Login to Access Video or Poster Abstract: MP83-01
Sources of Funding: Australian Prostate Cancer Research Centre New South Wales (APCRC-NSW to V.M.H.) with support to A.M.F.K., W.J. and A-M.H.; Chair support from the Petre Foundation and Sydney Medical School Foundation, Australia to V.M.H.; E.K.F.C. and D.C.P. are partially supported by the Movember Australia and the Prostate Cancer Foundation Australia (PCFA) Prostate Cancer Bone Metastasis (ProMis) Movember Revolutionary Team Award (MRTA) to the Garvan Institute of Medical Research (V.M.H. as a team lead).

Introduction

Prostate cancer is marked by clinical and pathological heterogeneity. Therefore, there is a critical need to identify biomarkers that predict clinical presentation and outcomes. While much attention has been given to the nuclear genome, less emphasis has been placed on the maternally inherited, circular mitochondrial genome. Alterations in mitochondrial DNA (mtDNA) copy number are common in many human cancers, distinguishing cancer from normal tissue. In contrast, preliminary studies in prostate cancer, have shown no relative difference in mtDNA copy number between matched tumor-normal prostate cells, suggesting a possible field effect. The aim of this study was to test if there is a difference in mtDNA copy number between prostate cancer patients, that could be used to differentiate disease stage and clinical outcome.

Methods

Fresh prostate cancer biopsies from 115 patients, with a median of 107 months clinical follow-up, were H&E stained and reviewed by pathologists for tumor grade and cellularity and marked for tumor excision. DNA was extracted from tumors and qPCR performed to establish relative mtDNA abundance. Multifocal biopsies were available and examined for seven of the patients.

Results

Multivariate linear models revealed a significant correlation of decreasing mtDNA copy number with estimated sample tumor cellularity (p = 0.049*) and increasing with tumor pathology ISUP score (p = 0.007**), but no correlation with disease relapse (p = 0.60) or age (p = 0.71). Data from single patient multifocal biopsies showed a similar mtDNA content (average of 1.68-fold difference, range 1.01-2.48), compared to between patient mtDNA content (upper and lower interquartile range 3.23-fold difference, total range 15.99-fold variation)

Conclusions

We report a positive correlation between mtDNA count and tumor pathology score. Lack of variation in mtDNA abundance between molecular lesions within the same patients adds to accumulating evidence of a field effect contributing to the multifocal presentation of over 70% of prostate cancers. An increase in mtDNA copy number may therefore provide an ideal predictive tool for pathological stage irrespective of tumor purity during prostate diagnostic biopsy.

Funding

Australian Prostate Cancer Research Centre New South Wales (APCRC-NSW to V.M.H.) with support to A.M.F.K., W.J. and A-M.H.; Chair support from the Petre Foundation and Sydney Medical School Foundation, Australia to V.M.H.; E.K.F.C. and D.C.P. are partially supported by the Movember Australia and the Prostate Cancer Foundation Australia (PCFA) Prostate Cancer Bone Metastasis (ProMis) Movember Revolutionary Team Award (MRTA) to the Garvan Institute of Medical Research (V.M.H. as a team lead).