expression is followed consecutively after initiation of different therapeutic regimes.
Acknowledgments
This study was supported by grants from "FoU Västra Götaland", "Stiftelsen Jubileumsklinikens Forskningsfond mot Cancer", and "Assar Gabrielssons Foundation" as well as the Alfried-Krupp-Förderpreis and the SFB 364 (TP A12).
Footnotes
1 current address: BC Cancer Research Centre, Terry Fox Laboratory, Vancouver, BC, Canada 
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  L. Teofili, M. Martini, F. Guidi, D. Venditti, G. Leone, and M. L. Larocca The PRV-1 gene expression in essential thrombocythemia Blood, November 1, 2004; 104(9): 2995 - 2996. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
