Fresh Look at Gene Fusion Yields Clues in Low-Risk Prostate Cancer

OncologyLive, Vol. 18/No. 24, Volume 24, Issue 24

Although gene fusions between the TMPRSS2 gene and ETS family of transcription factors in prostate cancer have been recognized for more than a decade, the clinical relevance of this fusion event continues to be debated among experts.

John C. Cheville, MD

Although gene fusions between the TMPRSS2 gene and ETS family of transcription factors (most commonly the ERG gene) in prostate cancer have been recognized for more than a decade, the clinical relevance of this fusion event continues to be debated among experts. However, the findings of a recent study suggests that the mechanism through which this fusion occurs may be more important than the presence of the fusion for identifying low-risk prostate cancers, since retention of the interstitial genes between the TMPRSS2 and ERG genes post fusion occurred more frequently in prostate cancers classified as very low or low risk.1

If these larger trials confirm these findings, testing for the retention or deletion of this segment could help predict which patients would benefit from active surveillance, according to senior author John C. Cheville, MD, consultant in the Division of Anatomic Pathology, Department of Laboratory Medicine and Pathology, and professor of laboratory medicine and pathology at Mayo Clinic in Rochester, Minnesota.

Cheville also stated that development of efficient, clinically relevant assays will be important for testing for this fusion event in large groups of patients first in the clinical trial setting and perhaps later in the clinic. These assays may also help determine the dynamic behavior of this fusion event in the primary and metastatic settings. The finding comes amid ongoing interest in developing molecular assays that would help stratify risk for men with newly diagnosed prostate cancer considering active surveillance.

Knowledge About Fusion Event

Although tissue-based molecular assays have not been incorporated into National Comprehensive Cancer Network guidelines, several tests are likely to be covered by CMS.2 Additionally, the Michigan Prostate Score is being developed as an early detection test for prostate cancer combining serum prostate-specific antigen with urinary PCA3 and TMPRSS2-ERG expression.3Fusion between the TMPRSS2 gene and the ETS transcription family is thought to be an early event that occurs in approximately half of all prostate cancers, although the mechanisms for this fusion are not completely clear. ETS family members are involved in the regulation of cell growth, proliferation, differentiation, and apoptosis via activation or repression of its target genes. In prostate cancer, the positioning of androgen-responsive promotors in frame is thought to initiate overexpression of the members of the ETS family. For example, the androgen-responsive gene TMPRSS2 is fused near its first exon with its promoter in frame with the 5’ exons of the ERG gene, which retains the characteristic functional domains of the ETS family.

The high prevalence of the androgen-regulated overexpression of ETS family fusions suggests their key role in the development of prostate cancer. However, results from in vivo mouse studies4 showed that overexpression of ERG led to development of prostatic intraepithelial neoplasia but not cancer, suggesting that although the fusion event may contribute to initiation of prostate cancer, other subsequent driver mutations (eg, PTEN loss) may be necessary to drive progression to cancer.

Association With Low-Risk Disease

Figure. Two Ways to Form a TMPRSS2-ERG Gene Fusion1

The clinical relevance of the different fusion events in the ETS family is still unclear. Some studies5,6 suggest an association between ETS fusion and response to hormonal therapy; however, one study’s findings suggested that TMPRSS2-ERG fusion status was not predictive of response to radiotherapy,7 and another study’s results showed that ERG overexpression was not associated with biochemical recurrence or disease-specific mortality after prostatectomy.8Two mechanisms of TMPRSS2-ERG fusion lead to fusion products that are structurally identical but occur through either deletion or recombination of the interstitial segment, resulting in the absence or presence, respectively, of genes in the interstitial region between the 2 genes (Figure1). Because 1 of the 17 identified protein-encoding genes in this interstitial region is thought to be a tumor-suppressor gene, Cheville and colleagues hypothesized that the retention of the interstitial segment with the TMPRSS2-ERG fusion is more common in low-risk prostate cancers, perhaps suggesting that the fusion mechanism is a better predictor of tumor aggressiveness and patient outcomes than presence of the fusion alone.To test their hypothesis, Cheville et al used mate-pair next-generation sequencing (NGS) on frozen prostate cancer specimens from 133 patients to characterize presence, diversity, and structure of ETS family rearrangements. The specimens were classified into low-volume Gleason 6 tumors (very low risk for progression; n = 53), high-volume Gleason 6 tumors (low risk for progression; n = 26), Gleason 7 tumors (intermediate risk for progression; n = 29), and Gleason 8 or higher tumors (high risk for progression; n = 25).

The fusion event occurred in 43%, 49%, 52%, and 24% of patients in the very low-, low-, intermediate, and high-risk groups, respectively. Of the 60 prostate cancers with TMPRSS2-ERG fusions, 21 had retention of the interstitial segment and 18 of these were in the very low—risk and low-risk groups. “The retention of the genes is tied closely to the Gleason score, which is tied closely with outcome,” said Cheville. “We need more patients to determine what the significance of the retention is and [if] it is an independent marker of good behavior. We need to look within those 17 genes to see if there is any gene that is playing the role of suppressing tumor growth or aggressive behavior.”

Univariate analysis (but not multivariate) showed that the incidence of biochemical recurrence was significantly lower if the prostate cancer had retained the interstitial segment than if the segment was deleted. “There is potential utility for determining the status of interstitial genes in stratifying men with prostate cancer into more well-defined risk groups, but this will require further study before it can be incorporated into clinical practice,” Cheville stated in a press release.9

Future Research Directions

Cheville stated that the small number of patients in each subgroup was the main limitation to the study and emphasized the need to test these findings in larger groups of patients. “Our numbers are too small to draw conclusions,” he said. “We’re looking at 20 or 30 patients [in each subgroup] when we should be looking at 400 to 500 patients to understand the relative risks.”

However, Cheville noted that testing large numbers of patients using NGS can be expensive, highlighting the need for more efficient methods to identify the interstitial segment, such as a fluorescence in situ hybridization (FISH) probe. Cheville and his colleagues are currently experimenting with a FISH probe that they developed for the genes in the interstitial region, which will enable them to study tumor behavior in large groups of patients over time in the primary and metastatic settings. “Because we can do FISH on a lot more cases than we can sequence, we’ll be able to apply this FISH probe to cases that have the TMPRSS2-ERG fusion,” said Cheville. “[We can] really understand how often [the interstitial segment] is retained or lost in primary diagnosis, and when the patient develops metastases, how does metastasis look relative to primary [cancer]? This would also be an important step forward in implementing more efficient clinical testing for patients.”

If larger studies confirm the association between retention of the interstitial segment after TMPRSS2- ERG fusion and lower-risk cancer, development of a sensitive FISH probe will also be important for its implementation in the clinic and for wider accessibility for all patients. Cheville noted that most laboratories can perform FISH but fewer have the resources for mate-pair NGS.

Cheville also noted that identifying the deletion or retention of the interstitial segment with TMPRSS2-ERG fusion could be used as part of a panel of genes, along with the Gleason score, to predict whether a needle biopsy missed a higher-grade component, which is the primary reason that 30% to 40% of patients with a Gleason score 6 ultimately need therapy. “If you take a needle biopsy, and it’s a Gleason score 6, and you show a deletion of TMPRSS2-ERG fusion or PTEN deletion or another marker of aggressive behavior, you can imagine that you missed a higher-grade component in the prostate,” said Cheville. “That patient may still be in active surveillance, but you’re going to watch them a lot more closely. Or if the markers indicate that this is aggressive, you may even want to treat based on the molecular profile. That’s where our research is focusing.”

Cheville noted that current data are insufficient to warrant choosing more drastic treatments, such as radical prostatectomy or radiation, for Gleason 6 tumors with a TMPRSS2-ERG fusion and deletion of the interstitial segment, which typically undergo surveillance. However, he stated that the TMPRSS2-ERG fusion with a deleted interstitial segment may be similar to the PTEN deletion, which is detected in about 10% of Gleason 6 cancers and 40% to 50% of higher-grade cancers, in terms of predicting aggressive disease. “If you have PTEN deletion on Gleason 6 cancer, there’s a high likelihood that the patient has a higher-grade cancer,” said Cheville. “We need to determine if that deletion of the interstitial genes is as strong as PTEN [deletion].”

Cheville stated that he and his colleagues will need to analyze the fusion events in a large number of cases with a wide spectra of Gleason scores and clinical outcomes to determine whether the TMPRSS2-ERG fusion mechanisms are as strong as some of the other markers for predicting clinical outcomes. He concluded that to establish the clinical relevance of TMPRSS2-ERG gene fusion mechanisms, further studies need to confirm that they are independent predictors of aggressive disease and whether they can be used to predict higher-grade cancer in patients with Gleason score 6 cancer on needle biopsy.


  1. Murphy SJ, Kosari F, Karnes RJ, et al. Retention of interstitial genes between TMPRSS2 and ERG is associated with low-risk prostate cancer. Cancer Res. 2017;77(22):6157-6167. doi: 10.1158/0008-5472.CAN-17-0529.
  2. NCCN Clinical Practice Guidelines in Oncology: Prostate Cancer Version 2.2017. National Comprehensive Cancer Network website. Updated February 21, 2107. pdf/prostate.pdf. Accessed December 4, 2017.
  3. Lebastchi AH, Russell CM, Helfand AM, et al. Michigan Prostate Score (MIPS): an analysis of a novel urinary biomarker panel for the prediction of prostate cancer and its impact on biopsy rates [abstract PD03-03]. J Urol. 2017;197(suppl 4):e128.
  4. Lebastchi AH, Russell CM, Helfand AM, et al. Michigan Prostate Score (MIPS): an analysis of a novel urinary biomarker panel for the prediction of prostate cancer and its impact on biopsy rates [abstract PD03-03]. J Urol. 2017;197(suppl 4)
  5. Klezovitch O, Risk M, Coleman I, et al. A causal role for ERG in neoplastic transformation of prostate epithelium. Proc Natl Acad Sci USA. 2008;105(6):2105-2110. doi: 10.1073/pnas.0711711105.
  6. Minner S, Enodien M, Sirma H, et al. ERG status is unrelated to PSA recurrence in radically operated prostate cancer in the absence of antihormonal therapy. Clin Cancer Res. 2011;17(18):5878-5888. doi: 10.1158/1078-0432.CCR-11-1251.
  7. Karnes RJ, Cheville JC, Ida CM, et al. The ability of biomarkers to predict systemic progression in men with high-risk prostate cancer treated surgically is dependent on ERG status. Cancer Res. 2010;70(22):8994-9002. doi: 10.1158/0008-5472.CAN-10-1358.
  8. Dal Pra A, Lalonde E, Sykes J, et al. TMPRSS2-ERG status is not prognostic following prostate cancer radiotherapy: implications for fusion status and DSB repair. Clin Cancer Res. 2013;19(18):5202-5209. doi: 10.1158/1078-0432.CCR-13-1049.
  9. Hoogland AM, Jenster G, van Weerden WM, et al. ERG immunohistochemistry is not predictive for PSA recurrence, local recurrence or overall survival after radical prostatectomy for prostate cancer. Mod Pathol. 2012;25(3):471-479. doi: 10.1038/modpathol.2011.176.
  10. Common genetic fusion event may be associated with low-risk prostate cancer [press release]. Philadelphia, PA: American Association for Cancer Research; November 10, 2017. Accessed December 1, 2017.