Trained dogs can detect canine urothelial carcinoma of the bladder

Trained dogs can detect canine urothelial carcinoma of the bladder | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Trained dogs can detect canine urothelial carcinoma of the bladder Isabelle Desmas-Bazelle, Nicola Jane Rooney, Steve Morant, Rob Harris, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5022433/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Dog have been trained to detect a variety of human cancers. However, studies examining their ability to find canine cancer are sparse and show variable results. Whilst dogs were shown to be successful at detecting malignant tumors from saliva samples, they were previously unsuccessful at finding canine urothelial carcinoma (UC) from urine. We used advanced training methods to explore if dogs can accurately detect urothelial carcinoma (UC) from canine urine. Urine samples from healthy dogs without urinary tract abnormalities (control), with non-malignant urinary tract disease (control), and with UC (positive) were collected prospectively. Urothelial carcinoma was diagnosed using cytology and/or histopathology. All dogs were naïve of chemotherapy, underwent bladder ultrasound and urinalysis. Residual urine remaining after conventional diagnostic procedures was frozen in glass jars until analysis and used for dog training and testing. Three dogs previously trained to detect human cancer underwent up to 29 days of additional training. They were each presented with 20 control samples (from either 13 or 14 unique individuals), 10 UC positive samples (from between 5 and 7 individuals), in a double-blind trial and their behavioral responses recorded. Overall, 80% of UC were correctly indicated, and 91.7% of controls correctly ignored (68% and 87% when just considering only unique donor dogs). Individual dogs showed sensitivities ranging from 70 to 90%, and specificities ranging from 85 to 95%. This study suggests that, when using appropriate methods, dogs can be trained to detect canine UC from urine, and could serve as an additional rapid, non-invasive diagnostic test for the disease. Figures Figure 1 Introduction Cancer is a major cause of morbidity and mortality in pet dogs, who present at roughly the same rate as humans. Approximately one in four dogs will develop cancer during their lifetime, and, like humans, the incidence increases rapidly with age (Schwartz et al 2022 , Kraus et al 2023 ), so half of all dogs over the age of ten years will develop cancer during their lifetime AVMA 2022). Urothelial carcinoma (UC), also known as transitional cell carcinoma (TCC), is the most common form of bladder cancer accounting for approximately 2% of all reported canine cancers (Fulkerson and Knapp 2015 ). It is an aggressive, highly invasive, lower urinary tract tumor in dogs (Tagawa et al 2020 ). Clinical presentation of UC is shared with several other disorders, including non-malignant conditions, making diagnosis challenging. Definitive diagnosis requires histopathology, which is invasive, costly, and time consuming, thereby delaying the start of treatment. Cystocentesis and needle sampling of bladder masses should be avoided due to the risk of tumor seeding (Hayashi et al 2020 ). Cytology of the urine (obtained via free catch sample) however lacks sensitivity, since sediment reveals tumor cells in only 30% of cases (Norris et al 1992 ). Two other diagnostic tests are currently available: the bladder tumor antigen test which is poorly specific (Henry 2003 ; Borjesson and Christopher 2009) and the BRAF mutation detection test (Tagawa et al 2020 ). The latter utilises DNA from urine and has a sensitivity of 83% in patients since only 83% of UCs carry the mutation (Mochizuki et al 2015 ). At diagnosis, canine UCs are often advanced (Fulkerson and Knapp 2015 ), hence there is great potential value in a new early, cheap, rapid, and non-invasive diagnostic test for UC. Dogs have been trained to reliably detect a large number of human medical conditions ( Rooney 2016 ) including osteosarcoma (Ortal et al 2022 ), bladder (Wills 2004) and lung (Feil et al 2021 ) cancer. It’s logical from an evolutionary perspective that they would also have the ability to detect canine cancer. An initial study trialed the use of four laboratory trained dogs to detect canine urinary tract cancer, from urine (Dorman et al 2017 ). Three dogs failed to learn the task and the fourth performed at chance levels during double-blind testing. However, training was on a small number of samples (only eight positive samples), to which dogs were repeatedly exposed and hence they may have learnt to identify individual urine donors rather than to discriminate cancer from controls. The authors concluded that sample storage, confounding odours, and other factors need to be considered in the design of future studies. The importance of these issues is now well understood in the medical detection dog field (Guest et al 2020 ). Greater success was seen when Malone et al ( 2023 ) trained six pet dogs to distinguish saliva samples from dogs with malignant tumors from health controls, showing mean sensitivity of 90% and specificity of 98%. Here they used a considerably larger number of sample donors for training (78 positive and 109 controls). Urine poses a more practicable, less invasive medium to screen for canine cancer compared to saliva. What’s more since UC is located in the urinary tract with which urine makes contact, it seems logical that, it would be an ideal medium for detection. However, there is the possibility that dog urine is so innately interesting to other dogs (Rooney and Parr Cortes 2023), that natural curiosity in the semiochemical signal overrides any trained categorisation. We therefore wished to explore if the methodological challenges that impacted the Dorman (2017) study, could be overcome to enable reliable detection of UC by dogs. The objective of this study was to investigate whether trained dogs can accurately detect canine urothelial carcinoma (UC) from urine. Materials and methods Ethics This study was approved by the Royal College Veterinary Surgeon’s (RCVS) Ethics Review Panel. Subjects Dogs were recruited from canine patients undergoing procedures performed during investigations for urinary tract problems at a single referral veterinary centre. The owners of all dogs attending the clinic for suspected urinary problems over a 5-year period of time were asked to consent to participation in the study. All recruitment and sampling was done prior to the commencement of chemotherapy. Each case was diagnosed via routine procedures utilising the least invasive method. Typically, cytology of urine was performed first, but if this was inconclusive then urethral catheterisation was used for suction biopsy. If both diagnostic methods were inconclusive, a biopsy was performed either via catheterisation with endoscopic forceps, or via cystoscopy. The method chosen was deemed the best way of performing the clinical workup in order to achieve a definitive diagnosis by the attending clinician. Dogs were thereby classified as positive for urothelial carcinoma or for non-malignant urinary tract disease via cytology or histopathology. Urine samples from dogs with non-malignant urinary tract disease (control), and dogs with UC (positive) were collected prospectively. Whilst study participation did not alter dogs’ investigation or treatment path, it permitted researchers to access veterinary records and allowed superfluous urine collected during veterinary investigation to be stored and used for dog training and testing. Only leftover urine was used so no extra samples were taken for the purpose of the study. Urine samples were frozen in glass aliquots within 30 minutes of collection at -80C. Samples were allocated a unique code name to respect General Data Protection Regulation (UK Government 2018). They were then transported on dry ice to Medical Detection Dog (MDD). On arrival, the samples were spilt into a training set (17 positives and 22 controls) and a testing cohort (9 positives and 18 controls). Each sample was split into two aliquots which were stored separately. The testing group were coded so all staff remained blinded. All testing samples were from individuals not included in the training set, so were naïve to the dogs. Samples were stored at -80C until the day of training. Additional control training samples Twenty additional healthy control dogs were recruited from those belonging to Medical Detection Dogs and staff and volunteers of the charity. Inclusion criteria was absence of urinary signs, normal bladder ultrasound scan and urinalysis, no comorbidities, and never having had chemotherapy. After the owner’s consent, all dogs underwent a conscious ultrasound scan of their bladder (carried out by VV,) and a free catch urine sample was collected. Ultrasonography was performed using a microconvex probe (SC 3123, 3–10 MHz), in both sagittal and transverse planes (MyLab Twice Esaote and MyLab Sat Esaote, Genova, Italy). At least moderate bladder filling was required for control dogs for accurate assessment of the bladder wall. History and veterinary records were also assessed to confirm healthy status and an absence of urinary tract clinical signs. Non-steroidal anti-inflammatory drugs (NSAIDs) were permitted in controls and patients alike. Dog training Four dogs who had previously been trained to detect human bladder and prostate cancer were trained to distinguish positive canine cancer samples from controls, using positive reinforcement. The training protocol was increased in difficulty over time. Initially, the new target scent was introduced to the dog, and the dog rewarded for showing a behavioral change in response. All dogs in this study showed a sit and stare response. Correct responses were rewarded with food or ball-play, in conjunction with the use of an audible clicker (used as a secondary reward). Three dogs had been trained to walk along a line of four stainless steel stands, one metre apart, each with an arm holding a 40 ml glass jar containing a swab sample. The fourth dog was trained to search a single sample at a time. A stainless-steel grill was placed over the mouth of each jar to prevent the dog touching the sample The dogs were trained to sniff each stand and show their indication behavior in response only to positive samples. Dogs were rewarded for both indicating positive target samples and for searching a negative line (or single stand for dog 3) without an indication. Their behavior was recorded on a database as IND (indicated), HES (hesitated), INT (interest, but weaker than HES), NI (no interest) or NS (not searched, usually because a sample earlier in the line was indicated). Initially, the line trained dogs were presented with one positive sample and three negative samples during each run. As training progressed, sequences of four negative samples were also introduced. After each line was searched, the plates were removed for cleaning in an ultrasonic cleaner with an enzymatic solution (Reprozyme TMlook),. They were then rinsed thoroughly using a commercial glass washer (minimum 85°C) and left to air dry to avoid any contamination. After five sessions, one dog (one trained on the line arrangement) was deemed not to be learning the task, and was also suffering with lameness. It was therefore eliminated from the study. The remaining three dogs were one male Cocker Spaniel (11.5 years), one female Cocker Spaniel (8 years) and one male Labrador Retriever (4 years), all neutered. They were trained over a period of 148 days, during which they underwent between 23 and 29 training sessions, encountering between 602 and 1072 training samples from between 76 and 79 different individual donors. Double blind testing Table 1 here Table 1 Details of target and control samples used during double blind testing including dogs’ age, sex, health or disease status Diagnosis urothelial carcinoma (UC) or nonmalignant urinary tract disease (MUTD) Breed Sex Neutered status Age at Sampling Method of urine collection Site of UC (if appropriate) Method of diagnosis (cytology or histology) UC West Highland White Terrier F Neutered 11 y 6 mo Free Catch Bladder neck histology UC * Springer Spaniel F Neutered 7 y 9 mo Catheter Ventral and mid area of the bladder cytology UC Staffie F Neutered 11 y 1 mo Free Catch Bladder wall histology UC Labrador F Entire 13 y 1mo Catheter Bladder neck and urethra cytology and histology UC Maltese F Entire 8 y 2 mo Catheter Caudo dorsal bladder wall cytology UC West Highland White Terrier M Neutered 7 y 6 mo Free Catch Bladder neck and proximal urethra cytology UC Hungarian Vizsla M Neutered 11 y Catheter Cranial and mid third of bladder cytology UC Boston Terrier M Neutered 9 y 9 mo Catheter Bladder neck cytology NMUTD Tibetan Spaniel M Entire 5 y 1 mo Cystocentesis NMUTD English Springer Spaniel M Neutered 8 y 1 mo Cystocentesis NMUTD Crossbreed F Neutered 2 y 3 mo Free Catch NMUTD Jack Rusell Terrier M Entire 6 y 4 mo Free Catch NMUTD Border Terrier F Neutered 4 y 11 mo Free Catch NMUTD Border Collies F Neutered 7y 3mo Free Catch NMUTD Crossbreed F Neutered 6 yo Free Catch NMUTD Brittany Spaniel F Neutered 10 y 2 mo Free Catch NMUTD Labrador F Neutered 6 y Free Catch NMUTD Griffon M Neutered 4 y 4 mo Free Catch NMUTD Golden Retriever M Neutered 11.5 yo Free Catch NMUTD Crossbreed F Neutered 4 y 11 mo Cystocentesis NMUTD Miniature Schnauzer F Neutered 3 y 9 mo Cystocentesis NMUTD Crossbreed F Neutered 5 y 8 mo Free catch NMUTD Crossbreed F Neutered 2 y 4 mo Cystocentesis NMUTD Labrador Retriever M Neutered 12 y 2 mo Free Catch NMUTD Rhodesian Ridgeback M Entire 3 mo Cystocentesis NMUTD Greyhound M N 4 y 10 mo Cystocentesis Each dog was presented with a different set of 20 control samples (from either 13 or 14 unique individuals), 10 positive samples (from between 5 and 7 individuals) and 10 fillers (control samples previously encountered during training, and hence not included in the statistical analysis. The samples came from 27 different donor dogs (Table 1 ). Dog 3 was presented with samples one at a time, as this was the method which had been used in their training. For dogs 1 and 2, samples were randomised in blocks of four, each including two controls, one positive and one filler sample; using the same four stand system as during training. They worked along each line of four samples in order. If they showed their trained indication (alert) behavior to one sample the remaining samples in this line were recorded as “not seen” (NS) in that run, and the line was presented again in reverse order at a later time. If the handler deemed it necessary, dogs were permitted to the search the samples twice before they reported the behavioral response to be recorded. Any samples still not seen after the forward and reverse runs were assembled into new lines, until all samples were searched. Five control and two positive samples from the training set were used for calibration runs, to confirm the dog was focused on the scent. Calibration runs were requested at any time by the trainer, were unblinded and their results were not included in the analysis. A blinded experimenter entered the responses to each sample into the database. All trial samples had previously been given a “blind ID” which was read by the data collection software to determine whether responses were correct. The software revealed whether the response was correct, and if so, the trainer was asked to reward their dog. If any sample was incorrect, no reward was given. The database was password protected and only the trial statistician (SM) knew both passwords prior to trial completion. Amongst the NMUTD dogs, six had a follow up of more than 6 months with four dogs still alive at the time of writing, and 3 had a follow up between 1 and 5 months. None of these dogs were subsequently diagnosed with UC. Statistical methods The database was scanned for the first occasion each dog encountered each sample in a valid run, ignoring occasions when a sample was set out in a line but not seen ( NS ). Subsequent encounters were ignored. Only behaviors recorded as IND were regarded as positive responses (indications), and all other behaviors ( HES , INT and NI ) were classified as negative responses. We used the first encounter with each novel sample to calculate odds ratios for indicating target vs control samples) sensitivity (proportion of target samples found) and specificity (proportion of control samples that were not indicated) for each dog. Due to a shortage of samples, the testing sample cohort, although all unique, were not all from different dogs. Therefore, we also conducted a sensitivity analysis using only the first sample encountered from each subject dog by each detection dog. We compared the likelihood of dog alerting with sampling method cystocentesis vs free catch for controls and catheter or free catch for target samples. Results for calibration samples and filler samples were recorded but they were not included in the statistical analyses. For target and control samples, logistic models were fitted to test for differences in the proportions indicated. Odds ratios were estimated for each dog individually, and overall. Data processing and analysis were conducted using R (R Core Team 2020 ). Results Table 2 here Table 2 Number of presentations and odds ratios for indications to targets and controls during double blind testing Targets Controls Dog Analysis r/n % (95% CI) r/n % (95% CI) Odds ratio Probability Dog 1 All samples 9/10 90.0 (53.3,99) 1/20 5.0 (0.7, 28) 171.0 (9.6,3055) p < 0.001 Distinct dogs a 6/7 85.7 (41.9,98) 1/13 7.7 (1.1, 39) 72.0 (3.81,1362) p = 0.004 Dog 2 All samples 8/10 80.0 (45.9,95) 3/20 15.0 (4.9, 38) 22.7 (3.1,164) p = 0.002 Distinct dogs a 4/6 66.7 (26.8,92) 3/14 21.4 (7.1, 49) 7.3 (0.88, 61) p = 0.066 Dog 3 All samples 7/10 70.0 (37.6,90) 1/20 5.0 (0.7, 28) 44.3 (3.9,500) p = 0.002 Distinct dogs a 2/5 40.0 (10.0,80) 1/13 7.7 (1.1, 39) 8.0 (0.53, 121) p = 0.133 Overall All samples 24/30 80.0 (62.1,91) 5/60 8.3 (3.5,19) 44.0 (12.2,158) p < 0.001 Distinct dogs a 12/18 66.7 (42.9,84) 5/40 12.5 (5.3, 27) 14.0 (3.61, 54) p < 0.001 The difference in the proportions of control and target samples indicated was highly significant for all three dogs (Table 2 ), when all samples were considered. The average sensitivity (% of target samples alerted to) was 80% and the average specificity (% of controls correctly not alerted to) was 91.7% (i.e. 100- minus 8.3% controls indicated). When we calculated sensitivity and specificity based only on the first presentation of each patient dog, both sensitivity (66.7%) and specificity (87.5%) rates were reduced. Considering just the samples from distinct dogs, the differences in the proportion indicated from target (as compared to control) samples remained significant for the population overall and for Dog 1 individually. Whether the control samples were collected via cystocentesis or free catch did not significantly affect the odds of the dogs alerting (17.6% vs 5.9%: odds ratio: 3.4(0.5, 23), p = 0.203). Similarly for targets, the odds of an alert in response to a sample, collected via catheterization (81%) was not significantly different to one collected by free catch (77.8%; 1.21 (0.18,8.2), p = 0.842) Discussion Whilst previous studies failed to train dogs to detect UC from canine urine (Dorman et al 2017 ), our study shows that dogs can be trained to reliably detect urine from patients with UC with high sensitivity (80%) and specificity (91.7%). This is likely due to the increased number of samples used here in training which ensured that possible confounds were ameliorated (Guest et al 2020 ) and that dogs learnt to categorise the samples based on cancer status rather that learning individual samples. These detection rates were similar to those found when dogs were trained to detect malignant tumours from saliva (Malone et al 2023 : sensitivity 90% specificity 98%). Urine can be collected relatively easily during natural voiding, which is less invasive than saliva collection. This is especially useful for patients which are challenging to handle due to fear or other behavioral issues, making urine an ideal media for screening by medical detection dogs. Furthermore, it is easier to collect a large amount of urine compared to saliva. However, dogs have a strong drive to smell one another’s urine. It is a semiochemical signal used in inter-dog communication, to derive information about social (Lisberg and Snowdon, 2009 ) as well as reproductive status (Jezierski et al., 2019 , Woszczylo et al., 2020 ). It was feared that dogs’ innate interest may override any training. We have shown that with systemic and well-planned training, this is not the case and it is possible to condition dogs to respond reliably to cancer status, although the urine attractiveness may have plausibly contributed to some of the errors seen. When considering all samples, average and individual dog sensitivity and specificity levels were all well above chance and comparable to levels seen in dogs trained by the same training organisation for the detection of malaria parasites from worn socks (Guest et al 2019: sensitivity 70–73% and specificity 90.3–91%), SARS Covid-2 from sweat samples (Guest et al 2022 ; sensitivity: 82–94; specificity 76–92%), and Pseudomonas aeruginosa bacterial pathogens from supernatant (Davies et al 2019: sensitivity and specificity > 90% depending upon comparative control and dilution). Rates were above those obtained when using dogs to detect some human cancers, such as human bladder cancer (e.g. success rate 41% Willis et al 2004 ). We suggest that future refinements to training and sample presentation may lead to even further increases in detection accuracy. Due to a shortage of samples, the testing sample cohort although all unique, were not all from different dogs. When we calculated sensitivity and specificity based only on the first presentation of each patient dog, both sensitivity (mean = 66.7%) and specificity (mean = 87.5%) rates were reduced, in each dog. This may be due to the duplicated samples being easier for dogs to identify than some others, or a consequence of learning from the first encounter. However, even when just using first encounters, the three dogs together indicated cancerous samples significantly more often than controls. These results were still above chance and above previous published results using dog urine (Dorman et al 2017 ). Unlike the previous study (Dorman et al 2017 ), our population comprised two different control populations, dogs with no underlying urinary disease and those with non-malignant urinary condition. This seems more comparable and hence applicable to veterinary practice where diagnostic tests could be used both within a population of seemingly healthy dogs (in order to detect an underlying cancer) or for dogs showing urinary signs, in order to distinguish between cancer and inflammatory or infectious diseases. Within this study the method by which the urine was collected did not affect the likelihood of the dogs correctly responding to a positive sample, nor correctly ignoring a control sample. However, we used relatively few samples, and hence it would prudent to test this further on a larger population. It's noteworthy that the detection dogs in our study showed considerable individual variability, in spite of relatively standardised training methods. Whilst the sensitivity of Dog 1 was 90%, that of Dog 3 was only 70%. The methods of presentation of samples may have contributed to this, as Dog 3 was presented with samples one at a time, unlike Dogs 1 and 2 who searched lines of four samples at a time. The detection rate differences may also reflect personality differences between the dogs (Bistre Dabah 2023). Further analysis using the data stored during training will enable understanding of the extent to which these individual differences develop or whether they are consistent from early training. Additional studies with a larger number of detection dogs would also be beneficial. Overall, the dogs showed high detection rates, similar to that of current diagnosis techniques, with the BRAF mutation test for example yielding detection rates of 83% (Mochizuki et al. 2015 ). It would be very useful to study whether the samples missed by the dogs would also be missed by the other screening tests, and whether the presence of a BRAF mutation leads to a difference in canine detection rates. If not, then there may be great value in adding the dog to the suit of diagnosis tools available. Olfactory testing by trained dogs has the capacity to be very rapid, and it would be feasible to take canine urine samples to a medical detection dog facility for rapid screening, potentially more quickly than current laboratory facilities. However, the practicality of storing samples at -80 C, or transporting on ice may be challenging and hence research to ascertain the relative importance of different methodological aspects (Parr Cortes et al submitted) would be valuable. For example, exploring to what extent storage temperature of the urine samples affects detection rates, since no link was found between the efficacy of canine olfactory detection of male prostate cancer and the testing temperature within the range 2–23 ° (Urbanova et al 2023). If storage at -80 C is not required then this could increase the practicality of the dog as screening tool. Future work would ideally also use a larger cohort of donor dog samples for training to ensure detection dogs are exposed to variation in breed, sex, neuter status, diet, age, medication, environment (e.g smoking household), and non-urinary diseases status. These factors could all alter the scent of urine and add challenge to the dogs, hence future research should explore the effect of each of these factors on the dogs’ accuracy during testing. We also suggest methodical comparisons of the detection rates using different media (e.g. urine vs saliva) using the same training approaches. For bladder cancers such as UC, we expect volatile organic compounds (VOCs) to be especially concentrated in the urine, and hence it may be an ideal media for this cancer type. However VOCs associated with cancer outside of the urinary tract may also be concentrated or more readily detected in urine, Feil et al ( 2021 ) found higher accuracy when dogs used urine compared to breath samples to detect human lung cancer. The viability of urine for detection of other canine cancers is yet to be fully explored, as is the use of other media such as saliva for UC detection. Future trials should examine whether urine is the optimal media for all cancer types, or whether using the bodily fluid most closely associated with the affected site is optimal. Future work should also test dogs’? ability to distinguish UC from control samples from dogs with cancers located in other anatomical locations. Studies to explore detection of feline cancers are similarly ongoing (Pellin et al 2024 ). Longitudinal studies of dogs’ capacity to alert prior to formal diagnosis are also required. These should also examine the feasibility of using dogs for screening breeds predisposed to UC such as the West Highland White Terrier (Pantke 2018 ). Large prospective cohort studies would be ideal. In conclusion, the sensitivity and specificity of dogs trained to find canine UC samples is encouraging, suggesting that further research has the potential to help develop methods for the early diagnosis of UC, by dogs or by other technology such as electronic noses. This may in turn lead to opportunities for earlier intervention than is possible at present. Declarations Ethics This study was approved by the Royal College Veterinary Surgeons Ethics Review Panel. Consent The owners of all dogs attending the clinic for suspected urinary problems over a 5-year period of time were asked to consent for participation in the study. Availability of data and materials Due to the sensitivity of the data involved, these data are published as a restricted dataset at the University of Bristol Research Data Repository data.bris, at https://doi.org/10.5523/bris.1cq4ulhrjdmpf240uhjb2o6jov. The metadata record published openly by the repository at this location clearly states how data can be accessed by bona fide researchers. Requests for access will be considered by the University of Bristol Research Data Service, who will assess the motives of potential data re-users before deciding to grant access to the data. No authentic request for access will be refused and re-users will not be charged for any part of this process Competing interests We are not aware of any conflicts of interest Funding This study was unfunded Authors' contributions CG, RH ID conceived the study and devised the initial design RH trained and supervised the training of the dogs ID , VV, SA and MG contributed to data collection SA was responsible for data curation SM,NJR and RH conducted data analysis NJR and ID drafted the MS RH, CG, reviewed the MS All authors approved the final version Acknowledgements We would like to thank all the staff of Davies Veterinary Specialists and Medical Detection Dogs who helped facilitate recruitment and data collection. We are very grateful to all dogs and their owners who participated. We thank all MDD trainers who contributed to the training of the dogs This study was presented in poster form at VCS congress 4-6 November 2021 References R Core Team. 2020. R: A Language and Environment for Statistical Computing . Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/ . AVMA (American Veterinary Medicine Association) 2022. Cancer in pets. AVMA . Accessed October 30, 2022. https://www.avma.org/resources/pet-own Hayashi T, Hirokawa M, Higuchi M, Kudo T, Ito Y, Miyauchi A. Needle Tract Implantation Following Fine-Needle Aspiration of Thyroid Cancer. World J Surg. 2020;44(2):378–384. doi: 10.1007/s00268-019-05161-1 . PMID: 31515571.ers/petcare/cancer-pets Borjesson, D.L., Christopher, M.M., Ling, G.V. 1999. Detection of canine transitional cell carcinoma using a bladder tumor antigen urine dipstick test. Veterinary Clinical Pathology 28: 33–38. Bistre Dabah S. 2023. Investigating the effect of individual attributes on dog-performance in medical detection tasks. Unpublished PhD thesis. University of Bristol. Davies, J.C., Alton,E.,; Simbo, A., ; Murphy, R., ; Seth, I., Williams, K., ; Somerville, M., Jolly, L.,Morant, S., Guest, C. 2019. Training dogs to differentiate Pseudomonas aeruginosa from other cystic fibrosis bacterial pathogens: not to be sniffed at? European Respiratory Journal 54 (5) 1900970. Dorman, D.C., Foster, M.L., Fernhoff, K.E., Hess, P.R. 2017. Canine scent detection of canine cancer: a feasibility study. Vet Med (Auckl). 8:69–76 https://doi.org/10.2147/VMRR.S148594 Feil, C., Staib, F., Berger, M.R. et al. 2021. Sniffer dogs can identify lung cancer patients from breath and urine samples. BMC Cancer 21, 917 https://doi.org/10.1186/s12885-021-08651-5 Fulkerson, C.M., Knapp, D.W. 2015. Management of transitional cell carcinoma of the urinary bladder in dogs: a review. Vet J. 205(2):217 – 25. doi: 10.1016/j.tvjl.2015.01.017 . PMID: 25747698. Guest, C., Harris, R., Anjum, I., Concha, A. & Rooney, N. J. 2020. A lesson in standardisation -subtle aspects of the processing of samples can greatly affect dogs' learning: subtle aspects of the processing of samples can greatly affect dogs' learning : Frontiers in Veterinary Science. Guest, C., Pinder, M., Doggett, M., Squires, C., Affara, M., Kandeh, B., Dewhirst, S., Morant, S., V; D'Alessandro, U.,; Logan, J. G., Lindsay S.W., 20129. Trained dogs identify people with malaria parasites by their odour The Lancet Infectious Diseases , 19 (6,): 578–580, Guest, C., Dewhirst, S. Y., Lindsay, S.W. et al 2022. Using trained dogs and organic semi-conducting sensors to identify asymptomatic and mild SARS-CoV-2 infections: an observational study J. Travel Medicine , 29, 3, Hayashi T, Hirokawa M, Higuchi M, Kudo T, Ito Y, Miyauchi A. 2020. Needle Tract Implantation Following Fine-Needle Aspiration of Thyroid Cancer. World J Surg. 44(2): 378–384. doi: 10.1007/s00268-019-05161-1 . PMID: 31515571. Henry, C.J. 2003. Evaluation of a bladder tumor antigen test as a screening test for transitional cell carcinoma of the lower urinary tract in dogs. Am J Vet Res 64(8): 1017–1020. Jezierski, T., Dzieciol, M., Szumny, A., Nizanski, W., Woszczylo, M., Pieczewska, B., Godzinska, E. J. 2019. Discrimination of estrus odor in urine by male dogs in different experimental settings. Journal of Veterinary Behavior-Clinical Applications and Research, 29, 25–30. Kraus C, Snyder-Mackler N, Promislow DEL. 2023. How size and genetic diversity shape lifespan across breeds of purebred dogs. Geroscience;45(2):627–643. doi: 10.1007/s11357-022-00653-w Lisberg, A. E., Snowdon, C. T. 2009. The effects of sex, gonadectomy and status on investigation patterns of unfamiliar conspecific urine in domestic dogs, Canis familiaris . Animal Behaviour, 77, 1147–1154. Malone, L. A., Pellin, M. A., & Valentine, K. M. 2023. Trained dogs can accurately discriminate between scents of saliva samples from dogs with cancer versus healthy controls, Journal of the American Veterinary Medical Association (published online ahead of print 2023). Retrieved May 11, 2023, https://doi.org/10.2460/javma.22.11.0486 Mochizuki, H., Shapiro, S.G., Breen, M. 2015) Detection of BRAF Mutation in Urine DNA as a Molecular Diagnostic for Canine Urothelial and Prostatic Carcinoma. PLoS ONE 10(12): e0144170. https://doi.org/10.1371/journal.pone.0144170 Norris, A.M., Laing, E.J., Valli, V.E., et al. 1992. Canine bladder and urethral tumors: A retrospective study of 115 cases (1980–1985) J Vet Intern Med. ;6:145–153. Ortal, A., Rodríguez, A., Solis-Hernández, M.P. et al. 2022. Proof of concept for the use of trained sniffer dogs to detect osteosarcoma. Sci Rep 1: 6911 https://doi.org/10.1038/s41598-022-11013 Pantke P. 2018. Diagnostik und Therapie des Übergangszellkarzinoms des unteren Harntraktes beim Hund. Kleintierpraxis. 63: 76–92. Parr-Cortes, Z., Rooney N.J., Wheatstone M., Stock , T., Pesterfield, C., Guest, C., Müller C.T. submitted Evaluating and contextualising volatile organic compounds in dog training breath samples from a patient with type 1 diabetes. Journal of Veterinary Behavior Medicine: clinical applications. Pellin, M. A., Malone, L. A., & Ungar, P. 2024. The use of sniffer dogs for early detection of cancer: a One Health approach. American Journal of Veterinary Research , 85 (1), ajvr.23.10.0222. Retrieved Feb 9, 2024, from https://doi.org/10.2460/ajvr.23.10.0222 Rooney N.J. 2016. A medical detection role for dogs International Animal Health Journal. 3, (4), 42. Rooney, NJ., Parr-Cortes Z. 2023., Olfaction and Welfare In Lazarowski, L. (ed) Olfactory research in Dogs Springer Link https://doi.org/10.1007/978-3-031-39370-9 Schwartz SM, Urfer SR, White M, Megquier K, Shrager S, 2022. Dog Aging Project Consortium, Ruple A. Lifetime prevalence of malignant and benign tumours in companion dogs: cross-sectional analysis of Dog Aging Project baseline survey. Vet Comp Oncol . 2022;20(4):797–804. doi: 10.1111/vco.12839 Tagawa M, Tambo N, Maezawa M, Tomihari M, Watanabe KI, Inokuma H, Miyahara K. 2020. Quantitative analysis of the BRAF V595E mutation in plasma cell-free DNA from dogs with urothelial carcinoma. PLoS One. Apr 24;15(4):e0232365. doi: 10.1371/journal.pone.0232365 . PMID: 32330187; PMCID: PMC7182225. Urbanová L. Vyhnánková , V., Nečasová , A, Filipejová, Z., Srnec , R.,Staňková , L., Rizzo, I., Pacík, D., Nečas , A 2023. The effect of urine sample temperature on the efficacy of olfactory detection of prostate cancer in men by a specially trained dog. Acta Vet. Brno 2023, 92: 303–307 https://doi.org/10.2754/avb202392030303 UK Government 2018. The Data Protection Act Data protection: The Data Protection Act - GOV.UK (www.gov.uk) Willis, C.M., Church, S.M., Guest, C.M., Cook, W.A., McCarthy, N., Bransbury, A.J., Church, M.R.T., Church, J.T. 2004. Olfactory detection of human bladder cancer by dogs: proof of principle study. BMJ 329: 712. Woszczylo, M., Jezierski, T., Szumny, A., Nizanski, W., Dzieciol, M. 2020. The Role of Urine in Semiochemical Communication between Females and Males of Domestic Dog (Canis familiaris) during Estrus. Animals , 10. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 13 Oct, 2024 Reviews received at journal 25 Sep, 2024 Reviews received at journal 09 Sep, 2024 Reviewers agreed at journal 05 Sep, 2024 Reviewers agreed at journal 05 Sep, 2024 Reviewers agreed at journal 05 Sep, 2024 Reviewers invited by journal 05 Sep, 2024 Editor assigned by journal 05 Sep, 2024 Submission checks completed at journal 05 Sep, 2024 First submitted to journal 03 Sep, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5022433","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":359002460,"identity":"c083b9d5-9357-4a29-9972-a992e67021fa","order_by":0,"name":"Isabelle Desmas-Bazelle","email":"","orcid":"","institution":"Davies Veterinary Specialists","correspondingAuthor":false,"prefix":"","firstName":"Isabelle","middleName":"","lastName":"Desmas-Bazelle","suffix":""},{"id":359002462,"identity":"3eddd160-9037-4414-958f-f9878e1c50c3","order_by":1,"name":"Nicola Jane Rooney","email":"data:image/png;base64,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","orcid":"","institution":"University of Bristol","correspondingAuthor":true,"prefix":"","firstName":"Nicola","middleName":"Jane","lastName":"Rooney","suffix":""},{"id":359002463,"identity":"aed8fdc2-9161-409d-90e1-45e6c3f101d4","order_by":2,"name":"Steve Morant","email":"","orcid":"","institution":"University of Dundee","correspondingAuthor":false,"prefix":"","firstName":"Steve","middleName":"","lastName":"Morant","suffix":""},{"id":359002464,"identity":"44e493c9-5840-40cc-a13a-5a3df7d24b2a","order_by":3,"name":"Rob Harris","email":"","orcid":"","institution":"Medical Detection Dogs","correspondingAuthor":false,"prefix":"","firstName":"Rob","middleName":"","lastName":"Harris","suffix":""},{"id":359002466,"identity":"0be10916-e5cd-4df9-83e5-98c7c63b23eb","order_by":4,"name":"Veerle Volckaert","email":"","orcid":"","institution":"Davies Veterinary Specialists","correspondingAuthor":false,"prefix":"","firstName":"Veerle","middleName":"","lastName":"Volckaert","suffix":""},{"id":359002467,"identity":"6d2ec088-1d19-49ae-9040-a2ad0204b43a","order_by":5,"name":"Mark Goodfellow","email":"","orcid":"","institution":"Davies Veterinary Specialists","correspondingAuthor":false,"prefix":"","firstName":"Mark","middleName":"","lastName":"Goodfellow","suffix":""},{"id":359002468,"identity":"0b66d333-0ab0-43a7-871c-758fc839dabf","order_by":6,"name":"Sophie Aziz","email":"","orcid":"","institution":"Medical Detection Dogs","correspondingAuthor":false,"prefix":"","firstName":"Sophie","middleName":"","lastName":"Aziz","suffix":""},{"id":359002470,"identity":"5be30887-6021-467d-a490-274e77346de8","order_by":7,"name":"Claire Marie Guest","email":"","orcid":"","institution":"Medical Detection Dogs","correspondingAuthor":false,"prefix":"","firstName":"Claire","middleName":"Marie","lastName":"Guest","suffix":""}],"badges":[],"createdAt":"2024-09-03 06:39:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5022433/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5022433/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":66186713,"identity":"74112e3a-33bb-4cac-86fa-0a404d2b3217","added_by":"auto","created_at":"2024-10-08 13:33:50","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":803413,"visible":true,"origin":"","legend":"\u003cp\u003eMedical detection dog sniffing sample presented in glass vial in a four stand lineup arrangement\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5022433/v1/a62b6c636cbca6cec29406a0.png"},{"id":66187928,"identity":"46f94bb9-d4f5-4351-a3bb-586053b9f270","added_by":"auto","created_at":"2024-10-08 13:41:54","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1335148,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5022433/v1/8b50c49a-1f20-4491-9267-d9748589d570.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Trained dogs can detect canine urothelial carcinoma of the bladder","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCancer is a major cause of morbidity and mortality in pet dogs, who present at roughly the same rate as humans. Approximately one in four dogs will develop cancer during their lifetime, and, like humans, the incidence increases rapidly with age (Schwartz et al \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2022\u003c/span\u003e, Kraus et al \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), so half of all dogs over the age of ten years will develop cancer during their lifetime AVMA 2022).\u003c/p\u003e \u003cp\u003eUrothelial carcinoma (UC), also known as transitional cell carcinoma (TCC), is the most common form of bladder cancer accounting for approximately 2% of all reported canine cancers (Fulkerson and Knapp \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). It is an aggressive, highly invasive, lower urinary tract tumor in dogs (Tagawa et al \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Clinical presentation of UC is shared with several other disorders, including non-malignant conditions, making diagnosis challenging. Definitive diagnosis requires histopathology, which is invasive, costly, and time consuming, thereby delaying the start of treatment. Cystocentesis and needle sampling of bladder masses should be avoided due to the risk of tumor seeding (Hayashi et al \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Cytology of the urine (obtained via free catch sample) however lacks sensitivity, since sediment reveals tumor cells in only 30% of cases (Norris et al \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e1992\u003c/span\u003e). Two other diagnostic tests are currently available: the bladder tumor antigen test which is poorly specific (Henry \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Borjesson and Christopher 2009) and the BRAF mutation detection test (Tagawa et al \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The latter utilises DNA from urine and has a sensitivity of 83% in patients since only 83% of UCs carry the mutation (Mochizuki et al \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). At diagnosis, canine UCs are often advanced (Fulkerson and Knapp \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2015\u003c/span\u003e), hence there is great potential value in a new early, cheap, rapid, and non-invasive diagnostic test for UC.\u003c/p\u003e \u003cp\u003eDogs have been trained to reliably detect a large number of human medical conditions ( Rooney \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) including osteosarcoma (Ortal et al \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), bladder (Wills 2004) and lung (Feil et al \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) cancer. It\u0026rsquo;s logical from an evolutionary perspective that they would also have the ability to detect canine cancer. An initial study trialed the use of four laboratory trained dogs to detect canine urinary tract cancer, from urine (Dorman et al \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Three dogs failed to learn the task and the fourth performed at chance levels during double-blind testing. However, training was on a small number of samples (only eight positive samples), to which dogs were repeatedly exposed and hence they may have learnt to identify individual urine donors rather than to discriminate cancer from controls. The authors concluded that sample storage, confounding odours, and other factors need to be considered in the design of future studies. The importance of these issues is now well understood in the medical detection dog field (Guest et al \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eGreater success was seen when Malone et al (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) trained six pet dogs to distinguish saliva samples from dogs with malignant tumors from health controls, showing mean sensitivity of 90% and specificity of 98%. Here they used a considerably larger number of sample donors for training (78 positive and 109 controls).\u003c/p\u003e \u003cp\u003eUrine poses a more practicable, less invasive medium to screen for canine cancer compared to saliva. What\u0026rsquo;s more since UC is located in the urinary tract with which urine makes contact, it seems logical that, it would be an ideal medium for detection. However, there is the possibility that dog urine is so innately interesting to other dogs (Rooney and Parr Cortes 2023), that natural curiosity in the semiochemical signal overrides any trained categorisation. We therefore wished to explore if the methodological challenges that impacted the Dorman (2017) study, could be overcome to enable reliable detection of UC by dogs. The objective of this study was to investigate whether trained dogs can accurately detect canine urothelial carcinoma (UC) from urine.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cp\u003eEthics\u003c/p\u003e \u003cp\u003eThis study was approved by the Royal College Veterinary Surgeon\u0026rsquo;s (RCVS) Ethics Review Panel.\u003c/p\u003e \u003cp\u003eSubjects\u003c/p\u003e \u003cp\u003eDogs were recruited from canine patients undergoing procedures performed during investigations for urinary tract problems at a single referral veterinary centre. The owners of all dogs attending the clinic for suspected urinary problems over a 5-year period of time were asked to consent to participation in the study. All recruitment and sampling was done prior to the commencement of chemotherapy.\u003c/p\u003e \u003cp\u003eEach case was diagnosed via routine procedures utilising the least invasive method. Typically, cytology of urine was performed first, but if this was inconclusive then urethral catheterisation was used for suction biopsy. If both diagnostic methods were inconclusive, a biopsy was performed either via catheterisation with endoscopic forceps, or via cystoscopy. The method chosen was deemed the best way of performing the clinical workup in order to achieve a definitive diagnosis by the attending clinician. Dogs were thereby classified as positive for urothelial carcinoma or for non-malignant urinary tract disease via cytology or histopathology. Urine samples from dogs with non-malignant urinary tract disease (control), and dogs with UC (positive) were collected prospectively. Whilst study participation did not alter dogs\u0026rsquo; investigation or treatment path, it permitted researchers to access veterinary records and allowed superfluous urine collected during veterinary investigation to be stored and used for dog training and testing. Only leftover urine was used so no extra samples were taken for the purpose of the study.\u003c/p\u003e \u003cp\u003eUrine samples were frozen in glass aliquots within 30 minutes of collection at -80C. Samples were allocated a unique code name to respect General Data Protection Regulation (UK Government 2018). They were then transported on dry ice to Medical Detection Dog (MDD). On arrival, the samples were spilt into a training set (17 positives and 22 controls) and a testing cohort (9 positives and 18 controls). Each sample was split into two aliquots which were stored separately. The testing group were coded so all staff remained blinded. All testing samples were from individuals not included in the training set, so were na\u0026iuml;ve to the dogs. Samples were stored at -80C until the day of training.\u003c/p\u003e \u003cp\u003eAdditional control training samples\u003c/p\u003e \u003cp\u003eTwenty additional healthy control dogs were recruited from those belonging to Medical Detection Dogs and staff and volunteers of the charity. Inclusion criteria was absence of urinary signs, normal bladder ultrasound scan and urinalysis, no comorbidities, and never having had chemotherapy. After the owner\u0026rsquo;s consent, all dogs underwent a conscious ultrasound scan of their bladder (carried out by VV,) and a free catch urine sample was collected. Ultrasonography was performed using a microconvex probe (SC 3123, 3\u0026ndash;10 MHz), in both sagittal and transverse planes (MyLab Twice Esaote and MyLab Sat Esaote, Genova, Italy). At least moderate bladder filling was required for control dogs for accurate assessment of the bladder wall. History and veterinary records were also assessed to confirm healthy status and an absence of urinary tract clinical signs. Non-steroidal anti-inflammatory drugs (NSAIDs) were permitted in controls and patients alike.\u003c/p\u003e \u003cp\u003eDog training\u003c/p\u003e \u003cp\u003eFour dogs who had previously been trained to detect human bladder and prostate cancer were trained to distinguish positive canine cancer samples from controls, using positive reinforcement.\u003c/p\u003e \u003cp\u003eThe training protocol was increased in difficulty over time. Initially, the new target scent was introduced to the dog, and the dog rewarded for showing a behavioral change in response. All dogs in this study showed a sit and stare response. Correct responses were rewarded with food or ball-play, in conjunction with the use of an audible clicker (used as a secondary reward).\u003c/p\u003e \u003cp\u003eThree dogs had been trained to walk along a line of four stainless steel stands, one metre apart, each with an arm holding a 40 ml glass jar containing a swab sample. The fourth dog was trained to search a single sample at a time. A stainless-steel grill was placed over the mouth of each jar to prevent the dog touching the sample\u003c/p\u003e \u003cp\u003eThe dogs were trained to sniff each stand and show their indication behavior in response only to positive samples. Dogs were rewarded for both indicating positive target samples and for searching a negative line (or single stand for dog 3) without an indication. Their behavior was recorded on a database as IND (indicated), HES (hesitated), INT (interest, but weaker than HES), NI (no interest) or NS (not searched, usually because a sample earlier in the line was indicated). Initially, the line trained dogs were presented with one positive sample and three negative samples during each run. As training progressed, sequences of four negative samples were also introduced. After each line was searched, the plates were removed for cleaning in an ultrasonic cleaner with an enzymatic solution (Reprozyme TMlook),. They were then rinsed thoroughly using a commercial glass washer (minimum 85\u0026deg;C) and left to air dry to avoid any contamination.\u003c/p\u003e \u003cp\u003eAfter five sessions, one dog (one trained on the line arrangement) was deemed not to be learning the task, and was also suffering with lameness. It was therefore eliminated from the study. The remaining three dogs were one male Cocker Spaniel (11.5 years), one female Cocker Spaniel (8 years) and one male Labrador Retriever (4 years), all neutered. They were trained over a period of 148 days, during which they underwent between 23 and 29 training sessions, encountering between 602 and 1072 training samples from between 76 and 79 different individual donors.\u003c/p\u003e \u003cp\u003eDouble blind testing\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e here\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eDetails of target and control samples used during double blind testing including dogs\u0026rsquo; age, sex, health or disease status\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"9\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiagnosis\u003c/p\u003e \u003cp\u003eurothelial carcinoma (UC) or nonmalignant urinary tract disease (MUTD)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBreed\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c5\" namest=\"c4\"\u003e \u003cp\u003eNeutered status\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eAge at Sampling\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMethod of urine collection\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eSite of UC (if appropriate)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c9\"\u003e \u003cp\u003eMethod of diagnosis (cytology or histology)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWest Highland White Terrier\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e11 y 6 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBladder neck\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ehistology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC *\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpringer Spaniel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7 y 9 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCatheter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eVentral and mid area of the bladder\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ecytology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eStaffie\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e11 y 1 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBladder wall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ehistology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabrador\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEntire\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e13 y 1mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCatheter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBladder neck and urethra\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ecytology and histology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMaltese\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEntire\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8 y 2 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCatheter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCaudo dorsal bladder wall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ecytology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWest Highland White Terrier\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7 y 6 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBladder neck and proximal urethra\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ecytology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHungarian Vizsla\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e11 y\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCatheter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eCranial and mid third of bladder\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ecytology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBoston Terrier\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e9 y 9 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCatheter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eBladder neck\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e \u003cp\u003ecytology\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTibetan Spaniel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEntire\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5 y 1 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnglish Springer Spaniel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8 y 1 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrossbreed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2 y 3 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJack Rusell Terrier\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEntire\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6 y 4 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBorder Terrier\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4 y 11 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBorder Collies\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7y 3mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrossbreed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6 yo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBrittany Spaniel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e10 y 2 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabrador\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e6 y\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGriffon\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4 y 4 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGolden Retriever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e11.5 yo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrossbreed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4 y 11 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMiniature Schnauzer\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3 y 9 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrossbreed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5 y 8 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrossbreed\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eF\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2 y 4 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabrador Retriever\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNeutered\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e12 y 2 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eFree Catch\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRhodesian Ridgeback\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEntire\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e3 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNMUTD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGreyhound\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eN\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4 y 10 mo\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eCystocentesis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c9\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eEach dog was presented with a different set of 20 control samples (from either 13 or 14 unique individuals), 10 positive samples (from between 5 and 7 individuals) and 10 fillers (control samples previously encountered during training, and hence not included in the statistical analysis. The samples came from 27 different donor dogs (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Dog 3 was presented with samples one at a time, as this was the method which had been used in their training. For dogs 1 and 2, samples were randomised in blocks of four, each including two controls, one positive and one filler sample; using the same four stand system as during training. They worked along each line of four samples in order. If they showed their trained indication (alert) behavior to one sample the remaining samples in this line were recorded as \u0026ldquo;not seen\u0026rdquo; (NS) in that run, and the line was presented again in reverse order at a later time. If the handler deemed it necessary, dogs were permitted to the search the samples twice before they reported the behavioral response to be recorded. Any samples still not seen after the forward and reverse runs were assembled into new lines, until all samples were searched.\u003c/p\u003e \u003cp\u003eFive control and two positive samples from the training set were used for calibration runs, to confirm the dog was focused on the scent. Calibration runs were requested at any time by the trainer, were unblinded and their results were not included in the analysis.\u003c/p\u003e \u003cp\u003eA blinded experimenter entered the responses to each sample into the database. All trial samples had previously been given a \u0026ldquo;blind ID\u0026rdquo; which was read by the data collection software to determine whether responses were correct. The software revealed whether the response was correct, and if so, the trainer was asked to reward their dog. If any sample was incorrect, no reward was given. The database was password protected and only the trial statistician (SM) knew both passwords prior to trial completion.\u003c/p\u003e \u003cp\u003eAmongst the NMUTD dogs, six had a follow up of more than 6 months with four dogs still alive at the time of writing, and 3 had a follow up between 1 and 5 months. None of these dogs were subsequently diagnosed with UC.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStatistical methods\u003c/h2\u003e \u003cp\u003eThe database was scanned for the first occasion each dog encountered each sample in a valid run, ignoring occasions when a sample was set out in a line but not seen (\u003cb\u003eNS\u003c/b\u003e). Subsequent encounters were ignored. Only behaviors recorded as \u003cb\u003eIND\u003c/b\u003e were regarded as positive responses (indications), and all other behaviors (\u003cb\u003eHES\u003c/b\u003e, \u003cb\u003eINT\u003c/b\u003e and \u003cb\u003eNI\u003c/b\u003e) were classified as negative responses.\u003c/p\u003e \u003cp\u003eWe used the first encounter with each novel sample to calculate odds ratios for indicating target vs control samples) sensitivity (proportion of target samples found) and specificity (proportion of control samples that were not indicated) for each dog. Due to a shortage of samples, the testing sample cohort, although all unique, were not all from different dogs. Therefore, we also conducted a sensitivity analysis using only the first sample encountered from each subject dog by each detection dog. We compared the likelihood of dog alerting with sampling method cystocentesis vs free catch for controls and catheter or free catch for target samples.\u003c/p\u003e \u003cp\u003eResults for calibration samples and filler samples were recorded but they were not included in the statistical analyses. For target and control samples, logistic models were fitted to test for differences in the proportions indicated. Odds ratios were estimated for each dog individually, and overall.\u003c/p\u003e \u003cp\u003eData processing and analysis were conducted using R (R Core Team \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e here\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eNumber of presentations and odds ratios for indications to targets and controls during double blind testing\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eTargets\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e \u003cp\u003eControls\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c8\" namest=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDog\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAnalysis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003er/n\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e% (95% CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003er/n\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e% (95% CI)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eOdds ratio\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eProbability\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eDog 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAll samples\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9/10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e90.0 (53.3,99)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.0 (0.7, 28)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e171.0 (9.6,3055)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDistinct dogs\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e6/7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e85.7 (41.9,98)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1/13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.7 (1.1, 39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e72.0 (3.81,1362)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.004\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eDog 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAll samples\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8/10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.0 (45.9,95)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e15.0 (4.9, 38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e22.7 (3.1,164)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDistinct dogs\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e4/6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66.7 (26.8,92)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e3/14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e21.4 (7.1, 49)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e7.3 (0.88, 61)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.066\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eDog 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAll samples\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7/10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e70.0 (37.6,90)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5.0 (0.7, 28)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e44.3 (3.9,500)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.002\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDistinct dogs\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2/5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40.0 (10.0,80)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1/13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7.7 (1.1, 39)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e8.0 (0.53, 121)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.133\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eOverall\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAll samples\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e24/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e80.0 (62.1,91)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5/60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e8.3 (3.5,19)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e44.0 (12.2,158)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDistinct dogs\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12/18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e66.7 (42.9,84)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5/40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e12.5 (5.3, 27)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e14.0 (3.61, 54)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe difference in the proportions of control and target samples indicated was highly significant for all three dogs (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), when all samples were considered. The average sensitivity (% of target samples alerted to) was 80% and the average specificity (% of controls correctly not alerted to) was 91.7% (i.e. 100- minus 8.3% controls indicated). When we calculated sensitivity and specificity based only on the first presentation of each patient dog, both sensitivity (66.7%) and specificity (87.5%) rates were reduced. Considering just the samples from distinct dogs, the differences in the proportion indicated from target (as compared to control) samples remained significant for the population overall and for Dog 1 individually.\u003c/p\u003e \u003cp\u003eWhether the control samples were collected via cystocentesis or free catch did not significantly affect the odds of the dogs alerting (17.6% vs 5.9%: odds ratio: 3.4(0.5, 23), p\u0026thinsp;=\u0026thinsp;0.203). Similarly for targets, the odds of an alert in response to a sample, collected via catheterization (81%) was not significantly different to one collected by free catch (77.8%; 1.21 (0.18,8.2), p\u0026thinsp;=\u0026thinsp;0.842)\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWhilst previous studies failed to train dogs to detect UC from canine urine (Dorman et al \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), our study shows that dogs can be trained to reliably detect urine from patients with UC with high sensitivity (80%) and specificity (91.7%). This is likely due to the increased number of samples used here in training which ensured that possible confounds were ameliorated (Guest et al \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and that dogs learnt to categorise the samples based on cancer status rather that learning individual samples. These detection rates were similar to those found when dogs were trained to detect malignant tumours from saliva (Malone et al \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e: sensitivity 90% specificity 98%).\u003c/p\u003e \u003cp\u003e Urine can be collected relatively easily during natural voiding, which is less invasive than saliva collection. This is especially useful for patients which are challenging to handle due to fear or other behavioral issues, making urine an ideal media for screening by medical detection dogs. Furthermore, it is easier to collect a large amount of urine compared to saliva. However, dogs have a strong drive to smell one another\u0026rsquo;s urine. It is a semiochemical signal used in inter-dog communication, to derive information about social (Lisberg and Snowdon, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) as well as reproductive status (Jezierski et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, Woszczylo et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). It was feared that dogs\u0026rsquo; innate interest may override any training. We have shown that with systemic and well-planned training, this is not the case and it is possible to condition dogs to respond reliably to cancer status, although the urine attractiveness may have plausibly contributed to some of the errors seen.\u003c/p\u003e \u003cp\u003eWhen considering all samples, average and individual dog sensitivity and specificity levels were all well above chance and comparable to levels seen in dogs trained by the same training organisation for the detection of malaria parasites from worn socks (Guest et al 2019: sensitivity 70\u0026ndash;73% and specificity 90.3\u0026ndash;91%), SARS Covid-2 from sweat samples (Guest et al \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; sensitivity: 82\u0026ndash;94; specificity 76\u0026ndash;92%), and \u003cem\u003ePseudomonas aeruginosa\u003c/em\u003e bacterial pathogens from supernatant (Davies et al 2019: sensitivity and specificity\u0026thinsp;\u0026gt;\u0026thinsp;90% depending upon comparative control and dilution). Rates were above those obtained when using dogs to detect some human cancers, such as human bladder cancer (e.g. success rate 41% Willis et al \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). We suggest that future refinements to training and sample presentation may lead to even further increases in detection accuracy.\u003c/p\u003e \u003cp\u003eDue to a shortage of samples, the testing sample cohort although all unique, were not all from different dogs. When we calculated sensitivity and specificity based only on the first presentation of each patient dog, both sensitivity (mean\u0026thinsp;=\u0026thinsp;66.7%) and specificity (mean\u0026thinsp;=\u0026thinsp;87.5%) rates were reduced, in each dog. This may be due to the duplicated samples being easier for dogs to identify than some others, or a consequence of learning from the first encounter. However, even when just using first encounters, the three dogs together indicated cancerous samples significantly more often than controls. These results were still above chance and above previous published results using dog urine (Dorman et al \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eUnlike the previous study (Dorman et al \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), our population comprised two different control populations, dogs with no underlying urinary disease and those with non-malignant urinary condition. This seems more comparable and hence applicable to veterinary practice where diagnostic tests could be used both within a population of seemingly healthy dogs (in order to detect an underlying cancer) or for dogs showing urinary signs, in order to distinguish between cancer and inflammatory or infectious diseases. Within this study the method by which the urine was collected did not affect the likelihood of the dogs correctly responding to a positive sample, nor correctly ignoring a control sample. However, we used relatively few samples, and hence it would prudent to test this further on a larger population.\u003c/p\u003e \u003cp\u003e It's noteworthy that the detection dogs in our study showed considerable individual variability, in spite of relatively standardised training methods. Whilst the sensitivity of Dog 1 was 90%, that of Dog 3 was only 70%. The methods of presentation of samples may have contributed to this, as Dog 3 was presented with samples one at a time, unlike Dogs 1 and 2 who searched lines of four samples at a time. The detection rate differences may also reflect personality differences between the dogs (Bistre Dabah 2023). Further analysis using the data stored during training will enable understanding of the extent to which these individual differences develop or whether they are consistent from early training. Additional studies with a larger number of detection dogs would also be beneficial.\u003c/p\u003e \u003cp\u003eOverall, the dogs showed high detection rates, similar to that of current diagnosis techniques, with the BRAF mutation test for example yielding detection rates of 83% (Mochizuki et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). It would be very useful to study whether the samples missed by the dogs would also be missed by the other screening tests, and whether the presence of a BRAF mutation leads to a difference in canine detection rates. If not, then there may be great value in adding the dog to the suit of diagnosis tools available. Olfactory testing by trained dogs has the capacity to be very rapid, and it would be feasible to take canine urine samples to a medical detection dog facility for rapid screening, potentially more quickly than current laboratory facilities. However, the practicality of storing samples at -80 C, or transporting on ice may be challenging and hence research to ascertain the relative importance of different methodological aspects (Parr Cortes et al submitted) would be valuable. For example, exploring to what extent storage temperature of the urine samples affects detection rates, since no link was found between the efficacy of canine olfactory detection of male prostate cancer and the testing temperature within the range 2\u0026ndash;23 \u0026deg; (Urbanova et al 2023). If storage at -80 C is not required then this could increase the practicality of the dog as screening tool.\u003c/p\u003e \u003cp\u003eFuture work would ideally also use a larger cohort of donor dog samples for training to ensure detection dogs are exposed to variation in breed, sex, neuter status, diet, age, medication, environment (e.g smoking household), and non-urinary diseases status. These factors could all alter the scent of urine and add challenge to the dogs, hence future research should explore the effect of each of these factors on the dogs\u0026rsquo; accuracy during testing.\u003c/p\u003e \u003cp\u003eWe also suggest methodical comparisons of the detection rates using different media (e.g. urine vs saliva) using the same training approaches. For bladder cancers such as UC, we expect volatile organic compounds (VOCs) to be especially concentrated in the urine, and hence it may be an ideal media for this cancer type. However VOCs associated with cancer outside of the urinary tract may also be concentrated or more readily detected in urine, Feil et al (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) found higher accuracy when dogs used urine compared to breath samples to detect human lung cancer. The viability of urine for detection of other canine cancers is yet to be fully explored, as is the use of other media such as saliva for UC detection. Future trials should examine whether urine is the optimal media for all cancer types, or whether using the bodily fluid most closely associated with the affected site is optimal.\u003c/p\u003e \u003cp\u003eFuture work should also test dogs\u0026rsquo;? ability to distinguish UC from control samples from dogs with cancers located in other anatomical locations. Studies to explore detection of feline cancers are similarly ongoing (Pellin et al \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eLongitudinal studies of dogs\u0026rsquo; capacity to alert prior to formal diagnosis are also required. These should also examine the feasibility of using dogs for screening breeds predisposed to UC such as the West Highland White Terrier (Pantke \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Large prospective cohort studies would be ideal.\u003c/p\u003e \u003cp\u003eIn conclusion, the sensitivity and specificity of dogs trained to find canine UC samples is encouraging, suggesting that further research has the potential to help develop methods for the early diagnosis of UC, by dogs or by other technology such as electronic noses. This may in turn lead to opportunities for earlier intervention than is possible at present.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics\u003c/p\u003e\n\u003cp\u003eThis study was approved by the Royal College Veterinary Surgeons Ethics Review Panel.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConsent\u003c/p\u003e\n\u003cp\u003eThe owners of all dogs attending the clinic for suspected urinary problems over a 5-year period of time were asked to consent for participation in the study. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials\u003c/p\u003e\n\u003cp\u003eDue to the sensitivity of the data involved, these data are published as a restricted dataset at the University of Bristol Research Data Repository data.bris, at https://doi.org/10.5523/bris.1cq4ulhrjdmpf240uhjb2o6jov. The metadata record published openly by the repository at this location clearly states how data can be accessed by bona fide researchers. Requests for access will be considered by the University of Bristol Research Data Service, who will assess the motives of potential data re-users before deciding to grant access to the data. No authentic request for access will be refused and re-users will not be charged for any part of this process\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCompeting interests\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe are not aware of any conflicts of interest\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis study was unfunded\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAuthors\u0026apos; contributions\u003c/p\u003e\n\u003cp\u003eCG, RH ID conceived the study and devised the initial design\u003c/p\u003e\n\u003cp\u003eRH trained and supervised the training of the dogs\u003c/p\u003e\n\u003cp\u003eID , VV, SA and MG contributed to data collection\u003c/p\u003e\n\u003cp\u003eSA \u0026nbsp;was responsible for data curation\u003c/p\u003e\n\u003cp\u003eSM,NJR and RH conducted data analysis\u003c/p\u003e\n\u003cp\u003eNJR and ID drafted the MS\u003c/p\u003e\n\u003cp\u003eRH, CG, reviewed the MS\u003c/p\u003e\n\u003cp\u003eAll authors approved the final version\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003eWe would like to thank all the staff of Davies Veterinary Specialists and Medical Detection Dogs who helped facilitate recruitment and data collection.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eWe are very grateful to all dogs and their owners who participated.\u003c/li\u003e\n \u003cli\u003eWe thank all MDD trainers who contributed to the training of the dogs \u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eThis study was presented in poster form at VCS congress 4-6 November 2021\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eR Core Team. 2020. \u003cem\u003eR: A Language and Environment for Statistical Computing\u003c/em\u003e. Vienna, Austria: R Foundation for Statistical Computing. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.R-project.org/\u003c/span\u003e\u003cspan address=\"https://www.R-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAVMA (American Veterinary Medicine Association) 2022. Cancer in pets. \u003cem\u003eAVMA\u003c/em\u003e. Accessed October 30, 2022. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.avma.org/resources/pet-own\u003c/span\u003e\u003cspan address=\"https://www.avma.org/resources/pet-own\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e Hayashi T, Hirokawa M, Higuchi M, Kudo T, Ito Y, Miyauchi A. Needle Tract Implantation Following Fine-Needle Aspiration of Thyroid Cancer. World J Surg. 2020;44(2):378\u0026ndash;384. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00268-019-05161-1\u003c/span\u003e\u003cspan address=\"10.1007/s00268-019-05161-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 31515571.ers/petcare/cancer-pets\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBorjesson, D.L., Christopher, M.M., Ling, G.V. 1999. Detection of canine transitional cell carcinoma using a bladder tumor antigen urine dipstick test. Veterinary Clinical Pathology 28: 33\u0026ndash;38.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBistre Dabah S. 2023. Investigating the effect of individual attributes on dog-performance in medical detection tasks. Unpublished PhD thesis. University of Bristol. Davies, J.C., Alton,E.,; Simbo, A., ; Murphy, R., ; Seth, I., Williams, K., ; Somerville, M., Jolly, L.,Morant, S., Guest, C. 2019. Training dogs to differentiate Pseudomonas aeruginosa from other cystic fibrosis bacterial pathogens: not to be sniffed at? \u003cem\u003eEuropean Respiratory Journal\u003c/em\u003e 54 (5) 1900970.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDorman, D.C., Foster, M.L., Fernhoff, K.E., Hess, P.R. 2017. Canine scent detection of canine cancer: a feasibility study. Vet Med (Auckl). 8:69\u0026ndash;76 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2147/VMRR.S148594\u003c/span\u003e\u003cspan address=\"10.2147/VMRR.S148594\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeil, C., Staib, F., Berger, M.R. \u003cem\u003eet al.\u003c/em\u003e 2021. Sniffer dogs can identify lung cancer patients from breath and urine samples. BMC Cancer 21, 917 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12885-021-08651-5\u003c/span\u003e\u003cspan address=\"10.1186/s12885-021-08651-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFulkerson, C.M., Knapp, D.W. 2015. Management of transitional cell carcinoma of the urinary bladder in dogs: a review. Vet J. 205(2):217\u0026thinsp;\u0026ndash;\u0026thinsp;25. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.tvjl.2015.01.017\u003c/span\u003e\u003cspan address=\"10.1016/j.tvjl.2015.01.017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 25747698.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuest, C., Harris, R., Anjum, I., Concha, A. \u0026amp; Rooney, N. J. 2020. A lesson in standardisation -subtle aspects of the processing of samples can greatly affect dogs' learning: subtle aspects of the processing of samples can greatly affect dogs' learning : Frontiers in Veterinary Science.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuest, C., Pinder, M., Doggett, M., Squires, C., Affara, M., Kandeh, B., Dewhirst, S., Morant, S., V; D'Alessandro, U.,; Logan, J. G., Lindsay S.W., 20129. Trained dogs identify people with malaria parasites by their odour \u003cem\u003eThe Lancet Infectious Diseases\u003c/em\u003e, 19 (6,): 578\u0026ndash;580, Guest, C., Dewhirst, S. Y., Lindsay, S.W. et al 2022. Using trained dogs and organic semi-conducting sensors to identify asymptomatic and mild SARS-CoV-2 infections: an observational study J. \u003cem\u003eTravel Medicine\u003c/em\u003e, 29, 3,\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHayashi T, Hirokawa M, Higuchi M, Kudo T, Ito Y, Miyauchi A. 2020. Needle Tract Implantation Following Fine-Needle Aspiration of Thyroid Cancer. World J Surg. 44(2): 378\u0026ndash;384. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00268-019-05161-1\u003c/span\u003e\u003cspan address=\"10.1007/s00268-019-05161-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 31515571.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenry, C.J. 2003. Evaluation of a bladder tumor antigen test as a screening test for transitional cell carcinoma of the lower urinary tract in dogs. Am J Vet Res 64(8): 1017\u0026ndash;1020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJezierski, T., Dzieciol, M., Szumny, A., Nizanski, W., Woszczylo, M., Pieczewska, B., Godzinska, E. J. 2019. Discrimination of estrus odor in urine by male dogs in different experimental settings. Journal of Veterinary Behavior-Clinical Applications and Research, 29, 25\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKraus C, Snyder-Mackler N, Promislow DEL. 2023. How size and genetic diversity shape lifespan across breeds of purebred dogs. Geroscience;45(2):627\u0026ndash;643. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s11357-022-00653-w\u003c/span\u003e\u003cspan address=\"10.1007/s11357-022-00653-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLisberg, A. E., Snowdon, C. T. 2009. The effects of sex, gonadectomy and status on investigation patterns of unfamiliar conspecific urine in domestic dogs, \u003cem\u003eCanis familiaris\u003c/em\u003e. Animal Behaviour, 77, 1147\u0026ndash;1154.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalone, L. A., Pellin, M. A., \u0026amp; Valentine, K. M. 2023. Trained dogs can accurately discriminate between scents of saliva samples from dogs with cancer versus healthy controls, \u003cem\u003eJournal of the American Veterinary Medical Association\u003c/em\u003e (published online ahead of print 2023). Retrieved May 11, 2023, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2460/javma.22.11.0486\u003c/span\u003e\u003cspan address=\"10.2460/javma.22.11.0486\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMochizuki, H., Shapiro, S.G., Breen, M. 2015) Detection of BRAF Mutation in Urine DNA as a Molecular Diagnostic for Canine Urothelial and Prostatic Carcinoma. PLoS ONE 10(12): e0144170. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0144170\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0144170\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNorris, A.M., Laing, E.J., Valli, V.E., et al. 1992. Canine bladder and urethral tumors: A retrospective study of 115 cases (1980\u0026ndash;1985) J Vet Intern Med. ;6:145\u0026ndash;153.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOrtal, A., Rodr\u0026iacute;guez, A., Solis-Hern\u0026aacute;ndez, M.P. \u003cem\u003eet al.\u003c/em\u003e 2022. Proof of concept for the use of trained sniffer dogs to detect osteosarcoma. Sci Rep 1: 6911\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-022-11013\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-11013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePantke P. 2018. Diagnostik und Therapie des \u0026Uuml;bergangszellkarzinoms des unteren Harntraktes beim Hund. Kleintierpraxis. 63: 76\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eParr-Cortes, Z., Rooney N.J., Wheatstone M., Stock\u003csup\u003e,\u003c/sup\u003e T., Pesterfield, C., Guest, C., M\u0026uuml;ller C.T. submitted Evaluating and contextualising volatile organic compounds in dog training breath samples from a patient with type 1 diabetes. Journal of Veterinary Behavior Medicine: clinical applications.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePellin, M. A., Malone, L. A., \u0026amp; Ungar, P. 2024. The use of sniffer dogs for early detection of cancer: a One Health approach. \u003cem\u003eAmerican Journal of Veterinary Research\u003c/em\u003e, \u003cem\u003e85\u003c/em\u003e(1), ajvr.23.10.0222. Retrieved Feb 9, 2024, from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2460/ajvr.23.10.0222\u003c/span\u003e\u003cspan address=\"10.2460/ajvr.23.10.0222\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRooney N.J. 2016. A medical detection role for dogs International Animal Health Journal. 3, (4), 42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRooney, NJ., Parr-Cortes Z. 2023., Olfaction and Welfare In Lazarowski, L. (ed) Olfactory research in Dogs Springer Link \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-031-39370-9\u003c/span\u003e\u003cspan address=\"10.1007/978-3-031-39370-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchwartz SM, Urfer SR, White M, Megquier K, Shrager S, 2022. Dog Aging Project Consortium, Ruple A. Lifetime prevalence of malignant and benign tumours in companion dogs: cross-sectional analysis of Dog Aging Project baseline survey. \u003cem\u003eVet Comp Oncol\u003c/em\u003e. 2022;20(4):797\u0026ndash;804. doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/vco.12839\u003c/span\u003e\u003cspan address=\"10.1111/vco.12839\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTagawa M, Tambo N, Maezawa M, Tomihari M, Watanabe KI, Inokuma H, Miyahara K. 2020. Quantitative analysis of the BRAF V595E mutation in plasma cell-free DNA from dogs with urothelial carcinoma. PLoS One. Apr 24;15(4):e0232365. doi: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0232365\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0232365\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 32330187; PMCID: PMC7182225.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUrbanov\u0026aacute; L. Vyhn\u0026aacute;nkov\u0026aacute;\u003csup\u003e,\u003c/sup\u003e V., Nečasov\u0026aacute;\u003csup\u003e,\u003c/sup\u003e A, Filipejov\u0026aacute;, Z., Srnec\u003csup\u003e,\u003c/sup\u003e R.,Staňkov\u0026aacute;\u003csup\u003e,\u003c/sup\u003e L., Rizzo, I., Pac\u0026iacute;k, D., Nečas\u003csup\u003e,\u003c/sup\u003e A 2023. The effect of urine sample temperature on the efficacy of olfactory detection of prostate cancer in men by a specially trained dog. Acta Vet. Brno 2023, 92: 303\u0026ndash;307 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2754/avb202392030303\u003c/span\u003e\u003cspan address=\"10.2754/avb202392030303\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUK Government 2018. The Data Protection Act Data protection: The Data Protection Act - GOV.UK (www.gov.uk)\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWillis, C.M., Church, S.M., Guest, C.M., Cook, W.A., McCarthy, N., Bransbury, A.J., Church, M.R.T., Church, J.T. 2004. Olfactory detection of human bladder cancer by dogs: proof of principle study. BMJ 329: 712.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWoszczylo, M., Jezierski, T., Szumny, A., Nizanski, W., Dzieciol, M. 2020. The Role of Urine in Semiochemical Communication between Females and Males of Domestic Dog (Canis familiaris) during Estrus. \u003cem\u003eAnimals\u003c/em\u003e, 10.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"veterinary-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Veterinary Oncology](https://veterinaryoncology.biomedcentral.com/)","snPcode":"44356","submissionUrl":"https://submission.springernature.com/new-submission/44356/3","title":"Veterinary Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5022433/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5022433/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDog have been trained to detect a variety of human cancers. However, studies examining their ability to find canine cancer are sparse and show variable results. Whilst dogs were shown to be successful at detecting malignant tumors from saliva samples, they were previously unsuccessful at finding canine urothelial carcinoma (UC) from urine. We used advanced training methods to explore if dogs can accurately detect urothelial carcinoma (UC) from canine urine.\u003c/p\u003e \u003cp\u003eUrine samples from healthy dogs without urinary tract abnormalities (control), with non-malignant urinary tract disease (control), and with UC (positive) were collected prospectively. Urothelial carcinoma was diagnosed using cytology and/or histopathology. All dogs were na\u0026iuml;ve of chemotherapy, underwent bladder ultrasound and urinalysis. Residual urine remaining after conventional diagnostic procedures was frozen in glass jars until analysis and used for dog training and testing.\u003c/p\u003e \u003cp\u003eThree dogs previously trained to detect human cancer underwent up to 29 days of additional training. They were each presented with 20 control samples (from either 13 or 14 unique individuals), 10 UC positive samples (from between 5 and 7 individuals), in a double-blind trial and their behavioral responses recorded.\u003c/p\u003e \u003cp\u003eOverall, 80% of UC were correctly indicated, and 91.7% of controls correctly ignored (68% and 87% when just considering only unique donor dogs). Individual dogs showed sensitivities ranging from 70 to 90%, and specificities ranging from 85 to 95%.\u003c/p\u003e \u003cp\u003eThis study suggests that, when using appropriate methods, dogs can be trained to detect canine UC from urine, and could serve as an additional rapid, non-invasive diagnostic test for the disease.\u003c/p\u003e","manuscriptTitle":"Trained dogs can detect canine urothelial carcinoma of the bladder","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-08 13:33:45","doi":"10.21203/rs.3.rs-5022433/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-13T21:13:26+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-25T15:22:40+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-09T22:40:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"300144130940006885639979759627211094051","date":"2024-09-05T19:51:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"214514476735096020224485221582484204345","date":"2024-09-05T13:41:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"263223587015300131653757206991176511677","date":"2024-09-05T13:03:21+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-09-05T12:31:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-09-05T04:57:26+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-09-05T04:55:18+00:00","index":"","fulltext":""},{"type":"submitted","content":"Veterinary Oncology","date":"2024-09-03T06:38:04+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"veterinary-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Veterinary Oncology](https://veterinaryoncology.biomedcentral.com/)","snPcode":"44356","submissionUrl":"https://submission.springernature.com/new-submission/44356/3","title":"Veterinary Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c51f181c-b2fb-4791-80c6-2498dccaf653","owner":[],"postedDate":"October 8th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2024-11-06T13:53:13+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-08 13:33:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5022433","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5022433","identity":"rs-5022433","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}