Publication of the International Union Against Cancer Publication de l’Union Internationale Contre le Cancer Int. J. Cancer: Supplement 11, 66–68 (1998) r 1998 Wiley-Liss, Inc. BENEFITS OF NUTRITIONAL INTERVENTION ON NUTRITIONAL STATUS, QUALITY OF LIFE AND SURVIVAL Jan VAN EYS* Department of Pediatrics, Vanderbilt University School of Medicine, Nashville, TN, USA Most cancers in children are acute diseases. Therefore, the incidence of malnutrition, in general, is not different from the incidence in the referral population. Some specific tumors, such as neuroblastoma and those resulting in the diencephalic syndrome, can be exceptions. By contrast, malnutrition is a frequent problem during modern intensive cancer treatment as the result of the associated anorexia, altered taste sensations and catabolic effects of drugs. In addition, there are psychogenic factors and metabolic consequences associated with the tumor itself. Nutritional support does improve the feeling of well-being and performance status, while maintaining or improving the immune competence, thereby potentially affecting survival by limiting infectious episodes. There is no convincing evidence to date that nutritional support has an antineoplastic effect per se, but deficiency of a specific nutrient might be beneficial because of a differential requirement between tumor and normal cells. Theoretically, nutritional support might enhance tumor growth but also susceptibility to chemotherapy. In either case, nutrition is a support modality that must be given with appropriate tumor-directed therapy if curative intent is the goal of treatment. Nutrition remains a consideration after therapy is completed. This generates different challenges. If further tumor-directed therapy is futile, the decision to continue nutritional support is difficult, but if the child is well, nutritional rehabilitation must be pursued. Finally, the cured child continues to benefit from dietary advice. Nutrition should be viewed for what it is: supplying the most basic need of children. Int. J. Cancer Supplement 11:66–68, 1998. r 1998 Wiley-Liss, Inc. Malnutrition and cancer are linked very closely in the minds of many, professional and lay persons alike. Most would agree that nutritional support is in the best interest of the child afflicted with cancer, though there are many who worry that feeding the child feeds the cancer, and so nutritional support may be counterproductive. In either case, people usually have a simple picture of what malnutrition is and why it is linked to cancer. In this report, I would like to indicate that this simple link is not a valid view. Malnutrition and cancer are interrelated in varying ways, depending on the stage of cancer evolution and therapy. The stages of cancer evolution which one can distinguish are diagnosis, therapy, completion of therapy with success or failure and, finally, off-therapy cancer survival. Each of these stages has specific etiologies for malnutrition and may require different approaches to nutritional challenges. Table I summarizes these stages and approaches to nutritional support. I would like to elaborate on this theme in some detail. One caveat: a communication problem can arise when nutrition is viewed at the cellular level instead of at the level of the whole person. This may seem to make sense in our era of molecular medicine. However, it is analogous to treating a disease instead of treating a patient who is ill. This report stresses the feeding of a child, though I touch upon nutrition at the cellular level. THE CHILD NEWLY DIAGNOSED WITH CANCER Cancer in children is more often than not an acute disease. The impact of the cancer on nutritional status is likely to be minimal at the time of diagnosis, if the diagnosis is made in a reasonably timely manner. A cooperative study of the nutritional status of pediatric cancer patients showed that the incidence of malnutrition at the time of diagnosis did not differ from that seen in patients with benign tumors, referred to the same institutions (Donaldson et al., 1981). The conclusion, that children newly diagnosed with cancer were not particularly predisposed to malnutrition, was drawn also from our own experience at the University of Texas M.D. Anderson Cancer Center (Carter et al., 1983a,b). Nevertheless, a heavy tumor burden might induce a catabolic state. The largest tumor burdens are found in hematologic malignancies. However, in one study, the nutritional status of untreated children with acute leukemia was no different from that of children with benign acute diseases (Uderzo et al., 1996). On the other hand, solid tumors in adolescents can predispose to malnutrition because of the competing metabolic demand for body growth. It might be added that all too often we ignore incipient malnutrition but we discriminate against obesity. When we analyzed our study population in Houston (Carter et al., 1983b), we found that children with cancer who were obese at the time of diagnosis fared somewhat better than their non-obese counterparts, though the differences in event-free survival were not statistically significant (37 obese vs. 110 normal weight children, p ⫽ 0.178; data not shown). There are a few exceptions to these generalizations. Most spectacular is the diencephalic syndrome, wherein a tumor in the anterior floor of the third ventricle on the anterior portion of the hypothalamus results in a wasting syndrome in spite of seemingly adequate intake of food. Occasionally, neuroblastoma results in the secretion of a vasoactive intestinal peptide, resulting in profuse watery diarrhea. However, paraneoplastic syndromes are the exception rather than the rule in pediatric oncology. Likewise, malnutrition in children newly diagnosed with cancer is the exception and not the rule in societies in which malnutrition is uncommon in the general population. Nevertheless, malnutrition is common in children on therapy. Therefore, it follows that early and close attention to nutritional status is essential for optimal outcome. Intervention for the newly diagnosed child with cancer and malnutrition ought to include the raising of the caloric and nutrient intake to the age-appropriate demands for growth and development. ATTEMPTS AT CURATIVE THERAPY Therapy for pediatric cancers is intensive. While surgery and radiotherapy are used commonly, the vast majority of pediatric oncology regimens are based primarily on chemotherapy. In addition, over the years those regimens have become more and more intensive. As a result, children who receive such treatment need careful monitoring and early nutritional intervention. This is all the more important because malnutrition is a contributing factor to the immunological deficiencies encountered in children with cancer who are on antineoplastic therapy. Etiologic considerations As in all diseases or disorders, therapy should be directed to the cause. However, there are layers of etiology. Nausea and vomiting are major contributors to the development of malnutrition. These symptoms are most often induced by chemotherapy and therefore iatrogenic. Antiemetics may not be effective and the management of iatrogenic malnutrition is not always simple. First, the concerns *Correspondence to: 3504 Ruland Place, Nashville, TN 37215-1812, USA. Fax: (615) 298-5414. NUTRITIONAL SUPPORT IN CHILDREN WITH CANCER 67 TABLE I – STAGES OF CANCER EVOLUTION AND APPROACHES TO NUTRITIONAL SUPPORT Evolution of cancer Time of diagnosis Administration of therapy with curative intent Completion of therapy Remission Persistent or progressive disease Cure Nutritional issues Approach to nutritional intervention Malnutrition is generally independent of cancer Malnutrition is multifactorial but often iatrogenic Address malnutrition independent of cancer therapy Nutritional support must include attention to social and psychogenic factors Malnutrition is multifactorial but often iatrogenic Malnutrition is multifactorial but often iatrogenic Malnutrition is sometimes tumor-induced Malnutrition is usually independent of previous cancer Unusual dietary habits may occur Nutritional rehabilitation must address psychogenic and social factors Nutritional support may be desirable of the family may result in counterproductive control issues. Nutrition and nurture are synonymous in the minds of many parents. Second, anticipatory vomiting is a real phenomenon, conditioned by chemotherapy. Third, bribing with favorite foods (to which parents often resort), and the association of these foods with nausea and vomiting, result in the phenomena of learned food aversions and psychogenic food refusal. The classical studies of Bernstein (1978, 1982) brought these problems to the attention of pediatric oncologists. It is a truism that a malnourished child needs to be refed, even when that child is undergoing intensive cancer therapy. However, that may be a far more difficult task than merely supplying the requisite calories and nutrients. When you add to that task the tendency of families toward non-compliance with medical advice, the challenge becomes even more daunting. Just feeding without social and psychological support is often doomed to failure. Nutritional considerations in chemotherapy Much of chemotherapy is based on ‘‘nutritional’’ principles. When we talk about nutrition we should mean overall, whole body nutrition. However, chemotherapy often exploits nutrition at the cellular level. A specific nutrient, considered to be required in the overall intake, may be of greater importance for one cell than another. Some compounds are not dietary essentials because they are manufactured in one organ, but they are necessary for another. We exploit those observations in chemotherapy. Methotrexate (a dihydrofolate reductase inhibitor) and other antifolates are effective because cancer cells need more tetrahydrofolate than ‘‘normal’’ cells. If we overdo the methotrexate-induced block we can bypass the effect of the drug by giving tetrahydrofolic acid in the form of leucovorin (citrovorum factor). Asparaginase is an effective antineoplastic agent because of the dependence of certain malignant cells on preformed asparagine, in contrast to normal tissues, which can synthesize this ‘‘non-essential’’ amino acid. To translate this approach to chemotherapy into nutritional manipulation as anticancer treatment remains an area that requires much more research. For example, iron deprivation can result in tumor regression. But iron overload can promote infection, while iron deficiency predisposes to bacterial growth. Furthermore, during formative years, iron deficiency can cause significant developmental deficiencies. All this is complicated by the fact that measures of iron status in the child with cancer are difficult to interpret, because ferritin is a tumor marker and, like transferrin, is an acute phase reactant that is sensitive to protein/calorie malnutrition (Garcia et al., 1989). Dietary iron was the nutrient lowest in intake for 277 patients at referral to the University of Texas M.D. Anderson Cancer Center for both children with solid tumors and those with hemopoietic malignancies, with a range of 70–78% of the recommended dietary requirement (Carter et al., 1983b; Garcia et al., 1989). It is not immediately clear whether iron supplementation in such circumstances is warranted. Such considerations may seem very different from the use of the many supplements and special diets that parents are prone to give Social and psychogenic factors must be addressed TABLE II – RATES OF SEPSIS IN CHILDREN WITH CANCER ON HYPERALIMENTATION1 Patient group All well-nourished All malnourished All control All IVH3 Well-nourished on IVH Malnourished on IVH Days on protocol2 Cases with sepsis Sepsis rate per 100 protocol days2 p 99.5 ⫾ 89.8 4 0.12 ⫾ 0.44 72.7 ⫾ 57.5 10 1.07 ⫾ 0.74 ⬍0.001 113.5 ⫾ 97.2 1 0.01 ⫾ 0.05 74.7 ⫾ 62.3 13 0.71 ⫾ 1.44 ⬍0.01 79.5 ⫾ 67.0 3 0.25 ⫾ 0.63 71.2 ⫾ 58.3 10 0.98 ⫾ 1.75 ⬍0.05 1Derived from Van Eys et al. (1980).–2Mean ⫾ SD.–3IVH, intravenous hyperalimentation. their children. The notion that a special diet can cure cancer is alive and well among the laity. However, if physicians view nutritional support too much from the viewpoint of influencing the tumor, they are no better. Viewing nutritional support as a way of improving the well-being of a child is, in the final analysis, more helpful. Individual nutrients Finally, certain nutrients are especially beneficial to certain organs. The classical example is glutamine, which is considered valuable in the protection of the intestinal mucosa during radiation and chemotherapy (Wilmore, 1997). In the selection of nutritional support, one has to take into account that specific deficiencies of individual nutrients may exist. Again, a classical example may suffice: magnesium deficiency occurs frequently after the use of platinum-containing chemotherapeutic regimens. Intervention strategies There are 3 dogmas to remember in nutritional support: (1) Nutrition is not medicine in the eyes of the patient and his family. You are feeding a person. (2) Nutritional support must be tailored to the needs of the individual patient. This includes his psychosocial and specific nutrient needs. (3) Nutrition should be supplied in the least invasive and most physiological way. The gut should be used if at all possible. Management should always be proactive. The intervention should be predicated on the future health status of the patient and not just on his or her status before therapy starts. Support personnel often try to bargain with patients about feeding tubes or hyperalimentation, even when it is certain that these are going to be needed. Hyperalimentation does result in a heightened exposure to infection because of the venous access required. However, not to feed results in a heightened susceptibility to infection also. In a prospective randomized trial of hyperalimentation, as adjuvant to chemotherapy for children with a bone tumor relapse, the increased number of infections did not correlate primarily with intravenous VAN EYS 68 hyperalimentation but with the nutritional status of the patient (Van Eys et al., 1980). This experience is summarized in Table II. In pediatric bone marrow transplantation protocols, proactive initiation of intravenous hyperalimentation remains routine (Weisdorf et al., 1987). The conditioning regimen for bone marrow transplantation is a severe physiological insult and thermodynamic deaths will occur if nutritional status is not maintained. A thermodynamic death is a death caused by insufficient energy being available to sustain the function and/or survival of vital cells. There was a time when intravenous hyperalimentation was thought to be beneficial per se in cancer therapy. However, its use is generally limited now to patients with specific indications (Archer et al., 1996) for metabolic complications and infectious problems are still all too frequent. It has been suggested that intravenous hyperalimentation be used to improve the effectiveness of cytotoxic therapy by giving the intravenous nutrients intermittently, just before a chemotherapeutic cycle. It was reasoned that the stimulus nutrition might give the tumor cells may make the cells more susceptible to chemotherapy (Copeman, 1994). Again, this misses an important point. It is true that the more invasive the intervention, the more likely that complications will arise. However, it is also true that the more malnourished a child is, the more complications arise. Nutrition is essential to staying alive. In the context of cancer therapy, that reality does not change. The emphasis should be on the child, not the cancer. COMPLETION OF THERAPY Remission When a child has finished treatment, remaining in continuous complete remission, nutritional rehabilitation still may be needed. By this time, the family dynamics may be such that severe abnormal feeding and eating patterns exist. Nutritional guidance and graded intervention are extremely important in order to minimize the long-term effects of cancer and its therapy. Relapsed, progressive cancer Unfortunately, some children have unrelentingly progressive cancer. In that instance, tumor-induced cachexia can become a major issue. While malnutrition is often iatrogenic, the tumor burden can add an extra demand. If further therapy is planned, the approach ought not to be any different from that taken for a child undergoing therapy with curative intent. If a phase I evaluation of a potential new therapy is planned, other challenges arise. Nutritional status should be documented carefully. These reports should indicate whether nutritional support was instituted and, if so, what was given and how. This is especially important when pharmacokinetics are to be determined. When the child is considered beyond hope, the decision for nutritional support can become both agonizing and confused. It is extremely important that communication among caregivers is explicit and that values are clarified. Is nutritional support a medical intervention or an act of compassion and nurture? The American Medical Association has taken the position that nutritional support is a medical intervention. As an illustration of how confused the discussion can become, I recall the circumstances in Houston, where we cared for a child with a relentlessly progressive tumor of the cervical spinal cord which was unresponsive to radiation and chemotherapy and not amenable to surgery. The child was considered socially unacceptable for inclusion in a home hyperalimentation program, especially given his incurable status. Over staff objections, it was instituted nevertheless. The boy had 18 months of totally uneventful home hyperalimentation and was satisfied with his quality of life. The same staff, who objected to keeping him alive with intravenous nutrition, instituted totally futile cardiopulmonary resuscitation when he arrested in the hospital (Van Eys, 1987). THE CHILD CURED OF CANCER Once a child is cured of cancer, it would seem that he or she should revert to the status of the average member of the general population. However, long-term sequelae are frequent. The incidence of second malignancies exceeds that of primary tumors found in age-matched controls. Known dietary carcinogens should be avoided. Permanent malabsorption can persist. More importantly, psychosocial idiosyncrasies are very common in cancer survivors. Nutrition remains a challenge. The use of self-imposed special diets or nutritional supplements continues far beyond the time of cure. We need careful, long-term follow-up to gauge the late effects of our therapy. Dietary evaluation and nutritional advice should be integral parts of this process. CONCLUDING REMARKS Nutrition is basic to the care of any individual, let alone a sick child. Having cancer constitutes no exception. There is no magic diet that cures cancer. Nutritional therapy, in the sense of maintaining or restoring adequate nutritional status, is a form of supportive care. Were our primary therapy truly effective, such supportive therapy would not be needed. Nutrition should be viewed for what it is: supplying the most basic needs of children. No child has died from being fed appropriately, but many die of starvation. The practice of pediatric oncology should not contribute to that statistic. REFERENCES ARCHER, S.B., BURNETT, R.J. and FISCHER, J.E., Current uses and abuses of total parenteral nutrition. Adv. Surg., 29, 165–189 (1996). BERNSTEIN, I.L., Learned taste aversions in children receiving chemotherapy. Science, 200, 1302–1303 (1978). BERNSTEIN, I.L., Physiological and psychological mechanisms of cancer anorexia. Cancer Res., 42(Suppl.), 715s–720s (1982). CARTER, P., CARR, D., VAN EYS, J. and COODY, C., Nutritional parameters in children with cancer. J. Amer. diet. Assoc., 82, 616–622 (1983a). CARTER, P., CARR, D., VAN EYS, J., RAMIRAZ, I., COODY, D. and TAYLOR, G., Energy and nutrient intake of children with cancer. J. Amer. diet. Assoc., 82, 610–615 (1983b). COPEMAN, M.C., Use of total parenteral nutrition in children with cancer: a review and some recommendations. Pediatr. Hematol. Oncol., 11, 463–470 (1994). DONALDSON, S.S., WESLEY, M.N., DEWYS, W.D., SUSKIND, R.M., JAFFE, N. and VAN EYS, J., A study of the nutritional status of pediatric cancer patients. Amer. J. Dis. Child., 135, 1107–1112 (1981). GARCIA, M.E., VAN EYS, J. and WANG, Y.-M., Iron nutrition in childhood malignancy. J. parenter. enteral Nutr., 13, 162–167 (1989). UDERZO, C., ROVELLI, A., BONOMI, M., BARZAGHI, A., STRADA, S., BALDUZZI, A., PIROVANO, L. and MASERA, G., Nutritional status in untreated children with acute leukemia as compared with children without malignancy. J. pediatr. Gastroenterol. Nutr., 23, 34–37 (1996). VAN EYS, J., Extraordinary therapy and social status—an instructive example. J. psychosoc. Oncol., 5, 31–38 (1987). VAN EYS, J., COPELAND, E.M., CANGIR, A., TAYLOR, G., TEITELL-COHEN, B., CARTER, P. and ORTIZ, C., A clinical trial of hyperalimentation in children with metastatic malignancies. Med. pediatr. Oncol., 8, 68–73 (1980). WEISDORF, S.A., LYSNE, J., WIND, D., HAAKE, R.J., SHARP, H.L., GOLDMAN, A., SCHISSEL, K., MCGLAVE, P.B., RAMSEY, N.K. and KERSEY, J.H., Positive effect of total parenteral nutrition on long-term outcome of bone marrow transplantation. Transplantation, 43, 833–838 (1987). WILMORE, D.W., Metabolic support of the gastrointestinal tract: potential gut protection during intensive cytotoxic therapy. Cancer, 79, 1794–1803 (1997).