Nausea as a sentinel symptom for cytotoxic chemotherapy effects on the gut-brain axis among women receiving treatment for recurrent ovarian cancer: An exploratory analysis

Introduction

The concept of a gut-brain axis (GBA) originated with research showing the action of gastrointestinal endocrine factors and extrinsic nerve pathways, primarily the vagus, on the brain in the control of digestion, metabolism and food intake [1, 2]. The concept has recently been extended to include gut microbiome modulation of GBA signaling pathways that could impact emotion, pain, and stress, as well as nutritional balance [3]. The GBA is increasingly recognized to include both afferent and efferent pathways, which provide rich opportunities for bi-directional communication and control processes [4]. A consistent and comprehensive description of symptoms linked to GBA activation does not yet exist, but there is considerable evidence for effects on appetite, nausea and vomiting, weight changes, visceral pain, and bowel symptoms (e.g., diarrhea, constipation, and cramping) [5]. Accumulating evidence also implicates additional symptoms such as depression and anxiety [6–8], cognitive/memory problems [9, 10], sleep rhythms [11], and fatigue [12, 13]. Activation of the gut-brain axis can occur anywhere along the pathway, from direct stimulation of the gut to systemic effects on the central nervous system components of the GBA [4, 14].

Cytotoxic chemotherapy continues to be the mainstay of adjuvant treatment for many cancers [15–17], but has long been recognized to have profound gastrointestinal side effects, with virtually universal nausea and vomiting for some agents if antiemetic therapies are not used [18–21]. Even with standard-of-care prophylactic anti-emetic treatments, nausea continues to be a common experience for women receiving cytotoxic chemotherapy for recurrent ovarian cancer, with acute nausea (within 24 hours) affecting about a third of patients and delayed nausea (after 24 hours) affecting nearly two-thirds [22, 23]. To date, the biological pathways that characterize such complex nausea in relation to additional symptoms experienced by these patients have not been well characterized. As one of the most prevalent side effects experienced by patients receiving cytotoxic chemotherapy, nausea in these patients may serve as a sentinel symptom for a broader spectrum of symptoms engendered by alterations in GBA functioning.

As a first exploratory test of the possible role of the GBA in symptom experiences among patients receiving chemotherapy for recurrent ovarian cancer, we compared women reporting nausea to women not reporting nausea with regard to the severity of other commonly reported symptoms. If nausea is a marker of GBA activation, the severity of other GBA associated symptoms would be expected to be higher among women experiencing nausea compared to those who were not. To date, no other studies have systematically evaluated these relationships.

Methods

Results

Discussion

Consistent with the study hypothesis, the presence of nausea and nausea severity during cytotoxic chemotherapy for recurrent ovarian cancer were significantly related to the experience of multiple other common cancer and treatment-related symptoms, which may be related to GBA activation. Also consistent with the exploratory hypothesis, the presence of nausea and nausea severity were not significantly related to the severity of hair loss, numbness/tingling, weight gain, or sexuality concerns.

Among the set of 15 symptoms that were significantly related to nausea in both sets of analyses, several have strong evidence linking them to the GBA (e.g., abdominal bloating, bowel disturbances, vomiting) [35, 36], and others have emerging but weaker links to the GBA (e.g., sleep problems, fatigue, memory problems, and mood swings [37]). Several of the symptoms with less clear associations with GBA may be attributable to effects of chemotherapy on the gut microbiome. The microbiome is increasingly recognized to have a broad range of effects on the activities of the central nervous system, including emotional and cognitive effects that have led to a call to expand the concept of GBA to highlight a brain-gut enteric micro-biota axis [3].

Particularly in need of additional research is the role of nausea as a specific precipitating GBA-related symptom with downstream effects on other symptoms and side effects of treatment. One possible clinically significant pathway for example would link nausea as a prodrome to reduced appetite, which in turn may contribute to reductions in food consumption and weight loss, which increases risk of premature mortality [38–40]. Nausea may also be related to reductions in fluid consumption, which were not assessed in the present study. Dehydration secondary to nausea and vomiting could contribute to renal toxicities in this patient population, which can be life threatening. Rising creatinine concentrations indicating renal toxicity are one of the most common reasons for dose delays among women with ovarian cancer [41–43].

Another issue that the results of the present study highlights for additional research attention is the possible impact of chemotherapy on the functioning of the vagus nerve as a contributing factor in GBA-related symptoms. In what are now largely separate silos of research, the importance of vagal function has been implicated in studies of: microbiome influences on symptom experience [44]; appetite [45]; nausea and vomiting [46]; fatigue [47]; drowsiness and sleep problems [48]; and peripheral pain sensation [49]. It is also important to note that nausea does not require a precipitating impact on the gut; nausea can be provoked by effects in the central nervous system components of the gut-brain axis. For example, research supports the role of the area postrema in chemotherapy-induced nausea and vomiting [50]. Chemotherapy (and/or residual cancer) may produce circulating factors that act on central components of the gut-brain axis to produce nausea [51, 52].

Another interesting finding is that 70% of our sample received chemotherapy regimens of minimal to low emetogenic potential, yet 56.4% of women were still experiencing nausea an average of 16 days post-treatment. Here it is important to emphasize that emetogenic potential is based solely on incidence of vomiting – not on incidence of nausea [53, 54]. Multiple studies have documented that although current anti-emetic regimens have reduced the incidence of vomiting, the incidence of nausea, especially delayed nausea, has remained unacceptably high, and perhaps even increased [55, 56]. Indeed, the incidence of delayed nausea, even in the context of guideline consistent anti-emetic treatment remains in the range of 40–70% [57–61]. This rate is consistently underestimated by clinicians and may contribute to sub-optimal management for patients receiving low emetogenic regimens [57]. These findings suggest a chronic type of chemotherapy-induced nausea that requires additional research and increased awareness in clinical practice. One rodent model [62] suggests that chemotherapy-related nausea may persist in some individuals for up to 6 months post chemotherapy. Another explanation for the high rates of nausea in our study could be that the recurrent cancer itself contributes to the nausea. In women with recurrent ovarian cancer, chemotherapy-induced nausea may be further complicated by the presence of peritoneal or liver involvement that may cause additional nausea via direct mechanical effects on the gut, or via stimulation of biological responses (e.g., cytokines, prostaglandins) that may systemically activate the GBA [63]. This study cannot distinguish between these possible mechanistic pathways, however, it is important to note that the presence of cancer per se does not explain the variability in nausea in our sample; 42% of women with active disease were not experiencing nausea.

It is important to emphasize that our goal in this exploratory study was to provide initial evidence of GBA activation during chemotherapy for recurrent ovarian cancer and to highlight the potential importance of additional research specifically designed to better characterize the nature of these relationships and the mechanisms involved. As a secondary data analysis, the present study had several weaknesses that should be noted. Most notably, the cross-sectional nature of the original mailed survey study [23] substantially limits causal interpretation of the significant associations between nausea and other symptoms reported here. It should also be noted that the assessments included in the original survey study were not specifically designed to explore GBA symptoms, or to investigate downstream effects of such symptoms, and were intentionally kept brief to minimize respondent burden. The self-report survey did not allow determination of whether nausea was related to cytotoxic chemotherapy administration, or to the cancer itself. The women in this sample were younger (55 years) than the typical age of diagnosis (63 years old).

Younger women in this sample and in other studies [64] are at higher risk for nausea. Future studies should evaluate whether age has a general influence on GBA activation among patients with cancer. Finally, access to medical record data was not requested from respondents to corroborate their self-reports of treatments received. Findings should be interpreted cautiously since patients’ understanding of the details of their disease and treatment, while generally accurate, may be imperfect [65–67].

If future research confirms the results of this initial study and the hypothesized impact of cytotoxic chemotherapy on the GBA, it would have important implications for the development of novel supportive therapies to increase the tolerability and effectiveness of cancer treatment for this patient population. By addressing chemotherapy-induced effects on the GBA, it may be possible to eliminate an entire cluster of symptoms, which would not only improve recurrent ovarian cancer patients’ quality of life, but also might have a significant impact on morbidity and mortality.