Passive bioelectrical properties for assessing high- and low-grade prostate adenocarcinoma

Authors

  • Ryan J. Halter,

    Corresponding author
    1. Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire
    2. Norris-Cotton Cancer Center, Lebanon, New Hampshire
    3. Dartmouth Medical School, Hanover, New Hampshire
    • Dartmouth College, Thayer School of Engineering, 8000 Cummings Hall, Hanover, NH 03755.
    Search for more papers by this author
  • Alan R. Schned,

    1. Norris-Cotton Cancer Center, Lebanon, New Hampshire
    2. Dartmouth Medical School, Hanover, New Hampshire
    3. Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
    Search for more papers by this author
  • John A. Heaney,

    1. Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire
    2. Norris-Cotton Cancer Center, Lebanon, New Hampshire
    3. Dartmouth Medical School, Hanover, New Hampshire
    4. Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
    Search for more papers by this author
  • Alex Hartov

    1. Thayer School of Engineering, Dartmouth College, Hanover, New Hampshire
    2. Norris-Cotton Cancer Center, Lebanon, New Hampshire
    Search for more papers by this author

Abstract

BACKGROUND

The electrical properties of prostate tissues are dependent on cellular morphology and have been demonstrated to distinguish between benign and malignant formations. Because Gleason grading is also based on tissue architecture we explored the hypothesis that the electrical properties might also provide discriminating power between high- and low-Gleason grade cancers.

METHODS

Electrical properties (σ, ε, Δσ, σ, fc, and α) were gauged from 546 prostate tissue samples and correlated with histopathological assessment. Primary and secondary Gleason grades and a Gleason score were assigned to the tissues identified as cancer. We evaluated how well differently graded cancers were separable from benign tissues and from each other on the basis of these properties using ROC curves.

RESULTS

Of the 546 prostate tissue samples, 71 were identified as cancer and 465 as benign. ε, Δσ, σ, and fc provided the most discriminatory power with area under the curves (AUCs) ranging from 0.77–0.82 for detecting any cancer, 0.72–0.8 for low-grade cancer, and increasing to 0.87–0.9 for detecting high-grade cancer. Further, ε, Δσ, and σ, provided AUCs ranging from 0.74 to 0.75 for discriminating between low- and high-grade cancers.

CONCLUSIONS

Using the electrical properties to identify prostate cancer is improved when high-grade cancers are sought. These electrical properties can also discriminate between different grades of tumors. These findings suggest that technologies being developed to sense and image these properties in vivo may discriminate between aggressive and indolent lesions. Prostate 71:1759–1767, 2011. © 2011 Wiley Periodicals, Inc.

Ancillary