8392

Saturday 30 June 1984

CANCER CACHEXIA AND PROTEINMETABOLISM

MALAYAPPA JEEVANANDAMSTEPHEN F. LOWRY

GLENN D. HOROWITZMURRAY F. BRENNAN

Department of Surgery, Memorial Sloan-Kettering Cancer Center,1275 York Avenue, New York,

New York 10021, USA

Summary Metabolic abnormalities arising from mal-nutrition and malignancy are seen in

patients with cancer cachexia. To discriminate between theeffects of pure malnutrition and tumour-specific metabolicalterations, the kinetics of whole-body metabolism weredetermined in 7 untreated, malnourished cancer patients andin 11 patients with benign disease, also malnourished, bymeans of primed constant infusion of15 N-glycine. Patients inboth groups received 500 kcal/day as 5% glucose solutionwithout any nitrogen. Whole-body protein turnover was 32%and 35% higher in the cancer patients than in the non-cancerpatients and starved normal subjects respectively. Similarly,the rate of protein synthesis was 35% and 54% higher in thecancer patients than in the non-cancer patients and starvednormal subjects respectively. Our study supports the viewthat aberrations of host metabolism in patients with cancercachexia are different from those in non-cancer patientssuffering equivalent weight loss.

Introduction

CANCER cachexia comprises several conditions includinganorexia, asthenia, loss of lean body tissues, and the inabilityto alter some metabolic regulatory functions appropriate tothe level of stress and nutritional status. Several explanationshave been proposed for the pathogenesis of cachexia in cancerpatients.1-7 The biological importance of humoral mediators,and proteolytic and growth factors in the evolution of thecancer cachexia syndrome2 has yet to be determined. The

progressive weight loss observed in cancer patients cannot befully explained by diminished food intake alone.6 Moreover,it is not clear whether the malnourishment that accompaniescancer results in distinct metabolic alterations or more closelyresembles those patterns observed in unstressed or stressedmalnutrition. Malnutrition is not unique to cancer: it affects30-50% of hospitalised patients,8 and most chronic, fatal,non-neoplastic diseases terminate in a cachectic state. Thereis evidence that the normal adaptive mechanisms seen inpatients with simple starvation are impaired in the patientwith uncontrolled cancer growth. Abnormalities in energy,carbohydrate, lipid, and protein regulation may ultimately bemanifested in the host organism during progressive tumourgrowth.’

Protein serves as a critical reserve of metabolic fuel whichbecomes depleted during progressive malnutrition, andresearch efforts have been focused on the altered proteinkinetics in cachectic cancer patients. Standard nitrogenbalance studies have yielded conflicting results, which areinherently difficult to interpret in the cancer population.9 Re-modelling of body protein may be an important aspect ofmetabolism in cancer patients, since a positive nitrogenbalance does not necessarily result in maintenance or

repletion of lean body mass.Reports on whole-body protein turnover (WBPT) in

cancer patients have shown conflicting results.1O-16 Thesedifferences are probably explained by differences in the

patient populations studied. For appropriate comparisonsbetween groups, the nutritional status, weight loss, physicalactivity, age, and body weight should be matched.

In this study WBPT was measured in 7 malnourishedcancer patients, 11 malnourished patients with benigndisease, and 5 healthy normal subjects who were starved for10 days,17 by analysing Nitrogen- 15 in both urinary urea andammonia during a primed constant infusion of

15N-glycine.18,19 Protein synthesis and catabolism rates werecalculated from WBPT and nitrogen intake and excretion.’9

1424

Materials and Methods

Patients

7 untreated cancer patients with associated malnutrition and 11 Ipatients with varying degrees of malnutrition associated with

benign disease were studied before the start of treatment. All werejudged to be free of other metabolic or endocrine disorders, andthere was no clinical evidence of infection. None had experiencednutritional supplementation or had been treated with anti-

neoplastic drugs. All patients were admitted either to the MemorialSloan-Kettering Cancer Center or to the clinical research center ofthe New York Hospital-Cornell Medical Center. All patients wereinformed about the purpose and procedure of the study, and they allgave informed written consent. The protocol was approved by therespective institutional review boards of the Memorial Sloan-

Kettering Cancer Center and New York Hospital-Cornell MedicalCenter. The study was initiated on the first day of admission, andeach patient received continuous infusion of 5% glucose solutionproviding about 500 kcal/day. This level of infusion was maintainedthroughout the study. Under these conditions, the rate of WBPTequals the protein catabolic rate. There was no nitrogen intakeduring either the 24 h before isotope infusion period or the 30-36 hisotope infusion. Water was allowed ad libitum but no othernutrient was provided. WBPT in each patient group was comparedwith that in a group of 5 healthy volunteers who underwent a 10-dayperiod of starvation.

Experimental DesignWBPT was estimated in each patient by means of a modification 18

of the Picou and Taylor-Roberts method. 19 The technique consistsof the administration of an initial dose of 15N-glycine followed by acontinuous infusion for 30-36 h, and separate measurements ofisotope excretion in urinary urea and ammonia. Before isotopeinfusion a urine sample was collected so that baseline enrichment of15N in urea and ammonia could be measured. An initial dose of0-45mg 15N/kg as sterile, pyrogen-free 15N-glycine (99 atom percent15N) was administered intravenously, followed by the continuousinfusion of 15N-glycine (0 - 00025 mg 15N kg - 1 min -I) solution insaline by means of a peristaltic pump for 30-36 h. The prime toinfusion ratio was 1800:1. Urine was collected every 4 h for the first16 h, then every 2 h, in containers with antibacterial, urease-inhibiting sodium fluoride tablets. The samples were processed andstored at -20°C and then analysed.

Analytical MethodsA 10% portion of each urine sample was pooled and analysed for

total nitrogen in a Technicon Autoanalyser after Kjeldahl digestion.The urinary urea and ammonia in each sample were isolated,19,20and the 15N was determined in a 60°, Nier type, dual collectorisotope ratio mass spectrometer. The 15N atom percent excess(APE) of urea and ammonia in each sample were obtained bysubtracting 15N atom percent urea or ammonia in the controlsample taken immediately before isotope administration.

Calculation

The stochastic approach of Picou and Taylor-Roberts19 was usedin the calculation of WBPT. It is assumed that, in steady-state

conditions and at isotopic equilibrium, the proportion of the isotopeexcreted in an end-product of nitrogen metabolism (such as urea orammonia) is the same as the proportion of the flux excreted in thesame end-product. Since steady-state plateau values of APE in ureaand ammonia are not similar and their relative contributions to thecalculation of WBPT are not clearly established, it is reasonable togive then equal weighting.’ 11,22 Thus we have used an "end-productaverage" technique 11,14, 7,18,22 in the calculation, in which there isan averaging of the plateau values of APE in urinary urea andammonia. The isotopic plateau values in urea and ammonia werereached after 18±2 h in all cases. The average value of 4-9 samplestaken during the last 10-18 h of tracer infusion was used to calculatethe plateau values. The mean coefficient of variation (standarddeviation as percent of mean enrichment) of the end-productaverage measurement is 4%.The flux or nitrogen turnover rate Q (mg N kg- 1 min - I) is

defined as the flow of nitrogen into, and out of, the metabolicnitrogen pool under steady-state conditions and is calculated bydividing the isotope infusion rate I (mg 15Nkg-1 min-I) by theplateau 15N APE in the urinary end-product. Thus:

Q=100x I/APE (1)

This simplified approach has been proven valid for assessingWBPT in human beings of all ages, under a variety of clinicalconditions. In this model, under steady-state conditions, Q is thesame as the catabolic rate C, since there is no dietary nitrogen intake(IN) in these patients; and synthetic rate S equals Q minus theurinary nitrogen excretion Et Thus:

Q=(C+ IN)=(S+ Et) (2)All group data are presented as the mean±SEM. Statistical

significance was ascertained by using unpaired Student’s t test.23

Results

The clinical status of each patient is given in table I andtable II. Metastatic liver disease was not present at subsequentlaparotomy in any of the cancer patients. The mean age was63±4 years for cancer patients and 53±4 for non-cancer

patients. All the subjects were judged to be clinicallymalnourished at the time of their referral, and weight lossaveraged about 20% in both groups. This degree ofnutritional depletion is often associated with a decline incirculating serum albumin concentration, and this was

observed in both patient groups (3’3±0-3 3 g/dl), whichsuggests that they were about equally malnourished. Plasmalevel of glucose was significantly higher (103±5 mg/dl) in thecancer group than in the non-cancer group (89±2 mg/dl)(p=0 O1).The individual rates of WBPT (Q), synthesis (S),

catabolism (C), and the nitrogen balance measured during thestudy are presented in table III. Mean values of thesemeasurements for the two patient groups are presented withthe results for the starved normal subjects in table IV. Thisgroup represents an unstressed normal population whichunderwent acute weight loss through malnourishment and

TABLE I-DIAGNOSIS AND HISTORY OF CANCER PATIENTS

- I

*BUN= blood urea nitrogen.

1425

TABLE II-DIAGNOSIS AND HISTORY OF NON-CANCER PATIENTS

TABLE III-PROTEIN KINETIC MEASUREMENTS*

*g protein kg- I day-tmg kg- 1 daY_ 1..

adapted to the starvation state. WBPT in the cancer patientswas 3 - -18±0’ 29 g protein kg- day- I, which was 32% higherthan in the non-cancer group and 35% higher than in thestarved normals (p=0-025). Similarly the whole-bodyprotein synthesis rate of 2’65 t 0 - 32g protein kg-1 day-1

was 35% higher than in the non-cancer group and 5407o higherthan in the starved normal subjects (p=0-025). Thepercentage of endogenous nitrogen from protein catabolismwhich is recycled into protein synthesis is 8307o in cancerpatients, 80% in non-cancer patients, and 76% in starvednormal subjects.

Discussion ’

WBPT is the sum of the metabolism of many individual

proteins. This measurement may have a practical value in thestudy of nutrition, growth, and the physiology of diseaseswhich are known to affect protein metabolism adversely.Adaptation of protein metabolism may be a key mechanismwhereby the body economises on energy expenditure. Thehuman response to acute malnutrition is a reduction in

energy-requiring protein turnover and synthesisy,24 Manycancer patients are chronically malnourished and sufferconsiderable weight loss, and it follows that WBPT is

correspondingly reduced. However, the direction of theeffects arising from the presence of neoplastic disease is notknown.

Protein kinetics in starvation-adapted normal subjects andin chronically malnourished but otherwise unstressed

patients with benign disease are found to be similar. We havedocumented the achievement of isotopic equilibrium underthe conditions imposed by this study and therefore proposethat our methods accurately reflect the whole-body proteinkinetics in these populations.Since serial evaluation of patients during the evolution of

cancer is not feasible, we have addressed only those changesobserved after the onset of cancer cachexia. The cancer

patients evaluated in this study were found to be

TABLE IV-COMPARISON OF NITROGEN DATA FROM MALNOURISHED SUBJECTS*t T

*Mean (&plusmn;SEM).tsee Rose et al.1’:p=<0’25 versus non-cancer controls or starved normal subjects.Sp=<0-025 versus cancer patients.

1426

hyperglycaemic and also demonstrated an overallacceleration of protein kinetics.The mean urinary nitrogen loss during glucose infusion

was slightly greater in cancer patients than in the non-cancercontrols, but less than that in starved normal subjects (tableIV). The cancer patients utilise a higher fraction (S/Q) of theprotein flux for synthetic purposes, and have a potentiallyhigher rate of nitrogen utilisation in comparison withcontrols. In human tumours, the rate of protein synthesis issimilar to that in the normal tissues from which they arise.24This suggests that tumour growth alone is unlikely to cause agreat increase in the whole-body synthesis of protein. It canbe speculated that other systemic processes induced bytumour growth may lead to the acceleration of proteinkinetics in host tissues.

During the rapid growth phase of recovery from mal-nutrition in children25 and the catabolic phase of severetrauma,26 sepsis,27 burns,28 and chronic infection29 proteinsynthesis and breakdown rates are higher than normal. In thisstudy, accelerated protein synthesis and breakdown rateshave been shown to be accelerated in cancer cachexia; also achange in body nitrogen content is accompanied by a

simultaneous increase in both synthesis and breakdown rates,but one changes more than the other, resulting in a net gain orloss. This translocation of nitrogen which takes place throughformation and utilisation of amino-acids is an energy-demanding process and may result in an increased metabolicrate and an expansion of the metabolically active pool ofnitrogen as observed in severe trauma and sepsis.26,27 Thissupports the view that an increase in protein breakdown is thebody’s initial response to stressed states and is an attempt tomobilise body nitrogen. However, protein synthesis is

extremely labile, and alterations in synthesis rates are themajor determinants of changes in body protein, weight, andnitrogen balance.

Accelerated protein synthesis and enhanced proteolysis inthe cancer patient may produce the severe imbalance in thelevel, as well as the utilisation rate, of free amino-acids. Anincrease in peripheral tissue release of glycine duringprolonged starvation resulted in elevated levels of glycine inarterial plasma.30 Cancer patients with weight loss showed alow level of arterial blood glycine3I,32 in comparison withnon-cancer controls and starved normals. A large reduction(62%) in plasma alanine levels has been noted in cases ofprolonged starvation,30 However, the levels of this

gluconeogenic amino-acid in cancer patients did not differfrom those of non-cancer patients,31 but were up to 40% lessthan those in starved normal subjects,3I,32 Malnutrition ischaracterised by a reduction in arteriovenous amino-aciddifferences and reducd plasma levels of essential branched-chain amino-acids.33 Raised levels of valine, leucine, andisoleucine were noticed during the first week of starvation,but they had declined after 5-6 weeks.3&deg; The levels of theseamino-acids were considerably higher in the cancer patientsthan in the non-cancer patients31 or the starved normals.32These observations support the view that the metabolic

changes arising in cancer patients are distinct from thoseassociated with simple starvation or benign disease.This study distinguishes between the effects of WBPT

arising from malnutrition and those arising from WBPT incancer patients. Cancer cachexia, as distinct from simplestarvation or malnutrition imposed by benign disease, seemsto give rise to metabolic changes similar to those seen duringmajor injury, sepsis, or burns. These changes includedisturbances of whole-body carbohyrdate, lipid, and proteinmetabolism and result in accelerated gluconeogenesis and

oxidation, protein synthesis and degradation, fatty-acidmobilisation and oxidation, and metabolic rate.

We thank Ms Gail Kuhlman and Ms Kathy Tobin for their skilled technicalhelp.This work was supported by grants from The American Cancer Society and

the National Institutes of Health to the General Clinical Research Center,Cornell University-New York Hospital.

Correspondence should be addressed to M. J.

REFERENCES

1. Brennan MF. Uncomplicated starvation versus cancer cachexia. Cancer Res 1977, 37:2359-64

2. Lawson DH, Richmond A, Nixon DW, Rudman D Metabolic approaches to cancercachexia. Ann Rev Nutr 1982, 2: 227-301.

3 Stein TP Cachexia, gluconeogenesis and progressive weight loss in cancer patients. JTheo Biol 1978; 73: 51-59.

4. Theologides A. Pathogenesis of cachexia in cancer&mdash;a review and a hypothesis Cancer1972,29:484-88

5. Waterhouse C. How tumors affect host metabolism. Ann NY Acad Sci 1974; 230:86-93

6. Costa C, Bewley P, Aragon M, Siebold J Anorexia and weight loss in cancer patients.Cancer Treat Rep 1981; 65 (suppl 5). 3-7

7. Brennan MF Total parenteral nutrition in the cancer patient. N Engl J Med 1981, 305:375-82

8. Bistrian BR, Blackburn GL, Vitale J, Cochran D, Naylor J Prevalence of malnutritionin general medical patients. JAMA 1976, 235: 1567-70

9 Watkin DM. Nitrogen balance as affected by neoplastic disease and therapy Am J ClinNutl 1961, 9: 446-60.

10. Carmichael MJ, Claque MB, Keir MJ, Johnston ID A whole-body protein turnoversynthesis and breakdown in patients with colorectal carcinoma. Br J Surg 1980, 67:736-39.

11. Glass RE, Fern EB, Garlic PJ Whole-body protein turnover before and after resectionof colorectal tumors. Clin Sci (Lond) 1983; 64: 101-08

12. Heber D, Chlebowski RT, Ishibashi PE, Herrold JN, Block JB. Abnormalities inglucose and protein metabolism in noncachectic lung cancer patients. Cancer Res1982: 42: 4815-19.

13. Norton JA, Stein TP, Brennan MF. Whole-body protein synthesis and turnover innormal man and malnourished patients with and without known cancer Ann Surg1981, 194: 123-28.

14. Klein CL, Camitta BM. Increased whole-body protein turnover in sick children withnewly diagnosed leukemia or lymphoma Cancer Res 1983; 43: 5586-92.

15. Burt ME, Stein TP, Brennan MF. A controlled randomised trial evaluating the effectsof enteral and parenteral nutrition on protein metabolism in cancer-bearing man. JSurg Res 1983, 34: 303-14

16. Hermann VM, Garnick MB, Moore FD, Wilmore DW Effect of cytotoxic agents onprotein kinetics in patients with metastatic cancer Surgery 1981; 90: 381-87.

17. Rose D, Horowitz GD, Jeevanandam M, Brennan MF, Shires GT, Lowry SF. Whole-body protein kinetics during acute starvation and intravenous refeeding in normalman. Fed Proc 1983; 42: 1070.

18. Jeevanandam M, Brennan MF, Mihranian MH, Daly JM, Horowitz GD, Lowry SF.Tracer priming in human protein turnover studies with 15N glycine Fed Proc 1983;42: 825.

19. Picou D, Taylor-Roberts T. The measurements of total protein synthesis andcatabolism and nitrogen turnover in infants in different nutritional states and

receiving different amounts of dietary protein. Clin Sci 1969, 36: 283-96.20. Long CL, Jeevanandam M, Kinney JM. Metabolism and recycling of urea in man Am

J Clin Nutr 1978, 31: 1367-82.21 Jeevanandam M, Long CL, Kinney JM. Kinetics of intravenously administered

15N-1-alanine in the evaluation of protein turnover. Am J Clin Nutr 1978, 31:1367-82.

22. Fern EB, Garlick PJ, McNurlan MA, Waterlow JC. The excretion of isotope in ureaand ammonia for estimating protein turnover in man with 15N glycine. Clin Sci 1981;61: 217-28

23. Snedecor GW, Cochran WR. Statistical methods. Ames, Iowa- Iowa State Universitypress, 1971.

24. Mullen JL, Buzby GP, Gertner MH, Stein TP, Hargrove WC, Oram-Smith J, RosatoFF. Protein synthesis dynamics in human gastrointestinal malignancies Surgery1980;87: 331-38

25 Golden MHN, Waterlow JC, Picou D. Protein turnover, synthesis and breakdownbefore and after recovery from protein-energy malnutrition. Clin Sci 1977; 53:473-77.

26 Birkhahn RH, Long CL, Fitkin D, Jeevanandam M, Blakemore WS. Whole-bodyprotein metabolism due to trauma in man as estimated by L-15N-alanine Am JPhysiol 1981, 241: E64-E71

27. Long CL, Jeevanandam M, Kim BM, Kinney JM. Whole-body protein synthesis andcatabolism in septic man. Am J Clin Nutr 1977; 30: 1340-47.

28 Klein CL, Young VR, Rohrbaugh DK, Burke JF. Increased rates of whole-bodyprotein synthesis and breakdown in children recovering from burns. Ann Surg 1978,187: 383-91

29. Tomkins AM, Garlick PJ, Schofield WN, Waterlow JC. The combined effects ofinfection and malnutrition on protein metabolism in children Clin Sci 1983; 65:313-24

30. Felig P, Owen OE, Warren J, Cahill GF, Jr Amino acid metabolism during prolongedstarvation J Clin Invest 1969, 48: 584-94.

31. Clarke EF, Lewis AM, Waterhouse C. Peripheral amino acid levels in patients withcancer. Cancer 1978; 42: 2909-13.

32. Burt ME, Aoki TT, Gorschboth CM, Brennan MF Peripheral tissue metabolism incancer-bearing man. Ann Surg 1983, 198: 685-91.

33. Smith SR, Pozefsky T, Chhetri MK Nitrogen and amino acid metabolism in adultswith protein-calorie malnutrition. Metabolism 1974, 23: 603-18.