Spectral biomarkers for chemoprevention of colonic neoplasia: a placebo-controlled double-blinded trial with aspirin.

Objective: A major impediment to translating chemoprevention to clinical practice has been lack of intermediate biomarkers. We previously reported that rectal interrogation with low-coherence enhanced backscattering spectroscopy (LEBS) detected microarchitectural manifestations of field carcinogenesis. We now wanted to ascertain if reversion of two LEBS markers spectral slope (SPEC) and fractal dimension (FRAC) could serve as a marker for chemopreventive efficacy.
Design: We conducted a multicentre, prospective, randomised, double-blind placebo-controlled, clinical trial in subjects with a history of colonic neoplasia who manifested altered SPEC/FRAC in histologically normal colonic mucosa. Subjects (n=79) were randomised to 325 mg aspirin or placebo. The primary endpoint changed in FRAC and SPEC spectral markers after 3 months. Mucosal levels of prostaglandin E2 (PGE2) and UDP-glucuronosyltransferase (UGT)1A6 genotypes were planned secondary endpoints.
Results: At 3 months, the aspirin group manifested alterations in SPEC (48.9%, p=0.055) and FRAC (55.4%, p=0.200) with the direction towards non-neoplastic status. As a measure of aspirin's pharmacological efficacy, we assessed changes in rectal PGE2 levels and noted that it correlated with SPEC and FRAC alterations (R=-0.55, p=0.01 and R=0.57, p=0.009, respectively) whereas there was no significant correlation in placebo specimens. While UGT1A6 subgroup analysis did not achieve statistical significance, the changes in SPEC and FRAC to a less neoplastic direction occurred only in the variant consonant with epidemiological evidence of chemoprevention.
Conclusions: We provide the first proof of concept, albeit somewhat underpowered, that spectral markers reversion mirrors antineoplastic efficacy providing a potential modality for titration of agent type/dose to optimise chemopreventive strategies in clinical practice.
Trial Number: NCT00468910.
(Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/.)