The expertise of the pathologist assessing your prostate biopsy cores may make a major difference to the way you and your doctors think you need to be treated. Indeed the following data would suggest that many prostate biopsy samples should be examined by genitourinary pathologists who really specialize in prostate pathology.
Al-Hussain et al. carried out a retrospective review of data from biopsy specimens diagnosed as including Gleason pattern 5 prostate cancer by the prostate pathology consultation service at Johns Hopkins (i.e., by Dr. Jonathon Epstein) between 2009 and 2010. The total series included 300 consecutive cases diagnosed as including Gleason pattern 5 prostate cancer.
It is important to understand that this series of 300 cases were all from cases sent to Johns Hopkins for a second opinion, and that this second opinion was not being requested by the original pathology laboratory that had assigned an inital diagnosis. (Johns Hopkins does also receive requests for second opinions from outside pathologists where there is difficultt in making a diagnosis, but these cases were not included in the current analysis.)
The authors were able to compare the diagnosis assigned by Dr. Epstein to the diagnosis assigned by the original, external pathology laboratory, with the following findings:
- In 146/300 cases (48.7 percent) Gleason pattern 5 had not been identified at all by the original pathology laboratory.
- Of the 146 cases in which no Gleason pattern 5 was originally identified by the original pathology laboratory
- 61/146 (41.8 percent) had been characterized as having a Gleason score of ≤ 7.
- 85/146 (58.2 percent) had been characterized as having a Gleason score of 4 + 4 = 8.
- 63/300 cases (21.0 percent) were diagnosed as Gleason 5 + 5 = 10 by Johns Hopkins.
- Of the 63 cases characterized as Gleason 5 + 5 = 10 by Johns Hopkins
- 26/63 (41.3 percent) had also been characterized as having a Gleason score of 5 + 5 = 10 by the original pathology laboratory.
- 27/63 (42.9 percent) had been characterized as having a Gleason score of 9 by the original pathology laboratory.
- 10/63 (15.9 percent) had been characterized as having a Gleason score of ≤ 8 by the original pathology laboratory.
With a degree of restraint, the authors write that:
Considering the important prognostic and therapeutic implication of misdiagnosing Gleason pattern 5, efforts should be made by the pathology community to acknowledge this as a problem and improve on individual pathologists’ accuracy by diverse medical education programs. In addition, urologists should not hesitate in sending biopsies with high-grade prostate cancer for expert genitourinary pathology second opinions.
It has long been known that the skill and experience of the pathologist can profoundly affect the correct assessment of Gleason patterns and Gleason scores from biopsy specimens and from post-surgical prostate speimens too. Apparently this problem continues to exist.
Filed under: Diagnosis, Risk | Tagged: Gleason, pattern, under-diagnosis |
THE "NEW" PROSTATE CANCER INFOLINK 
One aspect of these studies which is rarely mentioned is:
“What about the majority of re-examinations?” Were they all confirmed as correct or were some found to be over-diagnosed?”
Of course under-diagnosis is much more important than over-diagnosis because of the importance of identifying the high-grade disease that is most likely to be aggressive, but other studies in the past have demonstrated that there is a significant degree of over-diagnosis too.
All of which underlines the importance of getting an expert opinion before making a decision.
According to a study reported in this article, matches between biopsy and prostatectomy Gleason score (on “GS 6 level patients”) was found in 46.9% of cases, and close to 40% of all GS values were upgraded to 7; seems then that about 13% of GS values were downgraded.
Of course the study by Al-Hussain et al. above just compares first and second opinions on biopsy specimens that contained Gleason pattern 5, but not Gleason scores from biopsy and prostatecty specimens from the same patients.