The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus

Medical Hypotheses xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Medical Hypotheses journal homepage: www.elsevier.com/locate/mehy

The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus Alan Pomerantz ⇑, Ruben Blachman-Braun, Javier Andrés Galnares-Olalde, Roberto Berebichez-Fridman, Marino Capurso-García Universidad Anáhuac México Norte, Facultad de Ciencias de la Salud, Edo. de México, Mexico

a r t i c l e

i n f o

Article history: Received 28 October 2014 Accepted 23 April 2015 Available online xxxx

a b s t r a c t In order to find better tools in the diagnosis of cancer in an earlier and more precise manner, researchers have explored the use of volatile organic compound (VOCs) as a way to detect this disease. Interestingly, the canine olfactory apparatus was observed to detect cancer in two anecdotal reports. After the description of these events, researchers began to study this phenomenon in a structured way in order to assess the ability of canines in detecting cancer-related VOCs. Due to the fact that some of these studies have shown that the canine olfactory apparatus is highly proficient in the detection of cancer-related VOCs, in this article we assess the possibility of constructing a bioelectronic-nose, based on canine olfactory receptors (ORs), for the purpose of diagnosing cancer in a more sensitive, specific, and cost effective manner than what is available nowadays. Furthermore, in order to prove the feasibility and the need of the proposed apparatus, we searched for the following type of articles: all of the studies that have examined, to our knowledge, the ability of dogs in detecting cancer; articles that assess the dog olfactory receptor (OR) gene repertoire, since a central part of the proposed bioelectronic nose is being able to recognize the odorant that emanates from the cancerous lesion, and for that purpose is necessary to express the canine ORs in heterologous cells; examples of articles that depict different devices that have been built for the purpose of detecting cancer-related VOCs, so as to assess if the construction of the proposed apparatus is needed; and articles that describe examples of already constructed bioelectronic noses, in order to demonstrate the existence of a technical precedent and thus the plausibility of the proposed device. Ó 2015 Elsevier Ltd. All rights reserved.

Introduction In medicine, the odor of patients has been a great tool in the diagnosis of some diseases. For example, it is known that ancient Greeks were the first to harness the odor of patients and their corporal fluids as a way to diagnose certain pathologies [1]. Based on this ancestral knowledge, in the last decades scientists have been searching more precise and less invasive ways for the detection of several diseases. Therefore, the research to detect and identify some volatile organic compounds (VOCs), which are a source of odors, as a way to develop new diagnostic methods, has been of great importance to medical sciences [2–4]. VOCs are molecules capable of volatilizing at room temperature and are the product of different metabolic pathways [5,6]. The National Aeronautics and Space Administration along with the Jet Propulsion Laboratory and Caltech were one of the pioneers in the development of a sensor ⇑ Corresponding author at: Av. Universidad Anahuaca No. 46, Col. Lomas Anáhuac, Huixquilucan, Edo. de México C.P. 52786, Mexico. Tel.: +52 5556270210; fax: +52 5555961938. E-mail address: [email protected] (A. Pomerantz).

that could detect these volatile organic compounds [7,8]. Furthermore, Krishna Persaud was one of the first individuals that proposed the utilization of these sensors in order to detect certain disease processes [9–11]. In cancerous cells, a change in the rate of oxidative stress, lipid peroxidation, and gene sequences leads to abnormalities in the biochemical pathways of these cells and thus to the production of specific VOCs [12]. Based on the last information published by GLOBOCAN in 2008, the cancer types with the highest incidence are breast cancer (10.9%), prostate cancer (PCa) (7.1%), lung cancer (LC) (12.7%) and colorectal cancer (9.8%), while the cancer types with the largest mortality are LC (18.2%), stomach cancer (9.7%), liver cancer (9.2%) and colorectal cancer (8.1%) [13]. In an effort to diagnose this disease in earlier stages and thus to reduce its mortality, several diagnostic tests have been used. Nevertheless, in the detection of cancer these tests present several drawbacks. For example, mammography in women ages 40–49 has a lower sensitivity and specificity when compared with older women [14]; in LC low dose spiral CT scan has not been able to reduce mortality in patients who are affected by this disease

http://dx.doi.org/10.1016/j.mehy.2015.04.024 0306-9877/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

2

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

[15]. Moreover, because the current algorithms and diagnostic methods in the diagnosis of LC are not very precise most patients are diagnosed in late stages of this disease and less than 20% of diagnosed individuals are eligible for curative surgery [16]. To increase the likelihood of detecting cancer in earlier stages, with a greater specificity, sensitivity, and efficiency to what is already mentioned in the literature, and due to the ability of dogs to detect cancer-related VOCs; we assess the possibility of inventing a bioelectronic nose (a combination of a biological recognition part, olfactory receptors (OR), and a non-biological sensing platform) based on the canine olfactory apparatus [5,6,17–25]. In order to prove the feasibility of the proposed bioelectronic nose, we searched throughout the literature for the following type of articles: (1) all or most of the articles that have assessed the ability of dogs in detecting cancer, so as to evaluate if this skill portrayed by canines is good enough in order to translate it into a bioelectronic nose. (2) Articles that assess the dog olfactory receptor (OR) gene repertoire, since a central part of the proposed bioelectronic nose is being able to recognize the odorant that emanates from the cancerous lesion; and for that purpose we consider that is necessary to express in heterologous cells the canine olfactory receptors that are responsible for detecting the cancer-related VOCs. In order to achieve such purpose is essential to know and understand the dog OR repertoire. (3) Examples of articles that depict different devices that have been built for the purpose of detecting cancer-related VOCs, so as to assess if the construction of the proposed apparatus is needed. (4) Articles that describe examples of already constructed bioelectronic noses, which have used ORs expressed in heterologous cells as the biological sensing element of the apparatus, in order to demonstrate the existence of a technical precedent and thus the plausibility of the proposed device. Theory Canine clinical findings The observation that dogs are proficient in smelling cancer was perceived by Sir Hywel Williams and Andres Pembroke. In a letter to The Lancet they reported that one of their patients sought a consultation with them due to the great interest that her dog showed in a skin lesion on one of her legs. It was later excised, analyzed and proven to be a malignant melanoma. Years later, it was published that another patient was diagnosed with skin cancer, this time basal cell carcinoma, when he sought a consultation with his primary care physician after he noticed that his dog was paying special interest to a lesion that was previously diagnosed as eczema [26]. After these events, the scientific community started to pay attention to the ability of dogs in identifying cancer-related odors and the possible impact that this fact could have in the clinical practice. Thus, scientists began to study this remarkable ability of dogs in a more meticulous way [5,6,17–24]. Bladder cancer In 2004 Willis et al. published a study looking to verify empirically the olfactory acuity of dogs in the detection of bladder cancer. In the study, six dogs were used in order to detect the cancerous compounds in urine samples of patients presenting recurrent transition cell carcinoma of the bladder against a control group of healthy patients. The study showed that the dogs correctly selected the urine from the patients with bladder cancer with a success rate of 41% [17]. Willis et al. in 2010 published another study in which he and his collaborators used four dogs in order to evaluate if the canine olfactory apparatus reacts to cancer-related VOCs or to

inflammatory-related odors. They accomplished the mentioned purpose, by separating the controls into those that were completely healthy and those that that had one or more minor dipstick findings or any other non-cancerous urological pathology. Furthermore, the authors did not exclude any patient from the study upon the presence of confounders (i.e., medications, menstrual cycle, alcohol consumption, smoking habits, etc.). For the dogs as a group the calculated sensitivity was 64%, while the calculated specificity ranged from 92% for the best-performing dog, down to 56% for the worst-performing dog. The investigators did not find any significant variation in the performance of dogs relative to tumor grade [18]. Melanoma Pickel et al. published a study that had the purpose of assessing the possible existence of certain VOCs that could have a usage in the clinical practice as biomarkers for the detection of melanoma; in order to detect these VOCs, the authors proposed the usage of a biological detector: the canine olfactory system. For the study, two dogs were used alongside seven patients presenting melanoma (the stage and type of the melanoma varied among the patients). One of the dogs sniffed all of the patients, and it localized the lesion in six out of the seven patients; while the other dog only sniffed four out of the seven patients, and it localized the lesion in three out of the four patients. Additionally, different sub-tests were performed which revealed that dogs were capable of identifying and recognizing the smell of melanoma samples even when presented with distractors commonly found in a clinical environment (i.e., adhesive bandages, gauze, latex gloves, etc.) [19]. Lung cancer McCulloch et al. performed a study, in which they tried to prove if canines could detect LC and breast cancer through their sense of smell. The authors performed the study with the aid of five dogs and breath samples. In regard to LC; the canines were capable of discerning among LC patients and controls with a sensitivity of 99% and a specificity of 99%, regardless of the stage of the disease [5]. Another study was performed by Ehmann et al. that evaluated the ability of dogs to identify LC. For the study, four dogs were used along with breath samples from patients and controls. Patients with chronic obstructive pulmonary disease (COPD) were used as controls and patients were not excluded from the study on the basis of tobacco smoking, food ingestion and the intake of drugs. The authors calculated a sensitivity of 90%, a specificity of 72% and a positive and negative predictive value of 86% and 78%, respectively. Furthermore, the staged and type of LC varied among patients and it was concluded that these factors did not have an influence in the ability of the dogs to detect LC. Additionally, the author’s state that LC was detected regardless of the presence of COPD, tobacco smoke and food odors; thus it can be established that inflammatory process does not intercede in the ability of dogs in detecting cancer-related VOCs. Nevertheless, 9/112 drugs were identified as potential confounders [23]. Breast cancer As mentioned before, McCulloch et al. performed a study in order to demonstrate if the canine olfactory ability could be helpful in detecting lung and breast cancer. For breast cancer, the calculated sensitivity was 88% (95% CI 75–100%) and the calculated specificity was 98% (95% CI 90–99%) [5]. By using urine samples and four dogs, Gordon et al. assessed whether dogs could identify breast and prostate cancer-related VOCs. The authors state that the dogs’ choices of breast

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

cancer-related samples against those samples given by healthy controls were not significantly better than chance [24]. Ovarian cancer In order to aid in the development of an inexpensive and simple method for the diagnosis of ovarian cancer, an article was published in 2008 that tested the canine scent in the detection of ovarian carcinoma. The authors tested, with the aid of one dog, the detection of ovarian cancer by examining if the dog could discern between ovarian carcinoma and healthy tissue samples, and between ovarian carcinomas and other gynecological carcinoma gathered samples. In a single blinded manner, ovarian carcinoma samples were tested against other gynecological carcinoma samples, and the calculated sensitivity and specificity was 100% and 91%, respectively. Additionally, in a double blinded manner, the authors tested if the dog could discriminate between ovarian carcinoma samples and healthy tissue samples. The calculated sensitivity and specificity for this part of the study was 100% and 97.5%, respectively [6]. In 2010 an article was published that had as an objective to assess if the canine scent, through the smelling of blood samples, could detect the presence of ovarian carcinoma in patients with this disease. The authors assessed the mentioned objective with the aid of two dogs. Additionally, they also evaluated the capacity of both dogs to discriminate between ovarian cancer samples and tissue samples. The authors calculated a sensitivity of 100% and a specificity of 95% for tissue sample tests, and a sensitivity of 100% and a specificity of 98% for blood sample tests [20]. It is important to state that in both studies, neither the histopathological type, nor the grade and stage of the cancer had an influence on how well the canines were able to identify the ovarian cancer. Colorectal cancer Sonoda et al. published a study that evaluated the accuracy of canine scent in the detection of cancer-related VOCs in the breath and watery stool samples from patients with colorectal cancer (CRC). One dog was used for the study; and in order to test if the ability of dogs to smell the cancerous compounds was affected by the presence of other cofactors, the authors made no exclusion on the basis of age, smoking, disease stage, cancer site, and inflammation or bleeding in patients with cancer or in control individuals. Comparison of the smelling acuity of the dog with a diagnosis made by a colonoscopy yielded sensitivity for canine scent detection of 91% and 97% for breath tests and watery stool tests, respectively; and specificity of 99% and 99% for breath tests and watery stool tests, respectively [21]. Prostate cancer Cornu et al. tried to demonstrate, by the usage of a dog, if the presence of some VOCs in the urine of patients could result, somehow, in a better diagnostic tool for the detection of PCa. It is important to notice that the grade of the cancer varied among patients and that patients were not excluded from the study because of their medical history, tobacco, alcohol, or drug consumption, or any other habit that they presented at the time of the study. The dog was able to discern the urine samples between patients with PCa and those in the control group with a sensitivity and a specificity of 91% [22]. As already mentioned, Gordon et al. assessed the canine olfactory apparatus in the detection of PCa. As in the case of breast cancer, the dogs were unable to detect PCa from the urine samples by a better proficiency than chance alone [24].

3

In Table 1 we summarize the mentioned articles by stating the dogs and samples used, as well as, the type of studies, results, and methodological errors that the different authors committed upon the realization of the studies. Genetics of canine olfaction The OR genes were discovered by Linda Buck and Richard Axel in 1991, a merit for which they were awarded the Nobel Prize in Physiology or Medicine in 2004. ORs are believed to be encoded by the biggest superfamily in the mammalian genome. The canine OR gene repertoire consist of approximately 1300 genes, of which 12–18% are pseudogenes [27]. Dogs, like most mammals, have two olfactory systems: the accessory olfactory system and the main olfactory system. The accessory olfactory system contains, the accessory olfactory bulb and the vomeronasal organ, and one of its main functions is the detection of pheromones. The main olfactory system is composed of the olfactory mucosa, and the olfactory bulb. The mucosa contains the respiratory epithelium and the neuroepithelium. Moreover, the neuroepithelium contains neurons that express in their cilia the ORs [28,29] (Fig. 1). The ORs are G protein couple receptors composed of seven transmembrane hydrophobic domains. These receptors are the first part of a series of events that transforms a mechanical stimulus into a bioelectrical signal [30] (Fig. 2). This signal travels from the cilia into the olfactory nerve and then into the olfactory bulb. The olfactory bulb is the only way station that allows the signal to travel from the peripheral olfactory apparatus to the brain [29] (Fig. 3). Two major classes of ORs have been identified in dogs: classes I and II. It is thought that class I ORs react only to soluble odorants and that class II ORs respond exclusively to volatile odorants. In the dog, approximately 200 OR genes are part of the class I OR repertoire and all of them belong to the chromosome 21, one next to the other and all of them with a size of 29–31 Mb [29,31]. In a study carried out by Quignon et al., in 2003, 817 new canine OR sequences were deciphered. At the time only 21 canine OR sequences were known. The authors located OR sequences over 24 out of the 40 canine chromosomes. According to the study two chromosomes, 17 and 29, contain only one OR gene, 20 chromosomes contain 3–65 OR genes, whereas 124 and 109 OR genes are located in the canine chromosomes, 18 and 21 respectively [31]. In 2005, Quignon et al. were able to expand the known canine OR gene repertoire, and it was estimated that the repertoire is composed of 1094 genes spread across 49 clusters in the genome. Furthermore, the authors of the study stated that the only encountered change, in regard to the number of chromosomes implicated in housing OR genes, was in the canine chromosome 2; which was found to contain a small cluster of two genes. Overall, the increment in number of genes only augmented the size of the known clusters. In the same study it was reported that the dog OR gene repertoire is composed of 23 families and 300 subfamilies. Furthermore, it was also mentioned that ORs from the same subfamilies tend to be clustered and that only 22 canine subfamilies, 134 OR genes, representing 7% of all subfamilies can be located in more than one chromosome [28]. In addition to studying the location of the different OR genes, some researchers have investigated the polymorphisms of the canine OR repertoire, a fact that affects the canine olfactory acuity [32–34]. For example, Robin et al. analyzed the nucleotide sequences of 109 OR genes, which are representative of different OR gene families and subfamilies, gathered from 48 dogs of six breeds

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

4

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

Table 1 Summary of the mentioned articles. Type of cancer

Dogs

Type of study

Samples

Result

Points to improve upon

Refs.

Bladder cancer

1 Labrador, 3 Cocker Spaniels, 1 Mongrel and 1 Papillon

Double blind

Urine

The success rate was 41% (95% CI 23– 58% under assumption of normality and 26–52% using bootstrap methods)

[17]

2 Cocker Spaniels, 1 Springer Spaniel and 1 Labrador

Double blind

Urine

The calculated sensitivity was 64% (95% CI 55–73%), while the calculated specificity ranged from 92% (95% CI 82–97%) for the best preforming dog down to 56% (95% CI 42–68%) for the worst preforming dog

Greater number of samples Better management of the samples The inclusion of a greater number of cofounders Greater number of samples Better management of the samples

Melanoma

1 Standard Schnauzer and 1 Golden Retriever

Blinded

Tissue

The Standard Schnauzer searched 7 patients with a single melanoma lesion, while the Golden Retriever searched only 4 patients out of those 7 patients. The Standard Schnauzer localized the lesion in 6 patients. While the Golden Retriever localized the lesion in 3 patients

Greater number of samples Assessing whether canines are capable of differentiating between different inflammatory processes, a greater number of confounders and the skin cancer The inclusion of more types of skin cancer

[19]

Lung cancer

3 Labrador Retrievers and 2 Portuguese Water dogs

Double blind

Breath

Double blind

Breath

Greater number of samples Patients with other inflammatory processes where not used as controls Greater number of samples Better training methods

[5]

2 German Shepherd dogs, 1 Australian Shepherd dog, 1 Labrador Retriever

The calculated sensitivity and specificity was 99% (95% CI 99–100%) and 99% (95% CI 96–100%), respectively The calculated sensitivity and specificity was 90% (95% CI 78–97%) and 72% (95% CI 51–88%), respectively

3 Labrador Retrievers and 2 Portuguese Water dogs

Double blind

Breath

The calculated sensitivity and specificity was 88% (95% CI 75–100%) and 98% (95% CI 90–99%), respectively

[5]

1 Aussie Cocker mix,1 Cocker mix, 1 German Shepherd, 1 Rhodesian Ridgeback, 1 Boxer, 1 Italian Greyhound, 1 Chihuahua mix, 1 Miniature Goldendoodle, 1 Pembroke Welsh Corgi, 1 Border Collie

Double blind

Urine

No better than chance

Greater number of samples Patients with other inflammatory processes were not used as controls Better choosing of the dogs Greater number of samples Better training methods Better management of the samples

1 Riesenschnauzer

Double blind

Tissue

The calculated sensitivity and specificity was 100% and 97.5%, respectively

[6]

2 Giant Schnauzer

Double blind

Blood and tissue

In the study the authors calculated a sensitivity of 100% and a specificity of 95% for tissue sample tests, and a sensitivity of 100% and a specificity of 98% for blood sample tests

Greater number of samples Inclusion of other inflammatory processes, besides cancer, and other confounders (i.e., smoking) in the control group Inclusion of a greater number of dogs in the study Greater number of samples Inclusion of other inflammatory processes besides cancer and other confounders (i.e., smoking) in the control group

Colorectal cancer

1 Labrador Retriever

Double blind

Breath and watery stool samples

The authors calculated a sensitivity for canine scent detection of 91% and 97% for breath tests and watery stool tests, respectively; a specificity of 99% for, both, breathe tests and watery stool tests

Greater number of samples Inclusion of more inflammatory processes and a greater number of confounders in the control group Inclusion of a greater number of dogs in the study

[21]

Prostate cancer

1 Belgian Malinois Shepherd

Double blind

Urine

The calculated sensitivity and specificity was 91%

[22]

1 Aussie Cocker mix,1 Cocker mix, 1 German Shepherd, 1 Rhodesian Ridgeback, 1 Boxer, 1 Italian Greyhound, 1 Chihuahua mix, 1 Miniature Goldendoodle, 1 Pembroke Welsh Corgi, 1 Border Collie

Double blind

Urine

No better than chance

Greater number of samples Inclusion of a greater number of dogs in the study Greater number of samples Better training methods Better management of the samples

Breast cancer

Ovarian

[18]

[23]

[24]

[20]

[24]

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

5

Fig. 1. The canine olfactory system. Schematic diagram of a dog head. Axons from the main olfactory epithelium project to the main olfactory bulb, while axons of sensory neurons in the vomeronasal organ project to the accessory olfactory bulb.

(Greyhound, Pekingese, German Shepherd, Belgian Milionis, Labrador Retriever, English Springer Spaniel). The authors found 732 mutations (single nucleotide polymorphisms (SNPs) + indels), distributed across all but four of the examined OR genes (the SNPs were present with a frequency of one to 22 SNP per OR gene). It was also pointed out that there is clear variation between the selected breeds and the number of OR genes without SNPs. 24 genes without SNPs were found in the German Shepherds, 21 in the Greyhounds, 14 in the Labrador Retrievers and 10 in the remaining of breeds. When comparing the whole population, and not only the different breeds, it was found that most OR genes are either slightly or highly polymorphic. Additionally, the investigators observed that of the 732 identified mutations in the analyzed OR genes, 152 led to pseudoalleles (alleles with an interrupted coding frame), 307 led to silent mutations and 273 to missense mutations. Nevertheless, the authors stated that the distribution of SNPs causing an inactive gene (pseudoallele), was something that did not occurred in all of the tested canine population, depending instead upon the breed, and even sometimes upon each dog. In the study the D0 value (a value that assess the degree of linkage disequilibrium, where 0 reflects complete independence between loci, whereas a value of 1 reflects complete dependence) was calculated for each pair of SNPs with a minor allele frequency (the frequency in which the more infrequent allele is encountered in a given population) greater than 2. In German Shepherds and Labrador Retrievers 100% of the calculated SNP pairs had a D0 > 0.8 [33]. A fact that according to Giuseppe Lippi et al. demonstrates the enrichment of OR genes in these breeds, thereby

explaining why these breeds are use more than other breeds as sniffer dogs [35]. In another study, were 16 OR genes from 95 dogs of different breeds were analyzed. It was noted that the Boxers that were part of the study had a higher percentage of pseudogenes (21%) than the tested Puddles, a fact that according to the authors may contribute to the lesser smelling acuity of Boxers when comparing it with the olfactory acuity of Poodles [32]. In a third study, 5 OR genes extracted from 35 dogs, 31 police dogs (mostly German Shepherds) and 4 dogs trained to detect cancer (3 German Shepherd mixes and 1 Labrador Retriever), were examined for polymorphisms and it was tested if somehow the olfactory acuity of the dogs varied with the encountered polymorphisms. Interestingly, the dogs that were trained to detect cancer had a statistical significant influence on the detection of cancer upon the encounter of a SNP. The SNP is a G to A transition at position 592 in the cOR9S13 gene causing a substitution, alanine to threonine, at position 198 in the second extracellular loop (EC2) of the receptor. The authors stated that the performance of the dogs varied among those dogs that were homozygous AA, in comparison to those that where AG heterozygotes and GG homozygotes. Additionally, it was noted that there was no statistically significant differences between the AG heterozygotes and GG homozygotes [34]. Cancer related VOCs and their detection In order to reveal the cancer-related VOCs and as a method that has been suggested as a possible tool in the diagnosis of cancer, gas

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

6

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

Fig. 2. Pathway of signal transduction in olfactory sensory neurons. The odorant receptor defines odorant responsiveness. It involves G protein receptors, adenylyl cyclase, Na/Ca channels and Cl channels in order to produce an electric potential in the olfactory neurons.

Fig. 3. The canine nose. Cross-section of the main olfactory epithelium. Axons of sensory neurons project to the main olfactory bulb.

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

7

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx Table 2 Examples of cancer related VOCs. Type of cancer

Subtype of cancer

Source

Volatile organic compounds (VOCs)

Refs.

Breast

Breath

2-Propanol 2,3-Dihydro-1-phenyl-4(1H)-quinazolinone 1-Phenyl-ethanone Heptanal Isopropyl myristate 3,3-Dimethyl pentane 2-Amino-5-isopropyl-8-methyl-1-azulenecarbonitrile 5-(2-Methylpropyl)nonane 2,3,4-Trimethyl decane 6-Ethyl-3-octyl ester 2-trifluoromethyl benzoic acid 2-Ethylhexanethiol 2-Mercaptobenzyl alcohol 2-Mercaptpbenzoazole Calixarene

[36,41,42]

Bladder

Urine

Melibiose Uridine 2-Propenoic acid Glycerol Valerate Fructose

[37]

L-Valine

Citric acid Ribitol Lung

Breath

Small cell lung cancer

4-Methyl-octane 2-Ethyl-1-hexanol 2-Ethyl-4-methyl-1-pentanol 2,3,4-Trimethyl-pentane 2,3-Dimethyl-hexane 3-Ethyl-3-methyl-2-pentanone 2-Methyl-4,6-octadiyn-3-one 2-Propyl-1-pentanol 6,10-Dimethyl-5,9-dodecadien-2-one Styrene Decane Isoprene Benzene Undecane Propyl benzene 1,2,3-Trimethyl benzene Heptanal Isopropyl alcohol 4-Pentenol Ethane, 1,1,2-trichloro-1,2,2-trifluoro Propane, 2-methoxy-2-methy 2,3-Hexanedione 3-Hexanone, 2-methyl Camphor Isomethyl ionone Pentanoic acid Benzophenone Furan Anthracene, 1,2,3,4-tetrahydro-9-propyl Isobutene Methanol Ethanol Acetone Hydrazine Toluene Carbonic dihydrazide 3-ethyl-pentene Pentadine Dimethyl ether Styrene Isoprene 2-Methylpentene Pentane Ethylbenzene Xylenes Trymethylbencene Toluene Decane Heptane Styrene

[38,43–46]

[47]

(continued on next page)

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

8

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

Table 2 (continued) Type of cancer

Subtype of cancer

Source

Volatile organic compounds (VOCs)

Refs.

Ovary

Urine

1-Methyladenosine 3-Methyluridine 4-Androstene-3 17-Dione

[39]

Colorectal

Flatulence, breath

Nitrosamines 1,10-(1-Butenylidene)bis benzene 1,3-Dimethyl benzene 1-Iodo nonane ((1,1-Dimethylethyl)thio) acetic acid 2-Amino-5-isopropyl-8-methyl-1-azulenecarbonitrile Decanal 2-Methybutane 1,2-Pentadiene Cyclohexane Methylcyclopentane 4-Methytoctane 1,4-Dimethylnenzene

[36,48]

Melanoma

Skin samples

4-Methyl decane 5-Methyl dodecane Dodecane Undecane

[40]

Prostate

Urine, breath

Formaldehyde Toluene 2-Amino-5-isopropyl-8-methyl-1-azulenecarbonitrile p-Xylene 2,2-Dimethyl decane

[36]

Decanethiol Octadecanethiol 2-Mercaptobenzimidazole Benzylmercaptan

[49]

Hepatocarcinoma

chromatography–mass spectrometry (GC–MS) and some of its variants have been used [1,36–40]. In Table 2 we provide some examples of cancer-related VOCs that have been elucidated by using these techniques. In a review by Horvath et al., GC–MS, solid phase microextraction–GC, and proton transfer reaction-MS were mentioned as technologies used in the detection of cancer-related VOCs, from the exhaled breath of patients, in an effort to diagnose lung cancer. The sensitivity of the mentioned studies ranged from 54% to 85.2%, while the specificities varied from 69.2% to 99% [46]. Automated thermal desorption, and GC–MS were used in another study as a way of detecting breast cancer-related VOCs through the breath of patients. This study predicted breast cancer with 93.8% sensitivity and 84.6% specificity [1]. GC–MS was also used in order to asses if VOCs encountered in the breath of patients could serve in the detection of colorectal cancer; the achieved sensitivity, specificity and accuracy were 86%, 83% and 76%, respectively [48]. Ovarian and prostate cancers, by using the serum of patients, have also been tested with variants of GC–MS. In the ovarian cancer study, the authors were able to detect this malignancy with a sensitivity of 100% (95% CI 93–100%), a specificity of 95% (95% CI 87–99%), and a positive predictive value of 94% (95% CI 84–99%) [50]. For the PCa study, the authors used two algorithms in order to interpret the data. With the aid of one algorithm, 100% sensitivity and specificity was achieved, while 97% sensitivity and specificity were accomplished with the other algorithm [51]. GC–MS was also used in an effort to distinguish melanoma samples from healthy skin samples. The authors were able to distinguish these two samples with 89% sensitivity and 90% specificity [40]. Frozen urine samples and high performance liquid chromatography/mass spectrometry were used by Issaq et al. in order to assess this method as a possible diagnostic tool in PCa. With the aid of a specific algorithm, the achieved sensitivity and specificity for this method was 100% [52].

Although, the sensitivity and specificity of some of these studies are quite high GC–MS requires skilled individuals in order to apply this test, is expensive and it needs the samples to be preconcentrated before being transported to the device; thus more efficient technologies have been researched [46,53]. Recently, an article was published that assessed the performance of a device based on GC and a unique metal oxide sensor in the identification of patients with bladder cancer vs. patients without bladder cancer. The researchers used urine headspace samples for the study, and with the aid of a specific statistical method the device correctly assigned 100% of cancer cases and 94.6% of controls. Interestingly, the authors claim that this machine is low-cost and is able to perform under the hands of semi-skilled personnel, a clear advancement over other GC based devices. However, the preconcentration of the samples is also necessary, thus diminishing its ability in being a point-of-care test [54]. Other kinds of VOCs detectors, known as electronic noses (e-noses), have been constructed in an attempt to overcome the mentioned issues. For example, Barash et al. used a gold nanoparticle (GNP) based electronic nose in order to identify LC from headspace samples. This electronic nose was able to identify the disease with 96% sensitivity, 86% specificity, and 93% accuracy [7]. In another study, the breath of patients with lung, breast, colorectal, and prostate cancers were analyzed using GNP sensors and it was shown by the investigators that the nanosensors could differentiate between the breath of patients with cancer and controls. Moreover, this e-nose was able to differentiate between the different kinds of cancer. Nevertheless, the sensitivities, the specificities, and the accuracies were not mentioned in the study [36]. A SnO2 semiconductor sensor-based e-nose was constructed in order to discern ovarian carcinoma samples from healthy myometrium and postmenopausal specimens. The authors assessed that the e-nose was able to discern the carcinoma samples from the controls with a sensitivity of 84.4%, and a specificity of 86.8% [55].

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

Another e-nose, the CyranoseÒ 320 e-nose, demonstrated good sensitivity towards VOCs emitted from skin portions of patients affected by melanomas [56]. Although the usage of electronic noses as detectors of cancerrelated VOCs could lead to the development of an economic, noninvasive and user-friendly electronic nose [36], this kind of technology is still beneath the ability of dogs in terms of specificity and sensitivity and require a large sample size for analysis (e-noses usually have concentration thresholds up to 0.1 ppb, while animal olfaction can detect the presence of a molecule in a concentration of 106 ppb, or sometimes even lower) [25,57]. Bioelectronic noses Due to the mentioned setbacks that are part of the technologies mentioned in the last section, several prototypes of a possible bioelectronic nose have been constructed. Different non-biological sensing platforms have been used in order to detect the OR-odorant binding activity such as quartz crystal microbalance (QCM), surface plasmon resonance (SPR), microelectrodes, and field effect transistors (FETs) [25,58]. Additionally, these bioelectronic noses have used ORs as the biological component of the apparatus. It is important to state that since ORs are G couple receptors they need to be embedded in their membrane in order to work. Therefore, the ORs in different bioelectronic noses have been harvested in two ways, by directly immobilizing olfactory cilia extracted from the olfactory apparatus of animals and by expressing the OR in heterologous cells [58,59]. For the purpose of the bioelectronic nose that we are proposing, utilizing the ORs of dogs by extracting the cilia directly from the olfactory apparatus of a dog is inhumane and not a very practical task. Thus, is necessary to obtain the ORs by expressing them in heterologous cells. ORs have been expressed in several heterologous cells such as Escherichia coli, yeast, and mammalian cells, including: Sf9, HeLa, HEK-293, PC12, and CHO. Although it is possible to express this kind of receptors in heterologous cells, there is some difficulties in transfecting ORs into this cells, thus when expressing ORs in heterologous cells, it is important to carry out the following steps: The insertion of an N-terminal leader sequence like a rho-tag, since ORs are absent of this kind of sequences and because it appears that this kind of sequences facilitate the expression and trafficking of ORs to the plasma membrane [25,60]; and the insertion of nonOR G protein-coupled receptors (GPCRs), receptor transporting protein (RTP), receptor expression enhancing protein (REEP), and heat shock protein (Hsc70t); because it appears that these proteins display the capacity to aid in the appropriate localization and functionality of ORs [61]. QCM is a method that has been used in e-noses and uses the phenomenon of piezoelectricity in order to detect the interactions between odorants and chemicals that have been coated onto the crystal surface of a quartz microbalance. This method has also been used in the construction of certain bioelectronic noses but instead of using chemicals, investigators have used ORs as a way to capture these odorants [25]. For example a rat receptor expressed (OR 17) in mammalian HEK-293 cells was coated, along with the HEK-293 cells, on the crystal surface of a quartz microbalance. The QCM responded only to the presence octanal in a dose-dependent manner. In another study, the crystals in a quartz microbalance were coated with crude insoluble membrane extracts of E. coli expressing ODR-10, which is an olfactory receptor from Caenorhabditis elegans. Like in the previous study, this bioelectronic nose responded only to its ligand, diacetyl, in a concentration dependent manner [25,58]. SPR is a photoelectronic-based technique, which measures changes in the refractive index of a small amount of a material adsorbed on a metal [62]. With the aid of this method; the C.

9

elegans ODR-10 was express in HEK-293 cells and placed, along the HEK cells, on a gold sensor chip surface. The activation of the OR by its ligand, diacetyl, prompted an increment in the intracellular calcium ion that induced a concentration dependent change in the SPR signal [60]. In another study SPR was also used as the basis for a bioelectronic nose. Vidic et al. started the assembling process of the biomachine by expressing in the Saccharomyces cerevisiae strain MC18 the rat OR17, and the human OR 1740, as well as Gaolf proteins (Gaolf proteins were used since most OR-odorant interactions cannot elicit an SPR signal, due to the low molecular weight of the odorants, thus the departure of the Gaolf subunit from the lipid bilayer was used as a way to elicit the signal). In order to avoid the possible contribution of the yeast transduction pathway upon OR-ligand binding the authors disrupted the membrane of the transfected cells into pieces with a diameter of 50 nm that contained the expressed proteins. These vesicles were denominated as nanosomes, and were the ones used as the biological component of the electronic nose. To assess if the bioelectric nose worked the authors analyzed the binding of the ORs to their known ligand (the rat OR17, octanal, and the human OR 1740, helional). Both ORs reacted when exposed in a concentration dependent manner to the previously mentioned ligands. Furthermore, there was an SPR response only when GTP was present, and the sensitivity of the apparatus increased about four fold when GTP was replaced by its non-hydrolysable analogue GTP?S. Additionally, it was stated that a lag phase of 30 min was needed for the system to recover its capacity to response to new stimulus and that the apparatus displayed the same level of activity up to 8 times one hour apart from each other [61]. Vidic et al. used a similar bioelectronic nose that the one that was previously described by her with the extent of using in the new bioelectronic nose rat extracted OdorantBinding Proteins (OBPs), OBPs are members of the lipocalin family and are thought to aid in the passage of odorants through the mucus barrier towards the ORs, along with the same type of human OR that it was previously used. Interestingly, with the presence of the OBPs the SPR signal changed in two ways. Firstly, the intensity of the SPR signal increased, and secondly the SPR signal shape changed from a bell shape curve to a sigmoidal curve with plateau at high odorant concentration. By taking advantages of the change in the electrical potential of cells that take place when an OR is activated, certain types of microelectrode-based bioelectronic noses have been constructed [63]. A microelectrode-based bioelectronic nose was constructed by using electrochemical impedance spectroscopy. The authors used the rat OR 17, which was expresses in S. cerevisiae, and then prepared as membrane fractions, as the biological component of the biomachine. This apparatus was able to detect successfully the activation of the mentioned receptor by its odorants, octanal and heptanal [25,64]. Hun Lee et al. developed a bioelectronic nose by using microelectrode arrays (MEAs); and by expressing the OR 17 and the gustatory cyclic nucleotide channel, as a way to increase the cationic influx, in HEK-293 cells. In addition to using the already mentioned components, the authors used a biphasic electrical current stimulation in order to assess whether this kind of stimulus could increase the signal that is normally elicited by the HEK-293 cells upon odorant (octanal) binding. Amusingly, upon ligand binding the electrical stimulation increased the amplitude of the electrical signal by up to fivefold [65]. Carbon nanotubes (CNTs) based devices have shown great promise in the area of biotechnology. For example, single wall carbon nanotubes (swCNTs), a type of CNT based on a nanometer– diameter cylinder consisting of a single grapheme sheet wrapped up to form a tube have been made into field effect transistors (FETs). Therefore, surface modifications (i.e., OR-ligand binding) of the swCNTs can be used in order to alter their conductivity and thus they have been used in the process of developing novel

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

10

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

electronic devices such as bioelectronic noses [66–68]. By using swCNTs-FETs and the human OR, hOR2AG1, expressed in E. coli, Hun Lee et al. built a bioelectronic nose that was able to detect the specific odorant (amyl butyrate) of the utilized OR down to a 1 fM concentration [66]. Since there is an increment of heptanal in the blood of patients with blood cancer, a bioelectronic nose was constructed by using swCNT-FETs with the intent of diagnosing LC by detecting heptanal from human blood samples. The biological component of the apparatus consisted of nanovesicles, produced from HEK-293 cells; containing the human OR1J2, Gaolf, and the receptor-transporting protein 1S (RTP1S), which was utilized as a chaperon. The apparatus was tested with plasma (spiked with various concentrations of heptanal) extracted from healthy individuals, and the apparatus was able to detect the heptanal at a concentration as low as 1  1014 M. Although, heptanal is indeed elevated in lung cancer, its presence it’s not exclusive of the disease, thus the usage of heptanal as an only biomarker for LC detection would not be very sensitive or selective [68]. Another CNT-FET based bioelectronic nose was created for the purpose of assessing food quality. The authors utilized and expressed in HEK-293 cells (later converted into nanovesicles) the canine OR, cfOR5269, which is a specific receptor for hexanal, a VOC that may be used as an indicator of the oxidation of food. The apparatus was capable of detecting hexanal down to 1 fM concentration, and when tested against a spoiled food (milk) the bioelectronic nose exhibited measurable changes in its conductance. Additionally, the response of the apparatus augmented as the days passed and the milk got more spoiled [69]. Lastly, a flexible and ultrasensitive FET-based bioelectronic nose was created by Joo Park et al. The bioelectronic nose was constructed by using plasma-treated bilayer grapheme conjugated with a human OR (hOR2AG1) expressed in E. coli. Furthermore, it was capable of detecting the specific ligand (amyl butyrate) of the used OR down to a concentration of 0.04 fM, thus proving the ultrasensitivity of it. Moreover, the researchers investigated the durability of the constructed bioelectronic nose and it was concluded that the apparatus could be stored for a long period of time under controlled environmental conditions [70].

Evaluation of the hypothesis Some of the mentioned studies that assessed the ability of dogs in detecting cancer-related VOCs are not very promising in their results, while some achieved very favorable results. Moreover, as noted on Table 1 all of them have some flaws in their methodologies. Nevertheless, we believe that they prove that some canines have the innate ability of detecting cancer in a highly prolific manner, even in its early stages. Furthermore, the previous statement can be corroborated by the already mentioned studies that establish genetic differences in the OR repertoire among breeds, and even between individuals of a particular breed, which allows some dogs to have a keener sense of smell than the sense of smell that other dogs may have. Since it is already established that is highly possible that some dogs have the ability of detecting cancer in an extremely capable manner, we think that the next step in doing the proposed bioelectronic nose is the choosing the dog that will provide the DNA, which will be used to express the necessary ORs in any of the already mentioned heterologous cells. To choose the best possible canine first we propose testing a group of dogs, preferentially from the mentioned breeds that have a predisposition to having a better olfaction (i.e., German Shepherds and Labrador Retrievers), for the mentioned SNP that allows canines to have a better affinity for the detection of cancer. Once the SNP is identified, and since there is no other polymorphisms that have been associated with how well a

canine can detect cancer, we think that the next step in order to confirm which dog is the indicated in donating its DNA is by training the chosen dogs in a prolific manner [71], and subsequently assessing the trained dogs in their ability to detect cancer directly from blood/plasma, urine or breath samples gathered from actual cancer patients. Furthermore, with the present article two things can be assumed: that the canine olfactory system, can distinguish between different kinds of malignancies, as implied by the two studies that evaluated the canine scent in the detection of ovarian carcinoma, and that regardless of the histopathology, grade, and stage of a certain kind of cancer (i.e., lung cancer), it has a common scent, and thus it can be hypothesized that a certain type of cancer activates a common group of ORs or the same OR. However, it is important to point out that some e-noses have been able to distinguish between different subtypes of a certain cancer (i.e., non-small-cell lung carcinoma vs. small-cell lung carcinoma) [7], between cancer precursors and cancerous lesions [42], and between metastatic cells and non-metastatic cells [72]. Consequently, it may be wise to evaluate if dogs can distinguish between all of these variables in order to assess if these elements have the potential of being translated into a bioelectronic nose. Given that the location of most of the canine ORs are already known, as already stated in this article, and once the dog from which the DNA will be collected is chosen, we think that the next step in doing the proposed bioelectronic nose is the expression of the canine ORs in heterologous cells in order to discern which of these receptors are activated by the cancer-related VOCs. This is important because trying to place all of the known receptors in the bioelectronic nose could make the proposed biomachine a less economical and a less reliable prospect. The reliability issue is because as we add more receptors to the biomachine we can encounter a problem known as the curse of high dimensionality, where sensors that are not detecting the target molecule only add noise to the system thus decreasing the specificity of the apparatus [73]. In order to achieve the task of deciphering which ORs are the ones detecting the cancer-related VOCs, we recommend following the protocol published by Zhuang and Matsunami that provides the necessary steps, in order to achieve a successful cell-surface expression of ORs, and in order to evaluate, with the aid of a luciferase assay, the interaction between an OR and its ligand. Furthermore, the authors state that a single type of ORs can be expressed against hundreds of different odorants or multiple odorants at different concentrations; or that, alternatively, a large panel of ORs can be expressed and tested against a single type of odorant [60]. Since we believe that it is plausible that the types of VOCs exemplified in Table 2 are the ones that are activating the ORs responsible for detecting cancer (for example, the presence of heptanal in the breath of patients, as seen in Table 2, can be attributed to breast and lung cancer, and some of the mentioned bioelectronics noses have used heptanal in order to prove their functionality, thus proving that animals can detect this type of compounds). Although, none of the bioelectronics noses that are mentioned use canine ORs in order to detect heptanal, we hypothesize that since humans and rats are able to detect the mentioned compound then also dogs are able to do it. Moreover, the bioelectronic nose that used canine ORs was able to detect hexanal, a similar compound to heptanal, thus from this fact it can be also assumed that dogs are able to detect heptanal and that the OR responsible for detecting heptanal is a member of the same subfamily that the one that is responsible for the detection of hexanal), hence these type of compounds may be used to test which ORs are being activated by cancer. Nonetheless, currently GC–MS based apparatuses, in regards to the identification of VOCs, recognize only 60–90% of

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

the total metabolites in a biological sample [44]. Furthermore, although unlikely, Balseiro and Correa suggested that the detection of human cancer by dogs is based on the major histocompatibility complex [74]. Therefore, before or instead of using the type of VOCs mentioned in Table 2 we think that it may be wiser to asses which ORs are reacting to cancer by using breath, urine, blood, or tissue samples directly gathered from patients with cancer and exposing them to the canine ORs. In order to avoid possible confounders that may activate other type of ORs that are not directly linked with the olfaction of cancer, the samples should be from patients that do not consume illicit drugs, alcohol, tobacco, and the least number of possible medications. Additionally, control samples should be gathered from healthy individuals in order to discern between the ORs that are being activated by the odor of the sample itself and by the odor of the cancer. Lastly, the samples should also be gathered from patients that have a different histopathological type, grade and stage of a type of cancer in order to assess if these factors activate a different set of ORs, and if that is the case in order to discern the ORs or OR that detect the common scent of the cancer from the ORs that detect these specificities. Although, most of the studies that evaluated the canine scent in the detection of cancer-related VOCs did not encounter an element that may have behaved as a confounder (except for nine medications in the study by Ehmann et al.), we think that once it has been proven that a groups of ORs or a single OR respond exclusively to a certain type of cancer, then it would be interesting to expose these ORs or OR to the urine, blood, or breath from patients that have other inflammatory processes, and consume illicit drugs, alcohol, tobacco, and at least the most common medications in order to evaluate if the consumption of these elements elicit an activation upon these receptors or receptor. If the scent of a type of cancer activates a group of receptors and if the mentioned substances and illnesses activate some of these receptors then these ORs should be excluded (only if the sensitivity of the cancer detection is not diminished). If by the exclusion of these receptors the performance of the proposed bioelectronic nose may become affected, or if only one olfactory receptor is responsible for detecting a specific type of cancer, then the only option would be to ask the patient to stop the intake of these substances before the application of the test. It is important to state that in order to reduce the time and cost of the research that may yield the knowledge of which canine ORs are being activated by the cancer-related VOCs, then the class II ORs should be studied before the class I ORs, since it has been proposed, as stated already in this article, that the II ORs respond exclusively to volatile odorants thus making these kind of ORs better candidates than the class I ORs for the ORs that are able to recognize the cancer-related odorants. Once the group of ORs or the single OR that is responsible for responding exclusively to a certain type of cancer is identified, then it would be the proper time to commence the development of the bioelectronic nose. For the electronic component of the bioelectronic nose, CNT-based technologies have shown great sensitivity towards detecting the changes that occur during OR-ligand binding, great portability, and a great durability. Thus, we think that this kind of technology is very promising as the possible platform in which the bioelectronic nose could be build [59]. Furthermore, we think that nanovesicles should be used, as the carriers of ORs, since they can be produced in large quantities and stored from prolonged periods of time [69]. Additionally, as stated in two previously mentioned studies, these vesicles have been incorporated with the non-hydrolysable analogue of GTP, GTP?S, as well as, OBPs, which increased the performance of the bioelectronics noses. Consequently, these two types of molecules should be

11

incorporated into the nanovesicles that could potentially be utilized in the proposed bioelectronics nose. Although Vidic et al. used rat OBPs along with human ORs with good results, it is important to notice that in the proposed bioelectronic nose preferentially canine OBPs should be used because it has been suggested that OBPs have the highest affinity to olfactory receptors that originate from the same species [75]. A canine OBP was purified and partially characterized in a study by D’Auria et al. [76]. Nonetheless, it has been demonstrated that different types of OBPs exist in a certain species, and that these OBPs bind and carry different kind of odorants [77]. Therefore, more research has to be invested into the characterization, binding affinities, and genetic location of these proteins in canines in order to make the proposed bioelectronic nose a more efficient and effective apparatus. Lastly, it should be mention that when making the bioelectronic nose the number of a certain type of OR should be something that it is taken into account. For example, the researchers of the ultrasensitive FET-based bioelectronic nose reported that the sensitivity of the bioelectronic nose increased according to the following order 1OR < 2OR < 4OR [70].

Conclusion In light of the given information and with the right team of researchers (encompassing for example, physical engineers, electronic engineers, and geneticists), we consider that the realization of a bioelectronic nose for the detection of cancer (through the breath, urine, or blood from patients), based on of the canine olfactory apparatus is a realistic possibility. Furthermore, we envision an apparatus were the electronic component and the biological one (the nanovesicles) would be sold separately, as if the biological component were ‘‘spare parts’’, which could be incorporated (maybe with the aid of an auto sampler) with the electronic component of the bioelectronic nose at the doctor’s office in order to aid in the diagnosis of a specific type of cancer and if possible (if the canine OR repertoire can detect the differences between the subtypes of a specific type of cancer and if the curse of high dimensionality is not infringed) in order to differentiate, between the different subtypes of the cancer. In 2011, in the United States 88.7 billion dollars were spent on cancer [78]. If our hypothesis is correct we believe that this technology could reduce the economic cost of cancer by decreasing the use of techniques that are costly and not very sensitive or specific in the detection of cancer, and by detecting in an earlier manner this disease, thus reducing treatment costs. Additionally, this biomachine will improve the quality of life of those who suffer this disease, thus making this apparatus a more cost effective tool for the diagnosis of cancer than what is available nowadays.

Conflict of interest The authors declare that there is no conflict of interest. References [1] Hill D, Binions R. Breath analysis for medical diagnosis. Int J Smart Sens Intell Syst 2012;5:401–40. [2] Castro M, Kumar B, Feller JF, Haddi Z, Amari A, Bouchikhi B. Novel e-nose for the discrimination of volatile organic biomarkers with an array of carbon nanotubes (CNT) conductive polymer nanocomposites (CPC) sensors. Sens Actuators, B 2011;159:213–9. [3] Probert CS, Ahmed I, Khalid T, Johnson E, Smith S, Ratcliffe N. Volatile organic compounds as diagnostic biomarkers in gastrointestinal and liver diseases. J Gastrointest Liver Dis 2009;18:337–43. [4] CDC. Indoor environmental quality. [accessed 3.03.14].

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

12

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx

[5] McCulloch M, Jezierski T, Broffman M, Hubbard A, Turner K, Janecki T. Diagnostic accuracy of canine scent detection in early- and late-stage lung and breast cancers. Integr Cancer Ther 2006;5:30–9. [6] Horvath G, Järverud GA, Järverud S, Horváth I. Human ovarian carcinomas detected by specific odor. Integr Cancer Ther 2008;7:76–80. [7] Ryan MA, Zhou H, Buehler MG, Manatt KS, Mowrey VS, Jackson SP, et al. Monitoring space shuttle air quality using the Jet Propulsion Laboratory electronic nose. IEEE Sens J 2004;4:337–47. [8] Ryan MA, Homer ML, Buehler MG, Manatt KS, Zee F, Grat J. Monitoring the air quality in a closed chamber using an electronic nose. SAE Trans 1997;106(1):1014–9. [9] Persaud KC. Medical applications of odor-sensing devices. Int J Low Extrem Wounds 2005;4:50–6. [10] Persaud KC, Wareham P, Pisanelli AM, Scorsone E. ‘Electronic nose’ – new condition monitoring devices for environmental applications. Chem Senses 2005;30(Suppl. 1):i252–3. [11] Persaud K, Dodd G. Analysis of discrimination mechanisms in the mammalian olfactory system using a model nose. Nature 1982;299:352–5. [12] Barash O, Peled N, Tisch U, Bunn Jr PA, Hirsch FR, Haick H. Classification of lung cancer histology by gold nanoparticle sensors. Nanomedicine 2012;8:580–9. [13] World Health Organization. GLOBOCON 2008. Estimated cancer incidence, mortality, prevalence and disability-adjusted life years (DALYs) worldwide in 2008. Available: [accessed 3.03.13]. [14] van Ravesteyn NT, Miglioretti DL, Stout NK, Lee SJ, Schechter CB, Buist DS, et al. Tipping the balance of benefits and harms to favor screening mammography starting at age 40 years: a comparative modeling study of risk. Ann Intern Med 2012;156:609–17. [15] Ghosal R, Kloer P, Lewis KE. A review of novel biological tools used in screening for the early detection of lung cancer. Postgrad Med J 2009;85:358–63. [16] Bastarrika G, García-Velloso MJ, Lozano MD, Montes U, Torre W, Spiteri N, et al. Early lung cancer detection using spiral computed tomography and positron emission tomography. Am J Respir Crit Care Med 2005;171:1378–83. [17] Willis CM, Church SM, Guest CM, Cook WA, McCarthy N, Bransbury AJ, et al. Olfactory detection of human bladder cancer by dogs: proof of principle study. BMJ 2004;329:712. [18] Willis CM, Britton LE, Harris R, Wallace J, Guest CM. Volatile organic compounds as biomarkers of bladder cancer: sensitivity and specificity using trained sniffer dogs. Cancer Biomark 2010–2011;8:145–53. [19] Pickel D, Manucy GP, Walker DB, Hall SB, Walker JC. Evidence for canine olfactory detection of melanoma. Appl Anim Behav Sci 2004;89:107–16. [20] Horvath G, Andersson H, Paulsson G. Characteristic odour in the blood reveals ovarian carcinoma. BMC Cancer 2010;10:643. [21] Sonoda H, Kohnoe S, Yamazato T, Satoh Y, Morizono G, Shikata K, et al. Colorectal cancer screening with odour material by canine scent detection. Gut 2011;60:814–9. [22] Cornu JN, Cancel-Tassin G, Ondet V, Girardet C, Cussenot O. Olfactory detection of prostate cancer by dogs sniffing urine: a step forward in early diagnosis. Eur Urol 2011;59:197–201. [23] Ehmann R, Boedeker E, Friedrich U, Sagert J, Dippon J, Friedel G, et al. Canine scent detection in the diagnosis of lung cancer: revisiting a puzzling phenomenon. Eur Respir J 2012;39:669–76. [24] Gordon RT, Schatz CB, Myers LJ, Kosty M, Gonczy C, Kroener J, et al. The use of canines in the detection of human cancers. J Altern Complement Med 2008;14:61–7. [25] Hun Lee S, Hyun Park T. Recent advances in the development of bioelectronic nose. Biotechnol Bioprocess Eng 2010;15:22–9. [26] Church J, Williams H. Another sniffer dog for the clinic? Lancet 2001;358:930. [27] Rouquier S, Giorgi D. Olfactory receptor gene repertoires in mammals. Mutat Res 2007;616:95–102. [28] Quignon P, Giraud M, Rimbault M, Lavigne P, Tacher S, Morin E, et al. The dog and rat olfactory receptor repertoires. Genome Biol 2005;6:R83. [29] Quignon P, Rimbault M, Robin S, Galibert F. Genetics of canine olfaction and receptor diversity. Mamm Genome 2012;23:132–43. [30] Mombaerts P. Genes and ligands for odorant, vomeronasal and taste receptors. Nat Rev Neurosci 2004;5:263–78. [31] Quignon P, Kirkness E, Cadieu E, Touleimat N, Guyon R, Renier C, et al. Comparison of the canine and human olfactory receptor gene repertoires. Genome Biol 2003;4:R80. [32] Tacher S, Quignon P, Rimbault M, Dreano S, Andre C, Galibert F. Olfactory receptor sequence polymorphism within and between breeds of dogs. J Hered 2005;96:812–6. [33] Robin S, Tacher S, Rimbault M, Vaysse A, Dréano S, André C. Genetic diversity of canine olfactory receptors. BMC Genomics 2009;10:21. [34] Lesniak A, Walczak M, Jezierski T, Sacharczuk M, Gawkowski M, Jaszczak K. Canine olfactory receptor gene polymorphism and its relation to odor detection performance by sniffer dogs. J Hered 2008;99:518–27. [35] Lippi G, Cervellin G. Canine olfactory detection of cancer versus laboratory testing: myth or opportunity? Clin Chem Lab Med 2012;50:435–9. [36] Peng G, Hakim M, Broza YY, Billan S, Abdah-Bortnyak R, Kuten A. Detection of lung, breast, colorectal, and prostate cancers from exhaled breath using a single array of nanosensors. Br J Cancer 2010;103:542–51. [37] Pasikanti KK, Esuvaranathan K, Ho PC, Mahendran R, Kamaraj R, Wu QH, et al. Noninvasive urinary metabonomic diagnosis of human bladder cancer. J Proteome Res 2010;9:2988–95.

[38] Peng G, Tisch U, Adams O, Hakim M, Shehada N, Broza YY, et al. Diagnosing lung cancer in exhaled breath using gold nanoparticles. Nat Nanotechnol 2009;4:669–73. [39] Woo HM, Kim KM, Choi MH, Jung BH, Lee J, Kong G, et al. Mass spectrometry based metabolomic approaches in urinary biomarker study of women’s cancers. Clin Chim Acta 2009;400:63–9. [40] Abaffy T, Duncan R, Riemer DD, Tietje O, Elgart G, Milikowski C, et al. Differential volatile signatures from skin, naevi and melanoma: a novel approach to detect a pathological process. PLoS One 2010;5:e13813. [41] Phillips M, Cataneo RN, Ditkoff BA, Fisher P, Greenberg J, Gunawardena R, et al. Prediction of breast cancer using volatile biomarkers in the breath. Breast Cancer Res Treat 2006;99:19–21. [42] Shuster G, Gallimidi Z, Reiss AH, Dovgolevsky E, Billan S, Abdah-Bortnyak R, et al. Classification of breast cancer precursors through exhaled breath. Breast Cancer Res Treat 2011;126:791–6. [43] Chen X, Cao M, Hao Y, Li Y, Wang P, Ying K, et al. A non-invasive detection of lung cancer combined virtual gas sensors array with imaging recognition technique. Conf Proc IEEE Eng Med Biol Soc 2005;6:5873–6. [44] Kouremenos KA, Johansson M, Marriott PJ. Advances in gas chromatographic methods for the identification of biomarkers in cancer. J Cancer 2012;3:404–20. [45] Machado RF, Laskowski D, Deffenderfer O, Burch T, Zheng S, Mazzone PJ, et al. Detection of lung cancer by sensor array analyses of exhaled breath. Am J Respir Crit Care Med 2005;171:1286–91. [46] Horváth I, Lázár Z, Gyulai N, Kollai M, Losonczy G. Exhaled biomarkers in lung cancer. Eur Respir J 2009;34:261–75. [47] Poli D, Carbognani P, Corradi M, Goldoni M, Acampa O, Balbi B, et al. Exhaled volatile organic compounds in patients with non-small cell lung cancer: cross sectional and nested short-term follow-up study. Respir Res 2005;6:71. [48] Altomare DF, Di Lena M, Porcelli F, Trizio L, Travaglio E, Tutino M, et al. Exhaled volatile organic compounds identify patients with colorectal cancer. Br J Surg 2013;100:144–50. [49] Amal H, Ding L, Liu BB, Tisch U, Xu ZQ, Shi DY, et al. The scent fingerprint of hepatocarcinoma: in-vitro metastasis prediction with volatile organic compounds (VOCs). Int J Nanomed 2012;7:4135–46. [50] Petricoin EF, Ardekani AM, Hitt BA, Levine PJ, Fusaro VA, Steinberg SM, et al. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 2002;359:572–7. [51] Qu Y, Adam BL, Yasui Y, Ward MD, Cazares LH, Schellhammer PF, et al. Boosted decision tree analysis of surface-enhanced laser desorption/ionization mass spectral serum profiles discriminates prostate cancer from noncancer patients. Clin Chem 2002;48:1835–43. [52] Issaq HJ, Nativ O, Waybright T, Luke B, Veenstra TD, Issaq EJ, et al. Detection of bladder cancer in human urine by metabolomic profiling using high performance liquid chromatography/mass spectrometry. J Urol 2008;179:2422–6. [53] Barash O, Peled N, Hirsch FR, Haick H. Sniffing the unique ‘‘odor print’’ of nonsmall-cell lung cancer with gold nanoparticles. Small 2009;5:2618–24. [54] Khalid T, White P, De Lacy Costello B, Persad R, Ewen R, Johnson E, et al. A pilot study combining a GC-sensor device with a statistical model for the identification of bladder cancer from urine headspace. PLoS One 2013;8:e69602. [55] Horvath G, Chilo J, Lindblad T. Different volatile signals emitted by human ovarian carcinoma and healthy tissue. Future Oncol. 2010;6:1043–9. [56] Wilson AD, Baietto M. Applications and advances in electronic-nose technologies. Sensors (Basel) 2009;9:5099–148. [57] Vidic J. Bioelectronic noses based on olfactory receptors. In: Somerset VS, editor. Intelligent and biosensors. Rijeka: InTech Europe; 2010. p. 377–86. [58] Glatz R, Bailey-Hill K. Mimicking nature’s noses: from receptor deorphaning to olfactory biosensing. Prog Neurobiol 2011;93:270–96. [59] Vidic JM, Grosclaude J, Persuy MA, Aioun J, Salesse R, Pajot-Augy E. Quantitative assessment of olfactory receptors activity in immobilized nanosomes: a novel concept for bioelectronic nose. Lab Chip 2006;6:1026–32. [60] Zhuang H, Matsunami H. Evaluating cell-surface expression and measuring activation of mammalian odorant receptors in heterologous cells. Nat Protoc 2008;3:1402–13. [61] Bush CF, Hall RA. Olfactory receptor trafficking to the plasma membrane. Cell Mol Life Sci 2008;65:2289–95. [62] Pattnaik P. Surface plasmon resonance: applications in understanding receptor–ligand interaction. Appl Biochem Biotechnol 2005;126:79–92. [63] Vidic J, Grosclaude J, Monnerie R, Persuy MA, Badonnel K, et al. On a chip demonstration of a functional role for Odorant Binding Protein in the preservation of olfactory receptor activity at high odorant concentration. Lab Chip 2008;8:678–88. [64] Hou Y, Jaffrezic-Renault N, Martelet C, Zhang A, Minic-Vidic J, Gorojankina T, et al. A novel detection strategy for odorant molecules based on controlled bioengineering of rat olfactory receptor I7. Biosens Bioelectron 2007;22:1550–5. [65] Lee SH, Jeong SH, Jun SB, Kim SJ, Park TH. Enhancement of cellular olfactory signal by electrical stimulation. Electrophoresis 2009;30:3283–8. [66] Lee SH, Jin HJ, Song HS, Hong S, Park TH. Bioelectronic nose with high sensitivity and selectivity using chemically functionalized carbon nanotube combined with human olfactory receptor. J Biotechnol 2012;157:467–72. [67] McEuen PL, Fuhrer MS, Park H. Single-walled carbon nanotube electronics. IEEE Trans Nanotechnol 2002;1:78–85.

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024

A. Pomerantz et al. / Medical Hypotheses xxx (2015) xxx–xxx [68] Lim JH, Park J, Oh EH, Ko HJ, Hong S, Park TH. Nanovesicle-based bioelectronic nose for the diagnosis of lung cancer from human blood. Adv Healthc Mater 2013. [69] Park J, Lim JH, Jin HJ, Namgung S, Lee SH, Park TH, et al. A bioelectronic sensor based on canine olfactory nanovesicle–carbon nanotube hybrid structures for the fast assessment of food quality. Analyst 2012;137:3249–54. [70] Park SJ, Kwon OS, Lee SH, Song HS, Park TH, Jang J. Ultrasensitive flexible graphene based field-effect transistor (FET)-type bioelectronic nose. Nano Lett 2012;12:5082–90. [71] Williams M, Johnston JM. Training and maintaining the performance of dogs (Canis familiaris) on an increasing number of odor discriminations in a controlled setting. Appl Anim Behav Sci 2002;78:55–65. [72] Bajaj A, Miranda OR, Kim IB, Phillips RL, Jerry DJ, Bunz UH, et al. Detection and differentiation of normal, cancerous, and metastatic cells using nanoparticlepolymer sensorarrays. Proc Natl Acad Sci U S A 2009;106:10912–6. [73] Gardner JW, Boilot P, Hines EL. Enhancing electronic nose performance by sensor selection using a new integer-based genetic algorithm approach. Sens Actuators, B 2005;106:114–21.

13

[74] Balseiro SC, Correia HR. Is olfactory detection of human cancer by dogs based on major histocompatibility complex-dependent odour components? – a possible cure and a precocious diagnosis of cancer. Med Hypotheses 2006;66:270–2. [75] Ko HJ, Lee SH, Oh EH, Park TH. Specificity of odorant-binding proteins: a factor influencing the sensitivity of olfactory receptor-based biosensors. Bioprocess Biosyst Eng 2010;33:55–62. [76] D’Auria S, Staiano M, Varriale A, Scognamiglio V, Rossi M, Parracino A, et al. The odorant-binding protein from Canis familiaris: purification, characterization and new perspectives in biohazard assessment. Protein Pept Lett 2006;13:349–52. [77] Tegoni M, Pelosi P, Vincent F, Spinelli S, Campanacci V, Grolli S, et al. Mammalian odorant binding proteins. Biochim Biophys Acta 2000;1482:229–40. [78] American Cancer Society. Economic impact of cancer. ; 2013 [accessed 10.04.15].

Please cite this article in press as: Pomerantz A et al. The possibility of inventing new technologies in the detection of cancer by applying elements of the canine olfactory apparatus. Med Hypotheses (2015), http://dx.doi.org/10.1016/j.mehy.2015.04.024