1.3. New Vistas in Understanding Psychological Mechanisms of ANV: Negative Expectancies

In addition to demographic and clinical variables that contribute to ANV, psychological, cognitive, and social factors contribute to the conditioned response of ANV. For example,anxiety, self-absorption, and response expectancies have all been shown to be involved in ANV development (Montgomery et al., 1998; Morrow et al., 1998; Watson, 1993). Anxiety may affect the development of ANV at least in part through negative expectancies (Andykowski, 1990; Watson, McCarron, & Law, 1992), as expectancies affect development of other learning-based conditioning effects (Kirsch, 1997).

Current research on behavioral contributions relating to ANV focus on patients’ beliefs regarding symptom development. In the patient belief model of ANV development, healthcare professionals impart information about treatment side effects while listening and responding to patients’ fears and hopes. Patient and caregiver communication is crucial in the development of patients’ expectancies for treatment outcomes, termed “response expectancies,” which can influence physical, emotional and mental outcomes throughout the treatment process (Colagiuri, Roscoe, Morrow, et al., 2008). In a classic work on this topic, Kirsch (1985) suggests that anticipation or response expectancy for the occurrence of a physiological sensation (e.g., nausea) can generate corresponding subjective experiences and directly affect both physiological and psychological outcomes.

The effect of response expectancies can be considerable, and these expectancies account for variance in ANV above and beyond the emetogenicity of the chemotherapeutic agent and other known predictive factors for nausea such as gender and age (Colagiuri et al., 2008; Roscoe, Morrow, Colagiuri, et al., 2010; Roscoe, O’Neill, Jean-Pierre, et al., 2010). For example, Roscoe et al studied 194 breast cancer patients about to begin chemotherapy with a doxorubicin-based regimen and found that patients who believed that they were “very likely” to experience severe nausea were five times more likely to experience severe nausea than those who believed that they were “very unlikely” to do so (Roscoe, Bushunow, Morrow, et al., 2004).

Although expectancies may represent an acknowledgement of one’s propensity to develop nausea based on past experience (e.g., nausea during pregnancy or susceptibility to motion sickness), these expectancies are also influenced by socio-cultural factors and what patients are told to expect from the range of information they receive from clinicians, hospital staff, other patients, family, friends, and the world at large. Thus, nausea expectancies can be considered to have both internal and external sources, and it is these external sources that might provide a way to reduce chemotherapy-induced nausea. Unlike other risk factors for nausea, nausea expectancies are malleable and provide an opportunity for intervention.

A comprehensive review of the expectancy literature was undertaken by the British National Coordinating Centre for Health Technology Assessment. These authors screened 47,600 references for their relevancy and quality and chose the 93 highest-quality studies for their review. They reported that expectancies play an important role in symptom and side-effect development and further concluded that response expectancies constitute a central underlying mechanism of placebo effect (Crow, Gage, Hampson, et al., 1999). Similarly, the nocebo hypothesis, discussed by Hahn (1997) and Barsky et al. (2002), proposes that expectations of developing side effects can cause symptoms to appear. It is also likely that studies that have linked anxiety to the development of nausea are, to some extent, capturing the construct of negative expectancies (Roscoe, Jean-Pierre, Shelke, et al., 2006).

In sum, screening patients for the demographic, clinical, and psychological variables discussed above and for other cognitive/emotional vulnerabilities may help providers to identify patients likely to develop ANV. At present, there remains a need to develop a succinct and comprehensive screening process for risk factors related to ANV and to integrate such a process into cancer center treatment evaluations.

2.2. Overshadowing as a model of ANV

There is a paucity of animal models for CINV, and to date, there is no completely adequate laboratory model for ANV. Some research on ANV has been conducted using models such as body rotation as a stimulus to induce nausea and related symptoms in healthy humans (Stockhorst, Steingrueber, Enck, & Klosterhalfen, 2006). Body rotation results in a number of symptoms summed as motion sickness and of these, the most significant symptoms in humans are nausea and vomiting, pallor, and cold sweating (Yates, Miller, & Lucot, 1998). In this model, the afferent signals from the vestibular system constitute the unconditioned stimulus (Stockhorst et al., 2006). Studies using body rotation model suggest that techniques such as overshadowing procedure and latent inhibition could be helpful in the prevention and reduction of ANV (Hall & Symonds, 2006; Stockhorst, Enck, & Klosterhalfen, 2007b; Symonds & Hall, 1999).

Overshadowing, as originally described by Pavlov (Pavlov, 1960), is a technique used for reducing the association between the putative CS and the US. In a usual overshadowing experiment, the subject is conditioned in an adverse experimental setting to respond to a strong stimulus, and then the stimulus is withdrawn at the next exposure to the adverse experience (Stockhorst, Enck, & Klosterhalfen, 2007c). In a small study, 16 cancer patients were assigned to either the group with overshadowing (OV+) or the group without overshadowing (OV−). At the start of all infusions of two consecutive chemotherapy cycles A and B (acquisition), the OV+ group were made to drink a distasteful saline beverage, thereby overshadowing the CS, while the OV− group were made to drink water. All patients received water in cycle C (test). In this study, no patient from the OV+ group showed ANV in cycle C (test), while two patients from OV− group developed ANV (Stockhorst et al., 1998). Overshadowing is similar to latent inhibition which consists of pre-exposing the event to be used as the CS prior to the first administration of the US, thereby reducing the amount of classical conditioning that subsequently occurs (Lubow & Moore, 1959).

Besides the body rotation model, a conditioned gaping response in rats has also been used as a model for ANV. However, these animal studies examining the gaping response in rats with respect to overshadowing (Hall & Symonds, 2006; Sansa, Artigas, & Prados, 2007; Symonds & Hall, 1999), systemic treatment with lipopolysaccharide (Chan, Cross-Mellor, Kavaliers, & Ossenkopp, 2009), tetrahydrocannabinol (Parker & Kemp, 2001), and manipulation of the endocannabinoid system (Rock, Limebeer, Mechoulam, Piomelli, & Parker, 2008) have shown inconsistent results. Another animal model for ANV has shown cannabinoids to be effective in reducing ANV (Limebeer et al., 2008; Parker & Kemp, 2001; Parker, Kwiatkowska, & Mechoulam, 2006). In addition, some conditioning procedures in other animal model systems have been used to alleviate nausea and vomiting (Davey & Biederman, 1998; Lett, 1983). However, the incidence of ANV has remained the same (Morrow et al., 1998), despite the improvements in overall CINV with the introduction of 5-HT3 receptor antagonist medications. Overall, the best method to reduce or avoid development of ANV is to avoid both vomiting and nausea from the first exposure to chemotherapy.

3.2. Other pharmacological agents

Another class of agents that is of potential interest in the treatment of ANV are cannabinoids. Two synthetic, oral formulations, dronabinol (Marinol) and nabilone (Cesamet), are approved by the US Food and Drug Administration (FDA) for use in CINV that is refractory to conventional antiemetic therapy (Slatkin 2007). The side-effect profile of cannabinoids has limited their adoption into clinical practice especially with approval of newer effective drugs. However, there have been intriguing animal studies that suggest cannabinoids may have a role in the treatment of ANV. Parker and colleagues (Parker 2006, Parker 2011) utilized emetic reactions of the Suncus murinus (musk shrew) in their experiments. Following pairings of a distinctive contextual cue with the emetic effects of lithium chloride (toxin), the unique context acquired the potential to elicit conditioned retching (akin to ANV) in the absence of the toxin. The expression of this conditioned retching was completely suppressed by pretreatment with each of the principal cannabinoids found in marijuana. These results support the anecdotal use of cannabinoids for treatment of anticipatory symptoms although further research is needed to evaluate its potential effectiveness in the clinical setting.

The use of recommended pharmacotherapy to prevent and treat chemotherapy-induced acute and delayed emesis is critical in reducing the conditioned development of anticipatory symptoms. Advances in 5-HT3 antagonists and the development of NK-1 antagonists have dramatically improved control of acute and delayed CINV which in turn can reduce the incidence of ANV. Current guidelines (Roila 2010) recommend that for highly emetogenic chemotherapy agents, a three-drug regimen of a serotonin (5-HT3) receptor antagonist (palonosetron), dexamethasone, and NK-1 receptor antagonist (aprepitant or fosaprepitant) should be used before the beginning of chemotherapy to prevent acute CINV. Palonosetron plus dexamethasome is recommended for acute CINV from moderately emetogenic agents. For low risk chemotherapy regimens, dexamethasone or a 5-HT3 receptor antagonist or a dopamine receptor antagonist is recommended. For delayed CINV, dexamethasone and aprepitant, aprepitant alone, or dexamethasone alone can be used. To prevent ANV, the Multinational Association of Supportive Care in Cancer (MASCC) committee appropriately recommends the best possible control of acute and delayed CINV in conjunction with behavioral therapies.

4.3. Biofeedback, Imagery, and Relaxation

Other behavioral treatments for ANV center on relaxation and management of expectancies, the two principles targeted by SD and hypnosis. An early trial showed the efficacy of biofeedback in reducing severity of ANV through producing a relaxation response in patients (Burish, Shartner, & Lyles, 1981). Guided imagery, similarly, has been used to produce a relaxation response or to manage negative expectancies about treatment (Lyles, Burisch, Krozely, 1982). Guided imagery involves descriptive statements by a therapist or other provider to assist a patient in constructing a mental picture that would promote relaxation or a reduction in anxiety, such as a beach scene or an image of cancer cells shrinking. Progressive muscle relaxation (PMR) is often used as an incompatible response in SD interventions. PMR involves systematically tensing and relaxing muscle groups throughout the body until a full-body relaxation response is achieved (Arakawa, 1997; Yoo, Ahn, Kim, Kim, & Han, 2005). Used alone, PMR appears most effective in reducing severity of nausea, vomiting, and other side effects that develop after administration of chemotherapy (Molassiotis, Yung, & Yam, 2002). When combined with systematic desensitization or guided imagery, it has proven to be efficacious in reducing ANV (Yoo, Ahn, Kim, et al., 2005). Other forms of learned relaxation (Morrow & Morrell, 1982; Redd, Montgomery, & DuHamel, 2001) have been shown to be effective; a randomized controlled study of 48 patients undergoing chemotherapy found a relaxation intervention to be beneficial in controlling ANV in those patients who did not have an “information-gathering coping style”. (Lerman et al., 1990) Finally, shading into the realm of complementary treatment approaches, a study of a yoga intervention for CINV reported a significantly reduction in anticipatory nausea and vomiting in a sample of 62 early breast cancer patients (Raghavendra et al., 2007).

5.3. The Effect of Optimism on ANV

A positive attitude is often considered to be a beneficial attribute for patients who are confronted with a serious health concern. Popular authors, such as Bernie Siegel and the late Norman Cousins, have expounded on the virtues of positive thinking and suggest that it is related to better health outcomes. Scheier & Carver (1987) theorize that patient optimism has a significant positive effect both directly and indirectly on health and well-being. These researchers conceptualize optimism as a stable personality trait and use the term dispositional optimism to describe individuals with more global positive expectations. Optimism has been associated with increased motivation and persistence at tasks as well as an array of positive health-related outcomes, including lower postpartum depression, higher QOL 8 months after coronary artery bypass surgery, fewer physical symptoms of poor health and less emotional distress during and after radiation therapy. Pessimism, by contrast, is associated with depression and lowered disease resistance in breast cancer patients, poor health in middle and late adulthood, and more rapidly deteriorating health and increased morbidity in AIDS patients. Interestingly, a general tendency to be optimistic does not appear to predict degree of nausea from chemotherapy, even though specific nausea expectancies are strong predictors of this side effect (Roscoe et al., 2006). Treatment approaches that could tap into the construct of optimism to address negative expectancies, however, could prove efficacious in controlling ANV by enabling patients to view their chemotherapy side effects with optimism.

In conclusion, current anti-emetics are associated with bothersome side-effects in patients with cancer (Bergkvist & Wengstrom, 2006) and may not address the underlying mechanisms of ANV. ANV appears to be related to psychological, learning-related processes, and so may best be addressed by behavioral or psychological treatments. However, the factors of greater tolerability and fewer side effects of CAM techniques make them attractive alternatives to the less effective pharmacologic options available for the treatment of ANV. Behavioral and CAM techniques could also allow the patients to apply interventions to themselves (Rie Konno, 2010), allowing for greater patient self-efficacy and treatment control. Additional research is needed to confirm initial indication of the beneficial effects of some of these CAM treatments and to identify new, palatable approaches for the management of ANV.