| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Editorials |
Pain Research & Nuffield Department of Anaesthetics, University of Oxford, Oxford Radcliffe Hospital, The Churchill, Oxford OX3 7LJ, UK, Fax +44 1865 226978, e-mail andrew.moore{at}pru.ox.ac.uk
In this issue of Clinical Chemistry (pp 1588–94),
Junker et al. elegantly try to tease out the usefulness of the
percentage of free prostate-specific antigen (PSA) (that fraction not
bound to proteins like 
-antichymotrypsin) in making the differential
diagnosis of prostate cancer and benign prostatic hyperplasia (BPH).
Their approach was to select samples from well-defined patients, 30
with BPH and 50 with cancer, and subject them to analysis by four
commercial systems for analyzing the free and total PSA in serum. At a
sensitivity of 95%, negative predictive values and the proportion of
men with BPH who may be spared prostate biopsy were calculated.
There were differences between the assays. Absolute values reported for control materials with different amounts of complexed PSA varied by factors of 2 or more between the assays, and the proportion of free PSA calculated from control materials was consistently lower in the CIS and DPC methods than in the Boehringer and Hybritech methods. Negative predictive values at 95% sensitivity had confidence intervals that were wide (reflecting the limited number of samples), and those of all four methods overlapped. But the bottom line was that, on this set of samples, three of the four assay systems might prevent 40% or more of men with BPH from having unnecessary (and painful) prostate biopsies; the Hybritch method produced much lower estimates.
There is much interest in percentage of free PSA, even attracting a special supplement of the journal Urology (1). How many men might be spared unnecessary biopsy is addressed either directly or indirectly in at least 16 other papers published in the last few years on just this topic (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17). The one comparable report that used Hybritech assays (14) did not confirm Junker et al.'s finding of a lower number of preventable biopsies with this system. Together, the reports appear to confirm Junker et al.'s suggestion that a significant proportion of men might be spared unnecessary biopsy, apparently irrespective of method of analysis used. Nor did population studied affect the conclusion, whether reports analyzed a selected series of defined prostate cancer and BPH (2)(7)(9)(17) or larger numbers of consecutive patients attending urology clinics (14)(15). When screening men with total PSA <4 µg/L, a lower proportion appear to be spared unnecessary biopsies (5).
Junker et al. give us information not only across the whole range of total PSA, but also over the range of 4 to 25 µg/L. In men with total serum PSA in this range, but with palpably normal prostates, the cancer detection rate of sextant biopsy is ~25% (1). Others have sought to do subgroup analysis in truncated concentrations of total PSA, the so-called "gray area" of ~4–10 µg/L and also <4 µg/L where many cancers are found. There is limited evidence for better discrimination in the 4–10 µg/L range. The main importance is that men with PSA values of 10 µg/L or higher are likely to be investigated anyway, including having a prostate biopsy. Those with values of below some lower figure (2.5–4.0 µg/L) will not be investigated in any event. So knowing the performance of assays over ~4–25 µg/L covers the area of maximum interest.
Likelihood ratios for a positive test (defined as a percentage of free PSA below the threshold) calculated from sensitivity and specificity (18) are generally between 2 and 7. This may be modest, but given the prevalence of prostate cancer in men referred to urology clinics of ~30%, it may be a valuable part of the diagnostic algorithm.
The picture that emerges is that while it is a better diagnostic tool than total PSA, the percentage of free PSA fails to predict unequivocally those patients with cancer. On the basis of this test alone, preventing some unnecessary biopsies always results in some cancers missed, perhaps 10% or more. Concern has been expressed about this (4)(15), particularly because this failure to detect all cancers in the 4–10 µg/L range could be seen as a lower standard of care than present practice. The current view is that all men with total PSA in this range should undergo biopsy so that the 1 in 4 men who actually have cancer will not be missed.
Combining percentage of free PSA and prostate volume appears to have exciting prospects. Using the Tosoh system, Egawa et al. (7) showed that while only 11% of men with BPH and a prostate volume <30 mL had a percentage of free PSA <15%, 52% of men with prostate cancer did so. Even better results were reported in the same issue of the same journal by Stephan et al. (12). Using the DPC Immulite system, they found that men with a prostate volume <40 mL and a percentage of free PSA <15% misclassified only one of 16 men with BPH and only one of 26 men with prostate cancer, giving a sensitivity of 94%, specificity of 96%, and likelihood ratio of 23.5. We need these observations confirmed in larger prospective studies.
And we must not ignore other developments that are moving diagnosis of prostate cancer forward. The "gold standard" of the sextant biopsy, against which we measure serum tests, has recently been shown to underdiagnose cancer in as many as 35% of patients, particularly those with total PSA values <10 µg/L (19). Moreover, the Baltimore Longitudinal Study of Ageing has indicated, albeit in a small number of men, that the percentage of free PSA in serum is predictive of tumor aggression (20).
Diagnosing prostate cancer at an early stage is not easy. The percentage of free PSA is just one factor that helps make a diagnosis. There are others like total PSA, prostate volume, age, and family history, all of which have a part to play in helping a urologist make a final decision about diagnosis and treatment (18). Clinical chemists might be more proactive in producing systematic reviews to put all the new data into context. Given the limited number of groups that have published, and their research quality, it might even be possible to do single-patient metaanalysis as has been done for hemoglobin A1c (21). If it included all the clinical data available, a logistic analysis might inform us about which independent diagnostic factors are important, and how much weight to give them (22).
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. Peter, C. Unverzagt, T. N. Krogh, O. Vorm, and W. Hoesel Identification of Precursor Forms of Free Prostate-specific Antigen in Serum of Prostate Cancer Patients by Immunosorption and Mass Spectrometry Cancer Res., February 1, 2001; 61(3): 957 - 962. [Abstract] [Full Text]
|
|
|
|
 |
|
 J. Peter, C. Unverzagt, and W. Hoesel Analysis of Free Prostate-specific Antigen (PSA) after Chemical Release from the Complex with {alpha}1-Antichymotrypsin (PSA-ACT) Clin. Chem., April 1, 2000; 46(4): 474 - 482. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
