Japanese Journal of Clinical Oncology Advance Access published online on September 1, 2007
Japanese Journal of Clinical Oncology, doi:10.1093/jjco/hym090
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© 2007 Foundation for Promotion of Cancer Research
Relative and Combined Performance of Mammography and Ultrasonography for Breast Cancer Screening in the General Population: a Pilot Study in Tochigi Prefecture, Japan
1 Epidemiology Unit, Research Institute, Tochigi Cancer Center
2 Pediatrics Unit, Oomuta National Hospital, Omuta, Fukuoka
3 Department of Surgery
4 Department of Diagnostic Imaging, Tochigi Cancer Center
5 Medical Director
6 Department of Health Checkup and Examination, Tochigi Public Health Service Association
7 First Department of Surgery, Dokkyo Medical College, Shimotsuga-gun, Tochigi
8 Tochigi National Hospital, Utsunomiya
9 Health Promotion Division, Tochigi Prefectural Government, Utsunomiya
10 Tochigi Public Health Service Association, Utsunomiya
11 Tochigi Cancer Center, Utsunomiya, Japan
For reprints and all correspondence: Satoshi Honjo, Pediatrics Unit, Oomuta National Hospital, 1044-1 Tachibana, Omuta, Fukuoka 837-0911, Japan; E-mail: satoshihonjo{at}hotmail.com
Received January 13, 2007; accepted May 25, 2007
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
Background: Breast cancer screening by mammography is thought to be effective in reducing breast cancer mortality while ultrasonography is not accepted as a population screening modality, although the latter has been suggested to be useful in detection of cancer in the dense breast, relatively more typical for a younger woman.
Methods: Mammography with medio-lateral oblique view was offered on trial in 1999-2000 for 3453 female residents in Tochigi prefecture who also underwent clinical breast examination and ultrasonography. The municipalities that provided cancer screening were informed of the final diagnosis for women with positive findings in the screening trial by doctors who performed the diagnostic evaluation. Linkage was also made between the list of participants in the trial and registrations at Tochigi Cancer Registry for breast cancer cases diagnosed during 1999–2001.
Results: Thirteen cases with breast cancer were identified during a 2-year follow-up period: 10 were diagnosed subsequent to positive finding in the trial; two were negative in the trial and diagnosed 23 and 24 months after, respectively; and one had a positive finding at the trial but was undiagnosed at first and then diagnosed 18 months after the trial. Among the 11 cases judged as positive in the trial, four were judged only by mammography while three were judged only by ultrasonography. Those mammography alone-detected cases were relatively young, at 36, 40, 47 and 54 years of age, respectively, while the ultrasonography alone-detected cases were aged 50, 55 and 68, respectively.
Conclusions: Combined screening with mammography and ultrasonography may be feasible. A larger study is required to evaluate relative performance of mammography and ultrasonography in detail by characteristics of examinees and their breasts.
Key Words: follow-up studies • registries • public health • density, mass
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
Breast cancer screening by mammography is thought to be effective in reducing breast cancer mortality (1), while ultrasonography is not accepted as a population screening modality (2), although the latter has been suggested to be useful in detection of cancer in dense breast tissue (3,4). In the early 1980s, Kopans and colleagues conducted their triple-blind study with clinical breast examination (CBE), mammography and whole-breast water path ultrasonography among 1140 referred patients (5). Because of low sensitivity (64%) and the fact that no cancers developed in women with suspicious findings only by their ultrasonography (5), they concluded that ultrasonography should not be used as a routine modality for screening. However, the feasibility of screening by ultrasonography alone or in combination with other methods has been little examined for the general population.
Breast cancer has the highest estimated age-standardized incidence rate in Japan (36.0 per 100 000 females) with the peak (128.7 per 100 000) in the late forties in 1999 (6). It was estimated that only 10.5% of the target population, women aged 30 or older, received screening with CBE alone or in combination with ultrasonography, while only 1.2% underwent screening mammography provided by municipalities (cities, towns and villages) from April 2000 to March 2001 (7). The Ministry of Health, Labour and Welfare of Japan issued an amendment in 2000 to the guidelines for breast cancer screening to municipalities, instructing that mammography should be offered for women aged 50 or older annually, and another amendment in 2004 instructing that mammography should be carried out for women aged 40 or older biannually.
Under the circumstances, a pilot study for introducing mammography as a screening method was planned in 1999 in Tochigi prefecture, where approximately 16–17% of women aged 30 or older had participated in breast cancer screening by municipalities before 1999. Although the screening had been mostly with clinical breast examination (CBE) alone, 30–40% of the examinees had undergone ultrasonography as well since 1997 (8). Choice between the two schemes had been dependent on municipalities. In the pilot study of mammography in 1999, not only this modality but also CBE and ultrasonography were also offered simultaneously, as described below. Taking advantage of this opportunity, we studied the relative and combined performance of these modalities in mass screening for general population.
|
SUBJECTS AND METHOD |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
Breast Cancer Screening Trial with Mammography in 1999
Tochigi Public Health Service Association conducted screening with CBE and ultrasonography among 25 974 inhabitants based on consignments from the municipalities in the fiscal year 1999. To 3455 women who had undergone screening from October 1999 to March 2000, a free mammography was offered simultaneously. The aim of the mammography was explained with a leaflet, and none refused to undergo it.
Mammography was performed according to a manual (9) with a machine (MGU-200B, Toshiba Medical) using a screen and film system (Min-R2000/Min-R2000, Kodak). The mammogram was taken from a medio-lateral oblique view, and judged by two doctors independently, and findings were recorded as no abnormalities, benign calcification, benign mass, normal lymph node, mass, calcification, and others. A more cautious judgment was made for the final category when the judgments were discordant. Breast density according to the American College of Radiology's Breast Imaging Reporting and Data System (BI - RADS) (10) was recorded at the screening. Using 2943 women for whom findings by two doctors were available, agreement on the evaluation of breast density was examined and found to be fairly good (weighted kappa = 0.4760).
Ultrasonography was performed with a machine having a 7.5 MHz single probe mechanical scanner (Aloka SSD125, Aloka) or with a machine having a 7.5 MHz linear electronic scanner (Aloka SSD900). Technicians scanned entire breasts and recorded 2–40 still images in a magneto-optical disk in 3.5 min on average per woman. A single consultant radiologist (H.M.) judged all the recorded images. The technician who took images attended the evaluation by the radiologist. Findings included no abnormalities, cyst, and mass. During 1997–1998, 45 cases were detected in 43 210 women with CBE and ultrasonography (detection rate = 0.10%) (8). CBE was performed by surgeons and was reported to take 1.5–2 min on average per woman.
For each modality, findings were categorized into: (1) no abnormalities; (2) changes not suggestive of diagnostic examination; or (3) changes suggestive of diagnostic examination. Thus, the findings (1) and (2) were regarded as negative. Examination by each of the three modalities was performed independently of the other two. The most severe diagnosis was taken as the final one when diagnoses from the three modalities were discordant. For two women, mammography resulted in a poor study, and the final judgement was based on the other two modalities. Women with findings in the third category were recommended to undergo diagnostic examinations and tests by surgeon or gynaecologist. The choice of modalities for the diagnostic examination was dependent on the examining doctor.
Follow-up of Participants in the Trial
Municipalities that provided cancer screening were informed of final diagnoses for women with positive findings in the screening trial by the doctors who performed the diagnostic evaluation, although this notification system was not expected to work mandatorily.
In searching for other cases among women, especially those who were negative in the trial, a link was made for the present study between the list of participants in the trial and registrations at Tochigi Cancer Registry for breast cancer cases diagnosed during 1999–2001. The registry had not followed back cases that had come to the registry's attention via a death certificate mentioning cancer, and the proportion of the death certificate only (DCO) cases was 16.9% for breast cancer in this period. The linkage between the screening findings and the cancer registrations was approved by the institutional review board of Tochigi Cancer Center, and by the committee of Tochigi Cancer Registry.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
Mammography in the screening trial was positive for 279 women, who were recommended to undergo further examination (referral rate defined as proportion of examinees who were recommended for further examination = 8.1%). The corresponding figures for ultrasonography and clinical breast examination were 165 and 159 (4.8 and 4.6%), respectively. Thus, 530 women were judged positive by any of the three screening modalities (15.3%).
Ten cases with breast cancer were identified by 30 September, 2000, through 443 notifications by doctors in charge of diagnostic examination (Table 1). A further three cases were identified in May 2006, by the record linkage (cases 11–13 in Table 2) while five cases that had been identified by doctor's notification to the municipality were found not to have been reported to the cancer registry (cases 3–7 in Table 1). Case 11 had undergone screening with a negative result in October 1999, and breast cancer was diagnosed in November 2001 following a hospital visit due to a symptom, which was a palpable mass. Case 12 was found to have received the final diagnosis from one of us (J.A.) and detailed information was obtained. Although this case had been designated with a mass in her right breast by mammography and ultrasonography and a mass in her left by CBE at the screening, neoplasms had been ruled out by diagnostic examination at a cancer center. She visited the hospital again in 2001 due to a palpable mass detected by her in the left breast, and a cancerous lesion was detected. For case 13, cancer had also not been revealed in the next round of screening by CBE and ultrasonography in 2000 but was detected in 2001 by screening mammography and diagnosed as stage II (Table 2). Based on the condition that breast cancer screening was to be offered annually until 2004, no interim cancer was identified in the participants in the trial in 1999–2000 and the detection rate was 0.29%. The sensitivities of mammography, ultrasonography and CBE were 70.0, 60.0 and 20%, respectively.
|
|
Sensitivity and specificity were calculated again allowing for instruction from the Ministry of Health, Labour and Welfare of Japan in 2004 that the screening interval should be 2 years; case 12 was treated as non-interim cancer. Thus the two indices were calculated as follows on this assumption. Eight cases were detected by mammography among 3453 after excluding two women with poor studies, and the sensitivity and specificity were 61.5 and 92.1%, respectively. Ultrasonography detected seven cases, and the sensitivity and specificity were 53.8 and 95.4%, respectively. CBE detected two cases, and the sensitivity and specificity were 23.1 and 95.5% for CBE. There was no case detected exclusively by CBE alone. Combination of mammography and ultrasonography detected 11 cases, and the sensitivity and specificity were 84.6 and 88.4%, respectively. For the three cases judged negative by mammography but positive by ultrasonography in the screening trial (cases 5, 7, and 10), the diagnostic mammography was reviewed; case 10 was detected with a spot compression view but not with medio-lateral oblique and cranio-caudal views. Ultrasonography did not detect a cancer in the four youngest women in the present study (1–3 and 11).
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
The present study has several limitations. Firstly the sample size was small. Secondly the quality of the cancer registration was substandard as indicated by the high proportion of DCO cases. In fact, five cases (38%) in the present study were found not to have been reported to the cancer registry in the prefecture. This underreporting may possibly bias the estimation of sensitivity in both directions. In spite of these shortcomings, findings from the present study may be worthwhile, because relative and combined performance of simultaneous mammography and ultrasonography has been little examined in the general population elsewhere worldwide.
The Detection rate by mammography for breast cancer, 0.290%, for the present study was comparable with 0.264% for the trial in Miyagi (11) and 0.243% (12) for the study in Tokushima, and lower than 0.600% for the other study in the same prefecture (12). The sensitivity for mammography in the present investigation appeared comparable to those for the studies in Miyagi and Tokushima assuming that the screening interval was 1 year (P = 0.07, 0.12 and 0.70, respectively) (11,12). Because linkage with cancer registration to identify interval cancer was not mentioned in the studies in Miyagi (11) and Tokushima prefecture (12), interpretation of the direct comparison of sensitivities between the studies in the three prefectures should be limited.
Kopans summarized as follows: ‘huge efforts were made to develop ultrasonography as a breast-cancer screening technique’ (13). Ultrasonography has been proved to be effective in the detection of clinically and mammographically occult cancers in some studies (3,4,14) but not all (5). The effectiveness of this modality has not been tested in general populations, and one impeding factor against randomized studies may be time and cost. Buchberger and colleagues examined 6113 women with clinical and mammography-normal findings using ultrasonography with a handheld probe in 10–15 min on average per patient (4). In a study by Kolb and colleagues examining 13 547 women with dense breasts using ultrasonography with a handheld probe, the mean time for ultrasonography was reported to be 4 min 39 s (14). Because studies requiring the direct participation of a radiologist may be expensive, as suggested by Kopans (13), ultrasonography by a well-trained technician followed by a radiologist's diagnosis as employed in the present study would be worth considering. However, we should be modest in translating our preliminary experience to Western populations because of their higher breast cancer incidence and larger breast size (15).
Mammography is relatively insensitive for breast cancer in women with dense breast tissue, and breast density usually declines with age (10). Among five women aged 49 or younger in the present study, screening mammography and ultrasonography detected four cases and one case, respectively. Interpretation of this finding should be modest because breast density was recorded only for four out of all 13. Although results from the present study alone cannot support or change the current breast cancer screening programme in our country, the present findings may enhance large population-based studies to examine relative and combined usefulness of these two modalities.
|
Conflict of interest statement |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
Tochigi Public Health Service Association (H.M., M.I., K.T., M.S.) currently practices breast cancer screening by each or both mammography and ultrasonography based on consignment from municipalities.
|
APPENDIX |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
Breast Cancer Screening Judgment Committee
Dr Hideo Ishige, Kamitsuga General Hospital; Drs Isamu Ito, Hiroko Endo, Etsuo Takada, Dokkyo Medical College; Dr Shoichi Oka, Ohtawara Red Cross Hospital; Dr Shigeto Ogura, Oyama Municipal Hospital; Dr Hideyuki Kobayashi, Saiseikai Utsunomiya Hospital; Dr Hikaru Tamura, Haga Red Cross Hospital; Dr Tadao Noguchi, Noguchi Clinic; Dr Junji Furukawa, Tochigi National Hospital; Dr Hirofumi Mizunuma, Jichi Medical College; and Dr Noriaki Wada, Ashikaga Red Cross Hospital. Four (J.A., T.T., H.M., M.I.) of the authors were also members.
|
Acknowledgment |
|---|
This study was supported in part by grants-in-aid for Clinical Cancer Research (H16-023) and for Cancer Control (research on development of appropriate evaluating methods for cancer screening) from the Ministry of Health, Labour and Welfare of Japan. We thank the women in the mammography screening trial, and the doctors and technicians involved in the trial on whose contribution this study was based. We thank members of the Breast Cancer Screening Judgment Committee (Appendix). We thank the staff of Health Promotion Division, Tochigi prefectural government and of cities, towns and villages in the prefecture for their cooperation. We thank Mr Noriyuki Yamane, Mr Yoshikazu Satoh and Mr Takayuki Kimura, Tochigi Public Health Service Association, for their help in logistics and data management.
|
Footnotes |
|---|
T. Hasegawa is currently working with Tokyo Health Service Association. 
|
References |
|---|
|
TOPAbstractINTRODUCTIONSUBJECTS AND METHODRESULTSDISCUSSIONConflict of interest statementAPPENDIXReferences |
|---|
1 Humphrey LL, Helfand M, Chan BK, Woolf SH. Breast cancer screening: a summary of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med (2002) 137:347–60.
2 Teh W, Wilson ARM. The role of ultrasound in breast cancer screening. A consensus statement by the European group for breast cancer screening. European Journal of Cancer (1998) 34:449–50.[CrossRef][Web of Science][Medline]
3 Gordon PB, Goldenberg SL. Malignant breast masses detected only by ultrasound. A retrospective review. Cancer (1995) 76:626–30.[CrossRef][Web of Science][Medline]
4 Buchberger W, DeKoekkoek-Doll P, Springer P, Obrist P, Dunser M. Incidental findings on sonography of the breast: clinical significance and diagnostic workup. AJR Am J Roentgenol (1999) 173:921–7.
5 Kopans DB, Meyer JE, Lindfors KK. Whole-breast US imaging: four-year follow-up. Radiology (1985) 157:505–7.
6 Ajiki W, Tsukuma H, Oshima A. Cancer incidence and incidence rates in Japan in 1999: estimates based on data from 11 population-based cancer registries. Jpn J Clin Oncol (2004) 34:352–6.
7 Statistics and Information Department Minister's Secretariat Ministry of Health Labour and Welfare. (2002) Tokyo: Health and Welfare Statistics Association. Report on local health services and health services for the aged fiscal year 2000 (in Japanese).
8 Morikubo H, Moriai M, Watanabe A, Hara H, Saitoh S, Yamane N, et al. Current status on breast cancer screening in Tochigi prefecture that changed to the system using ultrasonography for all subjects (in Japanese). Month J Med Image Info (Eizo-Joho Medical) (2000) 32:522–6.
9 Guideline Drawing Committee Japan Association of Breast Cancer Screening. [Guideline for Breast Cancer Screening System Introducing Mammography] (in Japanese). 1st edn. Tokyo: Shinohara New Publisher 1997.
10 Harris JR, Lippman ME, Morrow M, Osborne CK. Diseases of the Breast. (2004) 3rd edn. Philadelphia, PA: Lippincott Williams & Wilkins.
11 Ohuchi N, Yoshida K, Kimura M, Ouchi A, Shiiba K-i, Ohnuki K, et al. Comparison of false negative rates among breast cancer screening modalities with or without mammography: Miyagi Trial. Japanese Journal of Cancer Research (1995) 86:501–6.[CrossRef][Web of Science]
12 Morimoto T, Sasa M, Yamaguchi T, Kondo H, Sagara Y, Kuwamura S, et al. Effectiveness of mammographic screening for breast cancer in women aged over 50 years in Japan. Japanese Journal of Cancer Research (1997) 88:778–84.[CrossRef][Web of Science]
13 Kopans DB. Breast-cancer screening with ultrasonography. Lancet (1999) 354:2096–7.[CrossRef][Web of Science][Medline]
14 Kolb TM, Lichy J, Newhouse JH. Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations. Radiology (2002) 225:165–75.
15 Maskarinec G, Nagata C, Shimizu H, Kashiki Y. Comparison of mammographic densities and their determinants in women from Japan and Hawaii. Int J Cancer (2002) 102:29–33.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||