|
|
||||||||
Reviews |
1 Department of Clinical Biochemistry, Royal Infirmary of Edinburgh, Edinburgh EH3 9YW, United Kingdom. Fax 44-131-536-2765; e-mail C.Sturgeon{at}ed.ac.uk.
| Abstract |
|---|
|
|
|---|
Methods: Recent national and international guidelines relating to the use of tumor markers in germ cell, colorectal, breast, ovarian, prostate, lung, neuroendocrine, and thyroid cancers were identified from the scientific literature and other sources and tabulated.
Results: Guideline recommendations developed by national and international groups and relating to the use of tumor markers for specific cancers are reviewed and compared, considering the recommendations made for their use in screening, diagnosis, prognosis, and monitoring of therapy. Potential advantages and disadvantages of clinical guidelines, how best to implement them, and means of auditing their effectiveness are also considered.
Conclusions: Excellent clinical guidelines, including recommendations for the most appropriate use of tumor markers, are already available for many cancers. Many questions relating to optimal use of these important tests remain to be answered, but current guidelines already contain much valuable information and advice. Further dissemination and implementation of the guidelines should encourage better use of tumor markers in clinical practice. Careful audit studies are also required to establish the impact of these guidelines on the practice of evidence-based medicine.
| Introduction |
|---|
|
|
|---|
Oncology is one important area in which there is increasing interest in outcome- and evidence-based medicine, and several international, national, and regional expert groups have been working to develop guidelines with the ultimate aim of improving the standard of patient care. Because many different specialties contribute to patient care, a multidisciplinary approach to guideline development is always desirable. Inevitably, however, some guidelines focus primarily on clinical aspects (e.g., surgical or medical procedures), whereas others are more specifically laboratory oriented, but the same general principles apply to the development of both. In this review, I briefly outline these principles before reviewing a selection of currently available guidelines for some commonly measured serum tumor markers and considering how these guidelines might best be implemented and improved. The guidelines are of necessity described only in outline because of space constraints; for more detailed information, the reader should consult the original reports, which are often conveniently provided on the Web.
| Development of Practice Guidelines for Tumor Markers |
|---|
|
|
|---|
The availability of such a large literature base means that development of tumor marker guidelines can be particularly helpful because these provide condensed and relevant information in a format convenient for the practicing clinician or laboratorian. The subject area is also reasonably well defined, thus fulfilling one of the primary requirements for guideline development (4). A suitably diverse and multidisciplinary group of individuals with particular interest in the clinical application of tumor markers, a means of identifying and assessing available evidence, and an ability to distill conclusions into readily understood practice guidelines are also necessary (4). Small groups working under the auspices of professional organizations are generally involved, with group members and their affiliations being identified in the reports.
Some organizations have focused on applications of tumor markers in specific cancers, e.g., the American Society of Clinical Oncology (ASCO)1 with breast and colorectal cancer (6)(7)(8), whereas others have considered their use in a range of cancers, e.g., the National Academy of Clinical Biochemistry (NACB) (9)(10) and the European Group on Tumor Markers (EGTM) (10)(11). In France the "Standards, Options and Recommendations" (SOR) project, which started in 1993, has involved fruitful collaboration of the Federation of the French Cancer Centres (FNCLCC), the 20 French Regional Cancer Centres, several French public university and general hospitals, private clinics, and medical specialty societies, yielding impressively thorough guidelines for several malignancies (12)(13)(14)(15)(16). The French methodology (17), like that of other groups, is based on literature review, followed by critical appraisal by a multidisciplinary group of experts and finally validation of the draft guidelines by specialists in cancer care delivery.
Detailed review of published reports is fundamental to guideline development, and many reports presenting guidelines for tumor markers provide useful and comprehensive summaries of relevant literature. Although quantitative systematic reviews, in which articles are inspected for susceptibility to bias and metaanalyses are performed (4), may be preferable, there are as yet few systematic reviews specifically relating to tumor markers on the Cochrane Library database, and generally conventional literature searches have been carried out.
When reviewing the literature, some assessment of relevance and possible bias is important, and ideally, some statement should be included to indicate the strength of the evidence on which recommendations have been made (4). Available guidelines for tumor markers vary considerably in this respect, with the information provided by the SOR (17), ASCO (6)(7), and Scottish Intercollegiate Guideline Network (SIGN) (18) groups being among the more rigorous. Some groups make no statement about the level of evidence accepted.
| Guideline Recommendations |
|---|
|
|
|---|
germ cell tumors
-Fetoprotein (AFP), human chorionic gonadotropin (hCG), and lactate dehydrogenase are well-established tumor markers integral to the successful management of patients with testicular and other germ cell tumors (19), so it is not surprising that for these markers, guidelines generally agree well (Table 1A
). This may reflect longstanding close collaboration among clinicians treating patients with these relatively rare tumors, international consensus conferences including consideration of the use of tumor markers in germ cell cancers having been held for more than 20 years (20). The availability of effective and curative treatment and the fact that changes in marker concentrations reliably reflect and predict clinical response, to the extent that serology may predominate over histology in treatment decisions, makes pre- and posttreatment determination of tumor markers mandatory.
|
Possibly reflecting different practice in Europe, the EGTM also recommends measurement of serum placental alkaline phosphatase because it is increased in up to 80% of testicular seminomas, only 20% of which produce hCG (21). The measurement of serum placental alkaline phosphatase cannot be recommended in smokers, however, because concentrations are increased up to 10-fold relative to those of nonsmokers, with considerable interindividual variation. The European Association of Urology (EAU) regards measurement of placental alkaline phosphatase and neuron-specific enolase (NSE) as optional in seminoma cases (22).
Studies by the International Germ Cell Cancer Collaborative Group (IGCCCG) have confirmed the prognostic value of AFP, hCG, and lactate dehydrogenase, which are now included in the staging systems of the American Joint Committee on Cancer (AJCC) and the Union Internationale Contre le Cancer (UICC) (10)(23). Pretreatment marker concentrations contribute to the classification of metastatic germ cell tumors as having good, intermediate, or poor prognosis (23).
Posttreatment monitoring is also essential for optimal care, and tumor marker normalization is required to assess the response to chemotherapy. The rapidity of decreases in tumor marker concentrations in the first 6 weeks of chemotherapy can predict the potential for relapse months later, and weekly measurements during chemotherapy are recommended. Follow-up schedules clearly depend on the histology, stage, and postorchidectomy treatment chosen. Broadly, many guidelines recommend monthly measurements for the first year after treatment for advanced disease, with measurements every 2 months in the second year, every 2 or 3 months in the third year, and then every 6 months up to 5 years, but there are slight variations. A Working Group of the EAU has recently published comprehensive guidelines for the diagnosis, staging, treatment, and follow-up of germ cell testicular tumors (22). These include, in helpful tabular form, detailed recommendations about timing of posttreatment surveillance of patients with different stages of disease and histology, incorporating recommendations for tumor marker measurement with those for physical examination and imaging techniques (physical examination, chest x-ray, and abdominal computed tomography scans and ultrasound).
The decreases in tumor marker concentrations after orchidectomy contribute to the management of patients with germ cell tumors (24) and can be determined from the serial marker measurements recommended by most groups. The NACB and EGTM guidelines specifically state that the rate of decrease should be calculated and compared with the normal rates of disappearance of AFP (half-life <7 days) and hCG (half-life <3 days) (9)(10)(21).
In relation to laboratory provision, the NACB and EGTM guidelines emphasize that both intact hCG and its free ß-subunit should be recognized by hCG assays used in oncology and that good baseline stability is essential for AFP and hCG because treatment may be instituted on the basis of relatively small increases in marker concentrations (9)(10)(21).
colorectal cancer
Several groups have considered the management of colorectal cancer, for which carcinoembryonic antigen (CEA) is at present the most relevant serum tumor marker. Although early detection is the key to control of colorectal cancer, there is unanimous agreement that, because of a lack of specificity and sensitivity for early disease, CEA has no role in either screening or early diagnosis (Table 1B
). Improved early detection is most likely to result from greater awareness of symptoms by patients and their physicians (25)(26) and possibly through the implementation of fecal occult blood screening programs, such as those recommended by the American Cancer Society (27).
In patients with colorectal cancer, CEA measurement before surgery is recommended by the ASCO because the result may complement pathologic staging and aid surgical treatment planning (7), and both the AJCC (28)(29) and College of American Pathologists (CAP) (30)(31) have proposed that CEA be included in the staging system for colorectal cancer. These recommendations are in accord with most guidelines (Table 1B
) and are based on many studies indicating that abnormal preoperative CEA values are associated with a higher risk of recurrence. However, there is as yet no evidence that patients benefit from adjuvant therapy solely on the basis of abnormal preoperative CEA concentrations.
The role of CEA in early detection of recurrence also remains controversial, as reflected in the differing guideline recommendations (Table 1B
), probably because few studies indicate that early intervention improves outcome. This may change because results of two recent metaanalyses suggest a statistically significant difference in cumulative 5-year survival in patients undergoing intensive follow-up (32)(33), with the second study suggesting improvement only if CEA assays are included (33). For patients in whom resection of isolated hepatic metastases would be clinically indicated, the ASCO, EGTM, and NACB guidelines all recommend measurement of CEA postoperatively every 23 months for the first 2 years after diagnosis (Table 1B
), the ASCO guidelines specifying that such patients should have stage II or stage III disease. If an abnormal CEA is confirmed, additional evidence of metastatic disease should always be sought before initiating therapy. The same groups also recommend use of CEA in monitoring patients with advanced colorectal cancer and in monitoring response to treatment, with the ASCO and EGTM suggesting measurements at 2- to 3-month intervals for the first 2 years after initial diagnosis. At least two successive CEA results above baseline are required to document progressive disease. Although CEA measurements should not be used in isolation to select treatment, they can provide additional information that may, for example, identify ineffective therapy that can be discontinued (10) and can also be particularly helpful when disease is not readily assessable by other means (34).
breast cancer
Several groups have made helpful recommendations about the use of both tissue and serum breast cancer markers (Table 1C
). Not surprisingly, there is unanimous agreement that determination of estrogen and progesterone receptor status is essential in primary lesions from all patients (pre- or postmenopausal) to identify those likely to respond to endocrine therapy. In metastatic breast cancers, however, receptor concentrations should be determined only if results will influence treatment decisions. Although not yet incorporated in all guidelines, there is also general agreement that determination of HER-2/neu (c-erbB-2) overexpression is mandatory in selecting patients for Herceptin® therapy (7)(10).
Numerous studies have confirmed that CA15-3 and BR27.29 are the best available serum markers for breast cancer, but this is another disease where application of currently available serum tumor markers is limited by low sensitivity in early-stage disease, lack of specificity, and controversy about whether their measurement benefits outcome (10). These markers are therefore not recommended by any group for screening, diagnosis, or staging of breast cancer (Table 1C
). Although high concentrations usually indicate metastatic disease, there are as yet insufficient data to allow incorporation of tumor markers into the breast cancer staging system (28).
Both CA15-3 and BR27.29 have Food and Drug Administration approval for monitoring breast cancer patients with advanced disease, but such approval does not necessarily imply clinical value (6). Because the benefit of posttreatment monitoring is somewhat controversial (the merits of measuring any tumor marker in the absence of effective therapy can be debated), recommendations from different groups vary (Table 1C
). Major obstacles identified in relation to the use of CA15-3 as an indicator of asymptomatic recurrence include the low incidence of CA15-3 in early-stage disease, the lack of effective treatment options for recurrences detected, and the low efficiency of detection (6). Although not recommending routine use of CA15-3 or BR27.29 alone for monitoring response to treatment, the ASCO guidelines support use of these markers to suggest treatment failure where disease is not readily measurable (6)(7)(8). The NACB, EGTM, and SOR guidelines all recommend that CA15-3 or BR 27.29 determinations be used with caution as an aid in monitoring the clinical course of breast cancer patients (9)(10)(12)(35). Both the EGTM and SOR guidelines also recommend the use of CEA in breast cancer, although the SOR guidelines stipulate that CEA should only be measured if CA15-3 is not increased at presentation (12).
ovarian cancer
CA125 is the best available serum marker for epithelial ovarian cancer, and available guidelines generally agree well (Table 1D
). Although not sufficiently sensitive or specific to be used for screening or diagnosis, CA125 may have a role in screening for malignancy, preferably in a clinical trial setting in women with a strong family history of ovarian cancer (36), as recommended in an NIH Consensus Statement (37). CA125 can also contribute to the differential diagnosis of pelvic masses in postmenopausal women. It is interesting to note that the French SOR guidelines suggest measurement of CEA or CA 19.9, especially in mucinous or endometrioid tumors, but only if CA125 is not increased at presentation (13). The SOR guidelines also recommend measurement of AFP and hCG in younger women to exclude a germ cell tumor (13).
Numerous trials have established the prognostic value of the rate of decrease in CA125 after cytoreductive surgery and during cytotoxic chemotherapy (10), and some guidelines recommend its prognostic use. However, the AJCC traditionally has not incorporated responses to treatment in staging, so it does not recommend inclusion of CA125 in the current TNM staging system for ovarian cancer (28).
In the absence of curative treatment for relapsed disease, there is no consensus as to whether or how often CA125 should be measured, and there are no data defining the frequency of surveillance (13). The SOR guidelines (13) recommend that if there is an increase in a previously normal CA125 concentration, the assay should be repeated after 23 weeks to confirm the increase and to calculate the time of doubling (24). If the increase is confirmed or if there are clinical signs of progression, a computed tomography scan should be undertaken (13). It is important to be aware that disease can progress without an increase in CA125.
prostate cancer
Prostate-specific antigen (PSA) is undoubtedly the best-known tumor marker; its use in screening for prostate cancer has aroused major interest from public and health professionals alike. Continued debate about the benefits of screening [of nine major American organizations concerned with health, only three currently advocate routine testing for prostate cancer (38)] is reflected in the differing recommendations made (Table 1E
). Less controversial, and recommended in most guidelines, is the use of PSA in combination with digital rectal examination (DRE) as an aid to diagnosis, with definitive diagnosis always requiring biopsy. Although means of improving the diagnostic accuracy of PSA (e.g., by use of age-related references) are often described in the text accompanying published guidelines, they generally are not included in recommendations made.
Although not yet officially incorporated in the TNM staging system (28), PSA is widely used as a prognostic indicator in patient management and is generally recommended for monitoring patients with prostatic cancer (Table 1E
) (10)(38)(39)(40)(41). The use and interpretation of serial PSA data, however, present major challenges, e.g., not all patients with biochemical recurrence will develop metastatic disease (42), PSA may be unreliable in poorly differentiated tumors, and there is no effective therapy for endocrine-resistant disease. Further refinement of existing guidelines will be desirable as knowledge of how best to use this test improves. However, several published guideline include very helpful and detailed discussion of the appropriate interpretation of PSA under different clinical circumstances (38)(39).
lung, neuroendocrine, and thyroid cancers
Guideline recommendations for tumor marker use in some other malignancies are outlined in Table 1F
. Tumor markers are infrequently measured in lung cancer patients because of the limited therapeutic options available, but NSE has been recommended for the differential diagnosis of small cell lung cancer (43). Although follow-up of asymptomatic patients after primary therapy for lung cancer remains controversial, serial marker determinations can also provide evidence of complete resection and/or early recurrence (43).
Neuroendocrine tumors are rare, but tumor markers can contribute substantially to what are often difficult diagnoses, as reviewed in guidelines for the diagnosis and therapy of multiple endocrine neoplasia types 1 and 2 recently published by an international group of endocrinologists (44). NACB recommendations for pheochromocytoma and neuroblastoma are more laboratory oriented and include helpful information about interferences in relevant assay methods (Table 1F
) (10). The British Thyroid Association (45) and NACB (46) have also published practical guidelines relating to measurement of calcitonin and thyroglobulin, these complementing guidelines previously published by the American Thyroid Association (47).
| Future Directions for Tumor Marker Guidelines |
|---|
|
|
|---|
Guidelines on the management of specific cancers essentially provide valuable consensus documents that can be used as reference material for specialist groups drawing up local protocols, as recently emphasized (39) and exemplified in the Scottish recommendations (49). This approach, enabling integration of the best external evidence with individual clinical expertise and patients choice, is fundamental to the practice of evidence-based medicine (1) and helps to transfer "ownership" of guidelines to those who can implement them.
audit
Local ownership has been identified as critically important to the successful implementation of guidelines, which are useful only if well publicized, readily implemented, and widely adopted. Some indication as to whether the availability of guidelines relating to tumor marker use has had any beneficial effect can be obtained from well-designed local and national audit projects, but as yet few such studies have been reported. Limited surveys in the United Kingdom suggest that tumor marker test requesting is generally appropriate, but confirm a need for improved consensus, e.g., about appropriate timing of tumor marker measurements, what constitutes a clinically significant change, and reference ranges (52). Further work is undoubtedly required in these areas.
benefits and limitations
The general benefits and limitations of clinical guidelines described previously (3)(52) also apply to guidelines for tumor markers. Clinical guidelines are not a "magic bullet" for healthcare problems, but provide just one of the options for improving the quality of care, being most relevant where practitioners are unclear about appropriate practice and scientific evidence can provide an answer (3). In this respect, the need for improved dissemination of information about the appropriate use of tumor markers is evident to anyone providing a routine diagnostic service for these relatively specialized tests. Although many questions remain to be answered and further objective scientific studies are required, much relevant information is already available. Development and refinement of guidelines such as those described here provide the most convenient means of making this information readily accessible to those who can best use it.
| Conclusion |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
-fetoprotein; hCG, human chorionic gonadotropin; EAU, European Association of Urology; NSE, neuron-specific enolase; AJCC, American Joint Committee on Cancer; CEA, carcinoembryonic antigen; PSA, prostate-specific antigen; and DRE, digital rectal examination. | References |
|---|
|
|
|---|
deskenny/acbi.html (Accessed May 2002)..
The following articles in journals at HighWire Press have cited this article:
![]() |
C. M. Sturgeon, M. J. Duffy, U.-H. Stenman, H. Lilja, N. Brunner, D. W. Chan, R. Babaian, R. C. Bast Jr., B. Dowell, F. J. Esteva, et al. National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines for Use of Tumor Markers in Testicular, Prostate, Colorectal, Breast, and Ovarian Cancers Clin. Chem., December 1, 2008; 54(12): e11 - e79. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. P. Diamandis, B. R. Hoffman, and C. M. Sturgeon National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines for the Use of Tumor Markers Clin. Chem., November 1, 2008; 54(11): 1935 - 1939. [Full Text] [PDF] |
||||
![]() |
G. Cattozzo, E. Guerra, F. Ceriotti, C. Franzini, and on behalf of the Enzyme Working Group of the Itali Commutable Calibrator with Value Assigned by the IFCC Reference Procedure to Harmonize Serum Lactate Dehydrogenase Activity Results Measured by 2 Different Methods Clin. Chem., August 1, 2008; 54(8): 1349 - 1355. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Cao and R. Rej Are Laboratories Reporting Serum Quantitative hCG Results Correctly? Clin. Chem., April 1, 2008; 54(4): 761 - 764. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Soletormos, A. Semjonow, P. E.C. Sibley, R. Lamerz, P. H. Petersen, W. Albrecht, P. Bialk, M. Gion, F. Junker, H.-P. Schmid, et al. Biological Variation of Total Prostate-Specific Antigen: A Survey of Published Estimates and Consequences for Clinical Practice Clin. Chem., August 1, 2005; 51(8): 1342 - 1351. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Rej Clinical Chemistry through Clinical Chemistry: A Journal Timeline Clin. Chem., December 1, 2004; 50(12): 2415 - 2458. [Abstract] [Full Text] [PDF] |
||||
![]() |
Session 2: Molecular Diagnostics: Molecular Diagnostics: Chairman's Introduction JOHN F. LECHNER, Office of Technology for Diagnostics Systems, Bayer Diagnostics, Berkeley, California 94702 Toxicol Pathol, January 1, 2004; 32(1): 137 - 138. [PDF] |
||||
![]() |
E. Floyd and T. M. Mcshane Development and Use of Biomarkers in Oncology Drug Development Toxicol Pathol, January 1, 2004; 32(1_suppl): 106 - 115. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |