Rectal Optical Markers for In Vivo Risk Stratification of Premalignant Colorectal Lesions

Abstract

Purpose: Colorectal cancer remains the second leading cause of cancer deaths in the United States despite being eminently preventable by colonoscopy via removal of premalignant adenomas. In order to more effectively reduce colorectal cancer mortality, improved screening paradigms are needed. Our group pioneered the use of low-coherence enhanced backscattering (LEBS) spectroscopy to detect the presence of adenomas throughout the colon via optical interrogation of the rectal mucosa. In a previous ex vivo biopsy study of 219 patients, LEBS demonstrated excellent diagnostic potential with 89.5% accuracy for advanced adenomas. The objective of the current cross-sectional study is to assess the viability of rectal LEBS in vivo.

Experimental design: Measurements from 619 patients were taken using a minimally invasive 3.4-mm diameter LEBS probe introduced into the rectum via anoscope or direct insertion, requiring approximately 1 minute from probe insertion to withdrawal. The diagnostic LEBS marker was formed as a logistic regression of the optical reduced scattering coefficient [Formula: see text] and mass density distribution factor D.

Results: The rectal LEBS marker was significantly altered in patients harboring advanced adenomas and multiple non-advanced adenomas throughout the colon. Blinded and cross-validated test performance characteristics showed 88% sensitivity to advanced adenomas, 71% sensitivity to multiple non-advanced adenomas, and 72% specificity in the validation set.

Conclusions: We demonstrate the viability of in vivo LEBS measurement of histologically normal rectal mucosa to predict the presence of clinically relevant adenomas throughout the colon. The current work represents the next step in the development of rectal LEBS as a tool for colorectal cancer risk stratification.

©2015 American Association for Cancer Research.

Figures

Figure 1

Clinical LEBS probe and point-of-care instrumentation used for data collection. (A) Mobile point-of-care system housing the data acquisition instrumentation, calibration mechanism, and custom operating software. (B) Backscattering light spectrum measured by the LEBS probe at 3 collection angles: −0.6°, +0.6° & +1.12° relative to the incident direction. (C) Demonstration of the scale of the LEBS probe relative to the hand of a clinician. Rectal LEBS measurements are taken via direct insertion into the colon. (D) Schematic showing the 3.4 mm diameter fiber-optic LEBS probe. The inset shows an image of the linear optical fiber array, with sample illumination via one of two inner fibers and collection of backscattered light via the remaining fibers.

Figure 2

Diagram and categorization of all enrolled patients. AA = patients with advanced adenomas and non-AA = patients with non-advanced adenomas.

Figure 3

Optical alterations of the rectal mucosa in patients exhibiting CRC field carcinogenesis enables accurate detection of clinically relevant adenomas. (A) Shows a significant decrease in the reduced scattering coefficient μs∗ for patients harboring advanced adenomas. (B) Shows a significant increase in the mass density distribution factor D. (C) Shows the diagnostic LEBS marker for patients in the validation set. (D) Shows the receiver operating characteristic (ROC) curves for validation set patients. The area under the curve (AUC) is 82% for advanced adenomas and 73% for multiple non-advanced adenomas. The x-axis of panels A-C are ordered according to increasing aggressiveness from left to right. Control = healthy controls, Non-AA = patients with non-advanced adenomas, AA = patients with advanced adenomas.

Figure 4

Diagnostic LEBS marker separated according to gender (panel A), other polyp types (panel B), and advanced adenoma location (panel C). (D) Diagnostic LEBS marker evaluated in patients without colonic preparation. In panel a, the P-values displayed below the LEBS marker for male and female patients harboring advanced adenoma are both calculated with respect to the corresponding gender of healthy control.