Lipton et al. (1) have recently reported on the serum concentrations of OPG1 in a wide population of healthy subjects and patients affected by both solid and hematological malignancies. The authors have found significant OPG elevations in patients with colorectal or pancreatic cancer, liver or soft tissues metastases, and lymphomas, while they have confirmed a significant reduction of OPG serum levels in myeloma patients (2); the study population did not include prostate cancer patients in which elevated OPG levels have been reported by others (3, 4). Unresolved analytical issues make conclusions quite puzzling. The major point concerns the values of circulating OPG or, to put it in another way, what we are measuring in serum. OPG is currently measured by ELISA, with either an antihuman OPG monoclonal antibody or the OPG ligand rank ligand for capture (5). Table 1 resumes strikingly different values of serum OPG reported in healthy controls from studies on cancer patients (1, 2, 3, 4). Different antibody-based ELISAs have been used in these studies; however, it is worth noticing that quite discrepant results have been obtained by different groups when using presumably the same matched antibody pair provided by the same manufacturer: Seidel et al. (2) reported median levels of 9 ng/ml, whereas Ueland et al. (6) found 10-fold lower concentrations. Such differences seem hardly accounted for by different age of subjects because the age-dependent increase reported by Yano et al. (7) did not reach 4 ng/ml. Apart from affinity and specificity of the antibody pair, other technical aspects such as the choosing of the protein used as standard and the suitable calibrator diluent could account at least partly for such discrepancies. Therefore, although differences in OPG serum levels undoubtedly exist under different pathological conditions, the lack of unequivocal definition of the real range of OPG serum concentrations makes it difficult to speculate about osteoclast-suppressing activity and possible goals of therapeutic use of OPG in cancer. Indeed, multiple myeloma is currently the only malignancy in which an OPG deficit has been confirmed in humans both in serum and in bone microenvironment, giving a rationale for its therapeutic use (1, 2, 8, 9).

Preanalitycal issues concerning OPG measurements are completely unexplored. The authors report on the differences between plasma (anticoagulant not specified) and serum. We have studied the effects of sample handling on serum levels of OPG. A group of 18 healthy volunteers (age: mean, 29.6 years (range, 20–57 years); male/female: 8/10) recruited from our medical staff provided venous blood samples between 8 and 9 a.m. after an overnight fast. Blood samples were taken directly into 5 × 6-ml sterile vacuum blood collection tubes without additives (Becton Dickinson Vacutainer Systems, Plymouth, United Kingdom); one sample was allowed to clot for 30 min, then was separated by centrifugation at 1000 × g for 15 min, aliquoted and frozen at −20°C within 90 min. The other samples were left unseparated either at room temperature (20–30°C), or at 2–8°C for 150 and 300 min and then separated, aliquoted, and frozen within 210 and 360 min, respectively. OPG concentrations were measured in the same run by an ELISA developed in our laboratory using commercially available reagents (OPG DuoSet; R&D Systems, Abingdon, United Kingdom; range, 62.5–4000 pg/ml; detection limit, 31.3 pg/ml; mean intra-assay coefficient of variation, 4.8%). We have observed that in blood samples left unseparated at room temperature, there was an upward trend in the OPG levels, which became significant after 300 min, with a median increase of 16.7%; this increase was completely prevented by refrigeration (2–8°C; Fig. 1). Indeed, changes in cytokine levels as a consequence of different collection and storage procedures have been reported by others and are likely to be important, particularly in retrospective analyses of samples from previous studies (10).

In conclusion, we suggest caution in interpreting data from OPG serum measurements. Rigorous validation and comparison of different ELISAs is urgently needed, together with complete definition of preanalytical issues, to make OPG serum measurement an efficient tool to understand bone involvement in cancer and to define the rationale for therapeutic use of OPG.

1

The abbreviation used is: OPG, osteoprotegerin.

Fig. 1.

Changes in serum OPG concentration in unseparated samples, left either at room temperature (20–30°C) or at 2–8°C values are mean ± SD. ∗ versus baseline, P = 0.000549 (n = 18), by multiple measures ANOVA followed by Newman-Keuls posthoc comparison test; † versus refrigerator, P = 0.000029 (n = 18), by t test for dependent samples.

Fig. 1.

Changes in serum OPG concentration in unseparated samples, left either at room temperature (20–30°C) or at 2–8°C values are mean ± SD. ∗ versus baseline, P = 0.000549 (n = 18), by multiple measures ANOVA followed by Newman-Keuls posthoc comparison test; † versus refrigerator, P = 0.000029 (n = 18), by t test for dependent samples.

Close modal
Table 1

OPG serum concentrations in healthy controls

ReferenceNo. of controlsAge, years mean ± SD or median (range)Male/female ratioOPG, pg/ml mean ± SD or median (range)ELISA
1  112 40 ± 13.6 40/72 170a Amgen 
2  40 nrb,c nrb,c 9,000 (5,100–130,000)d R&D Systems 
3  36 51.4e 36/0 33.9 ± 15.9 Immundiagnostik 
4  32 (23–50) 6/0 22.39 (15.95–57.47)f R&D Systems 
ReferenceNo. of controlsAge, years mean ± SD or median (range)Male/female ratioOPG, pg/ml mean ± SD or median (range)ELISA
1  112 40 ± 13.6 40/72 170a Amgen 
2  40 nrb,c nrb,c 9,000 (5,100–130,000)d R&D Systems 
3  36 51.4e 36/0 33.9 ± 15.9 Immundiagnostik 
4  32 (23–50) 6/0 22.39 (15.95–57.47)f R&D Systems 
a

Median; range as drawn by Fig. 1, ∼50–300 pg/ml.

b

nr, not reported.

c

Age- and sex-matched with 225 multiple myeloma patients, age 66.6 ± 9.3 years (mean ± SD).

d

Samples were diluted to fit the linear range of the standard curve (between 200 and 2000 pg/ml).

e

Mean.

f

Results were confirmed by a second ELISA in which RANKL was used as capture element.

1
Lipton A., Ali S. M., Leitzel K., Chinchilli V., Witters L., Engle L., Holloway D., Bekker P., Dunstan C. R. Serum osteoprotegerin levels in healthy controls and cancer patients.
Clin. Cancer Res.
,
8
:
2306
-2310,  
2002
.
2
Seidel C., Hjertner Ø., Abildgaard N., Heickendorff L., Hjorth M., Westin J., Nielsen J. L., Hjorth-Hansen H., Waage A., Sundan A., Børset M., for The Nordic Myeloma Study Group. Serum osteoprotegerin levels are reduced in patients with multiple myeloma with lytic bone disease.
Blood
,
98
:
2269
-2271,  
2001
.
3
Jung K., Lein M., von Hösslin K., Brux B., Schnorr D., Loening S. A., Sinha P. Osteoprotegerin in serum as a novel marker of bone metastatic spread in prostate cancer.
Clin. Chem.
,
47
:
2061
-2063,  
2001
.
4
Brown J. M., Vessella R. L., Kostenuik P. J., Dunstan C. R., Lange P. H., Corey E. Serum osteoprotegerin levels are increased in patients with advanced prostate cancer.
Clin. Cancer Res.
,
7
:
2977
-2983,  
2001
.
5
Chen D., Sarikaya N. A., Gunn H., Martin S. W., Young J. D. ELISA methodology for detection of modified osteoprotegerin in clinical studies.
Clin. Chem.
,
47
:
747
-749,  
2001
.
6
Ueland T., Bollerslev J., Godang K., Müller F., Frøland S. S., Aukrust P. Increased serum osteoprotegerin in disorders characterized by persistent immune activation or glucocorticoid excess: possible role in bone homeostasis.
Eur. J. Endocrinol.
,
145
:
685
-690,  
2001
.
7
Yano K., Tsuda E., Washida N., Kobayashi F., Goto M., Harada A., Ikeda K., Higashio K., Yamada Y. Immunological characterization of circulatory osteoprotegerin/osteoclastogenic inhibitory factor: increased serum concentration in postmenopausal women with osteoporosis.
J. Bone Miner. Res.
,
14
:
518
-527,  
1999
.
8
Giuliani N., Bataille R., Mancini C., Lazzaretti M., Barille S. Myeloma cells induce imbalance in the osteoprotegerin/osteoprotegerin ligand system in the human bone marrow environment.
Blood
,
98
:
3527
-3533,  
2001
.
9
Pearse R. N., Sordillo E. M., Yaccoby S., Wong B. R., Liau D. F., Colman N., Michaeli J., Epstein J., Choi Y. Multiple myeloma disrupts the TRANCE/osteoprotegerin cytokine axis to trigger bone destruction and promote tumor progression.
Proc. Natl. Acad. Sci. USA
,
98
:
11581
-11586,  
2001
.
10
Flower L., Ahuja R. H., Humphries S. E., Mohamed-Ali V. Effects of sample handling on the stability of interleukin 6, tumor necrosis factor α and leptin.
Cytokine
,
12
:
1712
-1716,  
2000
.