Whole blood GRHL2 expression as a prognostic biomarker in metastatic hormone-sensitive and castration-resistant prostate cancer
Edmond M. Kwan,Heidi Fettke,Megan Crumbaker,Maria M. Docanto,Sarah Q. To,Patricia Bukczynska,Andrew Mant,Nicole Ng,Siavash Foroughi,Lisa-Jane K. Graham,Anne-Maree Haynes,Sarah Azer,Lisi Elizabeth Lim,Eva Segelov,Kate Mahon,Ian D. Davis,Phillip Parente,Carmel Pezaro,Niranjan Sathianathen,Christine Hauser,Lisa G. Horvath,Anthony M. Joshua,Arun A. Azad,Tilman Todenhöfer
DOI: https://doi.org/10.21037/tau-20-1444
2021-04-01
Translational Andrology and Urology
Abstract:BACKGROUND: As potent systemic therapies transition earlier in the prostate cancer disease course, molecular biomarkers are needed to guide optimal treatment selection for metastatic hormone-sensitive prostate cancer (mHSPC). The value of whole blood RNA to detect candidate biomarkers in mHSPC remains largely undefined.METHODS: In this cohort study, we used a previously optimised whole blood reverse transcription polymerase chain reaction assay to assess the prognostic utility [measured by seven-month undetectable prostate-specific antigen (PSA) and time to castration-resistance (TTCR)] of eight prostate cancer-associated gene transcripts in 43 mHSPC patients. Transcripts with statistically significant associations (P<0.05) were further investigated in a metastatic castration-resistant prostate cancer (mCRPC) cohort (n=119) receiving contemporary systemic therapy, exploring associations with PSA >50% response (PSA<sub>50</sub>), progression-free survival (PFS) and overall survival (OS). Clinical outcomes were prospectively collected in a protected digital database. Kaplan-Meier estimates and multivariable Cox proportional-hazards models assessed associations between gene transcripts and clinical outcomes (mHSPC covariates: disease volume, docetaxel use and haemoglobin level; mCRPC covariates: prior exposure to chemotherapy or ARPIs, haemoglobin, performance status and presence of visceral disease). Follow-up was performed monthly during ARPI treatment, three-weekly during taxane chemotherapy, and three-monthly during androgen deprivation therapy (ADT) monotherapy. Serial PSA measurements were performed before each follow-up visit and repeat imaging was at the discretion of the investigator.RESULTS: Detection of circulating Grainyhead-like 2 (<i>GRHL2</i>) transcript was associated with poor outcomes in mHSPC and mCRPC patients. Detectable <i>GRHL2</i> expression in mHSPC was associated with a lower rate of seven-month undetectable PSA levels (25% <i>vs.</i> 65%, P=0.059), and independently associated with shorter TTCR (HR 7.3, 95% CI: 1.5-36, P=0.01). In the mCRPC cohort, <i>GRHL2</i> expression predicted significantly lower PSA<sub>50</sub> response rates (46% <i>vs.</i> 69%, P=0.01), and was independently associated with shorter PFS (HR 3.1, 95% CI: 1.8-5.2, P<0.001) and OS (HR 2.9, 95% CI: 1.6-5.1, P<0.001). Associations were most apparent in patients receiving ARPIs.CONCLUSIONS: Detectable circulating <i>GRHL2</i> was a negative prognostic biomarker in our mHSPC and mCRPC cohorts. These data support further investigation of <i>GRHL2</i> as a candidate prognostic biomarker in metastatic prostate cancer, in addition to expanding efforts to better understand a putative role in therapeutic resistance to AR targeted therapies.
urology & nephrology,andrology