Molecular Signature Distinguishes Indolent From Aggressive Prostate Cancer

Mitchell C. Benson MD, PhD

George F. Cahill Professor of Urology

Chair, Department of Urology

Herbert Irving Comprehensive Cancer Center

Columbia University Medical Center

Prostate cancer is one of the most common forms of cancer in older men, with more than 200,000 new diagnoses per year. An increase in the aging population, along with widespread screening for prostate-specific antigen (PSA), has contributed to a substantial rise in diagnoses of early- stage prostate tumors. The majority of these newly diagnosed cancers are slow-growing and require no treatment. However, identifying the lethal ones remains a major challenge for clinicians.

A team of interdisciplinary researchers at the Herbert Irving Comprehensive Cancer Center at Columbia University Medical Center has begun to address that problem. The team has developed a way to classify low Gleason score prostate tumors into indolent and aggressive subgroups based on the expression of genes associated with aging and senescence. In the process, they have identified a new three-gene biomarker that may soon help us to determine who will develop advanced prostate cancer and who will not (Figure).¹

Current Standard of Treatment

Determining the appropriate course of treatment is presently dependent upon Gleason grading, a histopathological evaluation that lacks a precise molecular correlate. Patients presenting with tumor biopsies rating higher than 8 on the Gleason score are advised toundergo immediate treatment, yet clinicians have no unequivocal recommendation for those with biopsies of low (Gleason 6) scores, or even for some patients with intermediate (Gleason 7) status.

The lack of reliable and reproducible assays to identify tumors destined to remain indolent has contributed to the overtreatment of some patients who would not have died from these tumors. The practice of “watchful waiting” or active surveillance provides an alternative, allowing clinicians to monitor low-risk prostate cancer with the intention of avoiding treatment unless there is evidence of disease progression. The obvious advantage of this approach is that it minimizes the serious side effects of urinary incontinence and impotence associated with surgery and radiation.

The difficulty is that active surveillance may miss the window of opportunity for early intervention when tumors that are seemingly lowrisk turn out to be aggressive. Thus, there is a critical need to identify reproducible and validated biomarker panels that can distinguish between the majority of low Gleason score tumors that will remain indolent and the few that are truly aggressive. Further, it would be particularly beneficial for such biomarker panels to provide a molecular correlate that could be used in conjunction with Gleason scoring for classifying biopsies from patients on surveillance.

The Importance of Signaling Pathways

One of the most significant risk factors associated with developing prostate cancer is aging. The biological processes associated with that represent a balance between antitumorigenic and protumorigenic signals. Led by Cory Abate-Shen, PhD, researchers hypothesized that these biological processes may also distinguish indolence from aggressiveness. More specifically, they posited that tumors destined to become aggressive might be enriched for protumorigenic aging signals, while those destined to remain indolent may be enriched in antitumorigenic aging signals.

One of the principal antitumorigenic signals of aging is cellular senescence. Senescence is known to play a critical role in tumor suppression in general and has also been associated with benign prostate lesions in humans and in mouse models.

Using gene set enrichment analysis (GSEA), and under the direction of Andrea Califano, PhD, the team found a gene signature representing biological processes of aging and senescence that can distinguish indolent versus aggressive prostate tumors. This gene signature is downregulated in aggressive tumors but upregulated in indolent ones. Analyses of these enriched genes led to the identification of a 19-gene indolence signature. This signature was further interrogated using a decision-tree learning model to pinpoint three genes—FGFR1, PMP22, and CDKN1A—that could predict the outcome of low Gleason score tumors.

Validation of the Three-Gene Test

In a blinded retrospective study, researchers in Abate-Shen’s laboratory, working with members of the Department of Urology, tested the accuracy of this three-gene panel on initial biopsy specimens from 43 patients who had been diagnosed with lowrisk prostate cancer and had been monitored for at least 10 years with active surveillance at Columbia University Medical Center. Fourteen of them ultimately developed advanced prostate cancer. All were correctly identified by the panel. Investigators plan to test this assay in a large prospective clinical trial.

Many patients diagnosed with low-risk prostate cancer face a difficult decision-making process. The development of validated biomarkers will allow clinicians to identify low-risk disease with a higher rate of success, thus easing patient anxiety about choosing active surveillance over early treatment.

Figure. This illustration depicts how the three-gene panel would be used in clinical care

Courtesy of Genes & Development/Newswise

Reference

1. S. Irshad, M. Bansal, M. Castillo-Martin, et al. A molecular signature predictive of indolent prostate cancer. Sci Transl Med. 2013;5(202):202ra122. doi: 10.1126/scitranslmed.3006408.