Dr. Erik Goluboff, assistant professor of urology at Columbia University College of Physicians & Surgeons and the senior author of the study - which is being presented at the meeting of the American Urological Association in Orlando, Florida - said that there is a "cutoff point"-0.15-above which readings in the PSAD have a 40% likelihood of being correlated with confirmed cases of malignant disease in Caucasian men. That cutoff point is, however, not useful with Hispanic men. "We don't know what it is, but we should probably have a different cutoff point for Hispanic men in interpreting the PSAD, or some other way of managing the Hispanic PSADs," said Dr. Goluboff, who is Director of Urology at the Allen Pavilion of NewYork-Presbyterian Hospital.
The authors, who also include Drs. John S. Lam, James M. McKiernan, Aristotelis G. Anastasiadis, and Mitchell C. Benson, note that although many studies have looked at PSA tests in Caucasians and African-Americans, few have addressed Hispanics. They set out to compare the performance of the test in groups of Hispanics and Caucasians.
A total of 404 Hispanics and 341 non-Hispanic Caucasians who had "elevated" PSA or abnormal rectal exam underwent transrectal ultrasound and biopsies between 1996 and 2001 at the Hospital. (Men were classified as Hispanic if they identified themselves that way, or by their surname.) Prior to biopsy, all patients underwent volume measurements of the entire prostate.
Of these patients, 242 Hispanics and 255 Caucasians had a PSA between 2.5 and 10 ng./ml. For these patients, PSAD (that is, PSA as a ratio to the volume of the prostate) was calculated, and the data were broken down between the two ethnic groups and between the negative and positive biopsy groups.
Thirty-five percent of the Hispanics, and 25.5% of the Caucasians, had cancer. There was no significant difference in age between the Hispanics and the Caucasians. There was no difference in mean PSA between Hispanics and Caucasians, or between men with malignant versus benign disease.
When it came to PSAD, however, there was a significant difference between Caucasians with malignant and with benign disease. The mean for the former group was 0.143 and for the latter group 0.113. (Dr. Goluboff acknowledges that the 0.143 for the malignant group is below the traditional cutoff point of 0.15. He says that the cutoff point is "not absolute"-that some cases of cancer will occur below that, and that he biopsies everyone, while some other doctors rely on the cutoff point in trying to avoid unnecessary biopsies.)There was also a significant difference in mean PSAD between Hispanics and Caucasians with malignant disease: 0.194 and 0.143, respectively.
The study is the first to show that at similar levels of PSA, PSAD is higher in Hispanics than in Caucasians. But in interpreting PSAD readings for Hispanics, it was impossible to tell which would have malignant disease and which would have benign disease. The mean for the malignant group was 0.17, and for the benign group 0.12. "Although that might seem to be a big difference, the ranges of the two groups were such that the difference was not statistically significant," Dr. Goluboff said. Thus, while PSAD is able to predict, with some reliability, between malignant and benign disease in Caucasians, it is not able to do so in Hispanics.
Dr. Goluboff comments that this suggests that a different "cutoff point" for PSAD-rather than the traditional 0.15-may be called for with Hispanic men, or there needs to be some other way of analyzing PSAD in Hispanics. "Further study in this area is certainly warranted," he said.