Glucagon-like peptide-2 (GLP-2) is an intestinal hormone that is secreted from the intestinal mucosa after food ingestion. It has an important role in maintaining the function of the intestine. However, its role in the pathophysiology of gastrointestinal problems of children undergoing anti-cancer treatment is not known. GLP-2 is a 33-amino acid peptide intestinotrophic hormone that is secreted in a nutrition-dependent manner from enteroendocrine L-cells in the intestinal tract. It promotes nutrient absorption via expansion of the mucosal epithelium by stimulation of crypt cell proliferation and inhibition of villus cell apoptosis (1). Increased GLP-2 secretion stimulates disaccharidase activity and intestinal glucose transport (2). Conversely, intestinal permeability, local pro-inflammatory cytokine response (1) and gastrointestinal motility (3) decrease with increasing GLP-secretion. Furthermore, it has been shown that the villus atrophy and increase in intestinal permeability that occur as a consequence of parenteral nutrition are reduced by treatment with a GLP-2 analog in animal studies (4), and GLP-2 treatment reduces malabsorption in adult patients with short bowel syndrome (5). Thus, GLP-2 secretion may be an important factor for intestinal recovery after injury. Oral and gastrointestinal side effects of anti-cancer treatment for childhood cancer as well as the disease itself may cause reduced enteral intake. The subsequent clinical problems that may lead to a reduced oral intake are mucositis, nausea/vomiting, abdominal pain and constipation (6). Several cytotoxic agents used in treatment protocols for childhood malignancy affect the dividing cells in the intestinal crypts causing apoptosis, decreased differentiation and proliferation of normal epithelial cells and mucus producing cells (7,8). This may lead to malabsorption, mucosal atrophy (9), inflammation and increased intestinal permeability (10). Local infections may aggravate these effects (11). Furthermore, neurotoxic anti-cancer agents may lead to decreased intestinal motility (12). The enteroendocrine cells seem to be unaffected by anti-cancer agents (13) but may be damaged by local infection (14,15). The cascade of intestinal mucosal injury, reduced intestinal motility, presence of infectious agents and immune insufficiency are assumed to increase the risk of bacterial translocation (10), which is thought to play a central role in the pathophysiology of septicaemia (16), the major cause of morbidity and mortality in childhood cancer. No studies on GLP-2 secretion have so far been performed in children with malignancy.
We hypothesized that the GLP-2 response to enteral intake might be reduced due to mucosal damage and/or reduced oral nutrient intake in children undergoing anti-cancer treatment. The aim of the study is to evaluate secretion of GLP-2 in response to a test meal in children undergoing anti-cancer treatment.
PATIENTS AND METHODS
Twenty-four children participated in the study from May, 2000, to November, 2001. The children belonged to three groups: 1) Fifteen children with haematological malignancies including acute lymphatic leukaemia (ALL) or acute myeloid leukaemia (AML) treated according to protocols of the Nordic Organisation of Paediatric Haematology and Oncology (NOPHO) (17); 2) One child with high grade Wilms tumour treated according to Société Internationale d'Oncologie Pédia-trique (SIOP) 93-01 protocol; and 3) Nine children undergoing stem cell transplantation (SCT) for primary immune or cancer diseases but receiving conditioning therapy composed of either: a) busulphan and cyclophosphamide, or b) total body irradiation and high dose cyclophosphamide. All the meal tests were performed after anti-cancer treatment infusion was terminated and no tests were performed during acute infectious disease.
Blood samples were collected after an overnight fast and 60 minutes (range, 50-70 minutes) after intake of a mixed meal. The samples were placed on ice immediately after withdrawal from a central venous catheter. All blood samples were centrifuged at 3000 rpm for 10 minutes within 30 minutes after withdrawal. The serum was separated and stored at −20°C until analysis. GLP-2 immunoreactivity was measured with a specific NH2 terminal radioimmunoassay as described previously (18,19).
In order to simulate the every-day clinical situation, the children were offered a test meal according to the following principles: 1) the children were encouraged to eat a meal as large as their appetite allowed within a period of 15 minutes; 2) children who were unable to eat were given a tube bolus meal as large as possible if they had a feeding tube in position; and 3) children refusing to eat were given nothing if they did not have a feeding tube, but the 2nd blood sample was still taken. It was impossible to compose a standard test meal for these children because of their different ages and, consequently, different food preferences and aversions. The food intake during the test meal was recorded. Furthermore, the parents or the nurse registered all oral nutrient intakes by household measurements the rest of the test day and the following 3 days to estimate the child's current dietary intake. Energy and nutrient intake was calculated on a computer using the software Dankost 2 (Danish Catering Institute). Energy and macronutrient intake were used in the present analysis and expressed in grams and energy percentage and compared to the Nordic recommendations for energy intake of healthy children (20). Supplemental parenteral nutrition was not discontinued during the test, because only enteral intake stimulates p-GLP-2 rise (2). In healthy adults mean fasting p-GLP-2 is 25 pmol/l (range, 19-27 (21)) and p-GLP-2 increase is 2 to 5 fold in response to a normal mixed meal of more than 400 kcal (1680 kJ/ 70 kg = 24 kJ/kg). A p-GLP-2 peak is seen 60 minutes after intake of a mixed meal (3).
Gastrointestinal Toxicity of Anti-Cancer Treatment
On the test day and the following 3 days, the parents or the nurse registered the oral and gastrointestinal toxicity according to the WHO (World Health Organisation) toxicity scoring system (22). The following factors were registered: oral pain (0-4; normal-not able to eat anything), abdominal pain (0-2; no pain-severe pain), stool frequency (number per day), stool consistence (1-5; firm-watery), nausea (0-2; no nausea-severe nausea), vomiting (number per day) and dysphagia (0-2; no dysphagia-severe dysphagia) and the total mean score/d was calculated. Absolute blood neutrophile counts were performed on the test day.
The ethical standards of the Helsinki Declaration were followed. The protocol was approved by the Ethics Committee for Medical Research in Copenhagen, Denmark (KF 01-280/99). The parents and mature children gave oral and written consent.
Data were expressed as mean (SD) and medians (range). Students' t test and Pearson's correlation were used for normally distributed parameters and Mann-Whitney and Spearman's correlation for parameters not normally distributed. Manual backward multiple regression models were used for determinations of explanatory factors of GLP-2 increase. The analysis was repeated only including the first test of each child with more than one test, to exclude that including children with two tests resulted in a bias. The SAS version 8.2 for Windows was used for statistical calculations. A value of P < 0.05 was considered significant.
Forty-four meal stimulation tests were performed. Twenty-nine children were able to eat the test meal. Twelve of the test meals were given through a previously placed feeding tube and three of the tests were performed in children who were unable to eat anything and who did not have a feeding tube. Characteristics of the children are shown in (Table 1). For thirty-two tests, a 4-day dietary record was completed. Twenty-eight tests were performed in children with haematological malignant diseases, 14 in children undergoing stem cell transplantation, one in a child with Ewing sarcoma and one in a child with Wilm's tumour. The sex ratio (male/female) was 28/16. Eight of the tests were performed in children receiving partial parenteral and four in patients receiving total parenteral nutrition. All of these were receiving stem cell transplantation. Twelve of the children were treated with corticosteroids at the time the test took place.
The children's energy intake during the 4-day period (calculated per kg body weight per day) was compared with the Nordic recommended energy intake of healthy children (20). The mean intake was 62% of recommended intake with a range from eating nothing to 157%, which is significantly lower than the recommended intake in healthy children (P = 0.0001, Fig. 1). Eighteen percent (n = 6) of the children had a mean energy intake below 25% of recommended intake, and 60% (n = 20) below 75% of recommended intake (Fig. 1). The children receiving corticosteroid treatment had a significantly higher energy intake (median 100% of recommended intake; range, 59-134%) than children not receiving corticosteroid (median, 51%; range, 0-157%; P = 0.02). Furthermore, the children undergoing stem cell transplantation had significantly lower energy intake (median 14% of recommended intake; range, 0-94) than children treated for ALL (median 94%; range, 29-157; P = 0.002). Using multiple regression analysis, predictors of the energy intake (% of recommended intake) were examined. WHO GI toxicity score was a highly significant determinant of energy intake (r = 0.53, P = 0.002), while neutrophile count × 109/l was insignificant. Furthermore, a significant correlation was found between the size of the test meal and the energy intake in percentage of recommended values during the 4 days of registration (r = 0.61; P = 0.0002).
Data from the 44 meal stimulation tests are shown in Table 2. The pre p-GLP-2 level was significantly higher (P = 0.0001) in our patients (mean, 38 pmol/l; range, 7-80) than in healthy adults (mean, 25 pmol/l, (21)). A multiple regression model of predictors for the level of pre p-GLP-2 was established. The model included the mean energy intake (% of recommended), corticosteroid treatment (yes/no), total WHO toxicity score and neutrophile count. Corticosteroid treatment was the only explanatory factor in the model. The mean (SD) value for the 12 tests in children in corticosteroid treatment was 48 pmol/l (18) vs. 35 pmol/l (17) in the remaining 22 tests (P = 0.03). There was a highly significant positive association between pre p-GLP-2 and the log plasma value of post p-GLP-2 (r = 0.56; P < 0.0001).
The associations between test meal size and GLP-2 increase expressed as absolute increase or as fold increase are shown in Figure 2A (r = 0.62, P ≤ 0.0001) and Figure 2B (r = 0.53, P ≤ 0.0001). The GLP-2 increase was clearly dependent on the energy content of the test meal. Only 12 of the meal stimulation tests (28%) resulted in a GLP-2 increase >2 fold, which is assumed to be the lower limit of normal values (2). Among the 11 test meals with energy content below 10 kJ/kg body weight, none had a GLP-2 increased above 1.2 folds (median, 0.92; range, 0.6-1.2). Among the 20 tests with a test meal energy content between 10 and 50 kJ/kg body weight, the median GLP-2 increase was 1.4 fold (range, 0.7-3.2), and only 3 had an increase above 2 folds. Among the 12 tests with a test meal energy intake above 50 kJ/kg body weight, half had an increase above 2 folds (median increase was 2.4 (range 0.9-5.3).
Multiple regression models of explanatory factors for p-GLP-2 increase, both fold and absolute increase, included the test meal in kJ/kg, WHO toxicity score and neutrophil count × 109/l. The models identified only the energy content of the test meal as an explanatory factor for the increase in p-GLP-2.
Finally, an analysis was performed including only the first GLP-2 meal test in each participating child (n = 25). The results were compatible with those obtained from analysis of the whole group (data not shown).
The main result of the study was that the p-GLP-2 response to a test meal in children undergoing anti-cancer treatment remained at the same level as in healthy individuals, if the energy content of the test meal had a reasonable size. However, the size of the test meals the children could eat was in most cases low and the energy intake over a 4-day period was, for most of the children, considerably below the recommended intake.
In the present study, 60% of the children had an energy intake below 75% of the recommended value, and in more than half of the children the test meal was too small to stimulate a p-GLP-2 increase above 1.2 folds. Furthermore, there was a strong correlation between the 4-day recorded energy intake and the size of the test meal, meaning that the meal stimulation test was indeed representative of the daily energy intake of the child. Other studies have also shown that the energy intake is reduced in children undergoing cancer chemotherapy (6,23). In a study of children with malignant diseases, their oral intake was reduced more than 20% during the initial anti-cancer treatment in spite of anti-emetic treatment and symptomatic treatment of other side effects (6), and some children did not eat at all for shorter or longer periods. A low enteral intake has a negative impact on intestinal function. In healthy humans the absorptive surface area of the intestinal mucosa adapts continuously to the enteral energy intake (24). In animal studies mucosal atrophy can be detected 24 hours after start of total parenteral nutrition and fasting (25,26). However, in humans mucosal hypoplasia seems to develop later, 3 to 4 weeks after initiating parenteral nutrition or fasting (27). Recovery of the intestinal damage after anti-cancer drug treatment may be accelerated, and the risk of bacterial translocation from the intestinal tract during treatment-induced neutropenia may be decreased when enteral nutrition is used. Until now it has not been proven that enteral nutrition can protect against bacterial translocation and septic disease (27). The necessary amount of energy given enterally needed to maintain or recover the intestinal mucosa is not known, but studies in piglets have shown that at least 40 to 60% of total nutritional energy needs to be given enterally to maintain the mucosa intact (28). Enteral nutrition is the only nutritional route that stimulates a rise in p-GLP-2 (1). It underlines the importance of enteral nutrition for recovery of intestinal injury. These results show that the enteral intake must have a certain energy content to stimulate the intestinotrophic activity of the hormone GLP-2. Children with a low energy intake may have an increased risk of developing intestinal mucosal atrophy. The results of the subgroup study were similar to the main results.
Elevated fasting serum concentrations were found in ileum resected short-bowel patients with a preserved colon (21) and in patients with ulcerative colitis or Crohn's disease (29). In our study, the children had significantly higher fasting p-GLP-2 levels compared with healthy adults, which may indicate intestinal injury. However, the fasting p-GLP-2 levels did not correlate to the WHO toxicity, which in our study was chosen as an indicator of mucosal damage. Elevated levels of fasting p-GLP-2 could represent part of the normal physiological feedback mechanism part of the normal intestinal adaptive response to injury or inflammation in patients with inflammatory bowel disease (29,30). Children treated with corticosteroid had higher fasting level of p-GLP-2. We cannot explain this finding despite of corticosteroid as well as GLP-2 being known to suppress release of proinflammatory cytokines (31,32). Nutrition seems to be important for the immunologic defence of the intestinal tract (33), but the complex interaction between nutrient availability and enteric nervous system and hormones on mucosal immunity is not fully understood (34).
The GLP-2 increase, measured in fold or absolute increase, seems to be important for the intestinotrophic effect of GLP-2 (35). In this study only the energy content of the meal was a predictor for the subsequent rise in p-GLP-2. A minimal amount of energy was needed for a response to occur. High toxicity according to the WHO scale was an important predictor for low energy intake during the 4-day recording, but low blood neutrophile count had no influence. In adults it has been shown that a mixed test meal containing 400 kcal (1680 kJ), produce a 2 to 5 fold rise in GLP-2 (2), corresponding to 24 kJ/kg for a person of 70 kg. The median energy intake of the test meals in this study was exactly 24 kJ/kg but the median relative increase in p-GLP-2 was only 1.3. The data in this study indicate that the p-GLP-2 response after a sufficient test meal is similar to that seen in healthy adults. However, the necessary amount of energy in the meal needed to stimulate the intestinal mucosa may vary from child to child. In addition, the patients in this study were children with different cancer diseases and anti-cancer treatment associated complications.
Animal studies have shown that GLP-2 treatment initiated after chemotherapy administration enhances intestinal recovery (36) and reduces chemotherapy-associated mortality (37). Furthermore, in animal studies of tumour-bearing rats, GLP-2 treatment reduces intestinal permeability (38) and decreases bacterial translocation (39). Prevention of total parenteral nutritional-induced mucosal hypoplasia in rats can be obtained by co-infusion of GLP-2 (40). A study of patients with short bowel syndrome has shown that GLP-2 is well tolerated (5).
Our results suggest that a sufficient enteral intake will stimulate GLP-2 secretion and thereby possibly prevent injury or promote recovery. The most important strategy to improve gastrointestinal recovery may therefore be to optimise enteral nutrition. A more vigorous use of nasogastric or gastrostomy tubes, using continuous infusion, may be needed. Since in a number of patients it is impossible to maintain a sufficient enteral energy intake, GLP-2 may be a future therapeutic option for treatment or prevention of mucosal barrier injury and bacterial translocation. Therefore, human intervention studies in patients undergoing cancer chemotherapy and studies of the GLP-2 receptor activity in the enterocytes during chemotherapy and severe intestinal damage are needed.
1. Drucker DJ. Gut adaptation and the glucagon-like peptides. Gut
2. Xiao Q, Boushey RP, Drucker DJ, et al. Secretion of the intestinotropic hormone glucagon-like peptide 2 is differentially regulated by nutrients in humans. Gastroenterology
3. Wojdemann M, Wettergren A, Hartmann B, et al. Inhibition of sham feeding-stimulated human gastric acid secretion by glucagon-like peptide-2. J Clin Endocrinol Metab
4. Drucker DJ, Boushey RP, Wang F, et al. Biological properties and therapeutic potential of glu-cagon-like peptide-2. J Parenter Enteral Nutr
5. Jeppesen PB, Hartmann B, Thulesen J, et al. Glucagon-like peptide 2 improves nutrient absorp-tion and nutritional status in short-bowel patients with no colon. Regul Pept
6. Skolin I, Axelsson K, Ghannad P, et al. Nutritient intake and weight development in children during chemotherapy for malignant disease. Oral Oncol
7. Shaw MT, Spector MH, Ladman AJ. Effects of cancer, radiotherapy and cytotoxic drugs on intestinal structure and function. Cancer Treat Rev
8. Smith FP, Kisner DK, Widerlite L, et al. Chemotherapeutic alterations of small intestinal morphology and function: a progress report. J Clin Gastroenterol
9. Hyams JS, Batrus CL, Grand RJ, et al. Cancer chemotherapy-induced lactose malabsorption in children. Cancer
10. Blijlevens NMA, Donnelly JP, De Pauw BE. Mucosal barrier injury: biology, pathology, clinical counterparts and consequences of intensive treatment for haematological malignancy: an overview. Bone Marrow Transplant
11. Khan HA, Wingaard JR. Infection and mucosal injury in cancer treatment. J Natl Cancer Inst Monogr
12. McGuire SA, Gospe SMJr, Dahl G. Acute vincristine neurotoxicity in the presence of hereditary motor and sensory neuropathy type I. Med Pediatr Oncol
13. Pinkerton CR, Cameron CHS, Sloan JM, et al. Jejunal crypt cell abnormalities associated with methotrexate treatment in children with acute lymphoblastic leukaemia. J Clin Pathol
14. Cunningham D, Morgan RJ, Mills PR, et al. Functional and structural changes of the human proximal small intestine after cytotoxic therapy. J Clin Pathol
15. Gwavava NJT, Pinkerton CR, Glasgow JFT, et al. Small bowel enterocyte abnormalities caused by methotrexate treatment in acute lymphoblastic leukaemea of childhood. J Clin Pathol
16. Heine H, Rietschel ET, Ulmer AJ. The biology of endotoxin. Mol Biotechnol
17. Gustafsson G, Schmiegelow K, Forestier E, et al. Improving outcome through two decades in childhood ALL in the Nordic countries; the impact of high-dose methotrexate in the reduction of CNS irradiation. Leukemia
18. Hartmann B, Johnsen AH, Ørskov C, et al. Structure, measurement, and secretion of human glu-cagon-like peptide-2. Peptides
19. Orskov C, Holst JJ. Radio-immunoassays for glucagon like peptids 1 and 2 (GLP-1 and (GLP-2). Scand J Clin Lab Invest
20. Sandstroö m B, Aro A, Becker W. Nordiska naäringsrekommendationer. Nordiska Ministerrädet
21. Jeppesen PB, Hartmann B, Thulesen J, et al. Elevated plasma glucagon-like peptide 1 and 2 con-centrations inileum recected short bowel patients with a preserved colon. Gut
22. W.H.O. World Health Organization. Cancer therapy evaluation program. Cancer therapy evaluation program. 1999; CTC, version 2.0:10-13.
23. Bosaeus I, Daneryd P, Svanberg E. Dietary intake and resting energy expenditure in relation to weight loss in unselected cancer patients. Int J Cancer
24. O'Brian DP, Nelson LA, Huang FS, et al. Intestinal adaptation: structure, function and regulation. Semin Pediatr Surg
25. Langkamp-Henken B, Kudsk KA, Proctor KG. Fasting-induced reduction of intestinal reperfu-sion injury. J Parenter Enteral Nutr
26. Moore FA, Moore EE, Jones TN, et al. TEN versus TPN following major abdominal trauma-reduced septic morbidity. J Trauma
27. MacFie J. Enteral versus parenteral nutrition: the significance of bacterial translocation and gut-barrier function. Nutrition
28. Burrin DG, Stoll B, Jiang R, et al. Minimal enteral nutrient requirements for intestinal growth in neonatal piglets: how much is enough? Am J Clin Nutr
29. Xiao Q, Boushey RP, Cino M, et al. Circulating levels of glucagon-like peptide-2 in human subjects with inflammatory bowel disease. Am J Physiol Regulatory Integrative Comp Physiol
30. Jeppesen PB, Hartmann B, Hansen BS, et al. Impaired meal stimulated glucagon-like peptide 2 response in ileal resected short bowel patients with intestinal failure. Gut
31. Alavi K, Schwartz MZ, Palazzo JP, et al. Treatment of inflammatory bowel disease in a rodent model with the intestinal growth factor glucagon-like peptide-2. J Pediatr Surg
32. Braun CM, Huang SK, Bashian GG, et al. Corticosteroid modulation of human, antigen-specific Th1 and Th2 response. J Allergy Clin Immunol
33. Johnson CD, Kudsk KA. Nutrition and intestinal mucosal immunity. Clin Nutr
34. Kudsk KA. Current aspects of mucosal immunology and its influence by nutrition. Am J Surg
35. Kato Y, Yu D, Schwartz MZ. Glucagonlike peptide-2 enhances small intestinal absorpotive function and mucosal mass in vivo. J Pediatr Surg
36. Tavakkolizadeh A, Shen R, Abraham P, et al. Glucagon-like peptide 2: a new treatment for chemotherapy-induced enteritis. J Surg Res
37. Broushey RP, Yusta B, Drucker DJ. Glucagon-like Peptide-2 (GLP-2) reduces chemotherapy associated mortality and enhances cell survival in cell expressing a transfected GLP-2 receptor. Cancer Res
38. Benjamin MA, McKay DM, Yang PC, et al. Glucagon-like peptide-2 enhances intestinal epithelial barrier function of both transcellular and paracellular partways in the mouse. Gut
39. Chance WT, Sheriff S, Foley-Nelson T. Maintaining gut integrity during parenteral nutrition of tumor-bearing rats: effect of glucagon-like peptide 2. Nutr Cancer
40. Chance WT, Foley-Nelson T, Thomas I, et al. Prevention of parenteral nutrition-induced gut hypoplasia by co-infusion of glucagon-like peptide-2. Am J Physiol