http://orcid.org/0000-0002-8917-2495
Figures
Abstract
Background
Despite the experimental evidence that certain dietary compounds lower the risk of radiation-induced damage to the intestine, clinical data are missing and dietary advice to irradiated patients is not evidence-based.
Materials and methods
We have previously identified 28 intestinal health-related symptoms among 623 gynaecological-cancer survivors (three to fifteen years after radiotherapy) and 344 matched population-based controls. The 28 symptoms were grouped into five radiation-induced survivorship syndromes: defecation-urgency syndrome, fecal-leakage syndrome, excessive mucus discharge, excessive gas discharge and blood discharge. The grouping was based on factor scores produced by Exploratory Factor Analysis in combination with the Variable Cutoff Method. Frequency of food intake was measured by a questionnaire. We evaluated the relationship between dietary intake and the intensity of the five syndromes.
Results
With the exception of excessive mucus discharge, the intensity of all syndromes declined with increasing intake of citrus fruits. The intensity of defecation-urgency and fecal-leakage syndrome declined with combined intake of vegetables and citrus fruits. The intensity of excessive mucus discharge was increased with increasing intake of gluten.
Conclusion
In this observational study, we found an association between a high intake of citrus fruits and vegetables and a lower intensity of the studied radiation-induced cancer survivorship syndromes. Our data suggest it may be worthwhile to continue to search for a role of the diet before, during and after radiotherapy to help the cancer survivor restore her or his intestinal health after irradiation.
Citation: Hedelin M, Skokic V, Wilderäng U, Ahlin R, Bull C, Sjöberg F, et al. (2019) Intake of citrus fruits and vegetables and the intensity of defecation urgency syndrome among gynecological cancer survivors. PLoS ONE 14(1): e0208115. https://doi.org/10.1371/journal.pone.0208115
Editor: John Green, University Hospital Llandough, UNITED KINGDOM
Received: April 10, 2018; Accepted: November 8, 2018; Published: January 2, 2019
Copyright: © 2019 Hedelin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This work was supported by the Swedish Cancer Society to GS, the King Gustav V Jubilee Clinic Cancer Foundation in Göteborg, Sweden to GS and the Swedish state under the ALF agreement in Göteborg. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Gunnar Steineck is PI for a study of probiotics sponsored by Probi AB. This does not alter our adherence to PLOS ONE policies on sharing data and materials. There exist no other competing interests.
Introduction
Radiotherapy is a common treatment for cancer in the pelvis and lower abdomen. Between 70% and 90% of patients receiving radiotherapy treatment have been reported to develop acute gastrointestinal symptoms [1, 2], and for many patients the treatment will induce lifelong survivorship diseases [3, 4]. An underlying cause is fibrosis, the end result of radiation-induced degeneration of the crypts in the gut wall, damage to the endothelium and inflammatory processes [5]. Possibly, the colonic wall attempts to repair itself by crypt fission [6], but insufficiently to prevent fibrosis.
Even though it is known that radiation induces intestinal fibrosis and that survivors suffer, few preventive or curative interventions have been successfully used in clinical practice [7–9]. A diet high in fruit and vegetables maintains a healthy function of the normal colon and the protective mucus layers are reduced by a diet without fiber [10]. There is no body of scientific evidence available today that shows that patients should avoid dietary fiber [7], but radiated patients are nevertheless often given the advice to do just that, avoid dietary fiber intake.
Fruits and vegetables contain non-starch polysaccharides such as pectin. Pectin is a highly complex polysaccharide fiber with soluble and viscous properties (so-called soluble fiber) and has been found to have chemo-preventive and radio-protective properties [11]. For example, pectin has been suggested to increase intestinal crypt stem cell survival following radiation injury [12] as well as reducing radiation-induced intestinal fibrosis [13]. Pectin also appears to prevent loss of microbial diversity caused by radiation [13].
Soluble fiber is fermented by colonic bacteria into short-chain fatty acids including butyrate [14] that have an anti-diarrheal, anti-inflammatory, anti-oxidative, anti-carcinogenic and anti-fibrogenic effects [13, 15, 16]. Butyrate acts as an energy source for the colonic epithelial cells, providing up to 70% of their energy requirements [17]. A diet low in fiber leads to starvation of the fiber-degrading gut microbiota. This promotes the expansion of bacteria specialized in degrading the host-secreted mucus glycoproteins. This in turn leads to erosion of the protective mucus barrier lining the gut wall [10]. Furthermore, our own preclinical studies indicate that a diet combining non-fermentable fiber with a larger volume of fermentable fiber is protective against radiation-induced gut wall damage (unpublished data). This protective effect probably results because the soluble fiber is transported with the insoluble fiber distally into the large intestine and rectum, where it is fermented and free fatty acids are produced [18].
We have previously documented patient-reported long-term symptoms one by one, using questionnaires specially developed for this approach to produce what we call atomized clinimetric symptom documentation. Self-reported information on long-term symptoms was collected from a large cohort of cancer survivors treated with radiation to the pelvic area and from matched control subjects [19].
Using a new approach, the documentation of intestinal related symptoms was analyzed using factor analysis [3] making it possible for us to identify five survivorship syndromes: defecation-urgency syndrome, leakage syndrome, excessive mucus discharge, excessive gas discharge, and blood discharge. These syndromes thus appear to underlie the nearly 30 intestinal-related symptoms reported by cancer survivors with a history of radiation-treatment focused on the pelvic area. In the population-based cohort study presented here, we examined the relationship between dietary intake of fruit and vegetables and intensity of the five radiation-induced survivorship syndromes three to fifteen years after treatment with radiotherapy.
Material and methods
Study population
The study design and assessment of patient-reported long-term symptoms have been described in detail elsewhere [19]. In brief; between 1991 and 2003, 1800 women were treated with external pelvic radiotherapy for a gynecological malignancy at two clinics in Sweden. The patients had received long-course external-beam radiation therapy, sometimes combined with brachytherapy, either as radical treatment or in addition to surgery [20]. This corresponded by and large to all the relevant gynecological cancer patients in two geographical regions. We excluded survivors born 1927 or earlier, as well as those who were illiterate in Swedish.
Based on the clinical epidemiological tradition for atomized patient-reported outcomes, we constructed a study-specific questionnaire with a section focusing on intestinal health. The atomized questions were based on semi-structured interviews and had wordings as close as possible to those used by the survivors in the transcribed interviews.
An introductory letter was sent to, 823 gynecological-cancer survivors and 478 matched (age, residence) population-based controls, explaining the objectives of our study and emphasizing that participation in the study was voluntary (detailed information is found in Supporting information in reference [3]). One week later, an interviewer phoned each informant. Those giving informed oral consent to participate received a postal questionnaire along with a letter and additional information. To maintain anonymity, each questionnaire contained a number for identification. Six hundred fifty (79%) survivors and 344 (72%) controls returned a questionnaire. An error in the matching procedure led to a younger control population (mean age 58.0) compared to the cancer survivors (mean age 64.4) which was adjusted for in the analyses. No children were included. The study was approved by the Regional Ethical Review Board (2005/1424-31/4), Stockholm, Sweden. The Ethical Review Board received all information and material before they gave their approval.
Radiation-induced survivorship symptoms
We utilized 28 different atomized symptoms related to intestinal health. For example, concerning toilet dependency we asked: “Have you had sudden defecation urgency that requires immediate toilet visits during the last six months?” Answers were given on a person-incidence scale (“No”, “Yes, occasionally”, “Yes, at least once a month”, “Yes, at least once a week”, “Yes, at least three times a week”, and “Yes, at least once a day”). Survey questions are presented in S1 File, also described in detail elsewhere [3, 19]. Through factor scores, five radiation-induced survivorship syndromes were identified (see statistical part below) and are described in detail elsewhere [3].
Dietary intake
The participants were asked to evaluate how often during the past six months they had eaten citrus fruits, beans and lentils, cabbage and broccoli, onion and garlic, grated vegetables, foods with gluten (the four cereals), chocolate, dairy products, spicy food or fatty food. The options given were “daily”, “weekly”, “monthly”, “less than monthly” or “never”. (Note: In Sweden, the term grated vegetables refers to raw carrot or raw beetroot and cabbage).
Total intake of vegetables and citrus fruits was grouped into five categories (at least daily, at least twice a week, at least weekly, 1–3 times a month, less than monthly or never).
The questionnaire included one question about changes in dietary habits; “Have you changed your diet after the completion of radiation therapy?” Those participants who chose the alternative “Not relevant, I have not changed the diet after completion of radiotherapy” were considered as not having changed their dietary behavior after radiotherapy; all other answers were considered as indicating that dietary behavior had changed after radiotherapy.
Statistical analysis
The factor scores on which the results presented here are based were produced using Exploratory Factor Analysis (EFA) in combination with a method called the Variable Cutoff Method (VCM). The former is an established dimension reduction method, and the latter aims at identifying spurious so-called factor loadings. Both methods are described in full detail elsewhere [3].
EFA models the covariance matrix of vector of stochastic variables X = (X1,…,Xn) in terms of a matrix L of factor loadings and a vector F = (F1,…,Fk), k < n, of unobserved, uncorrelated factors. The modelling assumption is: Xi—μi = li1F1 + … + likFk +εi, where the εi are unobserved error terms with mean zero and finite variances. Thus, in order to fit an EFA model to data, the lij and k have to be estimated. Performing EFA using standardized data is thus equivalent to performing EFA with correlation substituted for covariance. Before any calculations were performed, the data were subjected to two preliminary filtrations: individuals were excluded if they had a bowel stoma (n = 20 survivors, no controls), and if more than 30 percent of the data was missing for the 28 variables under study (n = 7 survivors, no controls). Patients who reported that they had an irritable bowel syndrome (n = 23) were not excluded from the analyses. In the final analyses 623 survivors and 344 controls were included.
In this particular application, the estimated Spearman correlation matrix of the data was used as input to the EFA model, since the data are ordinal. Pairwise deletion of missing values was applied when estimating the correlation matrix. The appropriate number of factors, k, was assessed using both parallel analysis [21] and a bootstrap version of Kaiser’s rule [22]. The two methods agreed on the number of factors. The EFA model was then fit using Maximum Likelihood Estimation [23].
The estimated matrix of factor loadings, L, was then subjected to the VCM. The VCM uses the parametric bootstrap to find factor specific cutoff values on factor loadings, (c1,…,ck), and sets factor loadings such that |lij|<cj to 0, resulting in a sparse or reduced version L’ of L. Before the calculation of the factor scores the data were imputed using a simple mode value imputation per variable, giving data matrix X’. The factor scores were then calculated using L’ as S = X’L’.
To evaluate the relationship between frequency of dietary intake and syndrome intensities (factor scores) of the five radiation-induced survivorship syndromes defecation-urgency syndrome, fecal-leakage syndrome, excessive gas discharge, excessive mucus discharge and blood discharge, we used Spearmans correlation coefficient. Further, the results were presented graphically as mean syndrome intensities plus-minus the standard error of the mean within certain diet intake frequency levels.
Results
Demographic data, clinical characteristics, and dietary behaviour of the study participants are presented in Table 1. Most of the cancer survivors were older than 60 years of age, and most had been treated for endometrial (58%) or cervical (23%) cancer. Fifty-two percent of the survivors were treated with external beam radiation plus Brachy therapy. Forty-three percent of the women had eaten citrus fruits daily and 3% were never-eaters. Fewer ate beans and lentils daily, 2% among survivors and 6% among control subjects.
We found a statistically significant positive correlation between frequency of citrus fruit intake and intensity of the radiation-induced survivorship syndromes, except for excessive mucus discharge (Tables 2 and 3). For intake of the different vegetable groups, there were statistically significant correlations between frequency of intake and intensity of defecation-urgency syndrome and fecal-leakage syndrome (Tables 2 and 3). Frequency of gluten intake was statistically significantly negative correlated with excessive mucus discharge (Tables 2 and 3). We found no statistically significant correlation between the frequency of intake regarding chocolate, dairy products, spicy food or fatty food, with any of the radiation-induced survivorship syndromes (data not shown).
In Figs 1 to 3, the relative measure of syndrome intensity (factor score; Y-axis), is plotted against increasing frequency of intake of the various food items; Fig 1) the intensity declined with increasing intake of citrus fruits, Fig 2) the intensity of defecation-urgency and fecal-leakage syndrome declined with increasing intake of vegetables and citrus fruits combined, Fig 3) for intake of gluten the trend was reversed, and the intensity of excessive mucus discharge was increased with increasing intake of gluten.
The figure depicts the relationships between frequency of citrus intake (x-axis) and intensities (factor scores) of the five radiation induced syndromes (y-axis: defecation-urgency syndrome, fecal-leakage syndrome, excessive gas discharge, excessive mucus discharge and blood discharge) among long-term gynecological cancer survivors and population controls. Red squares and green triangles denote the mean syndrome intensities within certain citrus intake frequency levels. Vertical black lines stretch between plus-minus the standard error of the mean. Asterisks correspond to significance levels of the Spearman correlations between syndrome intensities and frequency of citrus intake according to: p ≤ 0.001 - ***, 0.001 < p ≤ 0.01 - **. 0.01 < p ≤ 0.05 - *. Estimates of the Spearman correlations are found in Tables 2 and 3.
The figure depicts the relationships between frequency of combined intake of beans and lentils, cabbage and broccoli, onion and garlic, grated vegetables together with citrus fruits (x-axis) and intensities (factor scores) of the two radiation-induced survivorship syndromes (y-axis) defecation-urgency syndrome and fecal-leakage syndrome among long-term gynecological cancer survivors and population controls. Red squares and green triangles denote the mean syndrome intensities within certain vegetable and citrus intake frequency levels. Vertical black lines stretch between plus-minus the standard error of the mean. Asterisks correspond to significance levels of the Spearman correlations between syndrome intensities and frequency of vegetable and citrus intake according to: p ≤ 0.001 - ***, 0.001 < p ≤ 0.01 - **. 0.01 < p ≤ 0.05 - *. Estimates of the Spearman correlations are found in Tables 2 and 3.
The figure depicts the relationship between frequency of gluten intake (x-axis) and the intensity (factor scores) of the radiation induced syndrome (y-axis) excessive mucus discharge among long-term gynecological cancer survivors and population controls. Red squares/green triangles denote the mean syndrome intensity within certain gluten intake frequency levels. Vertical black lines stretch between plus-minus the standard error of the mean. Asterisks correspond to significance levels of the Spearman correlations between syndrome intensity and frequency of gluten intake according to: p ≤ 0.001 - ***, 0.001 < p ≤ 0.01 - **. 0.01 < p ≤ 0.05 - *. Estimates of the Spearman correlations are found in Tables 2 and 3.
In Fig 4, the relationship between radiation-induced survivorship defecation-urgency syndrome and citrus fruit intake is stratified by those who have changed and those who have not changed their dietary behavior after radiotherapy. The intensity of defecation-urgency syndrome declined with increasing intake of citrus fruits in both groups, although the absolute intensity score was higher among those who had changed their dietary behavior after treatment.
The figure depicts the relationship between frequency of citrus fruit intake (x-axis) and the intensity (factor scores) the defecation-urgency syndrome (y-axis) among long-term gynecological cancer survivors stratified according to having or not having made dietary changes after receiving radiotherapy (see the Methods section for detailed information). Red squares and green triangles denote the mean intensities of the syndromes within certain citrus intake frequency levels. Vertical black lines stretch between plus-minus the standard error of the mean. Asterisks correspond to significance levels of the Spearman correlations between syndrome intensity and frequency of citrus intake according to: p ≤ 0.001 - ***, 0.001 < p ≤ 0.01 - **. 0.01 < p ≤ 0.05 - *.
Stratified analysis was performed to identify, if smoking, age or time since treatment confounded the relationship between the radiation-induced defecation-urgency or fecal-leakage syndrome and citrus fruit intake. The analysis was stratified by those who were non-smokers and those who were former or current smokers, or time since treatment less than median (6.3 year) and above, or age less than median (65.1) and above. The results did not differ substantially between the strata for any of the analyses (data not shown).
Discussion
This is the first study evaluating the relationship between dietary intake and five different radiation-induced longstanding syndromes. Based on information from 623 women, three to fifteen years after radiotherapy, we found that frequent intake of citrus fruits and vegetables decreased the intensity of defecation-urgency syndrome, fecal-leakage syndrome, excessive gas discharge and blood discharge. The strongest relationship was seen for the defecation-urgency syndrome. In contrast, the intensity of excessive mucus discharge was increased with increasing intake of gluten.
Our results are supported by experimental data that show that compounds in fruits and vegetables such as soluble fiber protect against radiation-induced damage of the gut. However, the data from clinical studies are sparse [8, 9, 24–28]. Non-starch complex polysaccharides, found in high amounts in plant cell walls, seem to have several different functions that improve the colonic function (listed in Table 4).
In humans, dietary soluble fiber, like pectin, are not digested by endogenous enzymes in the small intestine but are degraded by the colon microbiota. The soluble fiber has a gelling effect in the digestive tract, slowing down digestion. By acting as prebiotics, the polysaccharides are fermented into short chain fatty acids in the large intestine [14]. A diet rich in non-starch polysaccharides is also believed to increase the transit rate through the colon, which in turn lowers the colonic pH. A lower pH 5.5 ameliorates production of short–chain fatty acids, in particular butyrate [30]. Short–chain fatty acids, especially butyrate, provide the colonic epithelial cells with up to 70% of their energy requirements [17]. Butyrate is suggested to have a function in the regulation of the gut wall barrier, an anti-inflammatory, anti-diarrheal and anti-fibrinogenic function, as well as preventing mucosal atrophy [15, 16, 36]. Pectin is suggested to improve colonic function [31] and has anti-inflammatory properties [32, 35]. In an experimental study on mice, pectin increased intestinal crypt stem cell survival after radiation [12]. An oral intake of dietary fiber increased levels of hydroxyproline, a component in collagen, in rats given radiotherapy [29]. A modified form of pectin has been found to increase the 137Cs clearance, a radio-isotope produced during uranium fission. When eaten for several days, toxic elements seem to be chelated by the pectin and then eliminated in the urine [11]. In a recent mouse experiment, dietary apple pectin ameliorated the radiation-induced increase in submucosa thickness and protected the terminal ileum against radiation-induced fibrosis [13]. In addition, the irradiated mice had a markedly decreased diversity of intestinal microbiota, while the pectin-treated mice had a smaller decrease in diversity. The pectin-treated mice also had markedly higher concentrations of short–chain fatty acids in their feces. However, extrapolating knowledge gained from animal models should be done with caution, since species differences, and even strain differences affect the response to radiation. In addition, the irradiation procedure in animal models is, for practical reasons, a simplification of pelvic radiotherapy. Nevertheless, the hall-marks of radiation-induced injury to the human intestines such as inflammation of the intestinal wall, an increase in serum cytokines, stem cell loss, crypt degeneration, angiogenesis and fibrosis are readily reproduced in animal models. By focusing on these similarities, animal models have provided the foundation of today’s understanding of how we react to radiation exposure, and how various external factors can modify this response [37].
In contrast to the beneficial effects of citrus fruits and vegetables on the defecation-urgency syndrome, increased intake of gluten promoted excessive mucus discharge. This may be a result of radiation-induced damage making the intestine more vulnerable, triggering a non-celiac gluten sensitivity. It can also be due to an interaction from, or confounding effect of, something else in the diet, correlated to gluten intake [38, 39]. The sensitivity to gluten may also be a result of an underlying genetic tendency towards coeliac disease, triggered by the radiation-induced immunosuppression [40]. However, we did not find any effect of gluten intake on defecation-urgency syndrome, a syndrome for which the symptoms are similar to those of coeliac disease. This suggests that the gluten sensitivity presented here is not related to coeliac disease.
The normal function of the gut is highly dependent on the type, diversity and the number of microbiota, which in turn depends partly on dietary habits [41, 42]. The concentration of soluble or insoluble fiber in the diet affects the composition and function of the gut microbiota [18, 43]. The colonic mucus barrier serves as a primary defense against pathogens and toxic compounds [10]. A diet low in fiber leads to starvation of the gut microbiota, which results in starting use of the host-secreted mucus glycoproteins as a nutrient source. This leads in turn to erosion of the colonic mucus barrier [10]. The erosion may facilitate migration of pro-inflammatory species such as Bacteroides fragilis through the gut wall [44].
Previously, we have identify that the symptoms exist in the background population [19], caused by other factors than radiation. At a certain intensity of symptoms, we can suspect chronic pathophysiological processes generated by the radiation therapy. Radiation causes degeneration of crypts in the gut wall and over time the gut wall may turn into rigid, dysfunctional fibrotic tissue. Crypts probably degenerate when the last stem cell has been depleted or made non-functional by lack of supportive cells [45, 46]. The damaged gut wall may regain normal function by recruiting non-dividing radio-resistant “reserve” stem cells that divide and repopulate the radiosensitive stem-cell population and new crypts can be formed by crypt fission [46, 47]. Probably, inflammation in the gut wall hinders intestinal restoration and results in fibrosis, and the inflammation may occur when the inner mucus layer fails to protect the gut wall from being exposed to pro-inflammatory bacteria [10]. Moreover, bacterial invasion may be enhanced when the epithelial tight junctions are injured by radiation [45, 48–50]. Such a mechanism could turn into a vicious cycle, self-propelling further inflammation, ischemia, gut wall leakage and ultimately degeneration of crypts and fibrosis of the gut wall [5, 45, 51]. Dietary components or probiotics may hinder or dampen this inflammation by two mechanisms: one by keeping the two mucus layers intact, the other by hindering a gut-wall starvation that makes the wall vulnerable to inflammatory agents [52].
Of the few randomized clinical studies investigating possible effects of soluble fiber on radiation-induced symptoms, both low and high fiber intake have been studied. In a recent study, 166 patients undergoing pelvic radiotherapy were randomly assigned to receive low-fiber, habitual-fiber or high-fiber dietary advice and treatment. The high-fiber diet resulted in reduced gastrointestinal toxicity both acutely and at 1 y compared with habitual-fiber intake [28]. The effect of fiber supplementation with Metamucil (Psyllium) on radiation-induced diarrhea was determined in a randomized trial of 60 patients. Metamucil was shown to significantly decrease the incidence and severity of diarrhea [8]. In another randomized trial [26], radiotherapy-induced gastrointestinal symptoms were compared between prostate cancer patients who were either advised to reduce insoluble dietary fiber and lactose, or to continue to eat their normal diet. The results showed that the dietary intervention was not superior to a normal diet in preventing the symptoms. In some cases, the symptoms, for example stool leakage and blood in stools, were worse among those in the intervention group than among those in the control group. Interestingly, the control group ate more high-fiber vegetables than the intervention group. If vegetables have a protective effect, this may explain the results of the study. In another study, patients with irritable bowel syndrome were given pectin powder daily or placebo in a six-weeks randomized trial. The pectin group experienced a greater reduction in composite symptom scores and Bristol stool scale scores, and pectin acted as a prebiotic and significantly enhanced fecal bifidobacteria and decreased total Clostridium sp. [27]. A just published Cochrane-review found a low-certainty evidence, suggesting that a high-fiber diet may lead to better intestinal health one year after radiation treatment [7]. The results were mainly based on Wedlake et al. which also used the highest fiber intake intervention, targeting of >18 g non-starch polysaccharide per day.
Today, there exists no consensus on dietary recommendations before or after radiotherapy, due to the lack of evidence. In Sweden, advice from clinical practitioners is, in general, given only to those patients who ask for it. Examples of such advice are to decrease the intake of lactose, fiber, cereals, or gas-producing food, take a prescription medication like Dimetikon and Loperamid, and if the patient is undernourished increase fat, protein and water intake. In descriptive studies where patients have been interviewed about the dietary practice they use to manage their gastrointestinal symptoms, some reported that elimination of brans, raw vegetables, brassica vegetables, or beans helped, whereas others reported that such interventions did not help [53, 54]. Vegetables like those in the genus brassica and beans or raw vegetables can cause in some individuals gastrointestinal symptoms and discomfort, even if there is no radiation-induced gastrointestinal damage, often due to increased gas formation [55]. Therefore, dietary advice to patients should be given on an individual level, recommending a diet high in soluble fiber if tolerated by the patient.
Strengths of our study include its large size, the population-based design, and the use of a novel metrics for the intensity of the specific syndromes reflecting specific radiation-induced survivorship diseases. Using these metrics instead of single symptoms or a score that nonspecifically reflects different radiation-induced survivorship diseases with varying pathophysiology, we will remove noise that otherwise would have diluted the effect measures. The ethnic homogeneity of our study population (only 8% of the women were born outside Northern Europe) reduces the risk of confounding by unmeasured factors, although it may limit the generalisability of the results to other populations or to men. We cannot exclude that the associations are different among survivors who did not participate. Since information on dietary intake was only collected once and after radiation, reversed causality could be argued. Even if information on symptoms and diet was collected at the same time, symptoms in the past could have changed the dietary behavior among the patients. However, the intensity of syndrome declined with increasing intake of citrus fruits both for those who had, or had not, changed their dietary behaviour after radiotherapy. In addition, if patients with a high burden of symptom had changed their diet, our results concerning gluten should have been similar to results for citrus fruits and vegetables. In contrast, those with a daily intake of gluten reported more often symptoms than those with no intake of gluten. There is a popular belief that gluten can increase problems from the gut and intestine. However, the same is not true for intake of fruit and vegetables; therefore, it is more likely that symptomatic patients will stop eating gluten than fruits and vegetables. In addition, the cancer survivors included in this study were treated between 1991 and 2003, when general dietary recommendations given by clinicians were few. Recommendations given did not include a recommendation to stop eating fruit and vegetables (interview with clinicians, unpublished data). Moreover, dietary habits are often life-long. Thus, we have reason to believe that the dietary behavior presented in this study most likely reflects dietary habits that were established well before radiotherapy was started.
One limitation is that we do not have information on total dietary intake. As a result, misclassification of fruit and vegetable intake may exist; even so, the misclassification is probably non-differential (the same question was answered by all participants regardless of symptom burden), which may lead to an attenuation of the risk estimate. The lack of total dietary intake measurement makes us unable to perform statistical adjustment for potential dietary confounders. However, the most likely confounding dietary factors such as intake of dairy products, spicy food or fatty food were estimated, and none of them was statistically significantly correlated to the intensity of the various radiation-induced survivorship diseases. Concerning smoking, age and time since treatment we found a small, if any, confounding effect concerning the relation between citrus and vegetables and intensity of the urgency syndrome. However, we cannot exclude residual confounding of unmeasured or unknown risk factors. With regard to the study design, our results needs to be confirmed in a clinical dietary intervention trial.
In conclusion, our results indicate that high intake of citrus fruits and vegetables lowers the intensity of radiation-induced cancer survivorship diseases. Evidence from experimental data clearly supports our findings. Our preliminary experiments in a mouse model, specifically developed to study the presence of degenerating crypts and fibrosis in distal colon and rectum after radiotherapy [56], suggest that different levels and type of dietary fiber affect the severity of radiation-damage in mouse colon. Therefore, it may be fruitful to identify individual dietary fiber, or groups of fiber, that could reduce the occurrence of radiation-induced intestinal damage. In future clinical trials, manipulation of diet before, during and after radiotherapy could provide us with new, relatively easy and affordable means to prevent or reduce radiation-induced survivorship diseases of the gut. A special emphasis should be placed on preventing the defecation-urgency syndrome, which is reportedly the most debilitating syndrome for the cancer survivor [19]. When it comes to optimizing dietary interventions with specific individual dietary fiber or groups of fiber, using criteria such as water solubility, molecular composition or other, more data from observational and pre-clinical studies are needed before clinical trials can be considered ethically defensible.
Supporting information
S1 File. Survey questions.
Questions regarding symptoms related to intestinal health, presented in original language and translation to English.
https://doi.org/10.1371/journal.pone.0208115.s001
(PDF)
Acknowledgments
The authors would like to thank all the women who participated in the study, as well as the staff at the Sahlgrenska University Hospital and Karolinska University Hospital who were involved in patient care.
References
- 1. Khalid U, McGough C, Hackett C, Blake P, Harrington KJ, Khoo VS, et al. A modified inflammatory bowel disease questionnaire and the Vaizey Incontinence questionnaire are more sensitive measures of acute gastrointestinal toxicity during pelvic radiotherapy than RTOG grading. International journal of radiation oncology, biology, physics. 2006;64(5):1432–41. pmid:16580497
- 2. Resbeut M, Marteau P, Cowen D, Richaud P, Bourdin S, Dubois JB, et al. A randomized double blind placebo controlled multicenter study of mesalazine for the prevention of acute radiation enteritis. Radiotherapy and oncology: journal of the European Society for Therapeutic Radiology and Oncology. 1997;44(1):59–63.
- 3. Steineck G, Skokic V, Sjoberg F, Bull C, Alevronta E, Dunberger G, et al. Identifying radiation-induced survivorship syndromes affecting bowel health in a cohort of gynecological cancer survivors. PLoS One. 2017;12(2):e0171461. pmid:28158314
- 4. Kuku S, Fragkos C, McCormack M, Forbes A. Radiation-induced bowel injury: the impact of radiotherapy on survivorship after treatment for gynaecological cancers. Br J Cancer. 2013;109(6):1504–12. pmid:24002603
- 5. Ferreira MR, Muls A, Dearnaley DP, Andreyev HJ. Microbiota and radiation-induced bowel toxicity: lessons from inflammatory bowel disease for the radiation oncologist. Lancet Oncol. 2014;15(3):e139–47. pmid:24599929
- 6. Baker AM, Cereser B, Melton S, Fletcher AG, Rodriguez-Justo M, Tadrous PJ, et al. Quantification of crypt and stem cell evolution in the normal and neoplastic human colon. Cell Rep. 2014;8(4):940–7. pmid:25127143
- 7. Lawrie TA, Green JT, Beresford M, Wedlake L, Burden S, Davidson SE, et al. Interventions to reduce acute and late adverse gastrointestinal effects of pelvic radiotherapy for primary pelvic cancers. Cochrane Database Syst Rev. 2018;1:CD012529. pmid:29360138
- 8. Henson CC, Burden S, Davidson SE, Lal S. Nutritional interventions for reducing gastrointestinal toxicity in adults undergoing radical pelvic radiotherapy. Cochrane Database Syst Rev. 2013(11):CD009896. pmid:24282062
- 9. Wedlake LJ, Shaw C, Whelan K, Andreyev HJ. Systematic review: the efficacy of nutritional interventions to counteract acute gastrointestinal toxicity during therapeutic pelvic radiotherapy. Aliment Pharmacol Ther. 2013;37(11):1046–56. pmid:23611411
- 10. Desai MS, Seekatz AM, Koropatkin NM, Kamada N, Hickey CA, Wolter M, et al. A Dietary Fiber-Deprived Gut Microbiota Degrades the Colonic Mucus Barrier and Enhances Pathogen Susceptibility. Cell. 2016;167(5):1339–53 e21. pmid:27863247
- 11. Leclere L, Cutsem PV, Michiels C. Anti-cancer activities of pH- or heat-modified pectin. Front Pharmacol. 2013;4:128. pmid:24115933
- 12. Sureban SM, May R, Qu D, Chandrakesan P, Weygant N, Ali N, et al. Dietary Pectin Increases Intestinal Crypt Stem Cell Survival following Radiation Injury. PloS one. 2015;10(8):e0135561. pmid:26270561
- 13. Yang J, Ding C, Dai X, Lv T, Xie T, Zhang T, et al. Soluble Dietary Fiber Ameliorates Radiation-Induced Intestinal Epithelial-to-Mesenchymal Transition and Fibrosis. JPEN J Parenter Enteral Nutr. 2017 Nov;41(8):1399–1410 pmid:27660288
- 14. Lupton JR. Microbial degradation products influence colon cancer risk: the butyrate controversy. The Journal of nutrition. 2004;134(2):479–82. pmid:14747692
- 15. Pacheco RG, Esposito CC, Muller LC, Castelo-Branco MT, Quintella LP, Chagas VL, et al. Use of butyrate or glutamine in enema solution reduces inflammation and fibrosis in experimental diversion colitis. World J Gastroenterol. 2012;18(32):4278–87. pmid:22969190
- 16. Leonel AJ, Alvarez-Leite JI. Butyrate: implications for intestinal function. Curr Opin Clin Nutr Metab Care. 2012;15(5):474–9. pmid:22797568
- 17. Pryde SE, Duncan SH, Hold GL, Stewart CS, Flint HJ. The microbiology of butyrate formation in the human colon. FEMS microbiology letters. 2002;217(2):133–9. pmid:12480096
- 18. Henningsson AM, Bjorck IM, Nyman EM. Combinations of indigestible carbohydrates affect short-chain fatty acid formation in the hindgut of rats. J Nutr. 2002;132(10):3098–104. pmid:12368401
- 19. Dunberger G, Lind H, Steineck G, Waldenstrom AC, Nyberg T, Al-Abany M, et al. Self-reported symptoms of faecal incontinence among long-term gynaecological cancer survivors and population-based controls. European journal of cancer. 2010;46(3):606–15. pmid:19926277
- 20. Lind H, Waldenstrom AC, Dunberger G, al-Abany M, Alevronta E, Johansson KA, et al. Late symptoms in long-term gynaecological cancer survivors after radiation therapy: a population-based cohort study. Br J Cancer. 2011;105(6):737–45. pmid:21847122
- 21. Buja A, Eyuboglu N. Remarks on Parallel Analysis. Multivariate Behav Res. 1992;27(4):509–40. pmid:26811132
- 22. Zientek LR, Thompson B. Applying the bootstrap to the multivariate case: bootstrap component/factor analysis. Behav Res Methods. 2007;39(2):318–25. pmid:17695360
- 23. Choi J, Zou H, Oehlert G. A penalized maximum likelihood approach to sparse factor analysis. Stat Interface. 2010;3(4):429–36.
- 24. McGough C, Baldwin C, Frost G, Andreyev HJN. Role of nutritional intervention in patients treated with radiotherapy for pelvic malignancy. British Journal of Cancer. 2004;90(12):2278–87. pmid:15162154
- 25. Touchefeu Y, Montassier E, Nieman K, Gastinne T, Potel G, Bruley des Varannes S, et al. Systematic review: the role of the gut microbiota in chemotherapy- or radiation-induced gastrointestinal mucositis—current evidence and potential clinical applications. Alimentary pharmacology & therapeutics. 2014;40(5):409–21.
- 26. Pettersson A, Nygren P, Persson C, Berglund A, Turesson I, Johansson B. Effects of a dietary intervention on gastrointestinal symptoms after prostate cancer radiotherapy: long-term results from a randomized controlled trial. Radiother Oncol. 2014;113(2):240–7. pmid:25467005
- 27. Xu L, Yu W, Jiang J, Feng X, Li N. [Efficacy of pectin in the treatment of diarrhea predominant irritable bowel syndrome]. Zhonghua Wei Chang Wai Ke Za Zhi. 2015;18(3):267–71. pmid:25809332
- 28. Wedlake L, Shaw C, McNair H, Lalji A, Mohammed K, Klopper T, et al. Randomized controlled trial of dietary fiber for the prevention of radiation-induced gastrointestinal toxicity during pelvic radiotherapy. Am J Clin Nutr. 2017;106(3):849–57. pmid:28679552
- 29. Kerem M, Bedirli A, Karahacioglu E, Pasaoglu H, Sahin O, Bayraktar N, et al. Effects of soluble fiber on matrix metalloproteinase-2 activity and healing of colon anastomosis in rats given radiotherapy. Clinical nutrition (Edinburgh, Scotland). 2006;25(4):661–70.
- 30. Oufir LE, Barry JL, Flourie B, Cherbut C, Cloarec D, Bornet F, et al. Relationships between transit time in man and in vitro fermentation of dietary fiber by fecal bacteria. European journal of clinical nutrition. 2000;54(8):603–9. pmid:10951507
- 31. Roth JA, Frankel WL, Zhang W, Klurfeld DM, Rombeau JL. Pectin improves colonic function in rat short bowel syndrome. The Journal of surgical research. 1995;58(2):240–6. pmid:7861779
- 32. Ye MB, Lim BO. Dietary pectin regulates the levels of inflammatory cytokines and immunoglobulins in interleukin-10 knockout mice. Journal of agricultural and food chemistry. 2010;58(21):11281–6. pmid:20945935
- 33. Sanaka M, Yamamoto T, Anjiki H, Nagasawa K, Kuyama Y. Effects of agar and pectin on gastric emptying and post-prandial glycaemic profiles in healthy human volunteers. Clinical and experimental pharmacology & physiology. 2007;34(11):1151–5.
- 34. Espinal-Ruiz M, Parada-Alfonso F, Restrepo-Sanchez LP, Narvaez-Cuenca CE, McClements DJ. Interaction of a dietary fiber (pectin) with gastrointestinal components (bile salts, calcium, and lipase): a calorimetry, electrophoresis, and turbidity study. Journal of agricultural and food chemistry. 2014;62(52):12620–30. pmid:25479310
- 35. Xiong Y, Chen D, Yu C, Lv B, Peng J, Wang J, et al. Citrus nobiletin ameliorates experimental colitis by reducing inflammation and restoring impaired intestinal barrier function. Molecular nutrition & food research. 2015;59(5):829–42.
- 36. Luceri C, Femia AP, Fazi M, Di Martino C, Zolfanelli F, Dolara P, et al. Effect of butyrate enemas on gene expression profiles and endoscopic/histopathological scores of diverted colorectal mucosa: A randomized trial. Dig Liver Dis. 2016;48(1):27–33. pmid:26607831
- 37. Williams JP, Brown SL, Georges GE, Hauer-Jensen M, Hill RP, Huser AK, et al. Animal models for medical countermeasures to radiation exposure. Radiat Res. 2010;173(4):557–78. pmid:20334528
- 38. Biesiekierski JR, Newnham ED, Irving PM, Barrett JS, Haines M, Doecke JD, et al. Gluten causes gastrointestinal symptoms in subjects without celiac disease: a double-blind randomized placebo-controlled trial. Am J Gastroenterol. 2011;106(3):508–14; quiz 15. pmid:21224837
- 39. Krigel A, Lebwohl B. Nonceliac Gluten Sensitivity. Adv Nutr. 2016;7(6):1105–10. pmid:28140327
- 40. Stewart AJ, Southcott BM. Coeliac disease following high-dose chemotherapy. Clin Oncol (R Coll Radiol). 2002;14(6):494–6.
- 41. Chassard C, Lacroix C. Carbohydrates and the human gut microbiota. Curr Opin Clin Nutr Metab Care. 2013;16(4):453–60. pmid:23719143
- 42. Sonnenburg ED, Sonnenburg JL. Starving our microbial self: the deleterious consequences of a diet deficient in microbiota-accessible carbohydrates. Cell Metab. 2014;20(5):779–86. pmid:25156449
- 43. Ghaffarzadegan T, Marungruang N, Fak F, Nyman M. Molecular Properties of Guar Gum and Pectin Modify Cecal Bile Acids, Microbiota, and Plasma Lipopolysaccharide-Binding Protein in Rats. PLoS One. 2016;11(6):e0157427. pmid:27315087
- 44. Wexler HM. Bacteroides: the good, the bad, and the nitty-gritty. Clin Microbiol Rev. 2007;20(4):593–621. pmid:17934076
- 45. Chang PY, Qu YQ, Wang J, Dong LH. The potential of mesenchymal stem cells in the management of radiation enteropathy. Cell Death Dis. 2015;6:e1840. pmid:26247725
- 46. Asfaha S, Hayakawa Y, Muley A, Stokes S, Graham TA, Ericksen RE, et al. Krt19(+)/Lgr5(-) Cells Are Radioresistant Cancer-Initiating Stem Cells in the Colon and Intestine. Cell Stem Cell. 2015;16(6):627–38. pmid:26046762
- 47. Mandir N, FitzGerald AJ, Goodlad RA. Differences in the effects of age on intestinal proliferation, crypt fission and apoptosis on the small intestine and the colon of the rat. Int J Exp Pathol. 2005;86(2):125–30. pmid:15810984
- 48. Fasano A. Zonulin and its regulation of intestinal barrier function: the biological door to inflammation, autoimmunity, and cancer. Physiol Rev. 2011;91(1):151–75. pmid:21248165
- 49. Shukla PK, Gangwar R, Manda B, Meena AS, Yadav N, Szabo E, et al. Rapid disruption of intestinal epithelial tight junction and barrier dysfunction by ionizing radiation in mouse colon in vivo: protection by N-acetyl-l-cysteine. Am J Physiol Gastrointest Liver Physiol. 2016;310(9):G705–15. pmid:26822914
- 50. Shim S, Jang HS, Myung HW, Myung JK, Kang JK, Kim MJ, et al. Rebamipide ameliorates radiation-induced intestinal injury in a mouse model. Toxicol Appl Pharmacol. 2017;329:40–7. pmid:28526636
- 51. Bentzen SM. Preventing or reducing late side effects of radiation therapy: radiobiology meets molecular pathology. Nat Rev Cancer. 2006;6(9):702–13. pmid:16929324
- 52. Schroeder BO, Birchenough GMH, Stahlman M, Arike L, Johansson MEV, Hansson GC, et al. Bifidobacteria or Fiber Protects against Diet-Induced Microbiota-Mediated Colonic Mucus Deterioration. Cell Host Microbe. 2018;23(1):27–40 e7. pmid:29276171
- 53. Gami B, Harrington K, Blake P, Dearnaley D, Tait D, Davies J, et al. How patients manage gastrointestinal symptoms after pelvic radiotherapy. Alimentary pharmacology & therapeutics. 2003;18(10):987–94.
- 54. Sekhon S. Chronic radiation enteritis: women's food tolerances after radiation treatment for gynecologic cancer. Journal of the American Dietetic Association. 2000;100(8):941–3. pmid:10955053
- 55. Winham DM, Hutchins AM. Perceptions of flatulence from bean consumption among adults in 3 feeding studies. Nutr J. 2011;10:128. pmid:22104320
- 56. Steineck G, Bull C, Kalm M, Sjoberg F, Alevronta E, Malipatlolla DK, et al. Radiation physiology—evidence for a higher biological effect of 24 Gy in four fractions as compared to three. Acta Oncol. 2017;56(9):1240–3. pmid:28379720