Detection of prostate cancer by radio-frequency near-field spectroscopy in radical prostatectomy ex vivo specimens

Prostate Cancer Prostatic Dis. 2013 Mar;16(1):73-8. doi: 10.1038/pcan.2012.34. Epub 2012 Sep 11.

Abstract

Background: The aim of radical prostatectomy (RP) is the complete removal of the prostate gland with negative surgical margins. The presence of cancer at the surgical margin is associated with higher probability of disease progression. Current methods of intraoperative margin assessment are inaccurate or time-consuming.The study goal was to evaluate the ability of a novel device (Dune Medical Devices) to differentiate between cancer and BPH.

Methods: A total of 49 patients undergoing RP in four medical centers between November 2007 and May 2008 were enrolled in this study.The device was applied to numerous intra- and extra-capsular sites of freshly excised RP specimens. Measurement sites were accurately marked and analyzed histologically. The ability of the device to differentiate between malignant and nonmalignant sites was assessed.

Results: A total of 15,156 measurements from 45 patients were analyzed. Differentiation of the intra-capsular malignant sites from extra-capsular nonmalignant sites (bladder neck and apex regions) depends on the cancer feature size. Differentiation was achieved with sensitivity and specificity of 93.6 (95% confidence interval (CI): 88-98) and 94.1 (95% CI: 93-95), respectively, at feature sizes at or >0.8 mm in diameter. The device was able to discriminate between all intra-capsular malignant (with feature sizes down to a few cells) and nonmalignant measurement sites, with sensitivity and specificity of 80.8 (95% CI: 73-87) and 68.4 (95% CI: 67-69), respectively.

Conclusions: First results from a radio-frequency near-field spectroscopy sensor look promising for differentiation between cancer and benign prostate tissue. The sensor's dimensions (radius of ~ 1 mm) and design enable use in open, laparoscopic and robotic RP to evaluate the surgical margins intraoperatively.

MeSH terms

  • Area Under Curve
  • Humans
  • Male
  • Neoplasm, Residual / diagnosis*
  • Prostatectomy
  • Prostatic Neoplasms / surgery*
  • ROC Curve
  • Radio Waves*
  • Signal Processing, Computer-Assisted