Racial difference in prostate cancer cell telomere lengths in men with higher-grade prostate cancer: a clue to the racial disparity in prostate cancer outcomes

Abstract

Background:

Black men have worse prostate cancer outcomes following treatment than White men even accounting for prognostic factors. However, biological explanations for this racial disparity have not been fully identified. We previously showed that more variable telomere lengths among cancer cells and shorter telomere lengths in cancer-associated stromal (CAS) cells individually and together (“telomere biomarker”) are associated with prostate cancer death in surgically-treated men independent of currently used prognostic indicators. Here, we hypothesize that Black-White differences in the telomere biomarker and/or in its components may help explain the racial disparity in prostate cancer outcomes.

Methods:

Black (higher-grade [Gleason ≥4+3]=34, lower-grade=93) and White (higher-grade=34, lower-grade=89) surgically-treated men were frequency matched on age, pathologic stage, and grade. We measured telomere lengths in cancer and CAS cells using a robust telomere-specific fluorescence in situ hybridization assay. Tissue microarray and grade-specific distributional cutpoints without regard to race were evaluated.

Results:

Among men with higher-grade disease, the proportion of Black men (47.1%) with more variable cancer cell telomere lengths was 2.3-times higher (p=0.02) than that in White men (20.6%). In contrast, among men with lower-grade disease, cancer cell telomere length variability did not differ by race. The proportion of men with shorter CAS cell telomeres did not differ by race for either higher- or lower-grade disease.

Conclusions:

A greater proportion of Black men with higher-grade disease have an adverse prostate cancer cell telomere phenotype than White men with higher-grade disease.

Impact:

Our findings suggest a possible explanation for the racial disparity in prostate cancer outcomes.

INTRODUCTION

Prostate tumorigenesis, including subsequent cancer progression, is a multi-step, evolutionary process. This process includes the acquisition of initiating genetic and epigenetic alterations that confer a proliferative advantage to pre-neoplastic cells, the development of progressive genomic instability with phenotypic variability, and finally the retention and expansion of these rare oncogenic variants by inhibition of apoptosis and the activation of telomerase (1). Increasing evidence supports that alterations in the surrounding tissue microenvironment promote prostate carcinogenesis (1,2). Thus, simultaneous assessment of the tumor and its microenvironment may yield important insights into the disease process, as well as identify novel molecular biomarkers and/or therapeutic targets.

Tissue-based measurements of telomeres, the repetitive DNA elements located at chromosomal ends that are essential for maintenance of genomic integrity, is one such potential biomarker (3,4). Since short, dysfunctional telomeres promote genomic instability (5), we previously demonstrated that tissue-based measurement of telomeres is useful in predicting prostate cancer death in a large prospective study of men surgically treated for clinically localized prostate cancer. Using a robust telomere-specific fluorescence in situ hybridization (FISH) assay that provides telomere lengths on a per cell basis, we observed in a cohort study that men with more variable telomere lengths among their prostate cancer cells and shorter telomere lengths in their prostate cancer-associated stromal (CAS) cells, especially when combined (the “telomere biomarker”), had a statistically significantly increased risk of developing distant metastases and ultimately dying from their disease (6). Importantly, these findings were independent of currently used prognostic indicators, and the telomere biomarker even performed well in men with intermediate risk disease (i.e. clinically-localized Gleason 7 prostate cancer). However, this study was conducted in a cohort in which the majority of the men were White.

Compared to White men, Black men have a higher age-adjusted prostate cancer incidence and mortality rates and have worse outcomes following primary treatment of their disease (7). Even among men with a very low-risk prostate cancer at diagnosis, Black men have more adverse pathologic features in their radical prostatectomy specimens than White men and are more likely to experience poor outcomes (e.g. Gleason upgrading and biochemical recurrence) (8). Furthermore, Black men presenting with very low-risk prostate cancer are more likely to also harbor a tumor in the anterior of the prostate, tumors that are often high grade and large in volume (9). However, only limited biological explanations for this racial disparity have been identified, for example somatic ERF mutations and differences in copy number alterations (10). Prostate tissue-based telomere length measurements have not been compared between frequency matched Black and White men.

Thus, given our prior findings and to address a possible biological explanation for the racial disparity in prostate cancer outcomes, we determined whether the prevalence of the telomere biomarker and its components differ in prostate tissue between age, stage, and grade frequency matched Black and White men surgically treated for higher- and lower-grade prostate cancer, overall and in restricting to men with tumors that with intact PTEN or negative for ERG, other prostate cancer biomarkers, that are known to differ by race (11–13).

MATERIALS AND METHODS

Tissue-based study populations

Two separate tissue microarray (TMA) sets designed to investigate whether prostate tissue-based biomarkers differ by race were obtained from The Prostate Cancer Biorepository Network (PCBN; http://prostatebiorepository.org/). The PCBN higher-grade TMA set was enriched for higher-grade disease and consisted of 4 cores each from tumor and benign-appearing areas for each of 57 self-reported Black men and 59 White men surgically treated for prostate cancer between 2014 and 2016, and frequency matched on pathologic grade, age at diagnosis, Gleason sum (Grade Group), and pathologic stage. The PCBN lower-grade TMA set was enriched for lower-grade disease and consisted of 4 cores each from tumor and benign-appearing areas for each of 77 Black men and 75 White men surgically treated for prostate cancer between 2000 and 2010, and frequency matched on age (within 3 years), prostatectomy date, surgeon, Gleason sum, stage, and margin status. The design of both TMA sets, measurement of telomere length, and analyses were approved by the Institutional Review Boards at the Johns Hopkins Bloomberg School of Public Health and Johns Hopkins University School of Medicine.

Telomere-specific FISH

Telomere length was assessed on TMA sections containing areas of adenocarcinoma using a telomere-specific fluorescence in situ hybridization (FISH) probe and 4’,6-diamidino-2-phenylindole (DAPI) for labeling total nuclear DNA as previously described (6,14). Each individual TMA spot on each TMA slide was imaged using the TissueFAXS Plus (Tissue Gnostics) automated microscopy workstation and Zeiss Z2 Axioimager microscope. The digitized fluorescent telomere FISH signals were quantified using the TissueQuest software module to analyze the fluorescent images with precise nuclear segmentation. For each case, we evaluated prostate cancer cells and prostate CAS cells. Based on their unique morphologic features, other cell types (e.g. infiltrating lymphocytes) were excluded from the digital image analysis. Men without a valid telomere measurement were excluded (PCBN lower-grade N=4; PCBN higher-grade N=8).

ERG and PTEN Immunohistochemistry

Immunostaining was previously performed on these TMAs using genetically validated rabbit monoclonal antibodies and staining protocols for ERG (EPR3864; Ventana) and PTEN (D4.3; Cell Signaling Technologies) on the Ventana Benchmark or Ventana Discovery Ultra (Ventana/Roche) (11). Visual scoring using a dichotomous and validated scoring system was performed (11).

Statistical analysis

Means and proportions for demographic and pathologic characteristics were calculated separately for Black and White men by TMA set and by prostatectomy Gleason sum (irrespective of TMA set). Median (telomere length) and standard deviation (telomere length variability) of the telomere signal normalized to DAPI were determined for each man for cancer cells and CAS cells. Using the telomere measurements that comprise the telomere biomarker – cancer cell-to-cell variability in telomere length and CAS cell telomere length, we identified TMA and grade-specific distributional cutpoints without regard to race as follows: more variable (top tertile) in telomere length among cancer cells versus less variable (bottom and middle tertiles), and shorter (shortest and middle tertiles) telomere length in CAS cells versus longer (longest tertile). Next, we categorized men by TMA- and grade- (lower; higher) specific tertiles of cancer cell length variability and TMA- and grade-specific tertiles of CAS cell telomere length. As in our prior study (6), we further categorized men as having more (top tertile) versus less variable (middle and bottom tertiles) cancer cell telomere lengths and as having shorter (shortest and middle tertiles) versus longer (longest tertile) of CAS cell telomere lengths. To form the “telomere biomarker”, we categorized the men jointly by variability and length as follows: less variable in cancer cells/long CAS, more variable in cancer cells/long CAS, less variable in cancer cells/short CAS, more variable in cancer cells/short CAS.

Given our prior observation that the prevalence of more variable telomere length among cancer cells differed by Gleason sum (6), a priori, we conducted the analysis separately by higher- and lower-grade disease. Only 6 men on the PCBN lower-grade TMAs had high-grade disease; thus, we were unable to establish biomarker cut-points for these men and they were excluded from the analysis. Thus, the higher-grade cases are all from the PCBN higher-grade TMA set.

Since White and Black men were well-matched on disease characteristics within each TMA set (Supplemental Table 1), we used the lower-grade cases combined from the PCBN higher-grade TMA set and the PCBN lower-grade TMA set. For the lower-grade cases, we calculated prevalences of the telomere biomarker and its components pooled over the two TMA sets by taking the weighted average. We compared the prevalences of the telomere measurements by race separately for higher- and lower-grade disease using the Chi-square test.

To be able to assess the independence of any racial differences in telomere measurements from the known racial differences in PTEN and ERG status, in sub-analyses, we determined whether any racial difference in the prevalence of the telomere biomarker or its components was observed among men with higher-grade disease who were PTEN intact or ERG negative. All analyses were performed using SAS v 9.4 (SAS Institute, Cary, NC). All statistical tests were two-sided, with p<0.05 considered to be statistically significant.

RESULTS

Characteristics of the Black and White men with prostate cancer

Table 1 provides the characteristics of the 68 men with higher-grade prostate cancer and the 182 men with lower-grade prostate cancer who were included in the study. White and Black men were frequency matched on age at diagnosis, Gleason sum score (i.e. Grade Group), and pathologic stage. For the higher-grade group, the mean age at diagnosis was 63 years, the majority of the men were Grade Group 5 (61.8%), the majority of men were pathological stage T3A or above, and some men presented with lymph node involvement. In contrast, for the lower-grade group, the mean age at diagnosis was 57 years, the majority of the men were Grade Group 1 (58%), the majority of men were pathological stage T2, and none of the men presented with lymph node involvement.

Table 1.

Pathologic characteristics of men whose prostatectomy tissue was represented on the TMA sets by race^*.

	Lower-grade		Higher-grade
	White	Black	White	Black
N	N=89	N=93	N=34	N=34
Mean age (sd)*	57 (6.0)	57 (5.9)	63 (6.6)	63 (6.8)
Prostatectomy Gleason (%)
≤6 (Grade Group 1)	58.4	58.1	-	-
3+4 (Grade Group 2)	41.6	41.9	-	-
4+3 (Grade Group 3)	-	-	23.5	23.5
8 (Grade Group 4)	-	-	14.7	14.7
9 (Grade Group 5)	-	-	61.8	61.8
Pathologic Stage (%)
T2	79.8	77.4	38.2	41.2
T3A	19.1	21.5	20.6	17.6
T3B	1.1	1.1	38.2	38.2
T4	0	0	2.9	2.9
Lymph Node Involvement (%)	-	-
N1	-	-	17.6	20.6
NX	-	-	0	5.9

^*Pooled over the PCBN lower-grade and PCBN higher-grade TMA sets. White and Black men on each TMA set were frequency matched on age, prostatectomy Gleason score, and pathologic stage.

^**Men with higher-grade disease whose prostatectomy tissue was represented on the PCBN lower-grade TMA set (n=6) were excluded because there was an insufficient number of cases to establish TMA- and high grade-specific biomarker cut-points.

Variability in cancer cell telomere lengths differs by race in higher-grade prostate cancer

As shown in Table 2, in the higher-grade cases, the prevalence of more variable telomere lengths among cancer cells was 2.3-times higher in Black (47.1%) than in White men (20.6%), a difference that was statistically significantly (p=0.02). In contrast, in the higher-grade cases, the prevalence of shorter CAS cell telomere lengths did not differ between Black and White men (p=0.44). While the prevalence of the telomere biomarker category associated with worse outcomes (more variable/short) appeared to be higher in Black (20.6%) than in White (8.8%) men, the overall distribution of the four telomere biomarker categories did not differ by race among men with higher-grade prostate cancer (p=0.16).

Table 2.

Prevalences of more variable telomere lengths among cancer cells, shorter telomere lengths in cancer associated stromal cells (CAS), and the telomere biomarker category by race and grade in the PCBN lower-grade and PCBN higher-grade TMA sets.^*

	Lower-grade**			Higher-grade***
	White	Black	P-value†	White	Black	P-value†
	(N=89)	(N=93)		(N=34)	(N=34)
Telomere Biomarker Components
More variable in cancer cells (%)	29.2	36.6	0.29	20.6	47.1	0.02
Shorter in CAS cells (%)	67.4	67.4	0.99	70.6	61.8	0.44
Telomere Biomarker
Less Variable/Long (%)	15.7	12.0	0.76	17.7	11.8	0.16
More Variable/Long (%)	16.9	20.7		11.8	26.5
Less Variable/Short (%)	55.1	52.2		61.8	41.2
More Variable/Short (%)	12.4	15.2		8.8	20.6

^*TMA- and grade-specific distributions used to determine biomarker cut-points.

^**Prevalences pooled over the PCBN lower-grade and PCBN higher-grade TMA sets using weights.

^***Men with higher-grade disease whose prostatectomy tissue was represented on the PCBN lower-grade TMA set (n=6) were excluded because there was an insufficient number of cases to establish TMA- and high grade-specific biomarker cut-points.

^†Chi-square test.

Telomere measurements do not differ by race in lower-grade prostate cancer

As shown in Table 2, in the lower-grade cases, the pooled prevalence of more variable telomere length among cancer cells (p=0.29), shorter CAS cell telomere lengths (p=0.99), and the telomere biomarker (p=0.76) did not differ between Black and White men.

The racial difference in more variable cancer cell telomere lengths in men with higher-grade prostate cancer is not explained by ERG or PTEN

Since racial differences have been reported in the prevalence of ERG or PTEN status, other prostate cancer biomarkers, we sought to exclude associations of ERG or PTEN with more variable cancer cell telomere lengths as the explanation for the observed racial difference in the latter in men with higher-grade prostate cancer. In the men with higher-grade disease (N=68), the Black and White difference in the prevalence of more variable cancer cell telomere lengths remained in men with ERG-negative (N=47 [69.1%]; B=48.2%, W=15.0%, p=0.03) and with PTEN-intact (N=50 [73.5%]; B=52.0%, W=20.0%, p=0.02) prostate cancers.

DISCUSSION

In this study of 127 Black and 123 White frequency matched men, we found that among men with higher-grade disease, a higher proportion of Black than White men had more variable telomere lengths in prostate cancer cells. We previously linked the presence of this adverse telomere phenotype at the time of prostatectomy with worse outcomes independent of currently used prognostic indicators stage and grade (6), and in this context, this racial difference may contribute to racial disparity in prostate cancer outcomes.

Previous studies have shown that extensive telomere shortening in cancer cells compared with normal epithelial cells in the vast majority of prostate tumors (15). Telomeric fusions, a marker for telomere dysfunction, have been detected at high frequencies in prostate cancers (16). Additionally, telomere shortening has been shown to occur in CAS cells (17,18) thereby leading to replicative senescence, an irreversible exit from the cell cycle (19). Senescent fibroblasts can promote SASP, a senescence-associated secretory phenotype (20), through the secretion of pro-inflammatory cytokines, growth factors, and matrix remodeling enzymes. In turn, this newly remodeled microenvironment can further promote prostate cancer initiation and progression. Consistent with this notion, we recently found that shorter telomeres in CAS cells, in combination with increased cell-to-cell telomere length variation among cancer cells, was strongly associated with progression to metastasis and prostate cancer death in men treated for clinically localized disease (6). However, it must be noted that given the demographics of populations, these previous studies were conducted almost exclusively in White men. The results from the current study suggest a difference in telomere length profiles among cancer cells (a higher prevalence of greater variability in length) between Black men and White with higher-grade disease. Assuming that the prognostic-factor adjusted hazard ratio for association between more variable telomere length among cancer cells and lethal prostate cancer in men with higher-grade disease (4+3) from our prior study (6) is the same as what we would observe in the Black and White men with higher-grade disease in the source population for the current study, we estimate that more variable telomere lengths accounts for a greater burden of the risk of progression to lethal disease in Black men (25%) than White men (13%). Thus, identifying men with higher-grade disease and more variable telomere length in their prostate cancer cells and providing optimized treatment given their higher risk of poor outcome should result in an even greater benefit in Black men than in White men, thereby helping to reduce the racial disparity in poor outcomes of this disease.

Recent clinical trials of prostate cancer treatment suggest Black men may have equivalent outcomes to White men (21,22). In these clinical trials, men were carefully selected and care is standardized. However, these observations do not exclude the possibility that Black men have biologically different or more aggressive prostate cancer. For example, previous studies have demonstrated that ERG and PTEN alterations in prostate cancer are less common in African-American than in European-American men (11–13). In our study, in the men with higher-grade disease, the Black and White difference in the prevalence of more variable cancer cell telomere lengths remained in men with ERG-negative or PTEN-intact disease. These results suggest that the observed difference racial variation in the cancer cell telomere length variability is not explained by racial differences in the prevalence of ERG or PTEN status.

There are a number of strengths of this study. We assessed telomere lengths in two different cohorts from Johns Hopkins, which are available through an important national resource – the Prostate Cancer Biorepository Network (PCBN). To our knowledge this is only the first study to investigate cell-specific telomere lengths directly in prostate tissues to determine if the prevalence of these telomere length measurements differs in prostatectomy tissues between age, stage- and grade-matched Black and White men. However, despite this strength, there are also limitations to our study. This is a retrospective study with cases coming from one large tertiary care hospital. Race was previously abstracted from the medical record by the teams that selected the men for the TMAs; we did not assess genetic admixture or social and lifestyle factors that correlate with race and that might influence telomere measurements. We previously observed that obesity and physical inactivity were associated with shorter CAS cell telomere length (23), and smoking was associated with more variable telomere length in CAS cells in men without major comorbidities (24), but none of these was associated specifically with more variable cancer cell telomere length (the prevalence of which differed by race in the current study). The sample size and time since prostatectomy was not sufficient to investigate the telomere biomarker and its components in relation to prostate cancer outcomes in the Black and White men whose prostatectomy tissue is represented on the two TMA sets. Future studies that include larger cohorts of matched Black and White men from multiple institutions with adequate clinical follow-up and information characterizing the tumor microenvironment (e.g. milieu of inflammatory cells, secretion of pro-inflammatory cytokines) are thus warranted.

In summary, our finding that among men with higher-grade disease, a higher proportion of Black than White men had more variable telomere lengths among prostate cancer cells suggests a possible explanation for the racial disparity in prostate cancer outcomes.