Editorial: Prostate Cancer

Abstract

The evolution of the use of prostate specific antigen (PSA) has truly been remarkable and has yielded a rapid growth of newly diagnosed prostate cancers. It would also appear that routine PSA measurements for early detection and screening have allowed for detection of greater numbers of cancers that are organanfined, a large percentage of which are only identified because of the elevated PSA blood test in the face of a normal digital rectal examination. Despite this fact, many unanswered problems persist, including the fact that many biopsies based on PSA elevations of 4 to 10 ng./ml. (Hybritech Tandem assay technique) do not reveal underlying cancer. Therefore, many different strategies have been developed in an attempt to decrease the number of negative (needless) biopsies while retaining an equal or greater number of positive biopsies. One such technique has been the use of PSA density testing for this purpose. PSA density is simply the PSA serum level divided by the volume of the prostate in cubic centimeters. In this issue of the Journal, Cookson et al (page 1070) evaluated the use of PSA density defined at a limit of 0.15 in patients with serum PSA levels of 4 to 10 ngJml. (Hybritech Tandem assay). They found that this technique of PSA density did not correlate well with the chance for a positive biopsy. Many others, including ourselves, would agree with this finding. Based on a large screening trial in which we participated, a PSA density of greater than 0.15 in men with serum PSA levels of 4 to 10 ng./ml. resulted in a positive predictive value of only 35 to 514b.l In contrast, however, data from our own institution combining transrectal ultrasound and PSA density have demonstrated a group of patients with normal transrectal ultrasound result and PSA densities of less than 0.12 in whom the chance for a positive initial biopsy was only 3.6%.% Should PSA density be used for initial diagnosis and indication for biopsy? The answer is probably not. Can PSA density along with ultrasonography be used to decrease the need for repeat biopsy in patients with initial negative biopsies? This answer is perhaps so. Another question that has arisen in the PSA era has been whether cancers determined by PSA testing are clinically significant. Most recent evidence would suggest that only a small percentage (approximately 10%) of PSA detected tumors (even in the face of a negative digital rectal examination) would be insignificant as defined by a volume of less than 0.5 cc.9 In this issue of the Journal Weldon et al (page 1074) demonstrated that even when focal biopsies are found histologically most patients will have significant cancer. In this report only 2 of 33 patients (6%) had insignificant cancer as defined. The difference between the results found in their series compared to others demonstrating more insignificant cancers in patients with focal positive biopsies may be explained by the patient populations studied (that is the number of patients with palpably abnormal prostates or elevated PSA levels). I believe that the most clinically useful conclusion of their study relates to the findings of focal cancers in patients with digitally abnormal prostates who had normal PSA levels. I agree that if the digitally abnormal area of the biopsy is negative, any other focal areas identified are more likely to be insignificant and that even a focal positive biopsy of any palpably abnormal area probably represents significant disease. Therefore, in patients with abnormal digital rectal examinations and normal PSA levels my practice has been to perform directed biopsies of the nodular area and ultrasonically abnormal areas only and not to perform systematic biopsies in these patients. The use of systematic biopsies to determine the volume and organ confinement of prostate cancer has been widely studied. Epstein et al recently suggested that criteria based on segmental biopsies can be derived that will define insignificant cancers with acceptable accuracy in patients with nonpalpable disease.4 In this issue of the Journal Loch et al (page 1078) argue that when 6 systematic biopsies reveal bilateral disease the total volume of cancer present can be estimated simply by determining the sum of the core cancer lengths on the side with the greatest core cancer length (ipsilateral side) and that contralateral biopsies do not contribute to a prediction of volume or distinguish bilateral spread from contralateral incidental cancem.6-6 In addition, they suggested that the presence of any Gleason pattern 4 or 5 cancer on a biopsy decreases the probability of a small cancer being found at prostatectomy. Unfortunately, they did not correlate their findings with organ confinement of the cancer or lymph node involvement, and they did not comment on the clinical followup of these patients after radical prostatectomy, including biochemical failure.