Gleason Grading of Prostate Cancer

A primer on grading of prostic adenocarcinoma. Thirteen pratical clues for Gleason grading in prostate needle biopsies are presented.
by David G. Bostwick, M.D., M.B.A.

Introduction

Histological grade is one of the most powerful predictors in prostatic adenocarcinoma. Accurate grading of cancer in contemporary 18-gauge needle biopsies is critical in planning appropriate treatment. More than 40 grading systems have been proposed in the last 75 years. The Gleason grading system, based on prospective study of more than 4000 patients between 1960 and 1975, is the standard method of grading prostate cancer throughout the world.

The Gleason grading system is based on the degree of architectural differentiation. A primary pattern is assigned for the dominant grade and a secondary pattern for the nondominant grade; the Gleason score is obtained by adding these two values. The secondary pattern should comprise at least 5% of the tumor. When a secondary pattern is not present, such as in small tumor foci in needle biopsies, the Gleason score is obtained by just doubling the primary grade.

The success of Gleason grading system is due to the following reasons:

  • It takes into account tumor heterogeneity by identifying primary and secondary patterns
  • A simplified and standardized drawing is available which has been extremely popular among pathologists (Fig.1)
  • It does not rely on morphogenetic or histogenetic models and simply based on the degree of acinar differentiation

Practical Clues in Grading Prostate Cancer

Following is a list of 13 practical clues that are helpful in assigning Gleason score to prostate cancer, especially in small foci in needle biopsies.

  1. Small foci of cancer do not necessarily mean low-grade cancer: With the advent of serum PSA and sextant biopsy, small foci of carcinoma are frequently found in needle biopsies. These small foci are not simply low-grade because of their size. A sampling of Gleason pattern 4 or 5 adenocarcinoma beneath the edge of the prostate may result in very few malignant acini in the specimen. (Fig.2)
  2. It probably isn't Gleason pattern 1: Gleason patterns 1 and 5 are the least common patterns of prostate cancer. Pattern 1 is usually present in the transition zone, an area infrequently sampled by needle biopsy (Fig.3). Further, these tumors are usually small.
  3. To identify Gleason pattern 1, the cancer must be circumscribed: The most important difference between Gleason pattern 1 and 2 is the presence or absence of circumscription respectively. Contemporary needle biopsies rarely provide the entire focus of cancer for evaluation, precluding evaluation of the periphery for completeness of circumscription. Consequently, the default grade for partially sampled low-grade cancers with uniform spacing is pattern 2 (Fig.4).
  4. Gleason pattern 2 should satisfy the three "R's": Round, Regularly spaced and Relatively uniform in size: Gleason pattern 2 cancer consists of predominantly round acini without sharp angulation or distorted shape (Fig.5). Nearly as important as acinar roundness is spacing - pattern 2 acini have relatively uniform spacing throughout the focus (Fig.6), unlike pattern 3, with variable spacing (Fig.7).
  5. Gleason pattern 2 acini may be close to one another but must have intervening stroma and no significant distortion of shape: If significant acinar crowding is present with some loss of intervening stroma between acinia, it may be more accurately considered as pattern 3. Any significant distortion of adjacent malignant acini constitutes Gleason pattern 3 (Fig.8).
  6. It's probably Gleason pattern 3: The 'default' grade for prostatic adenocarcinoma is pattern 3, recognizing that the great majority of cancers fall in this pattern, which encompasses the center of the normal distribution curve. More than 80% of Gleason's original series was pattern 3. Don't be hesitant about assigning pattern 3+3=6 to a needle biopsy simple because the previous five cases with small foci of cancer were the same grade (Fig.9)
  7. If there is a twofold or greater variation in acinar size, it's probably Gleason pattern 3 rather than pattern 2: When malignant acini are uniformly separated from one another, a twofold variation in acinar size distinguishes Gleason pattern 3 (Fig.10) from pattern 2 (Fig.11). Any variation in acinar size less than this may represent Gleason pattern 2 (exceptions to this exist; see Clue no. 8).
  8. Despite relative uniformity of acinar size, significant acinar angulation or distortion indicates Gleason pattern 3 rather than pattern 2: Significant acinar angulation violates Clue no. 4 above, precluding pattern 2. Some areas of Gleason pattern 3 may have relatively uniform acinar size with or without crowding. This pattern often has acini that are smaller than pattern 2. The lack of acinar roundness in such cases separates pattern 3 from pattern 2.
  9. Fusion is Fusion is Fusion (Gleason pattern 4): Acinar fusion separates most cases of Gleason pattern 4 and 3 (Fig.12). This is a critical cut-point in grading prostate cancer, as pattern 4 indicates poorly differentiated cancer. Fortunately, this is one of the most reproducible cut-points, because of the requirement for acinar fusion in pattern 4. If a line can be drawn around individual acini, no mattern how tightly packed, then the acini are not fused and it is pattern 3 (Fig.13).
  10. If a line can be drawn between acini that have no intervening stroma (fusion) for a length of at least four times the width of the acinus, this constitutes Gleason pattern 4: Tangentially cut tubular and tortuous acini of Gleason pattern 3 may mimic pattern 4, and such 'grade inflation' should be avoided. In difficult cases, if the length of 'fusion' of the acinus of concern is less than four times its width, we consider it pattern 3.
  11. If it's cribriform and nearly solid, it's probably Gleason pattern 4: Cribriform acini are usually pattern 3 (with comedonecrosis, pattern 5). However, when the sieve-like openings lose their round, rigid, punched-out appearance and become collapsed and nearly solid, it is best considered pattern 4 (Fig.14). Similarly, when the sieve-like masses lose their round countours, it often indicates transition to pattern 4 (Fig.15).
  12. The loss of most acinar lumens within fused acini indicates Gleason pattern 5: Most acinar lumens must be absent in order to separate Gleason pattern 5 from pattern 4. Tangential cutting and crush artifact may obscure or hide lumens. However, it most acini lack lumens, it constitutes pattern 5 (Fig.16). The presence of necrosis in any of the preceding patterns, usually pattern 3 or 4, also constitutes pattern 5 (Fig.17)
  13. If a line can be drawn between acini that have no intervening stroma (fusion) for a length of at least four times the width of the acinus, this constitutes Gleason pattern 4: Tangentially cut tubular and tortuous acini of Gleason pattern 3 may mimic pattern 4, and such 'grade inflation' should be avoided. In difficult cases, if the length of 'fusion' of the acinus of concern is less than four times its width, we consider it pattern 3.

Grading of Variants of Prostate Cancer

Variant
Gleason Primary Pattern
 
Ductal carcinoma (Fig.18)

3 (no necrosis); 5(necrosis)

 
Mucinous carcinoma (Fig.19A&B)

4

 
Signet ring cell carcinoma (Fig.20)

5

 
Sarcomatoid carcinoma (Fig.21A&B)

5

 
Neuroendocrine (Small cell) carcinoma (Fig.22)

5

 
Lymphoepithelioma-like carcinoma (Fig.23)

5

 
Squamous & Adenosquamous (Fig.24)

Variable; usually high-grade

 
Adenoid cystic carcinoma (Fig.25)

Variable; usually high-grade

 

Grading of Prostate Cancer After Therapy

The value of histological grading of prostate cancer after radiation therapy or hormonal therapy is controversial. A trend toward higher Gleason score after radiation therapy has been observed (Fig.26). A number of hypothesis have been suggested to account for this finding, including (1) cancer progression and tumor dedifferentiation following radiation therapy and (2) sampling variation in pre-irradiation biopsies and preferential progression of high-grade cancer. fig 26 The most important feature to be evaluated in post-radiation prostate biopsies is the extent of treatment effect. The Gleason grading system was originally based upon the architectural pattern of untreated cancer and is not applicable to histological changes resulting from treatment. Application of the Gleason grade to radiation treated prostates can only create confusion and must be avoided as has been recently recommended by the College of American Pathologists (1).

Summary

The issue of problems in accurate Gleason grading in needle biopsies has received a lot of attention lately. We and others have identified two major causes of errors in this area.

One is the presence of minimal focus of cancer in the biopsy which sometimes leads to a mistaken assumption that one is dealing with a well-differentiated cancer (Gleason score of 2+2=4 or less). Well-differentiated cancers are rare in the peripheral and central zones, sites that are preferentially sampled by needle biopsy. A review of 1355 needle biopsies and 88 transurethral resections sent to our consultation service between Oct. 1, 1996 and Sept. 31, 1997 confirmed our impression of undergrading of prostate cancer in needle biopsies ( 2 ). Two hundred twenty one out of 393 (56%) were assigned Gleason scores of 2-4 or described as "well-differentiated," or "low-grade" by the referring pathologists. We considered 5 (2%) of these cases as well-differentiated cancers.

The second common source of grading error is the identification of well-differentiated cancer in the needle biopsy, which may not correlate with tumor grade at radical prostatectomy due to tumor heterogeneity. The inter-observer reproducibility of Gleason scores 2-4 is poor. Many cancers thought to be well differentiated in the needle biopsy turn out to be under sampled high-grade cancers with extraprostatic extension. Thus, assigning a Gleason score 2-4 to small foci of cancer in needle biopsy may give a false sense of security to the urologist and the patient and may result in inappropriate action such as watchful waiting.

References

  1. Bostwick DG, Foster CS. Predictive factors in prostate cancer: Current concepts from the 1999 College of American Pathologists conference on solid tumor prognostic factors and the 1999 World Health Organization second international consultation on prostate cancer. Sem Urol Oncol 17(4): 222-272, 1999.
  2. Iczkowski KA, Bostwick DG. The pathologist as optimist: Cancer grade deflation in prostatic needle biopsies. Editorial. Am J Surg Pathol 22(10):1169-70, 1998.