Validation of Breast Cancer Biomarkers Identified by Mass Spectrometry

To the Editor:

Li et al. (1) should be congratulated for a valiant effort to validate 3 previously identified serum breast cancer biomarkers by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS).

Because there is considerable controversy on the value of this technology for cancer diagnostics (2)(3)(4)(5)(6)(7)(8)(9)(10)(11), it is important to comment on validation studies aiming to reproduce previously published data. Among 3 previously reported biomarkers, BC1, BC2, and BC3, one of these (BC1) was not confirmed, as it was previously shown to be decreased in breast cancer, whereas in the validation study by Li et al. (1), it was increased.

The other 2 candidate biomarkers, BC2 and BC3, were positively identified, by tandem MS, as complement C3a lacking its C-terminal arginine (C3a_desArg). BC2 was also identified as a truncated form of C3a_desArg.

In my opinion, the data presented in Fig. 4 of the article by Li et al. (1), showing the relative intensities of BC2 and BC3 in various groups of patients, are rather disappointing. For BC2, there is no difference between patients with benign breast diseases and patients with invasive carcinomas, although an increase was seen in ductal carcinoma in situ (DCIS). For BC3, there was no difference among patients with benign disease, DCIS, or invasive carcinomas.

The remaining question concerns the possible value of complement C3a_desArg and its fragment as candidate breast cancer biomarkers. The data provided by the authors (1) confirm my previous predictions that SELDI-TOF–identified biomarkers represent high-abundance proteins (in this case, C3, present in serum at concentrations of ∼1.2 g/L) that are produced mostly by the liver (3)(4)(5)(6). The proteolytic processing of peptides in the circulation by amino- and carboxypeptidases is well known, and it should not be surprising that the identified molecules represent modified and/or truncated forms of C3a.

I have previously speculated that a large number of SELDI-TOF–identified candidate biomarkers are acute-phase reactants (3)(4)(5)(6). C3, in accordance with my previous predictions, is also an acute-phase reactant whose serum concentration is increased or decreased in a wide variety of clinical conditions (12).

I conclude that the positive identification of previously described candidate serum biomarkers, BC2 and BC3, confirms my previous predictions that these are high-abundance proteins produced by the liver and that they represent nonspecific biomarkers of acute-phase reaction. Their performance as breast cancer biomarkers, as assessed by SELDI immunoassay, is not impressive and likely of questionable clinical value.