Japanese Journal of Clinical Oncology 31:263-269 (2001)
© 2001 Foundation for Promotion of Cancer Research
What Do Breast Cancer Patients Benefit from Staging Bone Scintigraphy?
Departments of 1Nuclear Medicine, 2Breast Surgery and 3Internal Medicine, Cancer Institute Hospital, Tokyo, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Background: A review and analysis of breast cancer treatment records were conducted to establish criteria for performing disease staging by bone scintigraphy in Japanese breast cancer patients.
Methods: Records from 5538 consecutive Japanese breast cancer patients from January 1988 to December 1998 were reviewed and analyzed to determine bone metastasis status at the time of initial treatment. Correlation between metastasis to bone and factors known before and after surgery was analyzed using logistic regression.
Results: The overall incidence of metastasis to bone was 2.13% [95% confidence interval (CI): 1.77–2.55%, 118/5538]. Multivariate logistic analysis revealed that tumor size, nodal involvement and histopathology correlated with metastasis to bone. Patients with tumors larger than 30 mm had a significantly higher probability of metastasis to bone, as did patients with lymph node evaluation results N 
1. The incidence of metastasis to bone was 0% in patients with stage 0 disease, 0.08% in stage I patients, 1.09% in stage II patients, 9.96% in stage III patients and 34.04% in stage IV patients. Stage II patients were sub-classified by tumor size T (small, 21–30 mm; and large, 31–50 mm), nodal involvement N and histopathology. The incidence of metastasis to bone in stage II patients was higher in patients with large tumors, scirrhous carcinoma or invasive lobular carcinoma or both.
Conclusion: Bone metastasis correlated with tumor size (T), lymph node involvement (N) and histopathology. Using the criteria that bone scintigraphy is not necessary in populations with a <1% incidence of bone metastasis, but is recommended at incidence >3%, the following conclusions were drawn. Staging by bone scintigraphy provided no benefit to patients whose disease was stage I or less, stage II with small tumors or stage II with large tumors marked by low-grade histopathology (papillotubular cancer). Bone scintigraphy is recommended in patients whose disease is stage II with large tumors marked by high-grade histopathology (scirrhous or invasive lobular cancer), stage III or stage IV. Consequently, staging by bone scintigraphy could be avoided in 71% (3943/5538) of Japanese breast cancer patients.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Breast cancer is a leading cause of female mortality in industrialized countries. One reason is that the incidence of breast cancer metastasis to bone is high, occurring in 58% of cases as shown at autopsy (1). However, many reports have shown that patients with stage I or II breast cancer have a lower probability of having bone metastasis in staging studies, indicating little benefit for these patients from the expensive procedure of bone scintigraphy for staging the disease (2–8). However, other authors have reported that patients with large stage II tumors or high-grade histopathology require staging by bone scintigraphy (9–11). Although most physicians accept the idea that staging bone scintigraphy of stage I breast cancer patients yields little clinically useful data, others regard it as a valuable diagnostic tool at any stage (12–15). Since these reports concern findings from European and American studies, it is not clear whether these data can be applied to Japanese patients, because variations in breast cancer incidence rates and biological expressions exist among different races (16).
Although staging by bone scintigraphy in early stage breast cancer may not be cost effective, our institute routinely performs a baseline bone scintigraphy on all new breast cancer patients. This provides a wealth of background data to assess both diagnostic procedures and treatment regimens. Using these data, a study to investigate the necessity of bone scintigraphy in patients with breast cancer at initial staging was conducted retrospectively.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Patients
Data from consecutive patients with breast cancer confirmed by histological examination, who were treated at the Cancer Institute Hospital, Tokyo, Japan from January 1, 1988 to December 31, 1998, were analyzed.
Primary and ipsilateral breast cancer patients were selected for this study. Patients exhibiting either of the following were excluded from this study: (1) bilateral breast cancer in the past or at the time of the surgery; (2) multi-focal breast cancer in the ipsilateral breast revealed by examination of resected tissue.
Factors Analyzed
Analysis using patients’ data was performed on factors that were available before and after surgery to determine which factors correlated positively with tumor metastasis to bone. These were age at the commencement of therapy, menstruation status, clinical stage assigned before bone scintigraphy, tumor size (T classification), clinical evaluation of lymph node involvement (N classification), pathological evaluation of lymph node involvement (pN classification), estrogen receptor status and histopathology determined biopsy. Clinical N3 (ipsilateral internal mammary node) was evaluated by magnetic resonance imaging and/or ultrasonography. In this study, the determination of stages was made using T and N (or pN) data regardless of the distant metastasis. The clinical stage (TN) and pathological stage (TpN) were classified according to UICC criteria (17). T2 was further classified as small T2 (more than 2 cm but not more than 3 cm) and large T2 (more than 3 cm but not more than 5 cm). This small and large T2 concept was proposed by Lagrange et al. (10) and is not included in the TNM system. If the estrogen receptor concentration of primary tumor was 5 fmol/mg total protein, estrogen receptor was regarded as positive.
Histopathological classification was made according to Japanese Breast Society guidelines (18,19). These classes include non-invasive cancer, papillotubular cancer, solid-tubular cancer, scirrhous cancer, invasive lobular cancer, other special types of invasive cancer except for invasive lobular cancer and unclassified cancer (because of small specimen sizes, cancer could be confirmed but could not be classified). Papillotubular, solid-tubular and scirrhous cancers were sub-classes of invasive ductal cancer (18,19).
Diagnosis of Bone Metastasis
A diagnosis of bone metastasis was defined as a positive finding by bone scintigraphy confirmed by a positive finding using another imaging technique, such as X-ray, computed tomography and magnetic resonance imaging. Patients without confirmed diagnoses of bone metastasis were enrolled in follow-up studies, had a biopsy, or both. Patients who had negative bone scan results but positive findings from other imaging techniques were regarded as having metastasis to bone. The sensitivity and specificity of bone scintigraphy in this study were calculated as 96 and 97%, respectively. These percentages may be high because experienced nuclear medicine physicians made the diagnosis.
Analytical Procedures and Statistical Methods
The factors that are available before surgery are regarded as more important than those available after surgery.
Analysis was carried out using the following protocol:
1. The incidence of bone metastasis at staging was calculated using age, menstruation condition (pre- or post-menopause), tumor size, disease T, N and pN status, estrogen receptor status and histopathology data.
2. The risk of bone metastasis (crude odds ratio) was calculated using a univariate logistic regression model to analyze each factor.
3. A multivariate logistic regression model was used to adjust the risk of bone metastasis (adjusted odds ratio) for each factor.
4. Factors that correlated significantly with bone metastasis were determined using a forward-stepwise method of logistic regression analysis.
5. The final incidence of bone metastasis was calculated according to the groups constructed from the results of 1–4. Because tumor size, lymph node involvement and histopathology were the factors related to metastasis to bone, stage was first used, then the sub-classes of T2 and the histopathological data were added.
A binomial distribution was calculated to determine the 95% confidence interval (CI) of bone metastasis incidence. Fisher’s exact test was used to evaluate the difference in incidence in each factor. Logistic regression analysis was used to calculate the crude and adjusted odds ratio for each factor. Forward-stepwise selection by logistic regression analysis was performed using an inclusion criterion of 0.1 and an exclusion criterion of 0.05. The SPSS data analysis software package, version 8.0 (Cytel, MA, USA) was used for statistical calculations. All p values were two-tailed. A p value of 0.05 was taken to be significant and the 95% confidence interval was calculated.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Characteristics of patients whose data were used in this study are shown in Table 1. The median age was 50 years (range: 23–88 years); five age categories were constructed. Menstruation condition was divided into three categories (pre-menopause, post-menopause and not available), two of which were used in the data analysis. Because male patients were included in this study and records on some female patients were incomplete, 41 patients had undetermined menstruation condition. Pathological lymph node evaluation was not available in 66 patients. Estrogen receptor status was not available in 1754 patients. Of the factors listed in Table 1, T, N, pN and histopathology show significantly different incidences of metastasis to bone. Age, menstrual condition and estrogen receptor status do not show significant differences.
|
The results of logistic regression analysis are shown in Table 2. Univariate analysis or crude odds revealed that age, menstruation and estrogen status did not correlate with metastasis to bone. T, N, pN and histopathology were significantly correlated. Age and menstruation status were used to adjust the odds, because these two factors did not correlate significantly with metastasis to bone. N and pN are very close factors. The N factor was used in multivariate analysis or to calculate adjusted odds, because this factor was available for all patients whereas the pN factor was not and the N factor was obtained before surgery. No bone metastasis was present in patients with non-palpable tumors; data from these patients were omitted from logistic regression analysis. Data from patients with 1–20 mm tumors were regarded as the baseline data. The chances of tumor metastasis to bone increased with increasing tumor size. The crude and adjusted odds of metastasis for 21–30 mm tumors were not significantly different from those for 1–20 mm tumors. Tumors larger than 30 mm gave results significantly greater than the baseline data. Tumors 21–50 mm in size are classified as T2 in the TNM system. Consequently, a statistically significant discrimination point for increasing bone metastasis rate occurs at 30 mm. No metastasis to bone was detectable in patients with non-invasive cancer or Paget’s disease (Table 1). Patients with unclassified cancer were regarded as having tumors with a complex histological composition. Therefore, patients with non-invasive cancer, Paget’s disease or unclassified cancer were omitted from logistic regression analysis. Data from patients with papillotubular cancer were used as baseline data. The crude odds of metastasis were greater for solid tubular and other special types of cancers, but the adjusted odds were not statistically significant. The crude odds of metastasis in patients with scirrhous and invasive lobular cancers were significantly higher than the baseline odds. Forward-stepwise analysis revealed that T, N and histopathology correlated positively with increased metastasis to bone.
|
The incidence of metastasis to bone for each clinical stage is shown in Table 3. The incidence of metastasis to bone in stages 0 and I was very low, 0 and 0.08%, respectively. Only one patient initially diagnosed with stage I disease had metastasis to bone. She was diagnosed with scirrhous cancer staged at T1N0. However, histopathological examination of the resected tissue following her surgery revealed lymph node metastasis up to the supraclavicular region. Consequently, the stage of this patient’s disease had been improperly determined. She should have been diagnosed with stage IV disease. The incidence of metastasis to bone in patients with stage II disease was 1.09% (95% CI: 0.76–1.52%, 34/3120). The incidence of metastasis to bone in patients with stage III disease was 9.96% (95% CI: 7.80–12.47%) and the incidence in patients with stage IV was 34.04% (95% CI: 20.86–49.31%). Table 4 shows the pathological stage (TpN) and the incidence of bone metastasis. The incidence of metastasis to bone was 0% for stages 0 and I, 0.64% for stage II, 7.25% for stage III and 13.73% for stage IV.
|
|
Table 5 shows the incidence of metastasis to bone cross-referred with T, N and histopathological classification data for patients with stage II breast cancer. The incidence of metastasis to bone exceeded 3% in patients with large T2 tumors, N1 and histopathologically confirmed scirrhous, invasive lobular or miscellaneous cancers. Additionally patients with T3, N0 and histopathologically confirmed solid tubular or scirrhous cancer also had a greater incidence of metastasis to bone.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
Many studies have reported that the incidence of bone metastasis differs according to the clinical stage of breast cancer (2–8,12–14). Consequently, accurate determination of disease stage is important to determine the proper treatment to improve the patient’s prognosis. The opinion that staging by bone scintigraphy is not necessary in patients with stage I or II breast cancer is primarily economic, although some physicians think that staging bone scintigraphy is valuable for all breast cancer patients (11–15). For example, although Coleman et al. (12) reported that bone scintigraphy was not necessary for diagnosis in stage I patients, they recommended a baseline scan in all patients with stage II–IV disease. At our hospital, all patients with breast cancer have staging done by bone scintigraphy despite the many reports mentioned previously. Most of the reported studies concern Caucasian patients. Consequently, these results from Europe and United States may not be applicable to Japanese patients with breast cancer, because variations in breast cancer incidence and biological expressions exist between different races (16). The results of the present study indicate that the incidence of metastasis to bone in Japanese patients with breast cancer is not different from that reported for Western patients.
Some studies analyzed metastasis of breast cancer to bone using data on either tumor size with a minor modification (T2 tumors were further divided into two subgroups, 21–30 and 31–50 mm) or tumor histopathology (10,11). Two T2 tumor subgroups and histopathological classification were used in the present study.
It is a difficult responsibility to establish an ‘acceptable’ incidence of metastasis to determine which patients require staging by bone scintigraphy. Since many new methods for treating metastasis to bone have been developed (20,21), the clinical importance of early and accurate diagnosis of metastasis to bone may greatly affect the therapeutic outcome. Consequently, the decision to perform baseline scans has traditionally lain with individual physicians or surgeons or with patients themselves. To aid these parties in making this decision, the data were analyzed assuming that (1) bone scintigraphy is not necessary in populations with an incidence of <1%, (2) bone scintigraphy is recommended in populations with an incidence >3%, because the therapeutic strategy would change dramatically if the tumor metastasized to bone, and (3) 1–3% incidence groups were considered to be in a boundary or ‘gray’ area. Another criterion was that any group with incidences of metastasis to bone that were statistically different from the baseline data were considered candidates for staging by bone scintigraphy.
These tentative criteria were applied to clinical stage and a scenario drawn. The incidence of metastasis to bone was very low, 0% (95% CI: 0.00–0.77%, 0/486) in patients with clinical stage 0 disease and 0.08% (95% CI: 0.00–0.46%, 1/1212) in patients with clinical stage I disease. Consequently, bone scintigraphy is not considered necessary in these groups as a routine staging procedure. The incidence of metastasis to bone in patients with clinical stage III disease was 9.96% (95% CI: 7.80–12.47, 67/673) and the incidence of metastasis to bone in patients with clinical stage IV disease was 34.04% (95% CI: 20.86–49.31%, 16/47); therefore, staging by bone scintigraphy is recommended for patients with clinical stage III and IV. However, the incidence of metastasis to bone in patients with clinical stage II disease was 1.09% (95% CI: 0.76–1.52%, 34/3120) in stage II, which was in the gray zone between the two previously cited criteria. When pathological lymph node involvement is apparent, the scenario is clearer. No bone metastasis is detected in patients with pathological stages 0 and I and a very low incidence (0.64%) of metastasis to bone is found in patients with stage II. The incidence of metastasis to bone is high in patients with stage III and IV, 7.25 and 13.73%, respectively. With the criteria in this study, staging bone scintigraphy is not necessary in patients with pathologic stages 0, I and II and staging bone scintigraphy is recommended in patients with stages III and IV. However, pN cannot be known before surgery.
Multivariate logistic regression analysis revealed that the incidence of metastasis to bone in patients with 21–30 mm tumors was not statistically different compared with the incidence in patients with 1–20 mm baseline tumors. However, the incidence of metastasis to bone in patients with tumors >30 mm in size was statistically different compared with the incidence in patients with baseline tumors. Papillotubular cancer was used as the baseline data for logistic regression analysis of histopathology data. Patients with solid tubular cancer had an odds ratio of 1.68 and patients with other special types of cancers had an odds ratio of 2.67. These results were not significantly different from the baseline incidence. Patients with scirrhous cancer had an odds ratio of 4.17 (95% CI: 1.81–9.57) and invasive lobular cancer had an odds ratio of 4.21 (95% CI: 1.35–13.15); these ratios were significantly different from the baseline incidence after adjusted odds analysis. Using both tumor size and histopathological classification, patients with clinical stage II breast cancer were further classified into subgroups as shown in Table 5. The incidence of metastasis to bone exceeded 3% in patients with large T2 tumors, N1 and scirrhous or invasive lobular cancers. Patients with miscellaneous cancers, patients with T3, N0 and solid tubular or scirrhous carcinoma also had a greater incidence of metastasis to bone. Therefore, bone scintigraphy is recommended in patients with scirrhous or invasive lobular cancers whose tumors are >30 mm in size. Patients with solid tubular cancer or other special types of cancers belong to the gray zone between criteria regarding staging bone scintigraphy and bone scintigraphy may benefit these patients. Using these criteria, 71% (3943/5538) of staging bone scans currently performed at our institute could be avoided.
It is well known that the tumor marker CA15-3 is a good marker to establish the extent of disease in patients with breast cancer. Measurement of CA15-3 may help to select the patients with a high incidence of metastasis to bone (22). Bone metabolic markers are expected to pick up bone metastasis (23). The presence of bone pain may also help in selecting the patients with a high incidence of metastasis to bone. If these data are available, the predictability may improve and the conclusion may change. Regrettably, these data are not available in this study.
In conclusion, the initial incidence of metastasis to bone was analyzed using factors available before surgery in patients with breast cancer. Bone metastasis possibility correlated with larger tumor size (T), lymph node involvement (N) and types of histopathology by logistic regression analysis. By using a criterion that bone scintigraphy is not necessary in populations with an incidence <1% and bone scintigraphy is recommended in populations with an incidence >3%, the following conclusions can be drawn. Staging by bone scintigraphy is not routinely necessary in patients with stage 0, I or II with either a small tumor (21–30 mm) or stage II with a large tumor (31–50 mm) and low-grade histopathology (e.g. papillotubular cancer). Bone scintigraphy is recommended in patients with stage II with large tumor and high-grade histopathology (scirrhous cancer or invasive lobular cancer), stage III and stage IV disease.
|
Acknowledgments |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
We thank Etsuji Nomura, Yasuhiko Yamada, Tomohiro Takiguchi for excellent work in taking bone scans and Ms Mihoko Tajiri for helping with data collection. This work was partly supported by the Vehicle Racing Commemorative Foundation.
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Mitsuru Koizumi, Department of Nuclear Medicine, Cancer Institute Hospital, 37–1 Kami-Ikebukuro, 1-chome, Toshima-ku, Tokyo 170-8455, Japan. E-mail: mitsuru@jfcr.or.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAcknowledgmentsREFERENCES |
|---|
1 Galasko CBS. The anatomy and pathways of skeletal metastasis. In: Galasko CBS, Weber DA, Gilbert HA, editors. Bone Metastasis. Boston: GK Hall 1981;49–63.
2 Anger K, Moller T, Zeimantz E. Skeletal scintigraphy in the care of breast cancer. Strahlentherapie 1983;159:745–50.[ISI][Medline]
3 Khansur T, Haick A, Patel B, Balducci L, Vance R, Thigpen T. Evaluation of bone scan as a screening work-up in primary and recurrence of breast cancer. Am J Clin Oncol 1987;10:167–70.[ISI][Medline]
4 Yeh KA, Fortunato L, Ridge JA, Hoffman JP, Eisenberg BL, Sigurdson ER. Routine bone scanning in patients with T1 and T2 breast cancer: a waste of money. Ann Surg Oncol 1995;2:319–24.[Abstract]
5 Wikenheiser KA, Silberstein EB. Bone scintigraphy screening in stage I–II patients: is it cost-effective? Cleve Clin J Med 1996;63:43–7.[ISI][Medline]
6 Gudgeon CA, Werner ID, Dent DM. A re-evaluation of isotope screening for skeletal metastases in node-negative breast cancer. S Afr Med J 1996;86:166–9.[ISI][Medline]
7 Ravaioli A, Tassinari D, Pasini G, Polselli A, Papi M, Fattori PP, et al. Staging of breast cancer: what standards should be used in research and clinical practice? Ann Oncol 1998;9:1173–7.
8 Samant R, Ganguly P. Staging investigation in patients with breast cancer: the role of bone scan and liver imaging. Arch Surg 1999;134:551–3.
9 Jacobson AF, Fogelman I. Skeletal scintigraphy in breast and prostate cancer: past, present and future. In: Freeman LM, editor. Nuclear Medicine Annual 1999. Baltimore: Lippincott Williams & Wilkins 1999;121–56.
10 Lagrange JL, Hoch M, Courdi A, Darcourt J, Benoiel J, Namer M, et al. Breast cancer: the role of scintigraphy in the initial evaluation and follow up of patients with small tumours (T less than or equal to 3 cm). Bull Cancer 1988;75:271–80.[ISI][Medline]
11 Koyama T, Hasebe T, Tsuda H, Hirohashi S, Sasaki S, Fukumoto T, et al. Histologic factors associated with initial bone metastasis of invasive ductal carcinoma of breast. Jpn J Cancer Res 1999;90:294–300.[ISI][Medline]
12 Coleman RE, Rubens RD, Fogelman I. Reappraisal of the baseline bone scan in breast cancer. J Nucl Med 1988;29:1045–9.
13 Ciatto S, Pacini P, Azzini V, Neri A, Jannini A, Gosso P, et al. Preoperative staging of primary breast cancer. A multicentric study. Cancer 1988;61:1038–40.[ISI][Medline]
14 Brar HS, Sisley JF, Johnson RH Jr. Value of preoperative bone and liver scans and alkaline phosphatase in the evaluation of the breast cancer patients. Am J Surg 1993;165:221–3.[ISI][Medline]
15 Bases R. Skeletal scintigraphy in breast cancer management. Q J Nucl Med 1998;42:43–8.[ISI][Medline]
16 Hunter CP. Epidemiology, stage at diagnosis and tumor biology of breast carcinoma in multiracial and multiethnic populations. Cancer 2000;88:1193–202.[ISI][Medline]
17 International Union Against Cancer. TNM Classification of Malignant Tumours, 5th ed., Sobin LH, editor. New York: Wiley-Liss 1997.
18 The Japanese Breast Society. General Rules for Clinical and Pathological Recording of Breast Cancer, 13th ed. Tokyo: Kinbara 1998.
19 Sakamoto G. Infiltrating carcinoma: major histological types. In: Page DL, Anderson TJ, editors. Diagnostic Histopathology of the Breast. London: Churchill Livingstone 1987;193–235.
20 Hortobagyi GN, Theriault RL, Lipton A, Pater L, Blayney D, Sinoff C, et al. Long-term prevention of skeletal complications of metastatic breast cancer with pamidronate. J Clin Oncol 1998;16:2038–44.[Abstract]
21 Serafini AN, Houston SJ, Resche I, Quick DP, Grund FM, Ell PJ, et al. Palliation of pain associated with metastatic bone cancer using samarium-153 lexidronam: a double-blind placebo-controlled clinical trial. J Clin Oncol 1998;16:1574–81.
22 Nicolini A, Ferrari P, Sagripanti, A, Carpi A. The role of tumour markers in predicting skeletal metastases in breast cancer patients with equivocal bone scintigraphy. Br J Cancer 1999;79:1443–7.[ISI][Medline]
23 Koizumi M, Takahashi S, Ogata E. Bone metabolic markers in bone metastasis of breast cancer. Int J Clin Oncol 1999;4:331–7.
Received November 30, 2000; accepted February 13, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Hamaoka, J. E. Madewell, D. A. Podoloff, G. N. Hortobagyi, and N. T. Ueno Bone Imaging in Metastatic Breast Cancer J. Clin. Oncol., July 15, 2004; 22(14): 2942 - 2953. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||