Race and Genetic Alterations in Prostate Cancer

Introduction

Black men in the United States are more likely to die of prostate cancer as compared with White men.^1,2 The relationship between these disparities, genetics, and social determinants of health is complex.³

DNA alterations include single-nucleotide variants, insertions and deletions, fusions, and changes in the copy number of genes. In prostate cancer, it is clear that DNA alternations contribute to gain, change, or loss of function of cancer-related genes. It is unknown how these changes relate to race, particularly as therapeutic decisions and prognosis can be mediated by genetic alterations.

Previous analyses have yielded conflicting results as to whether African American men have a higher frequency of actionable mutations and alterations in DNA repair genes.^4,5 A recent study using the Genomics Evidence Neoplasia Information Exchange (GENIE) cohort did not include analysis of copy number alterations (CNAs), a dominant mechanism of cancer gene function regulation in prostate cancer. Here, we present an analysis of differences in genetic alterations, including CNAs, among men of different self-reported races with primary and metastatic prostate adenocarcinoma using the updated GENIE 8 cohort.

Methods

Using the GENIE 8.0 registry, we identified men with primary or metastatic prostate adenocarcinoma who had targeted next-generation sequencing tumor genomic profiles as part of clinical practice at several institutions.⁶ We limited our data to those obtained from the Memorial Sloan Kettering Cancer Center (MSK)-IMPACT panel⁷ and the Dana-Farber Cancer Institute-Oncopanel.⁸ In contrast to a previous analysis, duplicate samples were excluded (n = 458).⁴

As these panels underwent various iterations, the genes included varied and the results are therefore presented as proportions with an alteration in a given gene with a changing denominator. Analysis of the overall number of alterations was limited to subjects profiled using the MSK-IMPACT 468 gene panel.

Differences among men of different self-reported races were evaluated applying a proportional test, and statistical significance (P < .05) was determined applying a two-sided Pearson chi-squared test. All statistical tests were performed using statistical package R.⁹

Results

Our cohort comprised 1,632 primary prostate adenocarcinomas and 898 metastatic adenocarcinomas. The number of copy number variations, fusions, and point mutations in select genes according to race is shown in Figure 1. An analysis of all profiled oncogenes by self-reported race is included with this paper as a Data Supplement.

FIG 1.

(A) Alteration profile of selected genes in primary and metastatic prostate cancer in the MSKCC and DFCI cohorts. Odds ratio with 95% CI is shown next to each column; White race was used as a reference. N values for analysis of each gene are included in the Data Supplement. (B) The proportion of subjects with at least one alteration in an actionable gene (ABL1, EGFR, ERBB2, BRAF, BRCA1/2, FGFR2/3, KIT, NTRK 1/2/3, PDGFRA, RET, ROS1, ALK, and PIK3CA) or DNA repair genes (ERCC5, MRE11, TP53BP1, POLE, RAD21, MSH2, MSH6, BRCA1/2, ATR, and ATM).⁴ DFCI, Dana-Farber Cancer Institute; NA, not available; MSKCC, Memorial Sloan Kettering Cancer Center; ref, reference.

Among men with primary prostate cancer, FOXA1 alterations were more common in Asian men as compared with Black (44.8% v 23.2%; P = .016) or White (44.8% v 19.8%; P < .001) men. TMPRSS2 alterations occurred more frequently in White men with primary prostate adenocarcinoma than Black or Asian men (39.8% v 25.2% [P = .001] and 20.0% [P = .018], respectively). Similarly, ERG alterations occurred more frequently in White men than Black men (29.5% v 15.3% [P < .001]). ZFHX3 alterations were also more common in Asian men (14.3% v 6.3% for White; P = .13 and 14.3% and v 3.4% for Black; P = .04). Mutations in the androgen receptor (AR) were rare among all races in men with primary prostate cancer, occurring in < 1% of Asian men, 2.5% of White men, and 2.8% of Black men.

Among men with metastatic prostate cancer, AR mutations had similar occurrences across races. Similar to primary disease, TMPRSS2 and FOXA1 alterations remained more frequent in White and Asian men in the metastatic setting, respectively. MYC alterations were more common in metastatic tumors from Black men as compared with White men (32.5% v 19.0%; P = .007). We did not observe any significant differences in BRAF between Black and White men or White and Asian men. We did observe a lower incidence of TP53 alterations in Black men as compared with men of other races.

There were more actionable mutations and total mutations in DNA repair genes in metastatic as compared with primary specimens (Fig 1). However, no statistically significant differences were noted in the prevalence of these alterations in men of different races.

Discussion

Race is a social and not biologic construct. In prostate cancer, there have been contradicting studies on the relationship between self-reported race and the frequency of genetic alterations in tumors of men with prostate cancer.^4,5 Previous studies excluded certain genetic features such as CNAs,⁴ thus limiting the interpretation of tumor genetic profiles by race. Here, we perform an updated analysis on men who underwent next-generation sequencing to better clarify the frequency of prostate cancer–associated alterations by race. Overall, we were unable to demonstrate clinically significant differences in actionable genetic alterations, including DNA repair genes between races.

These data contrast with the findings of a previous study.⁴ Although we used a similar cohort of patients, Mahal et al included more than 400 duplicates as distinct patients, which could have affected the results. Additionally, we expanded on their previous work to include CNAs. These observations, as well as an additional patient recruitment, could explain the differences in results.

Similar to previous studies, we demonstrate that FOXA1 mutations are seen in almost half of tumors of Asian men with primary prostate cancer, but only in 20% of men of other races. Li et al¹⁰ recently showed that FOXA1 mutations, which have been associated with a poor prognosis,¹¹ affected up to 40% of Chinese men with prostate cancer. Furthermore, we demonstrate that TMPRSS2 and ERG mutations are more frequent in White men. Mehra et al¹² have also shown that TMPRSS2 and ERG rearrangements can be found in 65% and 55% in men of European descent, respectively. Similar to Koga et al,⁵ with whom some samples used in this analysis overlap, we found that MYC mutations are more frequently found in metastatic tumors of Black men. Koga et al reported that ZFHX3 mutations are more common in Black men, whereas, in our series, we found that Asian men have significantly higher rates of ZFHX3 alteration in primary specimens with no differences noted between Black and White men. The abovementioned findings were also reported in a recent abstract presentation by Sivakumar et al,¹³ which included almost 12,000 men who had comprehensive genomic profiling using Foundation Medicine assay.

Importantly, we demonstrate that self-reported race does not appear to affect the proportion of patients who would benefit from targeted therapy or genetic profiling at present. Rather, alternative strategies to mitigate this large health disparity are warranted.

Although this is a relatively large study, the sample size cannot capture the population-level diversity in tumor biology by race. We focused intentionally on self-reported race and not genetic ancestry, and our results should be interpreted in that context. A targeted panel, rather than more comprehensive profiling, was used. The GENIE repository currently lacks data on treatment and common clinical and demographic information for further analysis.

In conclusion, contemporary next-generation sequencing of primary and metastatic prostate cancer did not reveal any significant differences in actionable mutations between self-reported races. Thus, to reduce and eventually eliminate cancer health disparities, a continuous focus on access and social determinants of health is critical. Although there are differences in tumor DNA in contemporary genetic panels among men of different races, these do not at present suggest a meaningful way to address the current disparity in health outcomes.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/po/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).