Non-invasive multiple cancer screening using trained detection dogs and artificial intelligence: A prospective double-blind study

Abstract Purpose: SpotitEarly Ltd. has developed a simple non-invasive self-administered screening method to detect cancer in humans. We conducted a double-blind study that evaluated the specificity and sensitivity of this method in detecting breast, lung, prostate, and colorectal cancer in a single test. The method's performance in detecting 14 additional malignancies was also evaluated. Experimental design: Breath samples of adults who underwent screening for cancer using gold-standard screening methods, or a biopsy for a suspected malignancy were collected. The samples were analyzed using a bio-integrated platform of trained detection canines and artificial intelligence (AI) tools. Specificity and sensitivity were analyzed. Results: Overall, 1386 participants (59.7% males, median age 56.0 years) were included. According to cancer screening/biopsy results, 1048 (75.6%) were negative for cancer and 338 (24.4%) were positive. Among the 338 positive samples, 261 (18.8%) were positive for the four cancer types that the canines were trained to detect, with an overall sensitivity and specificity of 93.9% (95% confidence interval [CI] 90.3%-96.2%) and 94.3% (95% CI 92.7%- 95.5%), respectively. The sensitivity of each cancer type was similar; breast: 95.0% (87.8%-98.0%), lung: 95.0% (87.8%-98.0%), colorectal: 90.0% (74.4%-96.5%), prostate: 93.0% (84.6%-97.0%). The sensitivity of other malignant tumors that the canines were not trained to detect was 81.8% (95% CI 71.8%-88.8%). The sensitivity of early-stage cancer detection (stages 0-2) was 94.8% (95% CI 91.0%-97.1%). Conclusions: A bio-hybrid multi-cancer screening platform, combining detection canines and AI tools using breath samples, demonstrated high sensitivity and specificity. This platform enabled early-stage cancer detection of multiple cancer types.