Article Text

Download PDFPDF

Impact of reporting rules of biopsy Gleason score for prostate cancer
  1. K Kuroiwa,
  2. H Uchino,
  3. A Yokomizo,
  4. S Naito
  1. Department of Urology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan
  1. Dr K Kuroiwa, Department of Urology, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan; humeiten{at}hotmail.com

Abstract

Aims: To investigate how the biopsy Gleason score (GS) and the clinical risk classification have been changed by the reporting rules.

Methods: 565 prostate biopsy specimens were reassessed. Each Gleason pattern, 1 to 5, was interpreted according to the modified Gleason grading system proposed by the International Society of Urological Pathology. The GS for each case was assigned by the previous reporting rules in the institute (OLD rules), applying the overall-scoring, and ignoring a pattern occupying less than 5% and the tertiary pattern. The GS was also assigned according to the NEW rules, applying the highest-core scoring and reflecting a pattern occupying less than 5% and the tertiary pattern.

Results: GS upgrading by the NEW rules was observed in 195 (35%) patients. Of these, 179 (92%) patients were upgraded only by applying the highest-core scoring. Of 198 patients with GS 6 by the OLD rules, 22 (11%) were upgraded to GS 3+4. Of 172 patients with GS 3+4 by the OLD rules, 59 (34%) and 28 (16%), respectively, were upgraded to GS 4+3 and ⩾8. Of 108 patients with GS 4+3 by the OLD rules, 63 (58%) were upgraded to GS ⩾8. As a result, the distribution of D’Amico’s clinical risk classification (low, intermediate and high risk) was changed from 26%, 43% and 31% to 23%, 35% and 41%, respectively.

Conclusions: Clinicians should be aware that the reporting rules, especially the highest-core scoring, contribute to a significant upward shift of the biopsy GS and risk classification.

Statistics from Altmetric.com

As in other malignancies, histological grading is one of the most important prognostic factors for prostate cancer and has been used in clinical practice for making treatment decisions.1 Prostate cancer has a unique and worldwide accepted histological grading system which was proposed by Dr Gleason, and which is determined only by the architectural pattern and categorised into five patterns.2

The Gleason score was originally defined as the sum of the most dominant pattern and the secondary dominant pattern. Reflecting the contemporary knowledge of prostate cancer and the changes in the manner of assigning the Gleason score, the International Society of Urological Pathology (ISUP) held a meeting and published a series of consensus statements for the Gleason grading of prostate carcinoma, including biopsy and prostatectomy specimens.3 Each Gleason pattern and the rare architectural variants were fully discussed and the schematic diagram of the modified Gleason system was presented. In addition, the reporting rules of the Gleason score for biopsy and prostatectomy specimens were discussed.

Regarding prostate biopsy specimens, the consensus was to assign individual Gleason scores to separate cores. Although the ISUP meeting did not actually decide that the highest-core score must be used for the Gleason score of each case, it is reported that the highest-core score would be selected as the Gleason score of the entire case by most clinicians in United States.4 In general, the secondary pattern occupying less than 5% of the tumour and the tertiary (third) pattern have been ignored for assigning the biopsy Gleason score. The consensus was to reflect the pattern occupying less than 5% of the tumour on the Gleason score if the pattern was higher than the primary pattern. The consensus regarding the tertiary pattern was that when a tumour includes Gleason patterns 3, 4 and tertiary 5 it should be classified as a high grade tumour (Gleason score 8–10) and should not be graded by simply listing the primary and secondary patterns.

To the best of our knowledge, the impact of these reporting rules for prostate biopsy specimens has not been well evaluated. In the present study, we investigated how the biopsy Gleason score and the clinical risk classification have been changed by the reporting rules of the Gleason score.

METHODS

Patient population

A total of 565 prostate biopsy sessions for 565 patients, which were performed from 2002 to 2007 and were positive for prostate cancer, constituted the cohort of the present study. A systematic ultrasound-guided biopsy taking six cores or more was performed in all patients.

Of these, 320 biopsies were performed in our institute. An additional 245 biopsies were performed in other hospitals; those patients visited our institute with pathological slides for consultation regarding treatment. Data on the patient’s age and serum prostatic specific antigen (PSA) before biopsy were available for all patients. In all cases, each biopsy core was placed in a separate container and processed for pathological examination. The procedure for this project was approved by the ethics committee of our institution, and conforms to the provisions of the Declaration of Helsinki.

Pathological assessment

All pathological specimens were reviewed by the consulting uropathologist (KK) blindly without looking at the previous pathological report of each biopsy session. Gleason patterns, from 1 to 5, were assigned for each positive core according to the schematic diagram of the modified Gleason grading system proposed by the ISUP.3 We assigned the Gleason score for each case by the previous reporting rules in our institute (OLD rules), which were based on the overall-scoring, ignoring a pattern occupying less than 5% of the cancer and the tertiary pattern. We also assigned the Gleason score according to the NEW rules, applying the highest-core scoring and reflecting a pattern occupying less than 5% of the cancer and the tertiary pattern (table 1). After the Gleason score of 7 was divided into 3+4 and 4+3, we graded the Gleason score of each biopsy session across the full spectrum of Gleason scores: 5, 6, 3+4, 4+3, 8, 9 and 10. We also grouped the cases into more clinically meaningful categories: ⩽6, 3+4, 4+3 and ⩾8 for the analysis.

Table 1 Comparison of the OLD and the NEW rules

With regard to the risk classification, the patients were classified into low risk (clinical stage T1c, T2a, and PSA level ⩽10 ng/ml, and Gleason score ⩽6), intermediate risk (other than low or high risk patients), and high risk (clinical stage ⩾T2c or PSA level >20 ng/ml, or Gleason score ⩾8) categories using the biopsy Gleason score, clinical stage and serum PSA as described by D’Amico et al.5

The Pearson correlation coefficient (r) on the relationship of the Gleason score between the OLD and the NEW rules was generated.

RESULTS

Table 2 presents the clinical characteristics of the 565 patients. Median age and serum PSA of the patients were 70 years (range 48–87 years) and 9.1 ng/ml (range 1.1–8544 ng/ml), respectively. A total of 329 (58.2%) patients had non-palpable disease on digital rectal examination; a median of 10 cores were taken in each patient.

Table 2 Clinical characteristics of patients (n = 565)

Overall, the Gleason score was upgraded in 195 (34.5%) patients and no patient was downgraded by the NEW rules when the Gleason score was divided into 5, 6, 3+4, 4+3, 8, 9 and 10. The average of the Gleason score was changed from 6.9 by the OLD rules to 7.2 by the NEW rules.

According to the OLD rules, the number of patients in each Gleason score (5, 6, 3+4, 4+3, 8, 9 and 10) was 2, 198, 172, 108, 46, 38 and 1, respectively. According to the NEW rules, the number of patients in each Gleason score (5, 6, 3+4, 4+3, 8, 9 and 10) was 1, 177, 107, 106, 93, 64 and 17, respectively. As a result, the distribution of each Gleason score (5, 6, 3+4, 4+3, 8, 9 and 10) by the OLD rules was 0.4%, 35.1%, 30.5%, 19.1%, 8.2%, 6.7% and 0.2%, which was changed to 0.2%, 31.3%, 18.9%, 18.8%, 16.5%, 11.3% and 3.0% by the NEW rules, respectively (fig 1).

Figure 1

Distribution of each Gleason score of 565 prostate biopsy sessions according to the OLD and the NEW rules.

Table 3 shows the correlation of each Gleason score between each rule. There was exact concordance of the Gleason score between the two rules in 370 (65.5%) patients and an agreement within one digit in 510 (90.3%) patients. The Pearson correlation coefficient (r) on the relationships of the Gleason score between the OLD and NEW rules was 0.88.

Table 3 Correlation of Gleason score of 565 prostate biopsy sessions according to the OLD and the NEW rules

Table 4 shows the relationship between the number of positive cores and each reporting rule. For the number of positive cores, the Gleason score was upgraded in 2 of 139 (1.4%) patients with one positive core, 29 of 107 (27.1%) with two positive cores, and 164 of 309 (53.1%) with more than two positive cores. Of 195 patients whose Gleason score was upgraded by the NEW rules, 179 (91.8%) patients were upgraded only by using the highest-core scoring, 0% (0/2) in patients with one positive core, 96.6% (28/29) in patients with two positive cores, and 92.1% (151/164) in patients with more than two positive cores.

Table 4 Number of upgrading of Gleason score by each reporting rule and positive biopsy core*

After combining all cases into more clinically meaningful groups (⩽6, 3+4, 4+3, ⩾8), the distribution of each Gleason score by the OLD rules was 35.4%, 30.4%, 19.1% and 15.0%, which was changed to 31.5%, 18.9%, 18.4% and 31.2% by the NEW rules, respectively (table 5). An exact concordance was observed between the two rules in 393 (69.7%) patients in the categorisation. Of 198 patients with a Gleason score of 6 by the OLD rules, 22 (11.1%) were upgraded to a Gleason score of 3+4. Of 172 patients with a Gleason score of 3+4 by the OLD rules, 59 (34.3%) and 28 (16.3%) were upgraded to a Gleason score of 4+3 and ⩾8. Of 108 patients with a Gleason score of 4+3 by the OLD rules, 63 (58.3%) were upgraded to a Gleason score of ⩾8.

Table 5 Correlation of Gleason score of 565 prostate biopsy sessions according to the OLD and the NEW rules, combining cases into more meaningful groups, Gleason score ⩽6, 3+4, 4+3 and ⩾8

Regarding D’Amico’s risk classification, the distribution of each group (low, intermediate and high risk) by the OLD rules was 25.5%, 43.2% and 31.3%, which was changed to 23.7%, 35.2% and 41.1%, respectively (fig 2). Table 6 shows the correlation of each category of D’Amico’s risk classification between each rule. Of 144 patients in the low risk category by the OLD rules, 10 (6.9%) were reclassified as intermediate risk by the NEW rules. Of 244 patients in the intermediate risk category by the OLD rules, 55 (22.5%) were reclassified as high risk based on the NEW rules.

Figure 2

Distribution of the D’Amico risk classification of 565 prostate biopsy sessions according to the OLD and the NEW rules.

Table 6 Correlation of D’Amico risk classification of 565 prostate biopsy sessions according to the OLD and the NEW rules

DISCUSSION

Since the Gleason grading system was derived in the late 1960s, the PSA test was introduced in the late 1980s and the ultrasound-guided sextant biopsy technique has been developed, which resulted in a dramatic stage migration to earlier stages because of the earlier detection of the disease.6 7

On the other hand, the Gleason score itself is reported to have migrated to a higher grade in recent decades due to changes in the interpretation of the pattern of prostate biopsy specimens, rather than a selective detection of more aggressive disease by the PSA test.811 For example, currently a Gleason score of 2–4 is not recommended for assignment to prostate biopsy specimens because of a significant upgrading in the matched prostatectomy specimens.12 In addition, the significance of the tertiary (third) pattern, which is higher than the primary/second pattern, has also been reported, and some pathologists may include a small amount of a high grade tertiary pattern in the biopsy Gleason score.13

Reflecting the above changes, the ISUP held a meeting in San Antonio with urological pathologists from throughout the world in order to achieve a consensus in controversial areas relating to the Gleason grading system. In addition to each Gleason pattern and the rare architectural variants, the rules of reporting the Gleason score for biopsy specimens were discussed.

The first rule concerns the Gleason score in the case of multiple positive cores with different grades. Should we assign an overall Gleason score for the entire case, or should we grade each core separately and select the highest score as the Gleason score of the entire case? It has been reported that patients with a Gleason score of 4+4 on one or more cores with a pattern of grade 3 in other cores should be given a final Gleason score of 4+4 instead of 4+3.14 Another opinion is that the highest Gleason score from all sites is significantly predictive of the final Gleason score on a matched radical prostatectomy specimen.15 In addition, all nomograms, including the Partin table, have used the highest-core scoring, and urologists have tended to use the highest Gleason score to determine their treatment plan.4 16 17 However, some clinicians may use the overall score, and the ISUP meeting did not actually decide that the highest-core score must be used for the Gleason score of each case.

The second rule concerns secondary patterns which present to a limited extent. In general, the secondary pattern occupying less than 5% of the tumour has been ignored for assigning the Gleason score. The consensus of the ISUP meeting was to include the pattern occupying less than 5% of the tumour within the Gleason score if the pattern was higher than the primary pattern, based on the concept that any amount of high-grade tumour on a biopsy specimen most likely indicates a more significant amount of high-grade tumour within the prostatectomy specimen.

The third rule is in regard to the tertiary pattern with more than two patterns in various portions on biopsy specimens. The consensus was that when a tumour includes Gleason patterns 3, 4 and tertiary 5 it should be classified as a high grade tumour (Gleason score 8–10) and should not be graded by simply listing the primary and secondary patterns. As a result of an ISUP consensus, in cases with patterns 3, 4 and 5, both the primary and the highest pattern should be recorded. It has also been reported that men with prostate cancer having a biopsy Gleason score of 7 and a tertiary grade of 5 have a higher risk of PSA failure in comparison to men with a Gleason score of 7 without a tertiary grade 5, and have a comparable risk with men with a Gleason score ⩾8.18

The data supporting the ISUP consensus are still not sufficient. As mentioned above, the Gleason score itself has migrated to a higher grade in recent decades due to changes in the interpretation of the pattern of prostate biopsy specimens. It has been reported that modified Gleason grading according to the ISUP consensus has also resulted in the upward grade shift.19 However, the impact of the reporting rules on the biopsy Gleason score and the clinical significance remain to be more fully evaluated. In addition, most studies regarding prostate cancer with the Gleason score do not mention the detailed rules of assigning a Gleason score.

We have demonstrated the significant upward shift of the Gleason score by applying the highest-core scoring and reflecting a pattern occupying less than 5% of the cancer and the tertiary pattern, even when each Gleason pattern was assigned by the most contemporary manner (ISUP consensus). A total of 34.5% of the patients were upgraded to a higher Gleason score, and no downgrading was observed.

To differentiate biopsy Gleason scores of 6 and ⩾7 is very important when an active surveillance is planned for the patient with prostate cancer, because such a management is mostly indicated for patients with a biopsy Gleason score ⩽6.16 20 In the present study, 11.1% of patients with a Gleason score of 6 by the OLD rules were upgraded to a Gleason score of 3+4. With regard to a biopsy Gleason score of 7, it has been reported that a Gleason score of 3+4 and 4+3 had different biological behaviour and should not be combined into one category.21 In the present study, 34.3% of patients with a Gleason score of 3+4 by the OLD rules were upgraded to a Gleason score of 4+3. In addition, 16.3% patients with a Gleason score of 3+4 and 58.3% of patients with 4+3 by the OLD rules were upgraded to a Gleason score of ⩾8.

This upward shift of the Gleason score also resulted in a significant change of the clinical risk classification.5 Of the patients in a low risk category by the OLD rules, 6.9% were reclassified as an intermediate risk category by the NEW rules. Of the patients in an intermediate risk category by the OLD rules, 22.5% were reclassified as a high risk category by the NEW rules. In a hospital where brachytherapy is performed only for low risk category patients, the change of risk classification would have a significant impact on decisions for treatment.

It has never been investigated which reporting rule for the biopsy Gleason score mentioned above has the largest effect on Gleason scoring. In the present study, the highest-core scoring, which solely accounted for 91.8% of all Gleason score upgrading, had the largest impact on the changes of the biopsy Gleason score. As a result, upgrading was frequently observed as the number of positive cores increased, since the highest-core scoring had an effect only on patients with multiple positive cores. The Gleason score upgrading by the NEW rules may be minimal in a setting in which the highest-core scoring is already used for the Gleason score as in previous nomograms.

In conclusion, the reporting rules, especially the highest-core scoring, are thus considered to result in a significant upward shift of the biopsy Gleason score and risk classification even when each Gleason pattern was assigned according to the ISUP consensus; this may affect the treatment planning of patients. Clinicians should also be aware of this upward shift by the NEW rules when either interpreting or comparing outcome data. Further study investigating the effect of this upgrading on prognosis is warranted.

Take-home messages

  • This is the first study to analyse the impact of reporting rules of the Gleason score with assigning each Gleason pattern by International Society of Urological Pathology consensus.

  • Gleason score upgrading was observed in about a third of the cases, according to the NEW rules which apply the highest-core scoring and reflect a pattern occupying less than 5% of the cancer and the tertiary pattern.

  • The distribution of D’Amico’s clinical risk classification also showed a significant upward shift by the NEW rules.

  • Among the reporting rules, the highest-core scoring had the largest impact on Gleason score.

REFERENCES

Footnotes

  • Competing interests: None.

  • Ethics approval: Ethics approval was obtained.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.