Scent-trained canine detects thyroid cancer in urine samples, could aid in diagnoses
Click Here to Manage Email Alerts
SAN DIEGO — Canines trained in scent detection could be a reliable addition to practices now used to diagnose thyroid cancer, according to research presented at The Endocrine Society annual meeting.
With an 88% accuracy rate based on the preliminary data, and no clinic or technology required, canines could play an important role in diagnosis, according to researchers.
“The gold standard in determining whether or not a nodule is cancerous is a fine needle aspiration biopsy … and that has about a 95% accuracy rate,” Donald L. Bodenner, MD, PhD, of the University of Arkansas for Medical Sciences, Little Rock, said in a press conference. “We’re not there, but we’re very close in terms of a noninvasive and inexpensive adjunct to current diagnostic practices.”
Donald L. Bodenner
Bodenner, with Andrew M. Hinson, MD, and colleagues collected urine samples from 59 individuals who presented to the institution’s thyroid clinic between February and November 2014 with more than one nodule appearing suspicious for cancer.
For the analysis, only metastatic thyroid carcinoma (conventional type) (n = 15) and benign thyroid disease (n = 19) as confirmed by final surgical pathology were included.
A canine imprinted with urine, blood and thyroid tissue obtained from patients with metastatic thyroid carcinoma, and trained over 6 months to discern between samples with the disease and samples that were benign, served as the detector.
In the lab, a gloved handler blinded to sample status presented the canine with samples in 3-mL cryotubes. If metastatic thyroid carcinoma was present, the canine laid down; if the sample was benign, the canine turned away. The handler verbally communicated the canine’s alert to a blinded study coordinator for recording.
Control samples, known as cancerous or benign, were interspersed with unknown samples. The canine received positive reinforcement for correct answers.
The canine’s alerts aligned with the final surgical pathology diagnosis in 30 out of 34 cases (88.2%), translating to a sensitivity of 86.6% (13/15) and specificity of 89.5% (17/19). The positive and negative predictive values were 73.3% and 95.3%, respectively.
In a previous study, the researchers demonstrated that scent-trained dogs can discriminate between body fluids from patients previously diagnosed with thyroid cancer vs. benign thyroid disease, Bodenner highlighted. The canines detected 97% of both urine and blood samples correctly, with only two false negatives.
“The two false negatives were from the same patient, and it was a tall cell variant of thyroid cancer; there are several variants in addition to routine papillary thyroid cancer,” Bodenner said.
“We think the dogs are even capable of telling the difference between standard papillary thyroid carcinoma and some of these other variants.”
Looking ahead, the investigators hope to monitor high-risk cancer patients, believed to be cured but have a probability for recurrence between 10% and 15%, to see if there is a clinically significant recurrence.
Further, they are interested to find out what the dogs are smelling. “Right now, nobody has any idea what scent profile is really triggering their response,” Bodenner said. We want to use proteomics or metabolomics, or some other methodologies to try to pinpoint what it is.”
The relatively simple diagnostic methodology, requiring only a urine sample and reliable transportation for the sample, could be useful in underserved communities or third-world countries, he said.
Collaboration with the veterinary department at Auburn University, which predominantly supplies the Department of Defense with animals for bomb detection, is an important next step.
“With all these animals coming home from Afghanistan and Iraq, they don’t know what to do with them,” Bodenner said. “We’re going to try to repurpose some of those. The potential is very great to look at a variety of different cancers.” – by Allegra Tiver
Reference: Hinson AM. Poster Board FRI-036. Presented at: The Endocrine Society Annual Meeting; March 5-8, 2015, San Diego.
Disclosure: Bodenner reports no relevant financial disclosures.