Page 17 - Read Online

Ding et al. J Transl Genet Genom 2021;5:50-61 I http://dx.doi.org/10.20517/jtgg.2020.01 Page 51

aged (56-70 years), and 0.69 (95%CI: 0.55-0.69) in old (> 70 years) patients. Metastasis-associated immune
responses in the tumor microenvironment were more pronounced in young and middle-aged patients than in old
ones, potentially explaining the difference in accuracy of prediction among the groups.

Conclusion: We developed a genomic classifier with high precision to predict early metastasis for younger CaP
patients and identified age-related differences in immune response to metastasis development.

Keywords: Differentially expressed gene, immune cell enrichment, metastasis, prostate cancer, tissue
microenvironment, age, prediction, patient stratification, clinical phenotype

INTRODUCTION
Prostate cancer (CaP) is primarily a disease of older men; only 9.2% of men develop CaP under age
[1]
[1]
55 years . Although the overall incidence of CaP is decreasing in the United States , the incidence
is increasing in younger (≤ 55 years) men compared to older men (> 70 years) [2-4] . Due to a longer life
expectancy, younger men with localized CaP are more likely to receive radical prostatectomy (RP)
[5,6]
treatment than older men . A recent long-term follow-up study demonstrated that only those patients
[7]
harboring lethal CaP and having a long-life expectancy benefited from RP treatment . Over the past
decade, several genomic signatures have been developed to predict CaP outcomes based on gene expression
[8]
in prostatectomy or biopsy tissues . However, no prognostic signature was tailored to predict aggressive
CaP in men younger than age 55 years. Converging data from clinical and molecular genetic studies
provide strong evidence that CaP in young men represents a distinct clinical phenotype with underlying
biological differences compared to older men [9-13] . We hypothesized that age-related differences in tumor
biology have implications for prognosis of early-onset CaPs. To test this hypothesis, we selected tumor and
matched benign prostatic tissue samples from men diagnosed with CaP at younger (≤ 50 years) and older
(71-75 years) ages with low (6), intermediate (7), and high (8-10) Gleason scores. We identified age-related
differentially expressed genes (DEGs) by comparing sample type (tumor versus matched benign) and
Gleason scores (low vs. high). Then we developed a genomic classifier using gene expression of age-related
DEGs and tested the classifier for accurate identification of young patients with aggressive CaP as defined
by metastasis within five years of RP.

METHODS
Patient characteristics, mRNA profiling, and identification of age-related DEGs
This study was approved by the City of Hope (COH) Institutional Review Board (IRB07244). Patients with
CaP and treated with RP between 1998 and 2013 at COH National Medical Center were selected based on
age at diagnosis and tissue availability [Table 1]. A total of 61 men diagnosed between the ages of 71 and 75
years (old) and 58 men diagnosed between the ages of 38 and 50 years (young) were used to identify age-
related DEGs for developing a gene expression classifier to predict metastasis following RP. Older cases
were matched to younger cases for cancer stage and Gleason score. Tissue processing and mRNA profiling
were performed as described . Follow-up data were abstracted from medical records and the COH cancer
[9]
registry. Age-related DEGs were identified from expression data using a mixed linear model implemented
[14]
in limma R [Supplementary methods].

iPAM classifier development and validation
Gene expression data (46,050 genes and 1,232 patients from RP) from the Decipher Genomic Resource
Information Database (GRID, Decipher Biosciences, San Diego, CA) [Supplementary Table 1] were used to
develop and validate a new genomic classifier. The study design is shown in Figure 1. Gene expression data
for the age-related DEGs were extracted from the Mayo Clinic (MC I) discovery cohort . A two-sample
[15]