A tumor marker has been defined as “any biological molecule produced either by a tumor cell itself or by a tissue of the body in response to tumor invasion that can be objectively measured in body fluids and tissues and used as an indicator of the tumor process” [1]. Serum tumor marker assays provide advantages such as being relatively low-cost, being minimally invasive and to be amenable to the development of automated tests. However, serum tumor markers often lack in both sensitivity and specificity. Sensitivity is particularly limited in detection of early-stage tumors and specificity may be affected by markers being elevated in various nonmalignant conditions such as inflammation. The clinical value of serum biomarkers can be increased when they are used together with other types of investigations such as imaging [[2], [3], [4]].
Protein biomarkers have a long history. Acid phosphatase was described as a marker for prostate carcinoma in 1938 [5], lactate dehydrogenase as a general tumor marker in 1954 [6] and cytokeratins originally described as tumor markers in 1957 [7]. Despite their long history and wide-spread use, protein tumor biomarkers are nevertheless somewhat controversial. Prostate specific antigen (PSA) is widely used in prostate cancer diagnostics with first decription by Wang et al. in 1979 [8]. The PSA test does, however, suffer from limited specificity, leading to overdiagnosis of low-risk cancers [9]. The positive predictive value (PPV) of PSA as a screening test was reported to be 22% in a recent meta study [10]. Large scale screening programs using PSA have therefore not been widely implemented [11]. Improved detection of high-grade prostate cancer can be achieved when additional protein biomarkers, genetic markers, family history and prostate examination are included in the clinical investigation [12]. CA15-3 is a commonly used marker for breast cancer but levels are rarely increased in patients with early disease and the main value of this marker is therefore for disease monitoring [13]. Similarly, CA125 is useful in the management of ovarian cancer patients but is of limited value for detection of early stage disease [14]. The power of CA125 can be increased if patients with increased levels are subjected to ultrasound investigations [15] (for further discussions, see below). A number of tests based on detection of cytokeratins/keratins are available (TPA®, TPS®, CYFRA21-1, M65® ELISA, M30 Apoptosense® ELISA) but, similar to other tumor markers, their sensitivities and specificities are limited and their main application is for monitoring purposes [16]. Cytokeratin markers will be discussed further below as bladder cancer markers using urine samples. In summary, serum tumor markers are generally not useful as tools for detection of cancer at early stages, due to either lack of sensitivity and specificity, or both but can be quite valuable for monitoring the efficiency of cancer therapy [17]. For a general discussion of this subject, see [[2], [3], [4]].
When tumor markers are not used as screening tools but instead used during clinical investigations of suspected malignancy, their value is increased. In a study involving 2711 consecutive patients with suspected cancer, a panel of tumor markers were used during subsequent clinical investigations. The use of tumor markers resulted in a sensitivity of 67% for detection of cancer with a specificity of 98% [18]. The sensitivity was 75% for patients with epithelial tumors and 81% in patients with cancer of unknown primary site. Involuntary weight loss is a condition that is associated with malignancy [19] and the value of serum tumor markers during investigations of the underlying reasons for such weight losses has been examined. Trapé et al. examined patients a cohort of 606 patients with involuntary weight loss according to tumor biomarker positivity [20]. Using a panel of 8 markers, an NPV of 96% and a PPV of 39% was observed for patients positive for at least one tumor biomarker. A more recent study included patients with involuntary weight loss and elevated levels of inflammation biomarkers (> 5 mg/L CRP and at least one additional marker) [21]. Including tumor marker panels in the follow-up of these patients showed a PPV of 55% and an NPV of 84%.
Comments (0)