Skip to main content

Archived Comments for: Differential profiling of breast cancer plasma proteome by isotope-coded affinity tagging method reveals biotinidase as a breast cancer biomarker

Back to article

  1. Is Biotinidase Really a Biomarker for Breast Cancer?

    Barry Wolf, Department of Medical Genetics, Henry Ford Hospital, Detroit, Michigan

    8 July 2010

    Drs. Un-Beom Kang and colleagues recently reported that biotinidase may be a breast cancer biomarker [1]. The authors used affinity chromatography to select cysteine-containing peptides/proteins to identify a group of proteins in plasma that show differences between populations of women with breast cancer and those who are unaffected. The authors found that by comparing biotinidase in 21 patients with breast cancer and 21 normal healthy controls, biotinidase was statistically down-regulated in the plasmas from the women with breast cancer. The authors thereby concluded that biotinidase is a “potential seriological biomarker for the detection of breast cancer.” In the last paragraph of the discussion, the authors indicate that biotinidase catalyzes the release and recycles endogeneous biotin, and that the enzyme is known to be secreted into blood, and is “highly active’ in serum, liver, kidney and adrenal glands. The authors also recognized that “it is, yet, difficult to explain how biotinidase is down-regulated in breast cancer plasma.” I too had similar concerns. However, the authors should consider various aspects of biotinidase metabolism and the properties of biotinidase in plasma/serum.

    First, plasma biotinidase is synthesized mainly in the liver. This has been confirmed by the markedly decreased biotinidase activity in sera of rats that have had partial hepatectomy, removing two-thirds of their liver, or in rats treated with carbon tetrachloride to induce necrosis [2;3]. Moreover, individuals with cirrhosis have decreased plasma biotinidase activity that correlates positively (r = 0.9) with the decrease in their serum albumin concentrations [4;5]. Therefore, even if the production of biotinidase is decreased in breast cancer tissue, it seems difficult, if not impossible, to explain how such a decrease in such a relatively small diseased tissue, especially without hepatic metastases, would affect biotinidase activity or its protein concentration in plasma in these women. The volume of blood in an average woman is 4.7 liters and their mean hematocrit is 38 %. Therefore, women have a serum/plasma volume of about 2.9 liters. Even if the synthesis of biotinidase is decreased by 50 % in the affected breast tissue as described, it is essentially impossible to conceive how this would alter the serum/plasma biotinidase activity or CRM! Even if the breast tissue produced a metabolite that was secreted from the tissue, was diluted in plasma, and made its way to the hepatocytes, how could it produce a decrease in synthesis of 50 %! Based on this, one should look for another, more reasonable explanation for the results reported in the paper.

    Through newborn screening of biotinidase deficiency and its confirmation by determining enzymatic activity in plasma/serum over the past 25 years, we have learned much about the enzyme’s stability in plasma [6;7]. Multiple children who were initially diagnosed with profound biotinidase deficiency (less than 10 % of mean normal adult activity) were subsequently determined to have normal biotinidase activity. These misdiagnoses were likely due to the fact that the enzyme in plasma is very sensitive to temperature. Serum/plasma samples must be stored at -80 ºC to retain their enzymatic activity. Storage at -20 ºC or higher results in marked loss of activity over various lengths of time. However, I am not aware whether the biotinidase protein (cross-reacting material, CRM, to antibody prepared against biotinidase) concomitantly decreased under these conditions. These considerations prompt querying the authors about the following:
    1. Were the plasma samples from both the individuals with breast cancer and the normal controls used for the determinations processed immediately after obtained and stored at -80 ºC?
    2. How were the various samples obtained and stored? Could they have remained at suboptimal temperatures for extended lengths of time before freezing, even if optimally frozen at -80 ºC?
    3. Were the plasma samples from the controls more recently obtained than those of individuals with breast cancer? I can only imagine that the samples from the women with breast cancer were obtained over a long period of time.
    4. Why didn’t the authors measure biotinidase activity rather than or in addition to CRM? Measuring activity is much easier, rapid and more specific than measuring CRM. Did the authors demonstrate that the concentration of CRM was linear across the range studied? Perhaps they should determine if the ratio of activity to CRM remains the same as normal if the enzyme is not being synthesized or is being degraded, or if the ratio of activity to CRM decreases compared to normal if the enzymatic activity is disproportionately decreased relative to the CRM.
    5. The range of normal activity in a population is very wide and may vary by age [8;9]. When examining the results of the plasma biotinidase CRM studies, such as in Figure 2, one sees hints of this variation in the normal individuals with half of them having CRM that is about the same as the mean of those with cancer. Moreover, because of the variation of the determinations and the small sample size shown in Figure 4C, it is likely that the differences between the cancer patients and the normal individuals are not significant with the exception of those with grade 1 cancer.

    Although it is important to identify biomarkers that can be useful in diagnosing presymptomatic individuals with a specific disorder, when a putative biomarker is identified, it is imperative to attempt to explain why the analyte would be a likely marker based on its metabolism or biochemical properties. Neglecting these factors may cause inappropriate assumptions about its significance.


    1. Kang U-B, Ahn Y, Lee JW et al. Differential profiling of breast cancer plasma proteome by isotope-coded affinity tagging method reveals biotinidase as a breast cancer biomarker. BMC Cancer 2010; : .
    2. Pispa J. Animal biotinidase. Ann.Med.Exp.Biol.Fenn. 1965; 43(Suppl.5): 1-39.
    3. Weiner DL, Grier RE, Watkins P, Heard GS, Wolf B. Tissue origin of serum biotinidase: Implication in biotinidase deficiency. American Journal of Human Genetics 1983; 34: 56A.
    4. Grier RE, Heard GS, Watkins P, Wolf B. Low biotinidase activities in the sera of patients with impaired liver function: Evidence that the liver is the source of serum biotinidase. Clinical Chimica Acta 1989; 186: 397-400.
    5. Pabuccuoglu A, Aydogdu S, Bas M. Serum biotinidase activity in children with chronic liver disease and its clinical implications. Journal of Pediatric Gastroenterology and Nutrition 2002; 34: 59-62.
    6. Hymes J, Fleischhauer K, Wolf B. Biotinidase in serum and tissues. Methods in Enzymology 1997; 279: 422-32.
    7. Wolf B. Clinical issues and frequent questions about biotinidase deficiency. Mol.Genet.Metab. 2010; 100:6-13.
    8. Wolf B, Grier RE, Secor McVoy JR, Heard GS. Biotinidase deficiency: A novel vitamin recycling defect. Journal of Inherited Metabolic Disorder 1985; 8 (Suppl 1): 53-8.
    9. Weissbecker KA, Nance WE, Eaves LJ, Piussan C, Wolf B. Detection of heterozygotes for biotinidase deficiency. American Journal of Human Genetics 1991; 39: 385-90.

    Competing interests