Once the decision to request any tumour marker measurement has been made, the quality of the result eventually reported will reflect events during the pre-analytical, analytical and post-analytical phases of analysis. Quality requirements will generally be similar to those for any other laboratory test, but it is useful to consider in turn each phase of the analytical process. All of these requirements have previously been well-discussed in many excellent reviews, some of which are listed at the end of this article. The purpose of the present review is to summarise very briefly those aspects which are of particular practical importance to those either providing or using a tumour marker laboratory service.

pre - analytical requiremants

Aspects of particular relevance to tumour marker testing are highlighted in Table 1. For most tumour marker measurements the timing of blood sampling is usually not critical. However, inappropriate timing of testing (e.g. measuring PSA in a patient with prostatitis, or taking specimens for hCG too soon after chemotherapy) can result in misleadingly high results which may cause undue alarm and distress to the patient, as well as decreasing confidence in laboratory testing.

For certain analytes, specimen type (e.g. whether plasma or serum) and storage (e.g. length of time on the blood clot) may be important, and kit manufacturers instructions should always be followed. Clinicians may not always be aware of these details, and it is the responsibility of the laboratory to keep them informed.

analytical requiremants

Satisfactory measurement of any analyte requires that the correct and appropriate specimen is analysed by a method meeting defined quality requirements for both Internal Quality Control (IQC) and External Quality Assessment (EQA). Some issues of particular importance for tumour marker measurements are listed in Table 2.

The laboratory should also be aware of analytical pitfalls specific to the given analyte. Three hazards which are of relevance to all tumour marker measurements, and which have the potential for causing serious clinical misdiagnosis, are shown in Table 3.

post - analytical requiremants

As analysis of tumour markers becomes more reliable, largely due to advances in automation, more time and attention can be focused on how to achieve most effective clinical use of these important tests. This requires attention to the post-analytical process: aspects of particular relevance to this are described in Tables 4 and 5.


It has only been possible here to touch upon the many topics of major importance to provision of a high quality analytical service for tumour marker measurement. It is intended that this article will form the basis of a much more detailed and well-referenced review of these issues.

background literature

  1. Ballesta AM, Torre GC, Bombardieri E, Gion M, Molina R: Updating on tumor markers in tissues and in biological fluids. Basic and clinical applications. 1993; Edition Minerva Medica: Turin.
  2. Fateh-Moghadam A, Stieber P: Sensible use of tumour markers. 1993; Jurgen Hartmann Verlag GmbH: Marloffstein-Rathsberg.
  3. Hayes DF, Bast RC, Desch CE, Fritsche H, Kemeny ND, Jessup JM, et al: Tumor marker utility grading system: a framework to evaluate clinical utility of tumor markers. J Nat Cancer Inst 88: 1456-1466, 1996.
  4. Klapdor R, ed. Tumor markers in clinical oncology: an overview. 1994; Sorin Biomedica SpA: Turin.
  5. Lamerz R, Dati F, Seller AC, Schnorr G. Tumordiagnostika. 1988; Behring Diagnostika: Behringwerke AG, Marburg.
  6. Mugan K, Carlson IH, Westgard JO: Planning QC procedures for immunoassays. J Clin Immunoassay 17: 216-222, 1994.
  7. Pannall P, Kotasek D: Cancer and Clinical Biochemistry. 1997; ACB Venture Publications: Cambridge
  8. Riesen WF, Keller H: Definition of the performance of tumor marker tests: Principal considerations. J Tumor Marker Oncol 8: 15-XX, 1993.
  9. Selley S, Donovan J, Faulkner A, Coast J, Gillatt D: Diagnosis, management and screening of early localised prostate cancer. Health Technology Assessment 1: 1-96, 1997.
  10. Seth J. Quality Assurance. In: Price CP, Newman DJ, eds. Principles and Practice of Immunoassay. 2nd ed. London: Macmillan, 1997, pp 209-241
  11. Sturgeon CM, Seth J: Why do immunoassays for tumour markers give differing results? -- A view from the UK National External Quality Assessment Schemes. Eur J Clin Chem Clin Biochem 34: 755-759, 1996.
  12. Tumor Marker Expert Panel (AACC): Tumor markers: Reclassification, reimbursement and recent advances. 1998; AACC: Washington DC.
  13. Tumor Marker Expert Panel (ASCO): Clinical practice guidelines for the use of tumor markers in breast and colorectal cancer. J Clin Oncol 14: 2843-2877, 1996.
  14. Van Dalen A: Quality control and standardization of tumour marker tests. Tumor Biol 14: 131-135, 1993.

Table 1. Pre-analytical considerations of relevance when measuring tumour markers

Timing of specimen collection Pre-treatment specimen always desirable.

No strong evidence of diurnal variation for most markers, so specimens can be taken at any time of day.

  Timing post-operatively (e.g. CA 125 may be increased by peritoneal trauma).

Avoiding sampling during menses (important e.g. for CA 125 when used in for screening those at high risk).

  Prostatic biopsy / transurethral resection of the prostate (TURP), catheterisation or acute painful urinary retention may markedly elevate serum PSA, but the magnitude of the increase in PSA levels is variable. Blood should be drawn before any manipulation of the prostate.

Prostatitis may also increase serum PSA levels. Specimens should be taken several weeks after resolution of prostatitis or urinary tract infections, and raised concentrations in patients with such infections should always be confirmed by repeat analysis post-treatment.

Digital rectal examination (DRE) may transiently elevate serum PSA levels; best to take blood prior to DRE. May also be increased post-ejaculation (note time post-ejaculation where relevant).

Effect of other treatment / medication Immunometric methods vulnerable to interference from human anti-mouse antibodies (HAMA); previous treatment with monoclonal antibodies should be noted on request form.
Effect of renal failure / impairment Not much documented evidence. Published reports suggest most likely to cause inappropriately elevated results for CEA, TPS and other cytokeratins.
Effect of cholestasis May markedly increase serum CA 19.9 concentrations.
Contamination with saliva May markedly increase apparent concentrations of CA 19.9, SCC, CEA, and TPS.
Type of specimen Serum or plasma generally are most appropriate and suited to most commercial assays, but requirements should be checked in kit inserts.

Differences between results in serum and EDTA plasma may be due to complement effects, but there is little documentation of this, or of the effects of gel tubes.

Stability of specimen on storage Generally stable, although separation of serum from the clot and storage at +4° C (short term) or -30° C (longer term) desirable as soon as possible. Heating of specimens usually undesirable (e.g. hCG, PSA).
  For PSA, separation of serum from the clot and storage at +4° C (short term) or -30° C (longer term) desirable as soon as possible, and preferably within 3 hours of sampling, especially if free PSA is measured. Samples may be stored at refrigerated temperatures for up to 24 hours, but if analysis is delayed beyond 24 hours, specimens should be stored frozen (at least at -20° C). For longer term storage, specimens should be stored frozen at -70° C. 

Table 2. Analytical considerations of particular relevance to the quality control of tumour marker measurements Requirements of Internal Quality Control (IQC)

Assessment of reproducibility Demonstration of intra-assay variability < 5%; inter-assay variability <10%.
Established criteria for assay acceptance Selection of appropriate criteria for acceptability of IQC, preferably based on logical criteria e.g. those of Westgard.
Specimens closely resembling authentic patient sera In general it is inadvisable to rely exclusively on QC materials supplied with the kit, and an authentic serum matrix control from an independent source should be included.
IQC specimens of concentration appropriate to the clinical application Negative and low positive controls should be included for all tumour markers, but there is also a need to cover the broader concentration range, and to assess accuracy of dilution steps required for high concentrations specimens.
Assessment of assay interferences Occasionally checking for interferences (eg from heterophilic and other antibodies, clotting agents in blood clotting tubes) is desirable.
Requirements of External Quality Assessment (EQA)
EQA specimens of appropriate analyte concentration Concentrations covering the working range of assays are adequate, although occasional inclusion of higher concentrations to check behaviour on dilution is desirable. Issue of normal analyte-free serum to check baseline security important for some analytes (e.g. AFP, hCG). 
Assessment of assay "stability" Assessment of the "stability" of results within a laboratory can readily be accomplished by issuing repeat specimens of the same pool and comparing results over time (eg 6-12 months).
Demonstrating accuracy and stability of target values These are usually consensus means, as reference methods are not available for these analytes. The validity of the consensus means should be demonstrated by assessing their:

Stability, by repeat distribution of the same pool.

Accuracy, by recovery experiments undertaken by supplementing pools with known amounts of the relevant International Standard (IS). [Such validation is possible for AFP, hCG and CEA, for which there are currently accepted IS (BS 72/225, IS 75/537 and IRP 73/601 respectively). The lack of IS for other tumour markers is currently being addressed.]

Interpretative exercises These provide a valuable means of comparing practice (eg reference ranges, cumulation of results) in different laboratories.

Table 3. Potential causes of erroneous tumour marker results

High dose hook effect Tumour marker concentrations routinely encountered range over several orders of magnitude. Protocols permitting identification of high dose ‘hooking’ are essential to avoid reporting misleadingly low results, particularly in patients for whom markers are being measured for the first time. [Hook effects can be minimised by using solid-phase antibodies of higher binding capacity, by assaying specimens at two dilutions, or by using sequential assays which include a wash step.]
Specimen carry-over Potentially a problem whenever very high concentration specimens assayed, so should occasionally be checked for.
Interference from heterophilic or human anti-mouse antibodies (HAMA) Falsely high or low results may be obtained for patient specimens containing anti-immunoglobulin G (IgG) antibodies capable of reacting with antibodies used in the assay. Presence of HAMA, frequently induced in cancer patients who have undergone treatment with mouse monoclonal antibodies for imaging or therapeutic purposes, may also give erroneous results. [Such interference can be detected by re-assaying the specimen after treatment with blocking agents (which are commercially available immobilised on tubes), by adding further non-immune mouse serum to the reaction mixture, or by re-assaying the specimen by a different method.]

Table 4. Post-analytical requirements of particular importance to provision of a comprehensive tumour marker service

Factual requirements

Clinical information from the requesting doctor Encouraging clinicians to provide very brief clinical information (eg "post-operative", "post-chemotherapy 5") is essential if any interpretation is to be provided, and may help to identify occasional laboratory errors (eg mis-sampling on an analyser). 
Availability of appropriate reference ranges Usually derived from an appropriately matched healthy population, reference ranges for tumour markers are usually most relevant for cancer patients pre- and immediately post-initial treatment. Subsequently, the patient’s own "baseline" provides the most important reference point for interpretation of marker results. If this is well-established, increases even within the reference range may be clinically significant.
Knowledge of what constitutes a significant or clinically relevant change Should include contributions of both biological variation and analytical variation. A confirmed increase or decrease of ± 25% is frequently considered to be of clinical significance, but further work is required in this area. [See e.g. Soletormos G et al. Interpretation of results for tumor markers on the basis of analytical imprecision and biological variation. Clin Chem 39: 2077-2083, 1993; Bonfrer JMG: Working Group on Tumor Marker Criteria (WGTMC). Tumor Biol 11: 287-288, 1990.]
Defined protocol when changing methods


It may be helpful to the laboratory to indicate changes of method on tumour marker reports, but it is more helpful if the laboratory highlights whether any change is likely to have affected interpretation of the trend in marker result. [This may necessitate eg analysis of the previous specimen by the new method, or requesting a further specimen to re-establish the baseline and/or confirm the trend in marker level.]
Knowledge of tumour marker half-lives Defined as the time to 50% reduction of circulating tumour marker concentrations following complete removal of tumour tissue (ie distinct from definition as used in other settings). Of most relevance to interpretation of serum concentrations of certain tumour markers, eg AFP and hCG. [See later detailed papers]
Objective comparison of tumour marker utility The need to obtain objective clinical information about tumour marker utility remains a priority, and is being considered by a number of professional organisations. [See eg Hayes DF et al, Tumor marker utility grading system: a framework to evaluate clinical utility of tumor markers. J Nat Cancer Inst 88: 1456-66, 1996.]

Table 5. Reporting requirements for provision of a comprehensive tumour marker service

Provision of fully cumulated results, since it is always the trend in marker concentration that is most informative. Minimal clinical details should appear on the report, to facilitate interpretation. Graphical representations may also be very helpful.
Recommendations as to the appropriate frequency of marker measurements. [See later detailed papers]
Recommendations as to when confirmatory specimens should be requested. An apparent rise in marker concentration should always be confirmed by repeat measurement.
Good communication between laboratory and clinical staff, which facilitates appropriate use of these (and other) tests.

Consult an expert

If you need more specific information on a topic on this page, please visit the forum-page and ask one of our professionals.