ARTHRITIS & RHEUMATISM Vol. 40, No. 3, March 1997, pp 461-468 0 1997, American College of Rheurnatology 461

RELATIONSHIP BETWEEN BIOCHEMICAL MARKERS OF BONE TURNOVER AND BONE SCINTIGRAPHIC INDICES IN

ASSESSMENT OF PAGET’S DISEASE ACTIVITY

LUISA ALVAREZ, PILAR PERIS, FRANCISCA PONS, NURIA G U A ~ B E N S , RAMON HERRANZ, ANA MONEGAL, JOSE L. BEDINI, RAMON DEULOFEU, M. JESUS MART~NEZ DE OSABA,

JOSE MUNOZ-GOMEZ, and ANTONIO M. BALLESTA

Objective. To evaluate the relationship between biochemical markers of bone turnover and bone scan indices of disease activity, as well as to analyze their variations based on skeletal involvement, in Paget’s disease.

Methods. Serum samples were obtained from 51 patients with Paget’s disease to determine the levels of total alkaline phosphatase (total AP), bone alkaline phosphatase (bone AP), propeptide carboxyterminal of type I procollagen (PICP), propeptide aminoterminal of type I procollagen (PINP), osteocalcin, tartrate- resistant acid phosphatase, and telopeptide carboxyter- minal of type I collagen. Urine samples were analyzed for levels of hydroxyproline (HYP), pyridinoline (PYR), deoxypyridinoline (DPYR), C-terminal telopeptide of type I collagen (CTx), and N-terminal telopeptide of type I collagen (NTx). In addition, 2 semiquantitative scintigraphic indices, disease activity (AI) and disease extent (EI), were obtained. Pagetic skeletal locations were evaluated individually, with special attention to skull involvement.

Results. All biochemical markers correlated with the AI and the EI. Serum PINP, bone AP, and total AP showed the highest proportions of increased values among the bone formation markers (94%, 82%, and 76%, respectively). Among the bone resorption markers, urinary NTx showed the highest proportion of increased values in patients with Paget’s disease (96%), compared

Luisa Alvarez, PhD, Pilar Peris, MD, Francisca Pons, MD, Nuria Guariabens, MD, Ramon Herranz, MD, Ana Monegal, MD, Jost L. Bedini, MD, Ramon Deulofeu, PhD, M. Jesus Martinez de Osaba, MD, Jost Mufioz-Gomez, MD, Antonio M. Ballesta, MD: Hospital Clinic i Provincial, University of Barcelona, Barcelona, Spain.

Address reprint requests to Luisa Alvarez, PhD, Servicio Bioquimica Clinica, Hospital Clinic i Provincial, Villarroel 170, 08036 Barcelona, Spain.

Submitted for publication July 8, 1996; accepted in revised form September 25, 1996.

with PYR (69%), DPYR (71%), CTx (65%), and HYP (64%). In patients with mild disease activity, serum PINP was the marker with the highest proportion of increased values (71%). In contrast, serum PICP and urinary CTx were the most discriminative markers for skull involvement. Except for higher values for most of the biochemical markers of bone turnover in flat bones, no major differences in other skeletal locations were observed.

Conclusion. The determination of serum PINP as a marker of bone formation and urinary NTx as a marker of bone resorption provided the best biochemi- cal profile to ascertain the extent and activity of Paget’s disease. In patients with skull involvement, serum PICP and urinary CTx were shown to be the most discrimi- native markers.

Paget’s disease of bone is a focal benign disorder that is characterized by high bone turnover. Morpholog- ically, the primary event is an intense focal bone resorp- tion that is rapidly followed by disordered new bone formation (1). At present, bone scintigraphy, usually reviewed by visual assessment, is the most reliable method for evaluating the distribution and extent of Paget’s disease. Biochemical markers of bone turnover have been proven to be of value in assessing the activity of the disease. Thus, determinations of serum total alkaline phosphatase (AP) or serum bone AP and urinary pyridinoline (PYR) excretion seem to provide the best biochemical approach for assessing Paget’s disease activity (2).

Nevertheless, few studies (3-5) have analyzed the correlation between biochemical markers of bone turn- over and scintigraphic assessment of Paget’s disease activity. Previous series included only a small number of markers, mainly, serum total AP, osteocalcin (BGP),

462 ALVAREZ ET AL

and urinary hydroxyproline (HYP). At present, how- ever, new assays have been developed for the detection of serum propeptide aminoterminal of type I procolla- gen (PINP) and for both the C-terminal (CTx) and the N-terminal (NTx) cross-linked telopeptides of type I collagen. The utility of these markers in evaluating Paget’s disease activity has yet to be established. Bio- chemical markers may reflect diverse aspects of bone metabolism and therefore could also be of value in distinguishing the composition and location of involved bone.

To determine the usefulness of the new biochem- ical markers of bone turnover in evaluating Paget’s disease activity and extent, most of the bone formation and resorption markers currently available were com- pared with bone scan indices of disease activity. A further aim of the present study was to analyze varia- tions in these measures based on skeletal involvement.

PATIENTS AND METHODS

Patients and controls. After approval was given by the ethics committee and informed consent was obtained, 51 patients with Paget’s disease (21 postmenopausal women and 30 men), ages 39-83 years (mean 5 SD 67 ? 11 years), were studied. In all patients, the diagnosis of Paget’s disease was documented by radiography and bone scintigraphy. Liver and kidney function test results were normal in all patients, and none had been treated with bisphosphonates, calcitonin, or plicamycin for a period of 6 months before the study. Paget’s disease was considered monostotic when the disease affected only a single bone, and polyostotic when 2 or more bones were involved.

Patients were classified into 3 groups according to the scintigraphic disease activity index (AI) (see below): mild disease included patients with an A1 t 1 8 (n = 7); moderate disease included patients with an A1 between 18 and 36 (n = 19); and severe disease included patients with an A1 >36 (n = 25). Because the average AI value in patients with monostotic disease was 18, this value was chosen as the upper range value for the low-activity pagetic involvement group (mild disease) as assessed by bone scan.

As controls, reference values were obtained from 59 healthy subjects who were matched for age and sex. These values were adjusted for body mass index and menopausal status. Only subjects who had no evidence of any disturbance of calcium metabolism or metabolic bone disease were in- cluded as controls.

Blood and 2-hour fasting urine samples were obtained from each patient and each control subject, on the same day between 8:00 and 10:00 in the morning. No dietary restrictions were imposed, either on the patients or on the control subjects, before collection of samples. Serum and centrifuged urine samples were kept frozen at -20°C until analysis.

Scintigraphic indices of disease activity and extent. Bone scintigraphy was performed on patients with Paget’s disease after intravenous injection with 740 MBq (20 mCi) of

Table 1. analyzed in the study patients

Glossary of the biochemical markers of bone turnover

Serum Total alkaline phosphatase Bone alkaline phosphatase Propeptide carboxyterminal of type I procollagen Propeptide aminoterminal of type I procollagen Osteocalcin Tartrate-resistant acid phosphatase Telopeptide carboxyterminal of type I collagen

Hydroxyproline Pyridinoline Deoxypyridinoline C-terminal telopeptide of type I collagen N-terminal telopeptide of type I collagen

Urine

’9mtechnetium-methylene diphosphate. Planar scans for the whole skeleton (anterior and posterior views) were obtained 2 hours postinjection. The scans were evaluated visually by 2 independent observers who were unaware of the patients’ status. To provide a more accurate interpretation, a semiquan- titative analysis was performed, and the scintigraphic AI was obtained according to the following calculation: A1 = sum of (bone extent X uptake ratio) for all scintigraphic lesions. The extent of the pagetic lesion within each bone was measured using the coefficients given by Coutris et al(6), which confer a value to each bone according to its volume. The extent coefficient for each bone was adjusted according to the scinti- graphic involvement (whole, three-fourths, one-half, or one- fourth). The sum of all coefficients resulted in the skeletal extent index (EI), which can vary between 0 (no bones affected) and 100 (all bones involved completely). A scinti- graphic uptake ratio was also obtained for each affected bone according to a 6-point scale (7). Normal bone served as a standard for comparison with areas of disease. The skeletal extent index was multiplied by the uptake ratio, thus obtaining an index of activity. Finally, the scintigraphic A1 for each patient was obtained as a result of the sum of the indices of activity for the affected bones.

Biochemical markers of bone turnover. In Table 1, the biochemical markers analyzed in this study are listed. Serum total AP activity was measured by a spectrophotometric kinetic assay, according to the recommendations of the Scandinavian Committee for Clinical Chemistry and Clinical Physiology, in a DAX 72 analyzer (Bayer Diagnostics Technicon, Tarrytown, NY). Serum bone AP and BGP were assayed using immuno- radiometric methods (Hybritech, Liege, Belgium and Cis Elsa- Osteo, Gif-sur-Yvette, France). Serum propeptide carboxyter- minal of type I procollagen (PICP), PINP, and telopeptide carboxyterminal of type I collagen (ICTP) determinations were made by radiometric methods using kits from Orion (Espoo, Finland). The activity of serum tartrate-resistant acid phosphatase (TRAP) was assessed by the modified Hillmann method using a kit from BioMerieux (Marcy-I’Etoile, France) in a Cobas Mira S analyzer. Urinary PYR, deoxypyridinoline (DPYR), CTx, and NTx were assayed by enzyme immuno- assays (Metra Biosysterns, Palo Alto, CA, Cis Bio Interna- tional, Gif-sur-Yvette, France, and Ostex, Seattle, WA). Uri-

EVALUATION OF PAGET’S DISEASE ACTIVITY 463

nary HYP was measured by high performance liquid chromatography.

Urine determinations were expressed in relation to creatinine excretion. Measurement of urinary creatinine was performed in the Cobas Mira S analyzer using an assay based on a modified Jaffe method (kit from Roche Diagnostics, Basel, Switzerland). The intraassay coefficients of variation were as follows: total AP 0.75%, bone AP 3.596, PICP 476, PINP 4.1%, ICTP 4.2%, TRAP 1.5%, BGP 3.S%, HYP 3%, CTx 5.276, NTx SS%, PYR 5%, and DPYR 5.3%. The interassay coefficients of variation for each of these assays were as follows: total AP 0.85%, bone AP S%, PICP 6%, PINP 6.396, ICTP 6.570, TRAP 3%, BGP 4.5%, HYP 6%, CTx 8%, NTx 7.876, PYR 6.5%, and DPYR 7%. The intraindividual variabilities (coefficient of variation) obtained from 7 un- treated patients with Paget’s disease were as follows: total AP 4.6%, bone AP 7.6%, PICP 8.1%, BGP 10.9%, PINP 19.5%, TRAP 9.276, HYP 9.S%, PYR 14.6%, DPYR 12.5%, ICTP 16%, NTx 17.7%, and CTx 24%.

Statistical analysis. Results were analyzed by nonpara- metric tests, because the distributions of the biochemical markers were skewed. Between-group comparisons were done by Mann-Whitney U test, and the Spearman’s rank correlation test was used for correlation studies. Differences between proportions were assessed by chi-square test. All statistical calculations were made using the Statgraphics Statistical soft- ware package (STSC, Rockville, MD).

RESULTS

Skeletal involvement. Thirty-four patients (18 men and 16 women) had polyostotic disease, with an average of 4 affected bones, and 17 patients (12 men and 5 women) had monostotic disease. The most frequently affected bones were the vertebrae (particularly, the lumbar spine), the iliac bone, and the femur. Most of these locations were affected in more than 50% of patients. In addition, the sacrum, skull, humerus, and facial bones were not unusual areas of involvement, and were found to be affected in more than 10% of patients. The vertebrae, femur, sacrum, and humerus were more frequently affected in men than in women. In contrast, the skull and facial bones were more frequently identi- fied as affected in women.

Bone scintigraphic indices. Values for the El and the A1 ranged from 1 to 20.25 (mean ? SEM 8.41 ? 0.695) and 4 to 80 (mean ? SEM 37.25 ? 2.8), respec- tively. There were no significant differences between the sexes (30 men versus 21 women), either in the A1 (mean f SEM 36.37 2 3.69 versus 38.52 ? 4.4; P = 0.71) or in the EI (8.46 2 0.98 versus 8.34 ? 0.97; P = 0.93). Moreover, age was not correlated with either index. Patients with monostotic disease had lower values for both the EI and the AI than did patients with polyostotic disease (Table 2).

Table 2. indices in patients with monostotic and polyostotic Paget’s disease*

Monostotic Polyostotic

Biochemical markers of bone turnover and scintigraphic

Controls disease disease Measure (n = 59) (n = 16) (n = 27)

Extent index Activity index Total AP (unitsiliter) Bone AP (ngiml) PICP (ngiml) PINP (ngiml) BGP (ngiml) TRAP (unitsiliter) ICTP (ngiml) HYP (“Mimg) DPYR (nMirnM) PYR (nM/mM)

Men Women

CTX (PdW NTx (“M BCEimM)

-

-

147 i 4.88 12.1 t 0.54 107 t 3.6 33 ? 2.3 17 t 0.9

2.9 ? 0.09 3.07 t 0.08

75 t 4.2 5.06 t 0.28

27 i 2.51 45 t 2.59

228 i 30 33 2 5.1

3.47 +- 0.42 17.31 t 2.16

312 t 49$ 42 t 7.4$

145 t 11$ 90 i 14$ 26 t 1.91

3.28 i 0.33 3.87 -t 0.357 124 t 13$

6.92 ? 0.667

50.33 t 5.63t 77.33 t 12.84$

267 2 55 238 i 44$

10.3 t 0.68t 45.83 +- 3.18t

759 t 121t 98 ? 169:

191 i 129: 293 t 56t 41 t 3.2t

4.32 i 0.195 6.23 i 0.765 292 t 371- 16.4 ? 2.lt

108 i 14.02t 147 t 34.02t 711 ? 635

1,022 ? 1889:

* Values are the mean 2 SEM. Patients with skull involvement are not included in these results. AP = alkaline phosphatase; PICP = propep- tide carboxyterminal of type I procollagen; PINP = propeptide aminoterminal of type I procollagen; BGP = osteocalcin; TRAP =

tartrate-resistant acid phosphatase; ICTP = telopeptide Cdrboxytermi- nal of type I collagen; HYP = hydroxyproline; DPYR = deoxypyr- idinoline; PYR = pyridinoline; CTx = C-terminal telopeptide of type I collagen; NTx = N-terminal telopeptide of type I collagen; BCE =

bone collagen equivalents. t P < 0.0001 versus patients with monostotic disease. $ P < 0.0001 versus controls. 5 P < 0.001 versus patients with monostotic disease. 7 P < 0.001 versus controls.

Biochemical markers of bone turnover. Skeletal involvement. Patients with polyostotic disease showed significantly higher values for all the biochemical mark- ers of bone turnover than did patients with monostotic disease (not including the patients with skull involve- ment) (Table 2). Except for the serum TRAP and urinary CTx values, biochemical marker values in pa- tients with monostotic disease were also significantly different from those in controls.

Bone scintigraphic indices and biochemical markers of bone turnover in patients with monostotic disease are shown, by specific areas of involvement, in Table 3. Patients with skull, facial bone, and pelvic (particularly, sacrum) involvement showed the highest EI and A1 values. In addition, these patients showed higher values for most of the biochemical markers of bone turnover than did patients with other skeletal areas of involvement. However, in spite of the high activity indices, patients with femoral and iliac bone involvement did not show significant differences in any of the bio-

464 ALVAREZ ET AL

Table 3. Biochemical markers of bone turnover and scintigraphic indices in patients with monostotic Paget’s disease, by area of involvement*

Lumbar spine Femur Iliac bone Sacrum Facial bones Skull Measure (n = 4) (n = 4) (n = 5) (n = 2) (n = 1) (n = 1)

Extent index Activity index Total AP (unitsiliter) Bone AP (ngiml) PICP (ngiml) PINP (ngiml) BGP (ngiml) TRAP (unitsiliter) ICTP (ngiml) HYP (“Mimg) DPYR (nM/mM) PYR (“MimM)

NTx (nM BCElmM) CTX ( F d W

1 5 0 5.13 f 0.52

196.5 f 14.9 22.2 i 3.38 126 2 10.35

56.2 f 11.7 18.3 2 2.73 3.1 2 0.33

3.95 i 0.5 97.5 2 21.3 6.29 f 0.41 53.2 i 6.94

153 119

3.13 t 0.36 15.6 2 2.07 269 f 40 34.7 f 10.7

144.5 t 22 59 f 5.85 22 t 1.15

2.73 2 1.01 3.14 i 1.01 111 2 18.8 5.04 f 0.83 41.3 f 3.71

35,187 91,285

5 2 0 23 f 1.22

249 f 55 36 f 9.8

145 2 6.12 62 f 5.9 27 f 2.66

3.04 i 0.25 3.02 f 0.13 148 f 34 7.21 t 1.07

67 2 20 373 f 40 256 f 60

5,5 25.30

473,913 53, 123

263,143 210

37,29 6.4, 3.8 5.9, 4.9 171,136

5.63, 8.8 63,84 373 390

4 24

499 87

114

36 3.8

7.67 108 13.4

452 35 1

-

-

6 30

2,076 293 380 767 30 3.4

4.99 673 21.7 323

1,152 3,841

~~

* Values are the individual determinations or the mean t SEM. See Table 2 for definitions.

chemical markers of bone turnover when compared with patients with vertebral involvement.

Skull involvement. When patients with skull in- volvement, with either monostotic or polyostotic disease, were evaluated individually (n = S), they showed higher values for all the biochemical markers of bone formation than did patients without skull involvement. However, among the markers of bone resorption, only higher urinary CTx and HYP values were observed in patients with skull involvement (Table 4). In addition, serum

PICP and urinary CTx were the most discriminating markers for skull involvement, since they were the only markers that showed significant differences between patients with skull involvement and those without skull involvement, even when patients with a similar A1 were compared.

Scintigraphic A1 and biochemical markers. Markers of bone formation. Mean 2 SEM values for serum total AP, bone AP, PICP, PINP, and BGP in the control group and in the 3 groups of patients with

Table 4. Biochemical markers of bone turnover and scintigraphic indices in patients with versus those without skull involvement*

Measure

Without skull involvement, but with similar activity

index to those with With skull Without skull Controls involvement involvement skull involvement (n = 59) (n = 8) (n = 43) (n = 23)

~

Extent index - 11.92 2 2.25 7.75 2 0.67 11.15 2 0.61 Activity index 48.2 f 6.7 35.2 ? 3 50.6 2 2.64 Total AP (unitsiliter) 147 i 4.88 1,516 t 477 592 -t 84t 836 +- 137 Bone AP (ngiml) 12.1 t 0.54 214 t 65 77 5 11.3t 109 2 19 PICP (ngiml) 107 2 3.6 433 f 84 172 5 9.2% 197 t 14$ PINP (ngiml) 33.3 t 2.29 443 t 97 206 2 38t 310 f 63

TRAP (unitsiliter) 2.89 rt 0.09 4.93 i 0.72 3.93 2 0.19 4.36 f 0.2 ICTP (nglml) 3.07 f 0.08 9.1 t 3.18 5.33 ? 0.51 HYP (nMimg) 75 t 4.17 525 f 167 228 5 267 311 5 40 PYR (nMimM)

Men 27 f 2.51 65.5 2 7.5 83 5 10.4 109 t- 14 157 f 38 Women 45 f 2.59 199 ? 65 128 5 26

DPYR (nMimM) 5.06 f 0.28 18.5 f 3.91 12.6 5 1.49 16.7 f 2.3 a x (cLgimM) 228 t 30 11,061 i 5463 547 5 67t 713 t 729

1,073 t 197 NTx (“M BCEimM) 33 f 5.21 1,703 2 1073 723 i 144

* Values are the mean f SEM. See Table 2 for definitions. 7 P < 0.001 versus patients with skull involvement. $ P < 0.0001 versus patients with skull involvement. § P < 0.05 versus patients with skull involvement.

-

BGP (nglml) 17 i 0.9 62 i 19 36 -t 2.481 44 2 3.44

6.54 t 0.87

EVALUATION OF PAGET’S DISEASE ACTIVITY 465

Table 5. Biochemical markers of bone turnover in the study patients, by disease group*

Controls Mild Moderate Severe Marker (n = 59) (n = 7) pt (n = 19) PS (n = 25) PS

Total AP (unitsiliter) 147 t 4.88 253 k 32 0.0001 493 t 98 0.0001; 0.05 1,059 t 194 0.0001; 0.0001; 0.05 Bone AP (ndml) 12.1 t- 0.54 33 t 6.45 0.0001 67 2 14.1 0.0001; 0.05 142 t 27 0.0001; 0.0001; 0.05

PINP (ng/ml) 33.3 2 2.29 76 i 12.4 0.0001 202 2 59 0.001; 0.05 365 t- 59 0.0001; 0.0001; 0.05

TRAP (unitsiliter) 2.89 t 0.09 3.37 t 0.47 NS 3.57 t 0.28 0.001; NS 4.68 t 0.27 0.0001; 0.05; 0.0001 ICTP (ndml) 3.07 t 0.08 3.58 i 0.34 NS 4.3 2 0.32 0.0001; NS 7.9 2 1.27 0.0001; 0.0001; 0.05

0.0001; 0.0001; 0.05 DPYR (nM/mM) 5.06 2 0.28 6.68 i 0.88 0.05 10.6 2 1.37 0.0001; NS PYR (nM/mM)

Men 27.5 t 2.51 41 i 3.51 0.05 78 t 10.8 0.0001; 0.001 111 t 20 0.0001; 0.0001; 0.05

PICP (ng/ml) 107 t 3.6 139 t 12.2 0.001 168 t 16 0.0001; NS 277 t 38 0.0001; 0.0001; 0.0001

BGP (ngiml) 17 t 0.9 21 t 1.92 NS 30 2 2.16 0.0001; 0.05 53 i 6.9 0.0001; 0.0001; 0.0001

HYP (nM/mg) 75 t 4.17 96 t 11.4 NS 189 t- 32 0.0001; 0.0001 389 t 63 0.0001; 0.0001; 0.0001 18 t 2.4

Women 45 k 2.59 70 t 5.17 0.001 119 t 27 0.0001; 0.0001 282 t 48 0.0001; 0.0001; 0.0001 CTX (I*gimM) 228 t 30 232 t 140 NS 1,824 2 1,280 NS 2,464 t 1,929 0.05; 0.05; NS NTx (“M BCElmM) 32.7 t 5.11 259 t 139 0.0001 690 t 325 0.05; NS 1,108 t 207 0.0001; NS; NS

* Values are the mean t SEM. Disease groups were determined according to the bone scintigraphic activity index. NS = not significant. See Table 2 for other definitions. f Mild disease versus controls. $ Moderate disease versus controls; moderate disease versus mild disease. 5 Severe disease versus controls; severe disease versus mild disease; severe disease versus moderate disease.

Paget’s disease, classified according to the scintigraphic AI, are shown in Table 5. Serum total AP, bone AP, and PINP levels differed significantly between the control group and the patients with mild disease, but no signif- icant differences were found in serum BGP values. Serum levels of BGP were, however, significantly higher in patients with either moderate or severe disease. Serum PICP was the only marker of bone formation that did not show significant differences between patients with mild disease and those with moderate disease.

Serum levels of PINP, bone AP, and total AP were increased in 94%, 82%, and 76% of patients, respectively, whereas only 34% of patients had increased serum BGP and only 50% had increased PICP values (Table 6). In addition, most of the patients with mild disease (71%) showed increased serum PINP values. Thus, we found that serum PINP was the marker of bone formation that better reflected Paget’s disease activity, in that this marker had the greatest proportion of increased values in all the patients and in the patients with mild disease.

Markers of bone resorption. Mean k SEM values for urinary HYP, PYR, DPYR, CTx, and NTx, as well as serum TRAP and ICTP, are shown in Table 5. Results for urinary PYR are expressed individually according to sex. Significant differences between the control group and the patients with mild Paget’s disease were observed only for levels of urinary PYR, DPYR, and NTx. Fur- thermore, in these patients, bone resorption markers were less frequently increased than were bone forma- tion markers (Table 6). Patients with either mild or

moderate disease differed only in their levels of urinary HYP and PYR.

Among the patients with Paget’s disease, urinary NTx was the most sensitive marker of bone resorption, since it showed the highest frequency of increased values (96%) (Table 6). Frequencies of increased values for urinary PYR (69%), DPYR (71%), HYP (64%), and CTx (65%) were slightly more altered than were those for serum TRAP (41%) and ICTP (54%). In the patients with mild disease, urinary NTx also seemed to be the

Table 6. bone turnover in the study patients, by disease group*

Frequency of increased values for biochemical markers of

Marker Mild Moderate Severe Total

Formation Total AP 317 (43)f 14/19 (74)$ 22/25 (88)t 39/51 (76)$ Bone AP 417 (S7)t 15/19 (79)$ 22/24 (92)t 41/50 (82)$ PICP 1/7 (14)t 5/18 (28)$ 18/23 (78)t 24/48 (50)$ BGP 0/6 (0)t 1/18 (6)$ 15/23 (65)t 16/47 (34)$ PINP 517 (71)f 11/11 (loo)$ 18/18 (loo)? 34/36 (94)$

HYP 1/7 (14) 9/18 (50)$ 22/25 (88) 32/50 (64)$ PY R 1/7 (14) 8/12 (67)$ 19/20 (95) 27/39 (69)$ DPYR 2/7 (29) 10/15 (67)$ 17/19 (89) 29/41 (71)$ ICTP 117 (14) 9/19 (47)$ 17/24 (71) 27/50 (54)$ TRAP 1/7 (14) 4/19 (2l)$ 16/25 (64) 21/51 (41)$ CTx 1/3 (33) 519 (56)$ 9/11 (82) 15/23 (65)$ NTx 2/2 (100) 11/11 (loo)$ 11/12 (92) 24/25 (96)$

*Values are the number (%) of patients. Disease groups were determined according to the bone scintigraphic activity index. See Table 2 for definitions. t P < 0.05 between markers within the group. $ P < 0,0001 between markers within the group.

Resorption

466 ALVAREZ ET AL

most valuable marker, although only 2 patients were analyzed.

Correlation studies. All biochemical markers of bone turnover were significantly correlated with both scintigraphic indices, the EI and the AI. Serum BGP was the marker that showed the highest correlation coeffi- cient with the A1 and the EI (r = 0.7228, P < 0.00001 and r = 0.667, P < 0.00001, respectively). However, when patients with skull involvement were not included in the study, serum PINP was the marker with the highest correlation coefficient with the A1 (r = 0.7784, P < 0.0001) (data not shown).

DISCUSSION

The results of this study show the usefulness of various biochemical markers of bone turnover in assess- ing the extent and activity of Paget’s disease of bone. The present data support our previous findings indicat- ing that serum total AP and bone AP are sensitive bone formation markers for assessing Paget’s disease. In the present study, serum PINP emerged as the most sensi- tive bone formation marker, whereas urinary NTx seemed to be the best bone resorption marker. More- over, although the biochemical markers of bone turn- over appeared not to be specific in discriminating the skeletal areas of involvement, some of the markers, such as serum PICP and urinary CTx, were shown to be good indicators of skull involvement.

In the present series, the distribution of skeletal involvement in Paget’s disease was consistent with that in previous studies, since the pelvis, vertebrae, and femur were the most frequently affected bones, and polyostotic disease was more prevalent than the mono- stotic forms. In addition, neither age nor sex was related to the disease extent or activity.

As expected, patients with polyostotic disease showed both higher values of biochemical markers of bone turnover and scan indices of disease activity than did patients with monostotic disease, confirming that biochemical markers of bone turnover reflect the extent of bone involvement (8).

All biochemical markers of bone turnover ana- lyzed in this study correlated with the indices of disease activity and extent. Among markers of bone formation, serum PINP was the most sensitive marker of disease activity, showing the highest proportion of increased values in all pagetic groups. Moreover, when patients with skull involvement were not included, serum PINP showed the highest correlation coefficient with bone scan indices of disease activity. Conversely, serum BGP

showed the lowest sensitivity for identifying active Paget’s disease, which probably reflects different aspects of bone metabolism (9). This marker did, however, show the highest correlation coefficients with indices of dis- ease activity, similar to results reported by Wilkinson et a1 (5). This finding does not imply that serum BGP is the best marker for evaluating Paget’s disease activity, since serum BGP levels tended to increase only when the A1 was very high. Although several hypotheses for explain- ing these discrepancies have been proposed (lo), our results suggest that serum BGP could be specifically related to the extent of the bone lesion.

As previously suggested (11,12) and in contrast with serum PINP, serum PICP was a relatively insensi- tive marker of disease activity. The reasons for the large discrepancies between serum PINP and PICP values need to be further studied. Theoretically, both propep- tides are cleaved from procollagen before mineralization of the bone matrix and are released into the circulation in a 1:l molar ratio (13). Therefore, their concentrations in blood represent the balance between propeptide cleavage and clearance rates (11).

Although the bone scan indicated bone forma- tion (l) , biochemical markers of bone resorption were also correlated with bone scan indices of disease activity. These data are not surprising, given the coupling be- tween formation and resorption in this disease (14). Among markers of bone resorption, urinary NTx showed the highest proportion of increased values (96% of patients). Recent data support the high sensitivity of NTx in controlling Paget’s disease activity (1546). Inter- estingly, urinary CTx, which is also derived from a telopeptidic region of type I collagen, the C-terminal, showed a lower proportion of increased values, with similar results for urinary HYP, PYR, and DPYR. It has been suggested that the greater sensitivity of urinary NTx is due to its being more bone specific than the other cross-links markers, with a significant amount of free PYR and DPYR coming from non-bone sources (15).

Randall et a1 (15) proposed that the proportion of peptide-bound cross-links such as NTx may be in- creased in high bone turnover states, partly due to a rate-limiting step in their degradation to free cross-links. In addition, preliminary results have suggested that degradation products derived from the C-telopeptide region of type I collagen may suffer a p-isomerization in the a1 chain (17), and only the isomerized forms of the epitope are recognized by the CTx CrossLaps enzyme- linked immunosorbent assay. Moreover, woven bone of Paget’s disease seems to be associated with an increased a-isoform production (1 8). Such differences could also

EVALUATION OF PAGET’S DISEASE ACTIVITY 447

explain the discrepancies between markers derived from the C-telopeptide and N-telopeptide regions, such as CTx and NTx.

When the skeletal location of the pagetic lesions was analyzed in order to identify some specific changes in the biochemical markers, no major differences were observed, except for higher values for most of the biochemical markers of bone turnover in flat bones. Early information shows that patients with skull involve- ment display high levels of serum total AP, for unknown reasons (3,19). In the present study, serum PICP and urinary CTx were the most discriminative markers for skull involvement. Although levels of urinary NTx and serum PINP were equally high in these patients, they were not discriminative for skull involvement, reflecting, in these cases, the activity of the disease. The reasons for such results are unclear, but they could be related to a higher osteoblastic activity in the skull, associated with an increased synthesis of type I collagen, with particular procollagen cleavage and degradation products. Hypo- thetically, this finding could be related to the anatomic differences of flat bones, which are derived from in- tramembranous histogenesis.

It seems clear that bone scanning is the method of choice for evaluating the extent, and probably the activity, of Paget’s disease, particularly at diagnosis. However, biochemical markers of bone turnover are preferred for controlling the disorder over a period of time, especially when monitoring the effects of therapy. Scintigraphy has limited repeatability because of expo- sure to radiation, whereas routine biochemical measure- ments are simple to perform. Recent studies suggest that quantitative bone scintigraphy may improve the manage- ment of patients with monostotic Paget’s disease who have relatively normal values for markers of bone turn- over (20). In this previous series, however, the evalua- tion of more sensitive markers for mild Paget’s disease, such as serum PINP, bone AP, or urinary NTx, was not performed. Since some of these markers have been demonstrated to be more sensitive than traditional markers in monitoring followup of the disease (16), application of the most appropriate measures needs to be established.

It seems reasonable to believe that the more sensitive a test, the more appropriate will be the control of the disease, especially in low disease activity forms. It is also interesting to speculate as to the kinds of infor- mation provided by a test when a specific skeletal site is affected, as well as the response to a particular therapy. Although followup studies are necessary to establish the indication of a specific marker for monitoring Paget’s

disease, based on disease activity, skeletal locations, or type of therapy, our present data indicate that serum PINP as a marker of bone formation and urinary NTx as a marker of bone resorption provide the best bio- chemical profile to ascertain the extent and activity of Paget’s disease. In patients with skull involvement, how- ever, serum PICP or urinary CTx could be more discriminative.

ACKNOWLEDGMENTS We are grateful to Montserrat Portas for her technical

assistance, to Pierre Garneau for reviewing the English manu- script, and to Lloren~ Quint6 and Pere Joan Ventura, Epide- miology and Biostatistics Unit, University of Barcelona, for revising the statistical analysis.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

REFERENCES Fogelman I: Bone scanning in Paget’s disease. In, Nuclear Medi- cine Annual. Edited by LM Freeman. New York, Raven Press, 1991 Alvarez L, Guanabens N, Peris P, Monegal A, Bedini JL, Deulofeu R, Martinez de Osaba MJ, Munoz-Ghez J, Rivera-Fillat F, Ballesta AM: Discriminative value of biochemical markers of bone turnover in assessing the activity of Paget’s disease. J Bone Miner Res 10:458-465, 1995 Meunier PJ, Salson C, Mathieu L, Chapuy MC, Delmas P, Alexandre C, Charhon S: Skeletal distribution and biochemical parameters of Paget’s disease. Clin Orthop 217:37-44, 1986 Rubini G, Lauriero F, Rubini D, D’Addabbo A: ‘)‘’Tcm-MDP global skeletal uptake and markers of bone metabolism in patients with bone diseases. Nucl Med Comm 14567-572, 1993 Wilkinson MR, Wagstaffe C, Delbridge L, Wiseman J, Posen S: Serum osteocalcin concentrations in Paget’s disease of bone. Arch Intern Med 146:268-271, 1986 Coutris G, Cayla J, Rondier J, Talbot JN, Bonvarlet JP, Milhaud G: Analyse des perturbations des voies principales du metabolism calcique dans la maladie de Paget: effets de l’administrdtion de calcitonine. Rev Rhum Ma1 Osteoartic 42:759-767, 1975 Vellenga CJLR, Pauwels EKJ, Bijvoet OLM: Comparison be- tween visual assessment and quantitative measurement of radio- activity on the bone scintigram in Paget’s disease of bone. Eur J Nucl Med 9533-537, 1984 Burckardt P: Biochemical and scintigraphic assessment of Paget’s disease. Semin Arthritis Rheum 23:237-239, 1994 Russell RGG, Colwell A, Hannon RA, Blumsohn A, Al-Dehaimi AW: Biochemical measurements in Paget’s disease of bone. Semin Arthritis Rheum 23:240-241, 1994 Delmas PD, Demiaux B: OstCocalcine et maladie osseuse de Paget. In, La Maladie Osseuse de Paget. Edited by L Simon, JL Sebert, CH HCrisson. Paris, Masson, 1989 Risteli J, Niemi S, Kauppila S, Melkko J, Risteli L Collagen propeptides as indicators of collagen assembly. Acta Orthop Scand 66 (suppl 266):183-188, 1995 Ebeling PR, Peterson JM, Kiggs L Utility of type I procollagen propeptide assays for assessing abnormalities in metabolic bone diseases. J Bone Miner Res 7:1243-1250, 1992 Sharp CA, Davie MWJ, Worsfold M, Risteli L, Risteli J: Discrep- ant blood levels of type I procollagen propeptides in active Paget’s disease of bone. Clin Chem 42:1121-1122, 1996 Walton RJ, Preston CJ, Barlett M, Smith R, Russell RGG: Biochemical measurements in Paget’s disease of bone. Eur J Clin Invest 7:37-39, 1977 Randall AG, Kent GN, Garcia-Webb P, Bhagat CI, Pearce DJ,

ALVAREZ ET AL

Gutteridge DH, Prince RL, Stewart G, Stuckey B, Will RK, Retallack RW, Price RL, Ward L: Comparison of biochemical markers of bone turnover in Paget disease treated with pami- dronate and a proposed model for the relationships between measurements of the different forms of pyridinoline cross-links. J Bone Miner Res 11:1176-1184, 1996

16. Pedrazzoni M, Alfano FS, Gatti C, Fantuzzi M, Girasole G, Campanini C, Basini G, Passeri M: Acute effects of bisphospho- nates on new and traditional markers of bone resorption. Calcif Tissue Int 57:25-29, 1995

17. Fledelius C, Kolding I, Bonde M, Cloos P, Christgau S, Qvist P, Byrjalsen 0, Christiansen C: Specificity of the CrossLapsTM

ELISA and the MAbA7 ELISA (abstract). Osteoporos Int 6 (suppl 1):193, 1996

18. Garnero P, Fledelius C, Bonde M, Christiansen C, Delmas PD: Impaired isomerization of type I collagen C-telopeptide in Paget’s disease: an index of bone quality (abstract). J Bone Miner Res 11 (suppl 1):S370, 1996

19. Harinck HIJ, Bijvoet OLM, Vellenga CJLR, Blanksma HJ, Frij- link WB: Relation between signs and symptoms in Paget’s disease of bone. QJM 58:133-151, 1986

20. Patel S, Pearson D, Hosking DJ: Quantitative bone scintigraphy in the management of monostotic Paget’s disease of bone. Arthritis Rheum 38:1506-1512, 1995