| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1
Laboratory of Molecular Biology, Department of Clinical Biochemistry, State Serum Institute, Artillerivej 5, 2300 Copenhagen, Denmark;
2
The Gynaecologic Clinic, The Juliane Marie Centre, Rigshospitalet, Copenhagen, Denmark DK-2100;
3
Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark DK-2100;
4
MD Anderson Cancer Center, Houston, Texas, 77030;
5
Department of Obstetrics and Gynaecology, Aarhus University Hospital, Skejby, Denmark DK-8200;
6
Gynaecology Cancer Research Unit, Department of Gynaecological Oncology, St. Bartholomew's Hospital, London, UK EC1A 7BE;
a author for correspondence: fax 0045 3268 3878, e-mail hogdall{at}dadlnet.dk
OVX1 is a tumor marker that may be of value in the management of ovarian cancer and appears to be complementary to CA125 measurement (1)(2)(3). The results of a study by Woolas et al. (2) suggested that OVX1 is increased in the serum of a major proportion of patients with stage I ovarian cancer who have serum CA125 concentrations within the health-related reference interval. We, therefore, had planned to include OVX1 measurement in a prospective study of screening for ovarian cancer. Because the screening study involved sending blood samples by post to a central laboratory, we undertook this preliminary study to assess the stability of OVX1 in the 2- to 3-day time period required for postal delivery.
All subjects in this study were apparently healthy premenopausal women. Venipuncture was performed on 20 subjects in the UK and 9 subjects in Denmark. The collection tubes, the storage temperature before centrifugation, and the length of storage before centrifugation were varied. All samples were then centrifuged at 2000g for 10 min, and the plasma or serum samples were stored at -20 °C until OVX1 analysis was performed. In addition, to investigate the effect of repetitive thawing and freezing, serum samples from nine subjects were divided into eight aliquots and frozen. After 24 h, seven aliquots from each subject were thawed at room temperature and refrozen after 1 h. This process was repeated daily, leaving one aliquot frozen each day, until the last aliquot was frozen and thawed after 8 day for OVX1 analysis.
The Wilcoxon test was used for paired data. Correlations were examined by linear regression.
The OVX1 double-determinant assay using a murine monoclonal antibody to detect an epitope on a high-molecular weight murine-like glycoprotein was performed as described previously (4). Monoclonal OVX1 antibody was labeled with Na125I, using the Iodogen method. In brief, 50 µL of phosphate buffer (0.5 mol/L, pH 7.4) was added to a 15 x 75-mm borosilicate tube coated with 10 µg of Iodogen. OVX1 (20 µg) was added in a volume of 150 µL of phosphate-buffered saline (PBS). Radioiodination was initiated by the addition of 0.5 mCi of Na125I (2.5 µL), and the mixtures were incubated for 20 min on ice. The protein-bound iodine was separated from free iodine by gel filtration on a PD-10 column (Pharmacia) that had been preequilibrated with PBS. A 3-µL sample of each fraction was counted in a Packard gamma counter to measure protein-bound radioactivity. Iodination efficiency was calculated by the formula: Iodination efficiency = (protein-bound cpm/total cpm) x 100%. The efficiency of iodination was between 70% and 80%.
Polystyrene beads (6.4 mm diameter; Precision Plastic Ball) were heated at 80 to 100 °C for 1 h, during which they were shaken every 15 min, and then cooled to room temperature. Purified OVX1 monoclonal antibody (100 mg/L) in PBS (0.15 mol/L sodium chloride, 50 mmol/L phosphate, pH 7.4) was added to cover the beads, and the mixture was incubated at 37 °C for 2 h and then at 4 °C overnight without shaking. The monoclonal antibody solution was aspirated, and 2.5 mL/L glutaraldehyde in PBS was added slowly to cover the beads for 10 min at room temperature. The beads were rinsed three times with PBS and blocked overnight with 20 g/L bovine serum albumin and 10 mL/L normal mouse serum in PBS at 4 °C. The beads were placed directly into enzyme immunoassay reaction trays (Abbott Laboratories), with one bead in each well. Five concentrations of HPLC-purified OVX1 antigen (20, 10, 5, 2.5, or 1.25 kilounits/L) and two control serum specimens, one with a low (5.3 kilounits/L) and one with a high (12.9 kilounits/L) concentration of OVX1 antigen, were run with each assay. For the high control, the intraassay CV was 9.5% (n = 6), whereas the interassay CV was 13% (n = 12). Duplicate 50-µL aliquots of each calibrator, control, or experimental sample were pipetted into their assigned wells. Portions (150 µL) of blocking solution (20 g/L bovine serum albumin and 10 mL/L normal mouse serum in PBS) were added to each well. The plates were incubated overnight at 4 °C and washed three times with PBS. 125I-labeled OVX1 diluted in 200 µL of blocking buffer (1000 cpm/µL) was added, and plates were incubated at 4 °C overnight. The plates were then washed three times with PBS, and the beads were transferred to tubes for gamma counting. The background was calculated from the number of counts associated with coated beads that had been incubated in blocking buffer.
A significant increase in serum OVX1 concentrations was observed in 20
samples allowed to clot at room temperature for 3 h
(P = 0.02) and 48 h (P = 0.00008),
compared with paired samples separated by centrifugation immediately
after venipuncture (Table 1
). This difference was not seen in the paired samples stored at
4 °C before centrifugation (P = 0.13; Table 1
).
Furthermore, serum OVX1 concentrations were stable in nine samples
stored at 4 °C in time intervals of 1 h (median, 5.6
kilounits/L; quartiles, 3.1–8.9 kilounits/L), and 2, 5, 7, 10, and
72 h (median, 6.1 kilounits/L; quartiles, 4.8–6.7 kilounits/L)
before centrifugation. Significantly lower OVX1 concentrations were
found in nine plasma samples collected in EDTA and in heparin-treated
samples than in paired serum samples collected in a plain tube
(EDTA-treated samples: median, 0.9 kilounits/L; quartiles, 0.6–1.1
kilounits/L; P = 0.006 after storage for 1 h; and
median, 1.5 kilounits/L; quartiles, 1.3–2.2 kilounits/L after storage
for 72 h; heparin-treated samples: median, 2.8 kilounits/L;
quartiles, 2.4–3.3 kilounits/L; P = 0.02 after storage
for 1 h; and median, 5.5 kilounits/L; quartiles, 3.7–10.8
kilounits/L after storage for 72 h; serum samples: median, 5.6
kilounits/L; quartiles, 3.1–8.9 kilounits/L after storage for 1
h; and median, 6.1 kilounits/L; quartiles, 4.8–6.7 kilounits/L after
storage for 72 h). Repetitive freezing and thawing produced only a
minor clinically unimportant decrease in OVX1 concentrations in stored
serum samples (median, 5.6 kilounits/L; quartiles, 3.1–8.9 kilounits/L
after one freezing cycle; and median, 5.0 kilounits/L; quartiles,
2.4–6.0 kilounits/L after eight freezing cycles; r =
-0.8; slope = -0.14; intercept = 6.0; P =
0.01).
|
The results indicate that the OVX1 radioimmunoassay is highly dependent on sample handling. First, OVX1 concentrations rose significantly if blood samples were allowed to clot at room temperature, but this effect was not seen if samples were kept at 4 °C. Second, there were significant differences between serum and plasma sam-ples. Serum samples once separated were stable on repetitive thawing and freezing. The OVX1 antibody is directed toward a high-molecular weight antigen with multiple epitopes (1). The cause of the increase in OVX1 on storage of blood at room temperature is uncertain, but one possibility is that cleavage of the antigen occurs, releasing smaller fragments, each with multiple OVX1 epitopes.
Different cutoffs at 7.2, 10.5, or 12.1 kilounits/L have been used in clinical studies (2)(4). The present findings, therefore, are clinically relevant because the assay range is narrow and the effect we have noted can cause an increase of OVX1 concentrations above these cutoffs. The findings have important implications for the collection of samples for OVX1 analysis by radioimmunoassay and for our ongoing ovarian cancer screening trial. It is clear that samples must be collected in plain tubes and the serum either separated immediately or the sample kept at 4 °C until centrifugation is performed. Unfortunately, our screening trial is based on postal transport of blood samples and immediate centrifugation is not possible. We therefore will not be able to incorporate OVX1 measurement into the study protocol unless a method for stabilizing the antigen during transport at room temperature is identified. Fortunately, repetitive freezing at -20 °C and thawing did not significantly affect OVX1 concentrations. This result is reassuring and suggests that earlier OVX1 studies based on serum samples that were separated and stored immediately are likely to be valid even if the samples analyzed had been frozen and thawed on several occasions.
Acknowledgments
This work was supported by The Danish Cancer Society and The Gynaecology Cancer Research Fund.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  I. J. Jacobs and U. Menon Progress and Challenges in Screening for Early Detection of Ovarian Cancer Mol. Cell. Proteomics, April 1, 2004; 3(4): 355 - 366. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
