The dietary habits are governed, in part, by oro-sensory detection of taste. In fact, the taste of nutrients leads human beings to decide quickly to accept or reject a food. Basically, there are five taste modalities, i.e., sweet, sour, bitter, salty, umami and perhaps, a sixth fat taste (Heinze et al. (2015). Beside the fact that taste is essential for life because it regulates food intake, taste also provides hedonic pleasure from eating. The taste perception also activates neuronal pathways, leading to the preparation for digestion, absorption, and storage of nutrients (Brondel et al., 2013). A dysfunction of taste perception (dysgeusia) may impair the quality of life by affecting appetite, body weight, and psychological well-being (Deems et al., 1991). There are several factors that may affect taste perception, including medication, nutrition, lesions in the oral mucosa, prolonged exposure to radiations and chemotherapy, smoking, chronic hepatitis, renal dysfunction, aging, and perturbation in hormonal secretions (Maffeis and Silva-Netto, 1989).

Changes in taste perception are especially important in diseases like cancer, which is one of the main causes of morbidity and mortality throughout the world. Altered taste perception in cancer subjects is usually ignored by clinicians as this aspect does not represent the life-threatening events. There are some indications that taste alteration might be an alarming early sign of tumor cell invasion in cancer patients (Sherry, 2001). Indeed, the most distressing symptom in patients with advanced cancer is gastrointestinal abnormalities, whereas the change in taste is the fourth most common symptom after dry mouth, weight loss, and early satiety (Komurcu et al., 2002). Some studies suggests that 15 to 100 percent of cancer patients may suffer from a taste change (Lockhart and Clark, 1989; Ripamonti et al., 1998).

Before discussing the relationship between cancer and taste changes, we must have an idea of oro-sensory perception of taste. Taste is a complex entity that interacts with other senses: Hearing, touch, smell, and vision. All the information from the sensory organs is finally analyzed by the central nervous system (CNS). During mastication, foods is mixed with saliva which is secreted by mandibular, sublingual, and parotid salivary glands. Saliva dilutes and disseminates palatable molecules to the taste receptors on the tongue, palate, larynx, pharynx, and the upper third of the esophagus (Matsuo, 2000). Taste receptors or taste receptor cells (TRCs) have been identified on tongue epithelium and throughout the digestive tract (Rozengurt and Sternini, 2007), but we will not discuss the latter part in this review article.

Human beings have around 5,000 taste buds. Of these, 30% are in the fungiform papillae, 30% in foliate papillae and 40% in circumvallate papilla (Suzuki, 2007). The filiform papillae contain no taste buds. Goblet or circumvallate buds are located in the posterior position of the tongue forming an inverted “V” which are nine in humans. The foliate papillae are located in posterior lateral position of the tongue. The fungiform and filiform papillae are present on the apical surface of the tongue. Each taste bud contains 50–100 taste cells, surrounded by supporting cells which are renewed after every 10 days (Wakisaka, 2005). The localization of taste buds and TRCs on human tongue are shown in Figure 1. Three cell types have been identified, based on morphological criteria (Takeda and Hoshino, 1975). Glial or Type I cells that assure homeostasis in the taste bud, are sensitive to salty substances and possess ATPase activity (Bartel et al., 2006), which is crucial to degrade high concentrations of extracellular ATP. The TRCs or type II cells are sensitive to sweet, bitter and umami substances via the receptor activation of T1R and T2R family. These cells secrete ATP in the interstitial medium through pannexin channel (Romanov et al., 2008). The ATP binds to the P2Y receptors on the presynaptic or type III cells. This mechanism results in the release of serotonin, thus activating the postsynaptic receptors which are involved in the transmission of the taste information to brain. The type III cells are sensitive to acidic substances (Yang et al., 2000; Brondel et al., 2013; Roland and and Rémi, 2013). The fourth type of taste cells are termed as type IV cells which are basal, non-polarized, presumably undifferentiated cells and serve as progenitor cells for other three types of taste cells (Miura et al., 2006). The comprehensive reviews on different taste modalities and the activation of different brain areas can be consulted elsewhere (Bermúdez-Rattoni, 2004; Chaudhari and Roper, 2010; Roper, 2013; Besnard et al., 2016). However, in brief, we would like to outline the implication of TRCs in different taste perception. Salt activates sodium channels, while the acidic compounds induce the depolarization via by blocking potassium channels (Kinnamon et al., 1988). Sweet, bitter, fat, and umami taste involve metabotropic receptors (Medler, 2011). CD36 and GPR120 play non-overlapping roles during orosensory detection of dietary fats (Ozdener et al., 2014). Two large families of receptors have been identified: T1R and T2R. The T2R receptors are involved in the perception of bitter taste (Chandrashekar et al., 2000). The heterodimeric T1R2/T1R3 detects the sweet taste (Montmayeur et al., 2001), whereas the heterodimeric T1R1/T1R3 is sensitive to umami taste (Chaudhari et al., 2000). The binding of sapid molecules to these taste receptors activates a G-protein, called Gustducin (Wong et al., 1996). The alpha-subunit of gustducin activates the PLCβ₂and generates inositol-trisphosphate (IP₃) and diacylglycerol. All these intracellular mechanisms lead to an increase in intracellular calcium and neurotransmitter release. A diagrammatical representation of various types of taste cells and receptors involved in taste perception are presented in Figure 2. The neurotransmitters released by the TRC activate the afferent nerve fibers that carry taste information to the CNS. These afferent nerves are: (i) the cord of the eardrum (gustatory branch of the facial nerve) which connects the fungiform papillae of the anterior two thirds of the tongue, (ii) the glossopharyngeal nerve (IX) that connects the goblet buds in the posterior third, and (iii) the superior laryngeal vagus nerve (nerve X) that transmits oropharyngeal sensitivity. Other nerves, such as the trigeminal nerve, are also incidentally involved in taste perception. Neurons conduct a first relay in the solitary nucleus, located in the dorsolateral part of the bulb, and then make a second relay in the ventromedial nucleus of the thalamus before they project into the cortical or other area, involved in identifying taste (type and intensity). The nucleus accumbens and the ventral tegmental area are also activated following food ingestion. They are involved in hedonic responses (pleasure) and the reward circuit (Kettaneh et al., 2002; Brondel et al., 2013; Roland and and Rémi, 2013).

One of the striking features of advanced cancer is the inflammatory state that is generally associated with an infection (Coussens and Werb, 2002). Rapidly proliferating cancer cells release a number of cytokines/chemokines which favor the recruitment of macrophages and neutrophils, which, in turn, produce a series of cytokines and cytotoxic mediators including prostaglandins (Kuper et al., 2000). It is mention worthy that 15% of cancer are associated with an infection that might trigger a sustained inflammatory state (Kuper et al., 2000). In neck cancer, the level of pro-inflammatory cytokines, i.e., IL-1α, RANTES, MIG, G-CSF, GM-CSF, INF-γ, TNF-a, IL-17, IL-4, IL-6, and IL-10, in the body is increased by several fold (Johnson et al., 2014). The cytokines like IL-12 and IFNγ have an anti-tumor role, while the cytokines like IL-6, IL-17, and IL-23 are pro-tumor (Lin and Karin, 2007). The cytokines IL-6 and IL-10 are associated with poor prognosis in all types of cancer (Lippitz, 2013). Inflammation plays a crucial role in cachexia (Epstein and Barasch, 2010). Conversely, cancer cachexia is associated with an increase in blood levels of C-reactive protein, cytokines (interleukin 1b, interleukin-6, TNF-α, and leukemia inhibitor factor, LIF) and other tumor derived factors like lipid mobilizing factor (LMF) and protein mobilizing factor (PMF). The LMF and PMF can directly mobilize fatty acids and amino acids, respectively, from adipose tissue and skeletal muscle. Ming-Hua et al. (2016) have recently confirmed that high concentrations of IL-1b, IL-6, TNF-α, and LMF are directly involved in cancer cachexia. Indeed, LMF initiates ubiquitin-dependent catabolic pathway and contributes to weight loss during cancer cachexia (Dimitriu et al., 2005) which is, somehow, associated with loss of taste perception as demonstrated by Maschke et al. (2017). The inflammatory markers via blood circulation may also exert their action in the brain and modulate the areas involved in the control of feeding behavior including smell and taste perception (Argilés et al., 2014). These observations suggest that alteration in taste in cancer patients might be controlled both at taste bud and brain levels. As regards the association between inflammation and taste bud dysfunction that might result into altered taste perception, we can cite the study of Wang et al. (2009). These investigators have reported the expression of Toll-like receptors (TLRs), type I and II interferon (IFN) receptors, and their downstream signaling components in taste tissue. Some TLRs appear to be selectively or more abundantly expressed in taste buds than in non-gustatory lingual epithelium. Immunohistochemical observations have confirmed the presence of these receptor proteins in taste bud cells, of which TLRs 2, 3, and 4 are expressed in type II cells. Administration of TLR ligands and lipopolysaccharides activated IFN-g signaling pathways, up-regulated the expression of IFN-g-inducible genes, and down-regulated the expression of c-fos in taste buds. Interestingly, systemic administration of IFNs triggered apoptosis of taste bud cells in mice and, consequently, contributed to the development of taste disorders. It has been shown that MRL/lpr mice have a high concentration of IFNγ and TNF-a, and INFγ would induce the apoptosis of TRCs. The perception of bitter, sweet and umami taste is also reduced in these mice. This alteration of taste perception is associated with a decrease in the number of gustducin positive cells and renewal of TRCs (Kim et al., 2012).

There also seems a relationship between inflammation and hyperglycemia in cancer since carbohydrate metabolism is altered in cancer patients (Duan et al., 2014), as is the case of type 2 diabetes. The high glucose metabolism in cancer cells might be due to high expression of glucose transporters (Walenta et al., 2000; Hauptmann et al., 2005). Gondivkar et al. (2009) have reported that the subjects with type 2 diabetes may also suffer from alterations in taste thresholds for different taste modalities. Hence, it is possible that the alteration in carbohydrate metabolism and taste perception may be a common mechanism between type 2 diabetes and cancer. However, this statement requires further studies in future.

Roughly, there are ~10¹⁴ microbes residing in human intestine with genetic content almost 100 times higher than that of the human genome (Ley et al., 2006). These microbiota are present throughout the gastrointestinal tract, starting from mouth till the terminal part of large intestine (Rozengurt and Sternini, 2007). Schmidt et al. (2014) have shown that the abundance of oral microbiota was varied in individuals during oral cancer. In another study, it was observed that some bacterial species in the buccal cavity of patients with oral squamous cell carcinoma differed from that of normal volunteers (Mager et al., 2005). Interestingly, the chnages in gut microbiota are closely related with alterations in taste preference in mice (Duca et al., 2012; Swartz et al., 2012). One possible mechanism might be that the modifications of gut microbiota might influence the expression of gut G-protein coupled receptors (Rousseaux et al., 2007), beside manipulating host feeding behavior through hormonal and neural mechanisms (Alcock et al., 2014). Oral mucositis, which is closely related with alterations in taste, is frequently reported in cancer patients undergoing chemotherapy (Knox et al., 2000). Wang et al. (2015) have proposed that disruption of oral microbiota could result in chemotherapy-induced inflammation through toll-like receptors (TLRs) and nucleotide oligomerization domain (NOD)-like receptors (NLRs). Ligands for these receptors like peptidoglycan, lipopolysaccharide, bacterial DNA, and protein flagellin are frequently provided by disrupted microbiota, which results in induction of inflammatory process. As cancer therapy is frequently associated with disrupted microbiota, while alterations of microbiota leads to inflammation and changes is taste perception, it would be natural to assume that disruption of gut microbiota might result in taste changes in cancer patients. However, this argument needs to be studied in future.

Different methods have been used for the measurement of taste impairment and detection thresholds in cancer patients. A detailed account of the methods will be out of scope of this article; hence, only a brief summary has been given. An electro-gustometer can be used for the detection of taste thresholds (Williams and Cohen, 1978). One electrode is the tongue electrode, while the other reference electrode is placed on dorsal side or the wrist. Electrical current is applied in different steps and the lowest current intensity, perceived by the subject, is taken as the detection threshold (Berteretche et al., 2004). Chemical detection involves using tastant solution with suprathreshold concentrations for each of the basic taste modalities (Trant et al., 1982; Bossola et al., 2007). The subject's mouth is rinsed with a sip of distilled water prior to testing each sample (Sánchez-Lara et al., 2010). The lowest concentration of solute at which the individual consistently recognizes the taste is considered as taste threshold (Gallagher and Tweedle, 1983; Steinbach et al., 2009). A 3-armed forced choice (3-AFC) method can be applied to monitor the subject's response for detection and reorganization of taste (Mossman and Henkin, 1978). Subjects can be asked to place a cross-hatch on a 10-cm line labeled at each end (0 = dislike extremely, no sweetness, sourness, saltiness, or bitterness, and 10 = like extremely, extremely sweet, sour, salty, or bitter) to indicate degree of liking and intensity of each sample (Trant et al., 1982). Alternatively, a scale ranging from 0 (total taste loss) to 3 (no taste loss) can be employed to classify the subjects, depending upon if they recognize no or all concentrations of a particular taste (Maes et al., 2002). Questionnaires have also been used by different researchers to determine the patient's self-reported taste changes and their effects on their quality of life (Chencharick and Mossman, 1983; Huldij et al., 1986; Bjordal et al., 1994; Bernhardson et al., 2008). The choice of a particular test depends on the practitioner.

The patients themselves can assess the changes in thresholds for different taste modalities. Consequently, they may change either their feeding habits or adopt appropriate palatable strategies. For example, breast cancer patients, undergoing docetaxel or paclitaxel chemotherapy, included a number of strategies like changing food habits by adding new recipes (Speck et al., 2013). Wickham et al. (1999) proposed to increase food palatability by adding artificial falvours. Rehwaldt et al. (2009) have reported that each patient should have its own particular strategy to adapt taste alterations. Hence, more than 50% of the patients reported that they tried selected one of the following strategies: More fats and sauces, eating smaller and more frequent meals, using more condiments, eating blander foods, adding something sweet to meats, sucking on hard candy, eating more boiled foods, and avoiding beef. These strategies were helpful for the majority (74–87%) of patients who tried them. Nonetheless, self-care strategies are difficult to be practized due to psychological constraints.

Patient counseling can be started in advance to prepare patient mentally before time. Rhodes et al. (1994) have demonstarted if the patients are prepared psychologically for taste alterations, they can tolerate taste changes easily. Ravasco et al. (2003) have stressed on counseling the cancer patients to face the altered changes in taste perception. Before chemotherapy, the patients can be encouraged to try new food products or supplements (Capra et al., 2001). Lemon juice and chewing gum could be used prior to meals to make the meals more pleasant. Small but frequent meals should be encouraged as they are better tolerated by the patients (Ravasco, 2005). Patients can also be asked to maintain good oral hygiene as it may also contribute to changes in taste. To assess whether changes in taste perception in cancer patients are causing taste aversion and, consequently, to malnutrition, the clinicians may use different approaches, like Malnutrition Screening Tool (Ferguson et al., 1999), Interdisciplinary Nutrition Care Plan (Capra et al., 2001), or the Patient-Generated Subjective Global Assessment (Ferguson, 2003). A close contact and relationship between health care professionals and patients is highly desirable in order to assess the nutritional status and improve the quality of life of the patients.

Zinc supplementation can be valuable for patients undergoing cancer chemotherapy. The results of a pilot study suggested that after 2 weeks of chemotherapy, the intake of Zinc protected the cancer patients against taste disorders (Yamagata et al., 2003). In another study, a Zinc containing formulation known as Polaprezinc was able to improve the taste alterations in 70 percent of the patients, though an early administration has been suggested (Mizukami et al., 2016). Furthermore, the results of a randomized and controlled clinical trial on the effects of Zinc sulfate have clearly demonstrated that exogenous Zinc improved taste acuity (Ripamonti et al., 1998). Similar findings were observed in a randomized placebo-controlled trial where Zinc sulfate prevented the radiation-induced taste changes in head/neck cancer patients (Najafizade et al., 2013). Halyard et al. (2007) conducted a clinical study on Zinc sulfate vs. placebo and concluded that intake of Zinc offered protection against alterations in taste modalities. Hence, the routine use of Zinc sulfate may be suggested for patients undergoing cancer therapy.

Amifostine, an organic thiophosphate, has been shown to protect normal tissues from damage caused by radiation and chemotherapy (Kouvaris et al., 2007). Amifostine may also play a role in the protection of salivary glands and, thus leading to improvement of xerostomia. In a phase II randomized trial, amifostine has been used to assess protection against the toxic effects of carboplatin (Bohuslavizki et al., 1998). Münter et al. (2007) also came to the similar conclusion that amifostine may prevent reduced salivary gland function in patients, subjected to radiotherapy.

The efficacy of two drugs, i.e., pilocarpine and bethanechol, that increase production of saliva was tested upon saliva secretion in cancer patients with hypo-salivation following radiation therapy (Gorsky et al., 2004). Though the patients reported improvement in saliva secretion by both the agents, only bethanechol improved taste perception.

A pilot investigation was conducted to assess the role of serotonergic blockade by using ondansetron, an antagonist of serotonin type-3 receptor (Edelman et al., 1999). This study included metastatic cancer patients who were not undergoing chemotherapy or radiotherapy. Utilizing an “extensive battery of taste tests,” it was concluded that ondansetron brought about significant improvements in patients to enjoy food. A major drawback in this study was that it was not a blinded trial, and, hence, placebo effects could not be fully excluded (Edelman et al., 1999). Moreover, it was not clear whether the enhanced enjoyment from eating was due to improvements in taste. In rodent model, it seems that 5-HT acts as a paracrine inhibitory feedback signal manifested by the inhibition of ATP secretion from TRCs. Taste buds express 5-HT_1A receptor and the use of its specific agonist, 8-OH-DPAT, inhibited taste-evoked Ca²⁺ release in TRCs and also curtailed ATP release. Conversely, blocking the action of endogenous 5-HT in taste bud cells by using WAY100635, a selective 5-HT_1A receptor antagonist, increased taste-evoked ATP release (Huang et al., 2009). However, in-depth studies aiming to reproduce the afore-mentioned observations in humans are awaited.

The pathophysiology of taste alterations differs with relation to the phase of the disease and specific treatment. Unfortunately, this aspect has been ignored in most of the publications. Researchers should take into account the therapy and stages of disease while collecting the data so that more specific patient care protocols can be devised in order to minimize the effect of a change in taste on patient's quality of life. Manufacturers must take into account the taste alterations that frequently accompany cancer and should design and provide food supplements accordingly to make them more palatable. An interesting idea can be to develop “taste enhancers” which, once added to food, could enhance its palatability by acting as agonist at specific taste receptor. Such agonists can be developed for each taste modality. After identification of particular taste impairment, the enhancers for that taste can be added to make food more pleasurable for patients.

NK and FG proposed the topic, supervised and finally reviewed. BM, AH, and AK participated equally in searching the data and writing of the review.