COMPARISON OF ASSAY OF TOTAL AND BONE-SPECIFIC ALKALINE-PHOSPHATASE IN THE ASSESSMENT OF OSTEOBLAST ACTIVITY IN PATIENTS WITH METASTATIC BONE-DISEASE

The evaluation of response of osseous metastases to systemic treatment s is often low as a consequence of the different radiologic appearance s that make objective assessment not only difficult but sometimes impo ssible. Radiographic evidence of recalcification, the UICC criterion o f response, is often evident for 6 months and sometimes may be delayed even more. This accounts for lower response rates in bone with respec t to other metastatic sites in clinical trials. A transient rise in bo ne formation indices may provide an early indication of bone healing a nd, along with measurement of symptomatic changes, could ameliorate th e response evaluation. Among the biochemical markers of bone formation , total alkaline phosphatase (TALP) is widely employed, but it lacks s pecificity. Estimation of bone isoenzyme (E-BALP) by electrophoretic t echniques is time consuming and semiquantitative. The immunoradiometri c assay (I-BALP) seems to overcome these limitations. In this study, w e compared the two methods of bone isoenzyme estimation with each othe r and with the levels of bone gla protein (BGP) and carboxyterminal pr opeptide of type I procollagen (PICP) in a group of 136 cancer patient s with bone metastases stratified as having lyric or mixed and blastic lesions at x-ray, and in 62 cancer patients without apparent bone inv olvement. The same indices were also evaluated prospectively in a pati ent subset submitted to chemotherapy associated with pamidronate. The aims of the study were to evaluate whether I-BALP is superior to E-BAL P and whether both methods of bone isoenzyme estimation are more advan tageous than TALP, BGP, and PICP in the assessment of osteoblast activ ity either in baseline conditions or in response to treatment. In bone metastatic patients with lytic appearances, values above the cut-off limit were observed in 32.1%, 23.3%, 48.9%, 32.9%, and 14% for, TALP, E-BALP, I-BALP, PICP, and BGP, while the corresponding percentages in those with blastic/mixed appearances were 74.0%, 84.8%, 76.9%, 51.9%, and 43.8%, respectively. In the patients without bone involvement, val ues within the normal range were 90.2%, 98.2%, 89.6%, 71.7%, and 90.2% , respectively. Levels of TALP, E-BALP, and I-BALP were reciprocally c orrelated in the three groups examined. In bone metastatic patients, h owever, the degree of correlation of the enzymes with PICP and BGP was weak. Liver isoenzyme of alkaline phosphatase (LALP) was found to cor relate with E-BALP, but not with I-BALP, in patients with mixed/blasti c lesions. Thirty-eight patients were submitted to pamidronate therapy (60 mg every 3 weeks, administered 4 times) in association with cytot oxic treatment. Osteoblastic markers were determined before any admini stration. Serum TALP, E-BALP, and I-BALP showed a transient rise in 9 cases, a progressive reduction in 12, no change in 3, and a progressiv e increase in 6. Changes in E-BALP and I-BALP from baseline were great er than these of TALP. A divergent pattern between TALP and both I-BAL P and E-BALP was found in 9 patients, whereas a divergent temporal pro file between the two methods of bone isoenzyme estimation was recorded in only 3 patients. Eight out of 38 cases obtained a partial recalcif ication of lytic and mixed lesions. Seven of them showed the concomita nt early increase in TALP, E-BALP, and I-BALP followed by a gradual de cline (osteoblastic flare), whereas 1 patient demonstrated the flare o f E-BALP and I-BALP but not of TALP. No relationship was found between response and temporal changes in in BGP and PICP serum levels. We con clude that I-BALP is a useful marker for detecting excess osteoblastic activity in patients who have at imaging ''pure'' lytic bone metastas es. In the longitudinal evaluation of patients receiving multiple pami dronate infusions plus chemotherapy, TALP, E-BALP, and I-BALP, but not BGP and PICP, appeared to be useful to identify responders in bone. A slight advantage of measurements of serum bone isoenzyme (by both tec hniques) over TALP is apparent, but this study fails to demonstrate a clear superiority of I-BALP over E-BALP.