A Supplement to Alkaline Phosphatase Fractionations: Utilization of Gamma- Glutamyl Transpeptidase and Hydroxyproline Assays

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 8, No. 2 Copyright 1978, Institute for Clinical Science A Supplement to Alkaline Phosphatase Fractionations: Utilization of Gamma- Glutamyl Transpeptidase and Hydroxyproline Assays D. SLAUNWHITE, Ph.D., R. L. TUGGEY, Ph.D., and GUSTAVO REYNOSO, M.D. Department o f Pathology, Charles S. Wilson Memorial Hospital, Johnson City, NY ABSTRACT Fractionation of serum alkaline phosphatase by either biochemical or electrophoretic techniques is often insufficient to determ ine the source of the elevated serum enzyme. In 31 unselected patients with high serum alkaline phosphatase a sim ultaneous determ ination of serum gamma glutamyl transpeptidase and urinary hydroxyproline excretion rates was performed, in addition to urea dénaturation and L-phenylalanine inhibition tests. Of 25 patients with definite diagnosis, the urea dénaturation test correctly identified the source of the elevated serum alkaline phosphatase (ALP) in 16 (64 percent). The serum gamma-glutamyl transpeptidase (GGTP) alone predicted the correct diagnosis in 64 percent of the cases and the urinary hydroxyproline (HOP) alone in 69 percent. W hen all three tests were performed simultaneously, the combination of results identified the source of the ALP in 88 percent of the cases. It is believed that this combination offers better sensitivity and specificity than any of the three tests used individually. Introduction Serum alkaline phosphatase (ALP) is a mixture of various tissue specific isoenzymes among which the principal ones are produced in the liver, bone, placenta and intestine. In order to ascertain the origin of an elevated ALP when either the liver or the skeleton may be involved by disease, laboratories have used a variety of procedures such as heat fractionatio n,13 urea d e n a tu ratio n 8 a n d e le c tro p h o re tic se p a ra tio n.3 All such techniques have b een criticized because of poor correlation betw een the results of the assay and the presum ed organ(s) of involvement and also because there is much overlap of results betw een groups o f patients w ith different organ involvem ent.5, 12,18 U rea d en atu ratio n, the p ro ced u re utilized in this study, classifies the results into three categories: liver pattern, bone pattern and mixed pattern. W hen a mixed pattern is obtained, the physician is unable to ascertain w hether bone or liver or both are responsible for the ele-

2 11 8 SLAUNW HITE, TUGGEY AND REYNOSO vation, and the test is therefore of no clinical utility. In order to substantiate the validity of the urea denaturation test in assigning a patient to the liver or bone categories, and to assist in resolving the origin(s) of the ALP in the mixed patterns, serum gamma-glutamyl transpeptidase (GGTP) and hydroxyproline (HOP) assays were utilized in all cases of elevated ALP w here fractionation by the urea m ethod seem ed indicated. GGTP was chosen because it is a sensitive indicator of hepatobiliary disease and, further, because it is not elevated in prim ary diseases of the skeleton, the placenta or the intestinal mucosa.3,10,15 Several investigators have th erefo re utilized a determ ination o f GGTP in order to ascertain whether, in a particular case, the elevation in ALP was secondary to hepatic or hepatobiliary disease.1,17 Although this additional assay aids in the differentiation of the source of the e le vated ALP, two unansw ered questions remain: (1) if the GGTP is elevated, can the ALP elevation still be the result of elevations of both liver and bone isoenzymes? and (2) if the GGTP is normal, is the ALP elevation necessarily due to bone involvem ent? Increased urinary HOP excretion rates, on the other hand, are observed only in conditions associated w ith increased collagen turnover rates and are, therefore, a good indicator of osteoblastic activity, since over 70 percent of the total body collagen is located in the bone matrix. In localized lesions such as metastatic disease and fractures, there is a clear relationship b etw een the urinary HOP excretion rate and the extent of bone involvement. In metabolic bone disease, patients with elevated HOP values have universally increased bone turnover rates, whereas in cases of osteoporosis in which the turnover rate is usually normal, the excretion of HOP is normal. Diseases in which elevated HOP values may be encountered include Paget s disease of the bone, Marian s syndrome, acromegaly, osteom alacia, hyperthyroidism and skeletal m alignancy, both primary and secondary.4 Several investigators have dem onstrated excellent correlation betw een urinary hydroxyproline excretion rates and the extent of pathological involvement in metastatic bone disease, in m any instances even before radiological evidence of bone m etastasis is observed.7,9,11 M ethods and Materials The population studied consisted of 31 unselected patients with elevated ALP of at least 200 IU per L. ALP and GGTP were m easured kinetically at 37 C on a G ilford M odel 2400-S sp e ctro p h o to meter.* The ALP and GGTP w ere assayed by the m ethod of Bow ers and McComb2 and Szasz,18 respectively. The m ethod for the urea denaturation of ALP was based on the procedure of Posen.8 Serum (0.2 ml) and 6 M urea (0.2 ml) were mixed, incubated for 18 minutes at 37 C and then 0.1 ml of the mixture was u tiliz e d in the ALP assay. T he L-phenylalanine (LPA) inhibition of intestinal ALP was based on the procedure of Fishm an.6 Serum (0.1 ml) and LPA/ AMP (2.7 ml) were incubated at r.t. for 20 minutes. Exactly 0.2 ml of p-nitrophenylphosphate substrate was then added and the reaction assayed kinetically at 37 C. The serum assays w ere perform ed on fresh serum or serum stored for less than two days at 4 C. Urinary hydroxyproline was determ ined according to the m ethod of Prockop and U denfriend,14 All patients were placed on a hydroxyprolinefree diet for 48 hours w ith the urine collected in HC1 during the second 24 hour period. The normal GGTP ranges for males and females were 0 to 45 IU per L and 0 to 40 IU per L while ALP and HOP were * Gilford Instrum ent Labs, Inc., O berlin, OH

3 SU PPLEM ENT TO ALP FRACTIONATIONS to 110 IU per L and 9 to 45 mg per 24 hours, respectively. ALP urea fractionations, which are expressed as percentage of activity remaining, are interpreted as follows: bone (0 to 15 percent), mixed (16 to 34 percent) and liver (35 to 60 percent). LPA rem aining activities below 80 percent are considered suggestive of intestinal origin of the ALP.15 Results The 31 patients were divided accordingly w ith the results of the urea dénaturation test in 3 groups: liver, bone or mixed patterns as shown in tables I, II and III. The discharge diagnoses were arrived at by a com bination of clinical, radiological and laboratory findings and were verified by biopsy, autopsy and/or com plete m etabolic w orkup w h en appropriate. The person interpreting the results of the tests which form part of this study had no knowledge of the results of the other studies or of the final diagnosis. O f the 31 patients w ith elevated alkaline phosphatase, the urea-resistant fraction was greater than 35 percent (liver pattern) in nine. Eight had definite and one had probable liver disease. A bone pattern (urea resistant fraction of 15 percent) was observed in seven patients. All seven had dem onstrable bone disease with osteoblastic activity. In 15 patients, the urea denaturation p attern was b e tween 16 and 35 percent (mixed pattern). O f these 15 patients, definite liver disease was present in five and in all five the elevated GGTP correctly predicted the diagnosis. In three patients w ith osteoblastic bone disease and no docum ented liver or biliary tract abnormalities, the GGTP was elevated in all three and the HOP in two. One patient with metastatic disease of both skeleton and liver had elevated GGTP but normal HOP. No definite diagnosis could be established in the remaining six cases. All six had mixed urea denaturation patterns. There were four instances of elevated GGTP and two of elevated HOP in this last group. Discussion Addition of the GGTP test to patients w ith elevated alkaline phosphatase contributed to an additional diagnosis of liver disease in five patients and addition of hydroxyproline excretion test contributed to an additional diagnosis of bone disease in two. Altogether there were nine elevations of GGTP in patients TABLE I Patients with Liver Pattern in the Urea Denaturation Test* U rea LPA C a se ALPf (P e r c e n ta g e ) (P e r c e n ta g e ) GGYPfl HOPx D is c h a r g e D ia g n o s is Small bowel obstruction L i v e r, metastatic adenocarcinoma Laennec's cirrhosis Laennec's cirrhosis Chronic passive congestion of liver Cirrhosis, micronodular ' Multiple liver granulomata Liver, metastatic adenocarcinoma Viral hepatitis Figures are percent of original activity remaining after denaturation with urea or inhibition with L-phenylalanine. GGTP values are in IU per L (normal up to 45) and HOP in mg per 24 hours (normal up to 45). For interpretation of denaturation pattern, see text. tserum alkaline phosphatase L-phenylalanine UGamma-glutamyl transpeptidase xhydroxyproline

4 120 SLA U N W H ITE, TU G G EY AND REYNOSO TABLE I I Patients with Osteoblastic Pattern in the Urea Dénaturation Test* U rea LPA% C a se ALPi ( P e r c e n ta g e ) ( P e r c e n ta g e ) GGTPf ffopx D is c h a r g e D ia g n o s is Paget's disease of bone Bone - metastatic adenocarcinoma Paget's disease of bone Acromegaly Ulnar fracture Paget's disease of bone Compression fracture of lumbar spine Figures are percent of original activity remaining after dénaturation with urea or inhibition with L-phenylalanine. GGTP values are in IU per L (normal up to 45) and HOP in mg per 24 hours (normal up to 45). For interpretation of dénaturation patterns, see text. tserum alkaline phosphatase L-phenylalanine UGamma-glutamyl transpeptidase xhydroxyproline without demonstrable bone disease. In one patient with colonic carcinoma and elevated CEA, the elevated GGTP is consistent with liver cell metastasis, even though the urea denaturation test gave a mixed pattern and the biopsy was negative. It is known that an elevated CEA is a more sensitive indicator of m etastatic liver disease than a single biopsy alone.16 Of 24 elevations of GGTP, liver disease was dem onstrated in 13 and probably dem onstrated in two (predictive value 15/24 or 64 percent). Of 13 elevations of HOP, bone disease was dem onstrated in nine (predictive value 9/13 or 69 percent). TABLE III Comparison of Apo A-I Levels in Diluted Sera Heated at 52 C for Three Hours Versus Delipidated Sera Apo A -I (mq per dl) Heated Sera Sample # Heated Sera* D elipidated Serai D elipidated Sera 1 (D.D.) (D.J.) (J.K.) (A.E.) " (J.S.) (A.2.) (J.C.) (T.C.) (T.M. ) {M.M.) (D.S.) Mean ± SEM 1.02 ± 0.03 *Fresh sera was diluted in the assay buffer. Five ml of the 10 5 dilution were heated at 52 C for three hours in siliconized test tubes and stored overnight at 4 C. tsera were delipidated, dissolved by gentle mixing in 0.9 percent NaCl for 48 hours at 4 C and then diluted in assay buffer. (Adapted from Karlin et al20 with permission.) Of the 25 patients w ith definite diagnosis, the urea denaturation test correctly identified the source of the elevated ALP in 16 (65 percent). A determ ination of GGTP confirmed the diagnosis in an additional five and a test for urinary HOP excretion rate contributed two more positive diagnoses. W hen all three tests are performed, the source of the ALP was identified in 22 of the 25 patients (88 percent). The com bination of the urea denaturation test plus a determ ination of GGTP and HOP offers a better combination of sensitivity and specificity than any of the three tests performed alone. References 1. Be t r o, M., O o n, R., and E d w a r d s, J.: Gamma-glutamyl transpeptidase in diseases of the liver and bone. Amer. J. Clin. Path. 60: , Bowers, G. and McComb, R.: A continuous spectrophotometric method for measuring the activity of serum alkaline phosphatase. Clin. Chem. 12:70-89, Canapa-Anson, R. and Row e, D.: Electrophoretic separation of tissue-specific serum alkaline phosphatase. J. Clin. Path. 23: , D u l l, T. and H e n n e m a n, P.: Urinary hydroxyproline as an index of collagen turnover in bone. New Eng. J. Med. 268: , Ew e n, L.: Separation of alkaline phosphatase isoenzymes and evaluation of the clinical usefulness of the determination. Amer. J. Clin. Path. 61: , 1974.

5 SU PPLEM EN T TO A LP FRACTIONATIONS F ish m a n, W., G r e e n, S., and I n g l is, M.: L-phenylalanine: An organ specific, sterospecific inhibitor of human intestinal alkaline phosphatase. Nature 198: , G u zzo, C., PACHAS, W., et al: Urinary hydroxyproline excretion in patients with cancer. Cancer 24: , H orn e, M., Co r n ish, C., and Posen, S.: Use of urea denaturation in the identification of human alkaline phosphatase. J. Lab. Clin. Med. 72: , HOSLEY, H., et al: Hydroxyproline excretions in malignant neoplastic disease. Arch. Intern. Med. 118: , L u m, G. and G a m b in o, R.: Serum gammaglutamyl transpeptidase activity as an indicator of disease of liver, pancreas or bone. Clin. Chem. 18: , Platt, W., D oo lit t le, J., and H artshorn. J.: Urinary hydroxyproline excretion in metastatic cancer of bone. New Eng. J. Med. 271: , POSEN, S.: Alkaline phosphatase. Ann. Intern. Med. 67: , Posen, S., Ne a l e, F., and Cl u b b, J.: Heat inactivation in the study of human alkaline phosphatase. Ann. Intern. Med. 62: , P rockop, D. and UDENFRIEND, S.: Specific method for the analysis of hydroxyproline in tissues and urine. Anal. Biochem. 1: , Re y n o so, G., E liza s, E. G., and M ittelm an, A.: The contribution of the intestinal mucosa to the total serum alkaline phosphatase activity. Amer. J. Clin. Path. 56: , R e y n o s o, G. and Ke a n e, M.: Carcinoembryonic antigen (CEA) in prognosis and monitoring of patients with cancer. Immunodiagnosis of Cancer. Herberman, R. and Mclntrye, R., eds. National Cancer Institute, in press. 17. R o s a l k i, S. and Ra u, D.: Serum gammaglutamyl transpeptidase activity in alcoholism. Clin. Chim. Acta 39:41-47, Sza sz, G.: A kinetic photometric method for serum gamma-glutamyl transpeptidase. Clin. Chem. 15: , Tsung, S. and Repk e, D.: Heat inactivation and determination of serum gamma-glutamyl transpeptidase as a profile in identification of human alkaline phosphatase. Amer. J. Clin. Path. 61: , WlNKELMAN, J., et al: The clinical usefulness of alkaline phosphatase isoenzyme determinations. Amer. J. Clin. Path. 57: , 1972.