Diagnostic Value of Urine Deoxypyridinoline for Detecting Bone Metastases in Breast Cancer Patients

Transcription

1 Annals of Clinical & Laboratory Science, vol. 33, no. 1, Diagnostic Value of Urine Deoxypyridinoline for Detecting Bone Metastases in Breast Cancer Patients Ming-Feng Hou, 1 Shwu-Bin Lin, 2 Shyng-Shiou F. Yuan, 3 Li-Yu Tsai, 4 Shih-Meng Tsai, 5 Jan-Shih Hsieh, 1 and Tsung-Jen Huang 1 1Department of Surgery, Obstetrics and Gynecology, Kaohsiung Medical University; 2 School of Medical Technology, National Taiwan University; 3 Department of Laboratory Medicine, National Taiwan University and University Hospital; 4 Clinical Laboratory, Kaohsiung Medical University Hospital; 5 Department of Public Health, Kaohsiung Medical University, Kaohsiung, Taiwan, Republic of China Abstract. Deoxypyridinoline (Dpd), a crosslink product of collagen molecules found in bone and excreted in urine during bone degradation, has been described as a marker of bone turnover in metastatic breast cancer. In this study, the urine deoxypyridinoline/creatinine (Dpd/Cre) ratio was determined by enzyme immunoassay in urine samples from 116 women with breast cancer. Bone metastases were confirmed by x-ray or CT scan, with follow-up >6 mo. The urine Dpd/Cre ratio was significantly higher in patients with bone metastasis, compared to those without bone metastasis (p <0.05). In patients with bone metastasis, ratios of urine Dpd/Cre were higher in those with multiple lesions, compared to those with a solitary lesion, and the values also reflected therapeutic response (p <0.05). Serial monitoring of urine Dpd/Cre revealed that an elevation was correlated with disease progression. Patients with stable bone disease under effective therapy had significant diminution of the urine Dpd/Cre ratios, compared to those with progression of bone disease (p <0.05). In conclusion, the urine Dpd/Cre ratio may be a useful marker for detecting bone metastases and evaluating their response to therapy. (received 18 June 2002; accepted 16 July 2002) Keywords: breast cancer, bone metastasis, deoxypyridinoline, tumor marker Introduction Bone metastasis is present in 69% of terminal breast cancer patients [1]. It is often difficult to detect early bone metastases and to assess their therapeutic responses solely on the basis of bone scans [2] and bone x-rays [3]. Bone scans are much more sensitive, but relatively non-specific, while bone x-rays are specific, but less sensitive [4,5]. Owing to the slow changes seen in bone during therapeutic response, bone scans should be re-checked after 3 mo [2]. It is important to find other methods, such as biochemical markers, for detecting bone metastases and monitoring their therapeutic responses. Address correspondence to Li-Yu Tsai, Ph.D., Clinical Laboratory, Kaohsiung Medical University, 100 Shih-Chuan 1st Road, Kaohsiung, Taiwan; tel ; fax ; mifeho@kmu.edu.tw. Pyridinoline cross-links contribute to the structural rigidity and integrity of collagen fibrils in bone. When bone resorption occurs, osteoclastic degradation of bone matrix releases pyridinoline and deoxypyridinoline (Dpd) into the circulation; these compounds are then excreted in urine [6]. Measurements of urinary elimination of these compounds may provide useful information on the change of bone resorption [7-9]. Dpd is a more sensitive marker, since it is found predominantly in bone, and it is well suited for diagnosis and monitoring of the breakdown of bone matrix by cancer cells [10,11]. Studies have shown that urine Dpd elimination is significantly increased in patients with bone metastasis, compared to normal subjects [12-14]. Monitoring urine Dpd elimination has revealed significant diminution after bisphosphonate therapy, according to previous reports [15,16] /03/0100/0055; $1.75; 2003 by the Association of Clinical Scientists, Inc.

2 56 Annals of Clinical & Laboratory Science In this study, we measured urine Dpd in patients with metastatic breast cancer and examined the role of Dpd as a biochemical marker of bone metastasis. Sequential examinations were performed to monitor the response of urine Dpd during treatment and to evaluate its correlation with bone metastasis activity.. Materials and Methods Patients and study design. The study population comprised 116 women (average age 48 ± 11.4 yr) with breast cancer who were treated at Kaohsiung Medical University Hospital from July 1995 to December Based on clinical, radiological, and scintigraphic evidence, patients with metastatic breast cancer, as well as those without it, were selected for this study. Patients with skeletal, endocrine, or hepatic diseases, or with previous medical conditions or drug therapies that affect bone metabolism, were excluded from this study. All patients were referred for 99m Tc-polyphosphate bone scan for evaluation of skeletal metastases. Bone metastases were confirmed by bone scintigraphy, x-ray, or CT scan. Radiological examinations were performed on metastatic bone lesions every 1 to 2 mo to evaluate the therapeutic response. The criteria for therapeutic response were defined according to WHO criteria, which were used to classify patients into stable and progressive bone disease groups. The category of stable disease (SD) included patients with partial response or lack of change that was sustained for at least 2 mo; the remaining cases, which included newly diagnosed, untreated cases, were assigned to the category of progressive disease (PD). The PD cases were categorized according to their responses to chemotherapy, endocrine treatment, and/or bisphosphonates. Bisphosphonate therapy consisted of two regimens: pamidronate (90 mg, iv, every 3-4 wk), or clodronate (1600 mg, po, daily). The two subsets (SD and PD) were analyzed for differences in urine Dpd levels. Sample collection and analysis. Non-fasting samples of urine (10 ml) and venous blood (5 ml) were collected between 9 and 11 am. Borate (1 g/l) was added to urine samples to prevent bacterial growth. The samples were stored at -20 C until analyzed. Prolonged exposure of the samples to light, especially direct sunlight, was avoided. The blood was allowed to clot and serum was separated immediately. Aliquots were taken for standard biochemical profiles. The biochemical profiles were performed within 24 hr after collection. Urine and serum creatinine (Cre) concentraions were assayed by standard laboratory methods using a Hitachi autoanalyzer with purchased reagents (Wako, Japan). Urine Dpd was measured by enzyme immunoassay on microtitre plates using reagents (Pyrilin R5 TM-D) supplied by the manufacturer (Metra Biosystems, France). Standards were included on each plate for calibration and results were calculated from the calibration curve using Softmax software and an IBM-compatible computer. High and low control samples were included in each assay. Each standard, control, and urine sample was assayed in duplicate. Urine Dpd results were expressed as urine Dpd/Cre ratio (nmol/mmol). The cutoff value for urine Dpd/Cre ratio was 8 nmol/ mmol, according to our previous study in healthy control subjects [14]. Stastistics. Results are reported as mean ± SD. The significance of differences of urine Dpd/Cre ratios in the stable disease group vs the progressive disease group was analyzed by t-test. For sequential monitoring data, paired t-tests were used to compare urine Dpd/Cre ratios before and after treatment. The criterion for statistical significance was p <0.05. Results Intra-assay and inter-assay data for quality control of Dpd and Cre analyses are shown in Table 1. Table 1. Analytical precision of Dpd and Cre assays, based on coefficients of variation (CV) of quality control (QC) samples. Assay (units) QC samples Intra- Interassayed assay assay (n = 20) CV (%) CV (%) Dpd (nmol/l) low level high level Cre (mmol/l) low level high level

3 Urine deoxypyridinoline in metastatic breast cancer 57 As summarized in Table 2, the patients were 23 to 76 yr old, with an average of 48 yr. Seventy of the patients had bone metastasis and 46 patients had extraskeletal metastasis. Among the 70 patients with bone metastasis, 50 had 2 skeletal lesions and 20 cases had solitary bone lesions. In 46 patients with extraskeletal metastasis, 12 cases had metastasis to liver, 25 to lung and 17 to soft tissues. Fortynine patients had received hormone therapy and 62 had received chemotherapy. Of 53 cases with bisphosphonate therapy, 27 were treated with pamidronate and 26 with clodronate. Thirty-nine patients showed progressive disease (PD) and 31 showed stable disease (SD). The mean urine Dpd/Cre ratio in all patients was 10.2 ± 4.5 nmol/mmol (Table 3). When the patients were categorized according to bone Table 2. Clinical characteristics of the patients with breast cancer that were included in this study. Clinical characteristics Number of patients All patients with breast cancer 116 Age (yr, mean ± SD & range) 48.3 ± 11.4 (23-76) Metastatic sites Bone 70 Soft tissue 21 Lungs 35 Liver 18 Brain 3 Multiple 41 Bone lesions Solitary metastasis 20 Multiple metastases 50 Extraskeletal metastasis 46 Prior therapy None 7 Hormone therapy 49 Chemotherapy 62 Radiation therapy 44 Pamidronate 27 Clondronate 26 Therapeutic Response Stable disease 31 Progressive disease 39 metastasis, Dpd/Cre ratios were higher in patients with bone metastases (12.5 ± 7.5 nmol/mmol) than in those without (7.6 ± 3.2 nmol/mmol) (p <0.05). Menopausal status was unlikely to be correlated with Dpd/Cre ratio, since the ratios in premenopausal patients (11.5 ± 8.1 nmol/mmol, n = 66) were comparable to those in postmenopausal patients (12.0 ± 7.9 nmol/mmol, n = 50). Urine Dpd/Cre ratios of 70 patients with bone metastases were further analyzed. When the 70 patients were subdivided into groups according to the extent of the disease in the bone, higher Dpd/ Cre ratios were observed in patients with 2 bone lesions (15.4 ± 7.2 nmol/mmol) compared to those with a solitary bone lesion (9.7 ± 3.4 nmol/mmol) (p <0.05). Thirty-one patients with bone metastases were classified as SD, and the remaining 39 cases were classified as PD at the beginning of sample collection. In 12 of these cases (3 SD cases and 9 PD cases), repeat samples were not collected, owing to loss to follow-up (n = 6) or death (n = 6). The initial urine Dpd/Cre ratios were significantly higher in the PD group (14.8 ± 7.8 nmol/mmol) than in the SD group (13.0 ± 6.5 nmol/mmol). After therapy, the Dpd/Cre ratios of patients reached lower levels (9.5 ± 7.8 nmol/mmol) than the initial levels in the SD group (p <0.05). The 17 patients treated with bisphosphonate in the SD group had lower urine Dpd/Cre ratios (8.0 ± 6.3 nmol/mmol) than the 11 patients without bisphosphonate (10.1 ± 5.8 nmol/mmol) and lower than than the initial levels (12.8 ± 6.4 nmol/mmol). In contrast, the urine Dpd/Cre ratios of the PD group reached much higher levels (19.2 ± 8.0 nmol/ mmol) than the initial levels (p <0.05). Even in PD patients treated with bisphosphonate (n = 22), the urine Dpd/Cre ratios were higher (18.9 ± 8.2 nmol/ mmol) than the initial levels (15.2 ± 7.7 nmol/ mmol). In the 46 patients without bone metastasis, 10 cases (22%) had elevated initial ratios of urine Dpd/Cre, and 7 of these cases developed bone metastasis within 12 mo. In patients with bone metastasis, urine Dpd/ Cre ratios did not differ in premenopausal (12.6 ± 12.5 nmol/mmol, n = 39) and postmenopausal patients (12.0 ± 7.9 nmol/mmol, n = 31) (p >0.05).

4 58 Annals of Clinical & Laboratory Science Table 3. Urine Dpd/Cre ratios (nmol/mmol) in patients with breast cancer, categorized according to clinical status. Clinical status Number Urine p of cases Dpd/Cre ratio All patients with breast cancer ± 4.5 Bone metastasis negative ± 3.2 positive ± 7.5 <0.05 Extent of bone lesions solitary metastasis ± 3.4 multiple metastases ± 7.2 <0.05 Therapeutic response stable disease before therapy ± 6.5 after therapy ± 7.8 <0.05 progressive disease before therapy ± 7.9 after therapy ± 8.0 <0.05 stable disease treated without bisphosphonate before ± 6.8 after ± 5.8 treated with bisphosphonate before ± 6.4 after ± 6.3 <0.05 progressive disease treated without bisphosphonate before ± 8.2 after ± 9.1 <0.05 treated with bisphosphonate before ± 7.7 after ± 8.2 <0.05 Sequential monitoring of Dpd/Cre ratios after initiation of bisphophonate therapy (pamidronate in 22 cases, clodronate in 21 cases) was performed in 43 patients with bone metastases. When the second urine samples were collected, significant elevation of Dpd/Cre ratio (18.9 ± 8.2 nmol/mmol, n = 22) was observed in the PD group when compared with their initial levels (p <0.05). The Dpd/Cre ratios of the SD group were lower (8.0 ± 6.3 nmol/mmol, n=17) than their initial levels (12.8 ± 6.4 nmol/mmol, n = 18) (p <0.05). This observation indicates that Dpd/Cre ratios reached nearly normal values under effective bisphosphonate treatment in the SD group, and that Dpd/Cre ratios reached much higher levels in the PD group. There were no significant differences in the Dpd/Cre ratios after pamidronate vs clodronate therapies in the SD or PD groups. This suggests that the Dpd change reflects the therapeutic response, rather than the different anti-resorptive therapeutic agents. Discussion Since bone destruction in most patients with metastatic breast cancer is caused by osteolysis followed by bone formation, urine Dpd may be a suitable marker for breast cancer metastasis, as was suggested in our previous study [14]. In the present study, urine Dpd levels were significantly higher in patients with bone metastases than in those without bone metastasis, suggesting that Dpd may be a useful marker for bone metastasis of breast cancer. However Dpd is unlikely to be an early tumor marker, for higher Dpd/Cre ratios were not observed in patients with a solitary bone metastasis. The extent of bone metastasis was previously reported to be correlated with Dpd/Cre ratios in breast cancer and other cancers [14]. In addition, a higher level of urine Dpd/Cre ratio was also found in patients with severe bone metastases, but was not correlated with more bone lesions, as was discussed in our previous report, which suggested that the initial bone destruction was followed by bone formation during bone metastatic status. Therefore, urine Dpd/Cre ratio seems to be an appropriate marker for multiple bone metastases, through imaging studies are needed to identify bone metastasis in its early stages. In this study, urine deoxypyridinoline was measured by an enzyme-linked immonosorbent assay (ELISA), which correlated well with the results obtained by high performance liquid chromatography with fluorometric detection [17,18]. In our previous study [14], urine Dpd/Cre ratios were around 5.6 nmol/mmol in a healthy group of Taiwanese women, similar to the findings in a study by Ohishi et al [19]. Although age-related changes of urinary Dpd/Cre ratios were noted in some reports [19,20], which showed significant increases in children and postmenopausal women, there has been no systematic study that indicates that the

5 Urine deoxypyridinoline in metastatic breast cancer 59 changes in Dpd excretion are caused by the menopause. In this study, there was no significant relationship between urine Dpd/Cre ratios and the menopausal status. This suggests that the bone destruction induced by the menopause is less than that caused by cancer metastases. In this study, 22% of cases with extraskeletal metastasis had elevated urinary Dpd/Cre ratios in the initial test, and developed bone metastasis during the 12 mo follow-up. Furthermore, those patients continuously had much higher urinary Dpd/Cre ratios. This reveals that bone metabolism is affected by breast cancer micrometastasis, which cannot be detected in clinical images, and that increased osteoclastic activity is frequently caused by cancer cells [21]. It may also be caused by humoral mediators that influence bone homeostasis via parathyroid hormone related protein, cytokines, prostaglandin, and transforming growth factors [22]. These findings may be correlated to the 25% of cases of stage III cancers that had elevated levels of Dpd/ Cre ratios in our previous study [14]. Bone micrometastasis was therefore detected by a bone turnover marker before the metastatic bone destruction could be detected by imaging studies. Radiography and bone scans have been generally accepted as routine examinations for bone metastasis [2,3]. Bone scans are more sensitive than bone radiography, whereas radiography is more specific, suggesting that diagnosis of bone metastasis generally needs a combination of the two images [4]. In this study, we found that urine Dpd/Cre ratios showed positive correlation with therapeutic response of bone metastatic lesions, and also that the elevation of urine Dpd/Cre ratios correlated with progression of bone metastasis. These results are consistent with the report by Lüftner et al [15] in a smaller number of cases, and other bone turnover markers which have been reported in other studies [23-25]. Imaging technics combined with monitoring of urine Dpd/ Cre ratios appears to offer additional information for the detection and evaluation of bone metastasis. Bone metabolism is characterized by both bone resorption and bone formation. Most metabolic bone diseases, including cancer metastasis, disrupt these two activities. Various bone resoption markers, such as the carboxyterminal propeptide of type I collagen, the carboxyterminal telopeptide of type I collagen [25-27], and bone formation markers such as bone-specific alkaline phosphatase, and osteoclaine, have been tested for detecting bone metastasis [14, 28]. Bone resorption markers were reported to be superior to bone formation markers for detection of bone metastasis [29]. Based on the findings of this study, although other bone turnover markers may be useful in detection and evaluation of bone metastasis, urine Dpd can be recommended for use in combination with imaging examinations. Patients in this study received various modalities of therapy, including several chemotherapeutic regimens, endocrine drugs, and different dosages of radiation. This shows that the therapeutic responses were derived from a heretogenous population. For example, cytoxic chemotherapy with drugs such as cyclophosphamide and methotrexate has been reported to affect mineralized tissue metabolism. Analysis of data regarding the responses to different modalities of therapy needs further evaluation and a greater number of cases. Lüftner et al [15] reported that patients with stable bone disease during iv pamidromate treatment had a significant fall of Dpdcrosslink, compared to those with progressive disease, and concluded that the net bone turnover is not increased at Dpd-crosslink elimination <8 nmol/mmol. In our study, the level of Dpd/Cre ratio decreased to normal in the SD group after therapy, especially after bisphosphonate therapy. These results indicate that bisphosphonate has an important role in inhibiting the destruction of bone, as has also been confirmed by other bone resorption markers. Otherwise, the two bisphosphonate therapies seem to have similar effects on bone destruction. In conclusion, this study revealed that the urine Dpd/Cre ratio is a useful marker for bone metastasis caused by breast cancer. Furthermore, Dpd can be useful in evaluating therapeutic response, especially in patients with progressive disease. Acknowledgment and Financial Disclosure The authors thank Ms Wen-Jen Lin for help in preparing this manuscript. The authors have no financial connection with any company or product mentioned in this article.

6 60 Annals of Clinical & Laboratory Science References 1. Coleman RE, Rubens RD. Bone metastases and breast cancer. Cancer Treat Rev 1985;12: Citrin DL, Hougan C, Zweibel W. The use of serial bone scans in assessing the response of bone metastases to systemic treatment. Cancer 1981;47: Blomqvist C, Elomma I, Virkkunen P. The response of bone metastases in mammary carcinoma. Cancer 1987;60: Hortobagyi GN, Libshitz HI, Seabold JE. Osseous metastases of breast cancer. Clinical biochemical, radiographic, and scintigraphic evaluation of response to therapy. Cancer 1984;53: Galasko CS, Doyle FH. The detection of skeletal metastases from mammary cancer. A regional comparison between radiology and scintigraphy. Clin Radiol 1972;23: Robins SP, Black D, Peterson CR. Reid DM, Duncan A, Seibel MJ. Evaluation of urinary hydroxypyridinium crosslink measurements as resorption markers in metabolic bone disease. Eur J Clin Invest 1991;21: Fujimoto D, Moriguchi T, Ishida T, Hayashi H. The structure of pyridinoline collagen crosslink. Biochem Biophys Res Commun 1978;87: Ogawa T, Ono T, Tsuda M, Kawanishi Y. A novel fluor in insoluble collagen: A crosslinking molecule in collagen molecule. Biochem Biophys Res Commun 1982;107: Eyre DR. Collagen cross-linking amino acids. Methods Enzymol 1987;144: Delmas PD, Schlemmer A, Gineyts E, Riis B, Christiansen C. Urinary excretion of pyridinoline crosslinks correlate with bone turnover measured on iliac crest bone biopsy in patients with vertebral osteoporosis. J Bone Miner Res 1991;6: Uebelhart D, Schlemmer A, Johansen JS, Gineyts E, Christiasen C, Delmas PD. Effect of menopause and estrogen treatment on the urinary excretion of pyridinium crosslinks. J Clin Endrocrinol Metab 1991;72: Coleman RE, Houston S, James I, Rodgen A, Rubens RD, Leonard RCF, Ford J. Preliminary results of the use of urinary excretion of pyridinium crosslinks for monitoring metastatic bone disease. Br J Cancer 1992;65: Paterson CR, Robins SP, Horobin JM, Preece PE, Cuschieri A. Pyridinium crosslinks as markers of bone resorption in patients with breast cancer. Br J Cancer 1991;64: Hou MF, Tsai LY, Tsai SM, Huang CJ, Huang YS, Hsieh JS, Chan HM, Wang JY, Chuang CH, Chen FM, Huang TJ. Biochemical markers for assessment of bone metastases in patients with breast cancer. Kaohsiung J Med Sci 1999;15: Lüftner D, Günthner S, Flath B, Müller C, Echtersff K, Mergenthaler HG, Wernecke KD, Possinger K. The course of the urinary DPD-crosslink secretion in metastatic breast cancer: A possibility for response assessment. Anticancer Res 1999;19: Body JJ, Dumon JC, Gineyts E, Delmas PD. Comparative evaluation of markers of bone resorption in patients with breast cancer-induced osteolysis before and after bisphosphonate therapy. Br J Cancer 1997;75: Seyedin SM, Kung VT, Daniloff YN, Hesley RP, Gomez B, Nielsen LA, Rosen HN, Zuk RF. Immunoassay for urinary pyridinoline: The new marker of bone resorption. J Bone Miner Res 1993; 8: Robins SP, Woitge H, Hesley R, Ju J, Seyedin S, Seibel MJ. Direct enzyme linked immunoassay for urinary deoxypyridinoline as a specific marker for measuring bone rebsorption. J Bone Miner Res 1994;9: Ohishi T, Takahashi M, Kawama K, Aoshima H, Hoshino H, Horiuchi K, Kushida K, Inoue T. Agerelated changes of urinary pyridinoline and deoxypyridinoline in Japanese subjects. Clin Invest Med 1993;16: Mole PA, Walkinshaw MH, Robins SP, Paterson CR. Can urinary pyridinium crosslinks and urinary oestrogens predict bone mass and rate of bone loss after the menopause? Eur J Clin Invest 1992;22: Nguyen-Pamart M, Bonneterre J, Hecquet B. Urinary excretion of deoxypyridinoline in patients with breast cancer. Anticancer Res 1995;15: Garet IR. Bone destruction in cancer. Sem Oncol 1993;20: Vinholes J, Coleman R, Lacombe D, Rose C, Tubian-Hulin M, Bastit P, Wildiers J, Michel J, Leonard R, Nortier J, Mignolet F, Ford J. Assessment of bone response to systemic therapy in an EORTC trial: Preliminary experience with the use of collagen cross-link excretion. Br J Cancer 1999;80: Walls J, Assiri A, Howell A, Rogers E, Ratcliffe WA, Eastell R, Bundred NJ. Measurement of urinary collagen cross-links indicate response to therapy in patients with breast cancer and bone metastases. Br J Cancer 1999;80:

7 Urine deoxypyridinoline in metastatic breast cancer Maemura M, NO Y, Yokoe T, Takei H, Horiguchi J, Koibuchi Y, Ikoda F, Ohwada S, Takeyeshi I, Morishita Y. Serum concentration of pyridinoline cross-linked caboxyterminal telopeptide of type I collagen in patients with metastatic breast cancer. Oncology Rep 2000;7: Bombardieri E, Martinetti A, Miceli R, Mariani L, Castellani MR, Seregni E. Can bone metabolism markers be adopted as an alternative to scintigraphic imaging in monitoring bone metastases from breast cancer? Eur J Nucl Med 1997;24: Delmas PD. Biochemical markers of bone turnover, J Bone Mineral Res 1993;8: Costa L, Demers LM, Gouveia-Oliveriea A, Schaller J, Costa EB, de Moura MC, Lipton A. Prospective evaluation of the peptide-bound collagen type I cross-links N-telopeptide and C-telopeptide in predicting bone metastases status. J Clin Oncol 2002;20: Shimozuma K, Sonoo H, Fukunaga M, Ichihara K, Aoyama T, Tanaka K. Biochemical markers of bone turnover in breast cancer patients with bone metastases: a preliminary report. Jpn J Clin Oncol 1999;28:16-22.