| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
1 Department of Chemical Endocrinology, Radboud University, Nijmegen Medical Center, Nijmegen, The Netherlands
aAuthor for correspondence.
To the Editor:
Mammaglobin is a protein with a strong association with breast tissue. As such, many reports have been published on the use of reverse transcription-PCR of mammaglobin mRNA for the detection of circulating breast tumor cells. In a recent article in Clinical Chemistry, Zehentner et al. (1) reported on the use of an ELISA to detect mammaglobin protein in blood from breast cancer patients. This strategy has much potential, as the only currently useful biomarker for breast cancer, CA15-3, has very limited specificity and sensitivity (2).
Given the importance of having a breast cancer-specific serum biomarker assay, it is a regret that the description of the mammaglobin serum assay given by Zehentner et al. (1) is so concise. Nothing is reported about sample handling, which could be important because the origins of patient and control samples differ. Furthermore, no information on assay characteristics and performance is shown. The data should therefore be considered with caution.
Our own efforts to date in setting up a validated ELISA for mammaglobin have been unsuccessful because all data on mammaglobin concentrations in blood from breast cancer patients were found to be attributable to nonspecific signals. Interference from human anti-mouse antibodies is a notorious source of false-positive signals. However, as our assays use avian antibodies (duck and chicken) in the preanalytical stage and mammalian (rabbit or mouse and goat) antibodies in the postanalytical stage of the ELISA, human anti-mouse antibody interference is excluded as a source of these false-positive signals (3). Another explanation might be that the biotinylated antibodies used by Zehentner et al. (1) for the serum mammaglobin ELISA led to false-positive signals (4).
Furthermore, the authors state that they quantify mammaglobin protein concentrations, but the antibodies were raised against purified native mammaglobin protein complex. Further analysis of their original report on these antibodies shows that the antibodies are, in fact, directed against mammaglobin-lipophilin B complexes and that the biotinylated antibody RO28 is reactive against lipophilin B peptides (5). This seems to us crucial information on the characteristics of the assay.
Finally, the authors state that the mammaglobin mRNA concentrations were only marginally associated with increased nodal status and not with other tumor or patient characteristics. We believe that this is caused by the fact that tumors from patients with a more favorable prognosis express higher numbers of mammaglobin transcripts per cell, as we described previously (6). Our results indicated that cells from estrogen receptor-positive, low-grade breast tumors are more likely to be detected when in circulation. These tumors express the highest number of mammaglobin mRNA molecules per cell. In contrast, estrogen receptor-negative and high-grade tumors express lower numbers of mammaglobin mRNA molecules per cell and are more likely to escape detection. The more than 10 000-fold variation in mammaglobin mRNA concentrations, coupled with its association with certain tumor characteristics, will lead to a divergence in detection probability for particular tumor types. Because the mammaglobin concentrations per cell and the likelihood to disseminate are inversely correlated, there might be no association of mammaglobin mRNA concentrations in blood with tumor or patient characteristics.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
