Should all “cribriform” prostate cancers be classified as Gleason 4 disease?


In recommendations made in association with a newly published study of the pathologiocal implications of “cribriform” prostate cancer tumors, Dr. Scott Lucia and colleagues are suggesting that all such tumors should be classified as Gleason grade 4 by definition.

The distinction between cribriform Gleason pattern 3 and 4 prostate cancer has been considered controversial for several years. The word “cribriform” means perforated with very small holes or “sieve-like.” So cribriform prostate cancer tumors have a sieve-like appearance.

Iczkowski et al., writing in a recent issue of the American Journal of Clinical Pathology, report that pathologic data from 153 men who underwent radical prostatectomy were carefully assessed. Of these 153 men, 76 were biochemical failures, with a PSA level of ≥ 0.2 ng/ml, and they were carefully matched with 77 men who did not undergo biochemical failure.

The results of the pathologic analysis were as follows:

  • High-grade tumor patterns in the post-surgical specimens occurred with the following frequencies:
    • Fused small acini, 83.7 percent
    • Papillary, 52.3 percent
    • Large cribriform, 37.9 percent
    • Small (≤ 12 lumens) cribriform, 17.0 percent
    • Individual cells, 22.9 percent
  •  A cribriform pattern was present in 46/76 biochemical failures (61 percent) but only 12/77 biochemical non-failures (16 percent) (P < 0.0001).
  • The cribriform pattern had the highest odds ratio (OR) for PSA failure (OR = 5.89; P < 0.0001).
  • The presence of large and small cribriform patterns was significantly linked to failure.
  • All 8 patients with a cribriform area sum of ≥ 25 mm2 underwent biochemical failure (range, 33 to 930 mm2).

Clearly this is not a large study, but there do appear to be good reasons to look more carefully at the clinical implications of a pathologic finding of cribriform prostate cancer (on biopsy as well as after radical prostatectomy).

One Response

  1. Epstein at Johns Hopkins thinks so. It was in the 2005 ISUP recommendations.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.