European Urology

Wang X, Yu J, Sreekumar A, Varambally S, Shen R, Giacherio D, Mehra R, Montie JE, Pienta KJ, Sanda MG, Kantoff PW, Rubin MA, Wei JT, Ghosh D, Chinnaiyan AM
N Engl J Med 2005;353(12):1224–35.
Looking for Other Markers Than PSA for Prostate Cancer Detection
Expert's summary:
In an era where PSA is increasingly challenged as an effective prostate cancer marker, a combined effort by the University of Michigan and the Harvard Medical School has used protein marker microarrays to identify auto-antibodies against tumour antigens in patients with prostate cancer. The authors constructed phage-protein microarrays in which peptides derived from the prostate cancer cDNA library were expressed as a prostate cancer phage fusion-protein. Phage-protein microarrays were then screened to identify phage-peptide clones that bind auto-antibodies in serum samples from patients with prostate cancer and controls.

They looked in total for a panel of 22 antigens derived of prostate cancer, found of potential interest at discriminating between patients with and without cancer.

In 60 patients and 68 controls, this 22 bio-marker assay reached a sensitivity of 81.6% and a specificity of 88.2%. The panel of peptides was far better than PSA for detecting prostate cancer.
Expert's comments:
This very important contribution opens the way for new avenues in prostate cancer detection.

One of the most promising approaches to cancer early detection, is not to look for cancer itself, but to look for immune responses to cancer. Interestingly, these new tests do not exclude using them in conjunction with PSA in order to increase their sensitivity and specificity. This panel of peptides was found to be far better than PSA for prostate cancer detection with an ROC curve which was close to perfection at 0.97. It is interesting to see that the area under the ROC curve of PSA was very elevated in the entire group of patients, close to 0.80, much better than many previous studies, although in the subgroup of patients with PSA between 2.5 and 10 ng/ml, the area under the ROC curve was within the generally observed value of 0.50.

In an Editorial O. J. Finn [1] makes the link between diagnostic and therapeutic possibilities, in other words using these auto-antibody signatures at the same time to establish the diagnosis, but also if the T cell response is investigated, to allow to know whether the tumour is likely or not to be destroyed by the immune system.

We, however, certainly have to be cautious about these new methodologies, because the results need to be verified in larger, unselected populations. Although there has been an explosion of knowledge in cancer biology and molecular biology, most of the recently detected marker substances have not found a place yet as a tool for diagnosis of cancer and for prostate specifically.

For instance, despite the enthusiasm about the results of proteomics, E.P. Diamandis [2] has recently warned that there are still many concerns. Using the auto-antibody signature for prostate cancer detection, large amounts of proteins do not need to be present, but rather very low levels of specific antigens, which indeed are much easier to detect given our available powerful immunological technology.