Heterogeneity of Gleason grade in multifocal adenocarcinoma of the prostate

Abstract

BACKGROUND

The Gleason grading system uniquely combines data from different areas of carcinoma in the same prostate specimen. Prostatic adenocarcinoma often is multifocal, and different Gleason grades may be present in different foci. The current study was undertaken to compare the Gleason grades of individual adenocarcinoma foci in a given specimen with the overall Gleason grades (primary and secondary) of that specimen.

METHODS

Data were obtained from 115 consecutive radical prostatectomy specimens via whole-mount processing and complete sectioning. Diagrams were constructed by tracing the outline of each whole-mount section, and tumor maps subsequently were generated. The largest focus was considered the index tumor. Each prostatectomy specimen was assigned primary and secondary Gleason grades, and each tumor focus was assigned its own primary and secondary Gleason grades. Tumor volume was measured using the grid method.

RESULTS

Two or more adenocarcinoma foci were present in 87% of all specimens (2 foci, n = 20; 3 foci, n = 33; 4 foci, n = 17; 5 foci, n = 13; > 5 foci, n = 17). Specimens (n = 15) containing a single tumor were excluded from further analysis. Among the remaining specimens (n = 100), all tumor foci had Gleason grades that were the same as the corresponding overall Gleason grades in only 9 cases (9%). The Gleason score (i.e., the sum of the primary and secondary grades) of the index tumor was correlated with the overall Gleason score in 68% of specimens. The primary grade of the index tumor was the same as the overall primary grade in 97 specimens, whereas the secondary grade of the index tumor was the same as the overall secondary grade in only 68 specimens. The primary and secondary grades of the index tumor, compared with the overall Gleason primary and secondary grades, were reversed in 17 specimens.

CONCLUSIONS

The findings of the current study demonstrated the histologic heterogeneity of multifocal prostate malignancies. Although the Gleason grading system was used to determine an overall score for prostate carcinoma within a specimen, the scores of individual tumors, including index tumors, often did not agree with this overall score. These findings may have implications with respect to future biomarker and tissue array studies. Cancer 2004. © 2004 American Cancer Society.

Prostate carcinoma is a heterogeneous disease with substantial histologic and anatomic variability and a wide spectrum of biologic behavior. Its clinical outcome is difficult to predict, especially if the initial specimen is a small biopsy section that may not be representative of the entire tumor. The presence of multiple foci of prostatic adenocarcinoma within the same gland is a common finding.1–6 Greene et al.6 showed that by the time of clinical diagnosis, > 80% of prostates with malignant disease contained 2 or more widely separated tumors. Multiple prostate tumors and foci of prostatic intraepithelial neoplasia (PIN) often arise independently within the same gland, indicating that each tumor may have different molecular characteristics and expression profiles.7, 8

The Gleason score is widely accepted as the standard for histologic grading of prostate malignancies. The Gleason grading system is based on the glandular architecture of the tumor specimen observed at low magnification.9 The specimens are given an overall Gleason score that combines the primary and secondary grades of the tumor based on defined architectural patterns, without attention to cytologic details. The current study of 115 whole-mount prostate specimens was undertaken to characterize different tumor foci and to analyze the association of the Gleason grades of individual adenocarcinoma foci with the overall Gleason grades of the specimen containing those foci.

MATERIALS AND METHODS

Patients

The study group consisted of 115 consecutive patients treated with radical retropubic prostatectomy with or without bilateral pelvic lymphadenectomy between January 2000 and June 2001 at Indiana University Hospital (Indianapolis, IN). None of the patients had received preoperative radiation or androgen deprivation therapy. Serum prostate-specific antigen (PSA) levels were measured using the DPC chemiluminescent assay (Diagnostics Products, Los Angeles, CA). The research was approved by the Indiana University Institutional Review Board.

Histopathologic Analysis

Each radical prostatectomy specimen was totally embedded and processed via the whole-mount method.10–14 Each specimen was weighed, measured, inked, and fixed in 10% neutral formalin. After fixation, the apex and base were amputated and serially sectioned at 3–5 mm intervals in the vertical parasagittal plane. The seminal vesicles were sectioned parallel to their junction with the prostate and entirely submitted for evaluation. The remaining specimen was serially sectioned perpendicular to the long axis of the gland from the apex to the base. Whole-mount sections were prepared at 5 μm sections and stained with hematoxylin and eosin for histologic evaluation.

Diagrams were constructed by outlining each whole-mount section, and tumor maps subsequently were generated (Fig. 1). Foci of adenocarcinoma were considered separate if they were separated by > 4 mm.3, 4, 15, 16 Each of these tumor foci was assigned a primary and secondary grade according to the Gleason system.9 These were then compared with the overall Gleason grades and the Gleason score of the specimen. The largest tumor focus was designated the index tumor.

Figure 1. Whole-mount section of a radical prostatectomy specimen showing three outlined tumor foci with corresponding individual Gleason scores. The Gleason score of the index tumor differs from the overall Gleason score.

Tumor volume, defined as the sum of the volumes of individual tumor foci, was determined using the grid method.13, 17–20 In this method, the sum of each area was multiplied by the thickness of the average slice, and the sum of these volumes was multiplied by a factor of 1.25 to account for tissue shrinkage during processing. Surgical margins were considered positive when the tumor cells were in contact with the inked margin.11, 12 The 1997 TNM system was used for pathologic staging.21

Statistical Methods

The number of patients in each category is reported for all categoric variables (e.g., Gleason score and number of tumor foci). Continuous variables (e.g., patient age) are reported as mean values with ranges. The correlation between the number of tumor foci and the other pathologic variables was tested using the Wilcoxon signed rank test. P values of < 0.05 were considered statistically significant.

RESULTS

Radical prostatectomy specimens from 115 patients were evaluated. Specimens with a single focus of adenocarcinoma were excluded (n = 15), yielding a final study population of 100 patients. Patient and specimen characteristics are summarized in Table 1. Two or more foci of adenocarcinoma were present in 87% of specimens (2 foci, n = 20; 3 foci, n = 33; 4 foci, n = 17; 5 foci, n = 13; > 5 foci, n = 17). All tumor foci in a given specimen had primary and secondary Gleason grades that were the same as the corresponding grades assigned to the specimen as a whole in only 9% (n = 9) of these cases (all with Gleason Grade 3 + 3). The Gleason grades of the index tumor were correlated with the overall Gleason grades in 68% of specimens. The primary grade of the index tumor was the same as the overall primary grade in 97 specimens, whereas the secondary grade of the index tumor was the same as the overall secondary grade in only 68 specimens (Table 2). The primary and secondary grades of the index tumor, compared with the overall Gleason primary and secondary grades, were reversed in 17 specimens.

Table 1. Patient and Specimen Characteristics for 100 Patients with Multifocal Prostatic Carcinoma

Characteristic	No. of patients (%)
PSA: prostate-specific antigen.
Mean age in yrs (range)	61 (45–76)
Preoperative PSA (ng/mL)
< 4.0	4 (4)
4.9–9.9	72 (72)
10.0–19.9	21 (21)
> 20.0	3 (3)
Pathologic findings
Primary Gleason score
2	6 (6)
3	76 (76)
4	16 (16)
5	2 (2)
Secondary Gleason score
2	12 (12)
3	46 (46)
4	37 (37)
5	5 (5)
Overall Gleason score
5	18 (18)
6	27 (27)
7	48 (48)
8	2 (2)
9	51 (5)
Mean largest tumor dimension in cm (range)	1.65 (0.30–3.50)
Mean total tumor volume in cm^3 (range)	2.32 (0.15–12.5)
No. of tumor foci per specimen
2	20 (20)
3	33 (33)
4	17 (17)
≥ 5	30 (30)
Extraprostatic extension
Negative	78 (78)
Positive	22 (22)
Seminal vesicle invasion
Negative	97 (97)
Positive	3 (3)
Surgical margin
Negative	67 (67)
Positive	33 (33)
Pathologic stage
pT2a	5 (5)
pT2b	72 (72)
pT3a	19 (19)
pT3b	3 (3)
Perineural invasion
Negative	28 (28)
Positive	72 (72)

Table 2. Correlation of Overall Gleason Grades with Grades in Individual Tumor Foci

Tumor	No. of cases	Primary grade (%)a	Secondary grade (%)b
a Primary Gleason grade matches primary overall Gleason grade of specimen. b ondary Gleason grade matches secondary overall Gleason grade of specimen.
Index tumor	100	97 (97)	68 (68)
Tumor 2	100	91 (91)	58 (58)
Tumor 3	80	73 (91)	51 (64)
Tumor 4	47	39 (83)	29 (62)
Tumor 5	30	25 (81)	18 (58)

Evaluation of the nonindex tumor foci revealed a weaker correlation between individual and overall Gleason grades as the number of tumor foci increased (Table 2). This trend was more pronounced in terms of the secondary grade compared with the primary grade (Table 2). The second largest tumor had a primary grade that was the same as the overall primary grade in 91 specimens, whereas its secondary grade was the same as the overall secondary grade in only 58 specimens.

High-grade components (Gleason Grade 4 or 5) were identified in specimens with overall Gleason grades of 2 + 3 (n = 1), 3 + 2 (n = 5), and 3 + 3 (n = 13). These high-grade elements were found predominantly in the index tumors, with only three specimens having the high-grade component in nonindex tumor foci. Gleason Grade 5 tumors were identified in specimens with overall grades of 3 + 4 (n = 5) and 4 + 3 (n = 7), with 3 of these Grade 5 foci found in nonindex tumors.

The number of tumor foci was correlated with overall tumor volume (P = 0.004; Fig. 2). The number of tumor foci was not correlated with age at prostatectomy, preoperative PSA level, overall Gleason score, largest tumor dimension, extraprostatic extension, seminal vesicle invasion, pathologic stage, perineural invasion, or high-grade PIN (P > 0.05).

Figure 2. Correlation between number of tumor foci and total tumor volume in radical prostatectomy specimens (P = 0.004).

DISCUSSION

The current investigation of 115 totally embedded, serially sectioned, whole-mount prostatectomy specimens demonstrated that multifocal prostate carcinoma is common (87%) in prostatectomy specimens and that there can be extensive histologic heterogeneity among foci within a given specimen. Individual tumor foci within the same prostate specimen often had at least one Gleason grade that was different from the corresponding overall grade of the specimen. The Gleason grades of the index tumor were the same as the overall Gleason grades in approximately two-thirds of all specimens. The correlation between the primary grade of an individual focus and the overall primary grade of the specimen containing it was stronger than the correlation between the secondary grade of a tumor focus and the overall secondary grade of the corresponding specimen.

The Gleason method is the most commonly used system for histologic grading of prostatic adenocarcinoma. It recognizes tumor heterogeneity and incorporates primary and secondary grades into a final score. However, this system does not address the issue of multifocality. Although prostate carcinoma is commonly a multifocal and heterogeneous disease, many studies are based on information obtained from assessing only the largest tumor foci.5, 16, 22, 23 Among the 100 specimens included in the current study, there were 290 separate tumor foci. The majority of separate tumors in the same prostate specimen had different Gleason grades. Only nine specimens had the same Gleason grades in all tumor foci (Gleason Grades 3 + 3 in all cases). These results are similar to those of Ruijter et al.,15 who found in a study of 61 radical prostatectomy specimens that only 28% contained a single tumor and that only 16% had a single histologic grade of malignancy. In many cases, the overall Gleason score was not correlated with the score of the index tumor. In addition, the overall secondary grades tended to be lower than the grades observed in the index tumors.

In the current study, a small number of moderately differentiated (Gleason score, 5–6) nonindex tumors also contained high-grade elements (Gleason Grade 4 or 5). A recent study24 has suggested that secondary tumors in the prostate do not contribute to preoperative PSA and are unlikely to be clinically significant. Others have found an absence of predictability regarding the localization of the Gleason grades, with a random distribution of high-grade elements within separate tumor foci and anatomic regions.5 In an assessment of index tumors, Pan et al.25 found that the existence of a high-grade component, even if it accounted for only a small percentage of a given tumor, had a marked adverse influence on the biologic behavior of the tumor; these investigators therefore proposed a modification of the Gleason system that included notation of the presence of a tertiary element along with the original Gleason score. Currently, it is our practice to report the overall percentage of Grade 4/5 elements together with the overall Gleason score for each specimen. Knowledge of the presence of poorly differentiated carcinoma, whether present in the index tumor or in secondary foci, is clinically useful. Although many larger tumors are poorly differentiated, poorly differentiated foci may also be observed in smaller tumors, which can exhibit more extensive genetic alterations than might be expected. In fact, several studies have shown that the dominant (index) tumor will only occasionally share genetic changes with lymph node metastases in patients with lymph node–positive prostate carcinoma.15, 26, 27 Nondominant tumor foci may give rise to lymph node metastases, which can result in death. The size and grade of a dominant tumor focus (i.e., index tumor) cannot be used reliably to predict its biologic behavior.15, 26, 27

Our findings demonstrate the histologic heterogeneity of multifocal prostate malignancies. Although the Gleason grading system is used to determine an overall score for prostate carcinoma within a specimen, the scores of individual tumors, including the index tumor, often do not agree with this overall score. The findings of the current study may have important implications with regard to future biomarker and tissue array studies.