Sniffing Out Bladder Cancer: Utilizing an E-nose Based on Eco-Friendly PANI Thin Films

Abstract

BlC is the seventh most common cancer and ninth leading cause of cancer related death. The
current standard procedure for BIC detection and monitoring tumor progression and
recurrence involves urine cytology, cystoscopy, and biopsy. However, its an expensive, low
sensitivity, invasive and painful test. [1] Thus, new diagnostic approaches that will improve the
diagnostic accuracy, less expensive, noninvasive and safe screening of BIC are needed. [2,3]
E-nose has been successfully used for diagnosis of health conditions via detection and
classification of VOCs associated with related disease in the human body, including exhaled
breath, skin/sweat, urine, saliva, feces and intestinal gas.
[4] This subject has been dealt in a
more modern context of biomedical research. The VOCs in fluids emerge at very early stages
of the cancer so their isolation and detection could serve as a pathway for early detection of
the cancer. Of these body fluids, urine is one of the most useful VOC sources for monitoring
body chemistry or state of health. Bladder urine is normally sterile and thus the presence of
pathogenic bacteria in urine is indicative of a urinary tract infection. [5] Matsumura et al.,
established the feasibility of using urinary volatiles to detect lung cancer for the first time. [6]
Roine et al., proved that both malignant and nonmalignant prostate cell lines have distinct
smell prints.[7] Asimakopoulos et al., reported a potential role of the e-nose in identifying
prostate cancer by smelling urine samples with high specificity.[8]
In the field of BIC, this study
trained dogs to dentify people with BIC based on urine odour with powerful results.[9] The main
success of this study was the demonstration that BIC gives an odor signature to the urine.
Bernabei et al., performed an e-nose based on eight quartz crystal microbalance (QCM) gas
sensors to prove the possibility of diagnosing urinary tract cancers. [10] Simian et al., outlined
a simple fluorometric optical sensor system to detect VOCs of BIC in patients, and also
achieved 66.67% sensitivity and 75.00% specificity for classification of high-grade and lowgrade bladder cancer patients. [11] However, the selective and cross-reactive sensing of BIC
VOCs have no reported and the sensitivity, accuracy and specificity needed improved
furthermore.
In this work, an e-nose based on eco-friendly polyaniline (PANI) thin films is successfully
applied to noninvasive diagnosis of BIC from urine sources. An e-nose is fabricated by 10
kinds of PANI thin films on the glass substrate covered with a pair of fluorine-doped tin oxide
(FTO) interdigitated electrode, and exhibits cross-reactive sensing for 11 types of BlC
biomarkers. Using urine samples are collected from BIC patients and healthy controls, then
feature extracted from origin response curves to select useful information and realize the
effective pattern recognition. The support vector machine (SVM) analysis shows excellent
discrimination between the BIC patients and healthy controls 100 % sensitivity and 100 %
accuracy. Overall, this study paves the way to use e-nose for simple, inexpensive, portable,

and noninvasive diagnosis of urological diseases. Figure 1 is the scheme illustrating the main
experimental procedure