| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
1
Irving Center for Clinical Research, and Departments of
2
Pathology, and
3
Medicine, Columbia University College, of Physicians and Surgeons, New York, NY 10032
a Address correspondence to this author at: Irving Center for Clinical Research, Columbia University College of Physicians and Surgeons, 630 W. 168th St., PH10-305, New York, NY 10032. Fax 212-305-3213; e-mail jfo1{at}columbia.edu
To the Editor:
Iles et al. (1) speculated on the value of development of a specific assay to measure human luteinizing hormone ß-core fragment (hLHßcf) without cross-reactivity with human chorionic gonadotropin ß-core fragment (hCGßcf). We wish to draw readers’ attention to the fact that we developed and reported such a specific assay for hLHßcf in 1995 (2) and reviewed the applications of this assay in 1996 (3). We wrote (3) that the assay "may have useful applications in tumor marker assays, pregnancy tests, and menopause". The hLHßcf assay (B505-B503) exhibits 1% cross-reactivity with hLH and 0.1% cross-reactivity with hCGßcf. We found that the assay detects the urinary form of hLHßcf in the urine of normally ovulating women (2). We applied this assay to the urine of postmenopausal women and found significant quantities of hLHßcf but not hCGßcf. The mean concentration of hLHßcf for 107 samples from postmenopausal women was 236 pmol/g creatinine (4). We also measured hCGßcf in the same samples and found that its concentration was one-tenth that of hLHßcf. For hCGßcf, we used our laboratory assay B210-B108 (5), which has 2% cross-reactivity with hLHßcf. These data suggest that to use hCGßcf as a cancer marker, an assay of very high specificity for this molecule is required because of possible cross-reaction with hLHßcf.
We observed in our work that hLHßcf cross-reacted with polyclonal antisera to hLH, suggesting that urinary ovulation test kits that measure intact hLH have the potential for similar hLHßcf cross-reactions if they are based on polyclonal antibodies or monoclonal anti-hLH antibodies that cross-react with hLHßcf. If such should be the case, then the day of ovulation as predicted by the kit could be in error.
The results presented by Iles et al. (1) supported our data published in 1995 (2), 1996 (3), and 1998 (4). It is important to emphasize the potential risk of hLHßcf cross-reactivity in urine when related gonadotropins (e.g., hLH, hLHß, hCG, hCGß, and hCGßcf) are measured.
|
References
Williamson Laboratory, St. Bartholomew’s and, The Royal London School, of Medicine and Dentistry, London EC1A 7BE, UK, E-mail R.K.Iles{at}mds.qmw.ac.uk
a Address correspondence to this author at: Irving Center for Clinical Research, Columbia University College of Physicians and Surgeons, 630 W. 168th St., PH10-305, New York, NY 10032. Fax 212-305-3213; e-mail jfo1{at}columbia.edu
To the Editor:
O’Connor et al. highlight the important potential for the glycoprotein hormone degradation products to cross-react in gonadotropin and thyrotropin assays. We actually pointed out (1) that increased urinary luteinizing hormone ß-core fragment (LHßcf) seriously compromises the use of hCGßcf as a tumor marker because in our assay it cross-reacted 100% and that urinary human chorionic gonadotropin ß-core fragment (hCGßcf) assays with very low cross-reactivity with LHßcf [such as those discussed in Refs. (2)(3)] need to be developed. Three important and related issues need to be investigated.
The dynamics of LHßcf release into urine and its relation to LH
metabolism.
In our 1993 study (4), we found
immunoreactive material resembling "beta-core" in the urine of
normally menstruating women after the surge in LH, peaking 2 days
afterward. Furthermore, whereas the LH surge was clearly discernible as
a single-day spike in urinary LH concentrations, the ß-core-like
material persisted over 6 days. We speculated that this ß-core-like
material was in fact a urinary degradation product of LH similar
to the ß-core molecule formed from the degradation of hCG
(5). The subsequent isolation of LH-ß-core from
pituitary tissue and the development of a LHßcf assay by Kovalevskaya
et al. (2) and Birken et al. (6) firmly
established that this indeed was an LH-core molecule. This has led to
an important new nomenclature for what had previously been referred to
as beta-core, namely LHßcf and hCGßcf. Follicle-stimulating
hormone ß-core fragment and thyroid-stimulating hormone ß-core
fragment probably also exist.
Kovaliskaya et al. (2) and O’Connor et al. (3) also confirmed that urinary LHßcf concentrations peaked 2–3 days after the LH surge and persisted for 6 days. Based on the studies of metabolism (7)(10), although some 22% of intact hCG is removed by direct glomerular filtration, the renal parenchymal cells appear to take up hCG and release the degradation products into the urine. This implies that a slow and specific mechanism exists for the removal, deactivation, and degradation of the glycoprotein hormones. Such an unusual mechanism of metabolic clearance is worthy of further investigation, not only because might it be relevant to understanding the clearance processes of other proteins but because it might have clinical relevance to such diverse areas as altered renal function during disease.
Cross-reactivity effects of the urinary metabolites in studies of
ectopic hCG expression.
Urinary LHßcf is a significant
confounding factor in the use of hCGßcf as a tumor marker in women
(1). The sheer magnitude of the problem is illustrated in
the estimates of LHßcf in the urine of pre- and postmenopausal women:
as with most biochemical measurements, the range of LHßcf is not
gaussian distributed but is skewed, leading to extremely high values at
the 95th centile. This problem is compounded further by the midcycle
surge in LHßcf up to 70–80 pmol/mol creatinine (600–700 fmol/mg
creatinine) and the considerable increase after menopause, with a mean
of 27 pmol/mol creatinine (236 fmol/mg creatinine)
(2)(3). We estimated that the 75th centile
values of urinary LHßcf increase from 410 to 640 pmol/mol creatinine
in postmenopausal women (at 40–80 years). Urinary hCGßcf assays used
in oncology usually operate in the pmol/L range and have detection
limits of 1–10 pmol/L. The concentrations of LHßcf encountered after
menopause have a significant cross-reactivity effect on urinary
hCGßcf assays, even those with a 0.5% cross-reactivity (Table 1
). Although a hCGßcf assay with a 2% cross-reactivity is
better than the hCGßcf assays I have used, the concentrations of
LHßcf are sufficiently high in >25% of postmenopausal
women as to still cause a significant false measure in the hCGßcf
assay (Table 1
).
Does LHßcf cross-react in ovulation prediction kits, and if so,
does this give a false timing of ovulation?
With the use of home ovulation kits measuring urinary LH surges, it is vital to know whether LHßcf can cross-react in the system. A consequence of such cross-reactivity would be the mistiming of ovulation by up to 3 days. What effect this has on couples trying to conceive is not known and needs investigation. Clearly the discovery of the urinary gonadotropin core fragments has wide implications for the clinical chemistry community. Preparing this letter has highlighted for me the need for international collaboration and standardization because on close examination it would appear that our estimates of the concentration of LHßcf are 10-fold higher than those of O’Connor et al. We need to know the cross-reactivities in gonadotropin assays, but to do this we also need international reference preparations. The laboratory of O’Connor, Kovalevskaya, and Birken appears to be unique in having highly purified preparations of both hCGßcf and LHßcf.
References
-subunit and -subunit dissociated from intact human chorionic gonadotropin, despite differences in sialic acid contents. Endocrinology 1987;121:1215-1220.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
