A study group “Quality control of breast cancer screening with mammography” was organized to investigate essential issues for the nationwide introduction of mammography in breast cancer screening in Japan.
Biennial clinical breast examination (CBE) with mammography demonstrated higher detection rate of early breast cancer, as well as higher sensitivity of breast cancer screening when compared to the current program, i. e., annual screening by CBE. The cost-effectiveness of screening with mammography was better than that of CBE alone. The life-years saved from breast cancer was calculated to much exceed the radiation-induced risk when mammographic screening would be conducted in women aged 50 yr and over.
The screening combined with mammograpy seems justified for Japanese women aged 50 yr and over from the aspects of cost-effectiveness and risk-effectiveness results. Utilization of mammography is a more appropriate system for breast cancer screening as compared to current program.

View full abstract

Clinical breast examination (CBE) is recommended as a modality for breast cancer screening in Japan. Now, the nationwide introduction of screening mammography (SMG) is being considered, since intervention trials using SMG have shown an improved detection rate of early breast cancer and a lower false-negative rate.In this study, we performed a cost-effectiveness analysis of breast cancer screening with or without SMG.
The total net costs per patient for CBE with SMG would increase by 1.15-1.40 as compared with CBE alone. The number of years of life saved per woman for CBE with SMG would be 2. 15 times higher than for CBE alone. Thus, the cost-effectiveness ratio (the cost per life-year saved) for CBE with SMG would be approximately 75% of that for CBE alone.
In comparison with the cost-effectiveness ratio of other forms of cancer screening, colon examination would have the lowest cost-effectiveness ratio, followed by stomach, cervical, breast (CBE with SMG), breast (CBE alone) and lung examination.
We conclude that breast cancer screening combined with mammography would be preferable to clinical breast examination alone in terms of cost-effectiveness in Japan. Further investigations, including the screening interval and the targeted age group, are required to reduce the total net costs.

View full abstract

For quality control in screening mammography, we have investigated the mammography classification performance of medical doctors in an image-reading study. In order to evaluate their performance in distinguishing breast cancer from benign tumor or normal breast by mammography, 25 medical doctors with varying levels of experience reported their breast cancer judgment by rating each mammogram. They interpreted the mammograms of 49 patients with breast cancer and 251 patients with benign tumor or normal breast. We conducted this image-reading study twice, with an interval of 3 months. From receiver operating characteristic (ROC) analysis and Brier score analysis, there was a statistically significant difference between each doctor's performance ; and the performance of medical doctors with a lower number of mammograms interpreted was significantly worse than that with a larger number of mammograms interpreted (p<0.05).
Therefore, this difference in mammogram interpretation among medical doctors was estimated to be due mainly to experience in reading mammograms. Furthermore, we investigated the interobserver agreement in judging the findings of mammography by kappa analysis ; the findings tested were calcification, circumscribed lesion, stellate lesion, and architectural distortion. The interobserver agreement for calcification was significantly highest among these mammography findings (p<0.05), and the interobserver agreement for architechtural distortion was very low.
From the above results, we confirmed that there was variability in the diagnostic accuracy of mammography among medical doctors, and that improvement of their performance was needed in order to improve the efficiency of screening mammography.

View full abstract

Two trials of mass screening for breast cancer using mammography were carried out in Tokushima prefecture. A total of 9 cases (0.24%) among 3, 707, and 14 cases (0.56%) among of2, 500 were diagnosed as breast cancer. In over half of the mammographic examinations, breast cancer was detected as a mass shadow. We consider that one of the quality control parameters for mass screening using mammography is imaging of mass shadow.

View full abstract

In 1991 and 1995, the quality of mammograms in the institutes participating in the Kawasaki Breast Cancer Screening Program were evaluated using questionnaires and the breast-phantom image test. In 1995 mammography equipment such as phototimers, and grid and focus size, were improved compared to those in 1991. In contrast, increased numbers of high-speed automatic film processors (45 seconds), which were used in six institutes in 1995, resulted in the use of low-contrast screen/film systems, an increase in radiation dose and relatively low scores in the breast-phantom image tests. In 1995, the number of institutes that passed the breast-phantom image test increased to 8/21 (38.1%), compared to 1/21 (4.8%) in 1991, if tests were performed on the basis of their own shooting conditions. The number of institutes that passed the tests was increased to 13/21 (61.9%) in 1995, when breast-phantom image test were carried out using the screen/film systems and shooting conditions recommended by us. Before the introduction of mammography for breast screening in Japan, quality assurance systems for mammograms should be properly established.

View full abstract

The success of screening mammography depends on the production of high-quality and low -dose images. Achieving high-quality studies at low dose requires vigilant attention to quality control for everyone involved in breast cancer screening. We evaluated 14 mammography units in Gunma Prefecture as outlined in the American College of Radiology Mammography Quality Control Manual (Revised Edition, 1994).
We selected five quality control programs from the ACR-QC-Manual.
1) Automatic exposure control (AEC) system performance assessment
2) RMI-156 phantom images
3) High-contrast resolution pattern measurement of limiting resolution
4) Beam quality assessment half value layer measurement
5) Breast entrance exposure average glandular dose
[Results]
1) In evaluation of the AEC system, the maximum difference in film density at each thickness of the acrylic slab (2, 4, 6 cm) was 0.71. Upon comparison of film density with the same thickness of acrylic slab in each unit, the maximum difference in film density was more than 1.0.
2) The RMI-156 phantom image of each unit passed 64.3% of the ACR mammography accrediation.
3) In measuring limiting resolution, the smaller focus size of the X-ray tube gave higher resolution. However, there was considerable difference even with the same type of unit.
4) All units with a Mo filter for HVL passed the ACR mammography accreditation.
5) The average glandular dose was regulated to lower than 3 mGy with the grid, but one of the units without such a grid had a value of over 1 mGy. The mean dose was 1.23 mGy with the grid, and 0.94 mGy without.

View full abstract

For the early detection of nonpalpable breast cancer, X-ray mammography is being considered as a breast cancer screening procedure. High-quality mammography screening requires stable and high-quality images which can display microcalcifications and low-contrast tissue structures with the minimum patient dose. To meet this requirement, mammography facilities have to use radiographic equipment specifically designed for mammography and to optimize the performance of the equipment (e.g., X-ray mammography unit, screen-film combination and film processor) and radiographic technique (e.g., patient positioning and breast compression). In order to achieve the required quality levels, facilities must establish and maintain quality assurance (QA) programs.
In addition, radiological technologists are required to master mammographic techniques and quality control (QC) methods before performing mammographic procedures, and must continue to make efforts to improve their radiographic techniques by participating in training and medical conferences related to mammography.
Finally, we propose that the Japan Association of Breast Cancer Screening establishes education programs related to mammography.

View full abstract

In incorporating mammography (MMG) into breast cancer screening, an essential issue is to establish a MMG QA/QC system by standardizing image quality and exposure dose, which are fundamental to MMG screening systems. At present, each subsystem related to QA/QC is gradually progressing toward completion through the efforts of those involved. Henceforth, it will be imperative to see how these subsystems can be organized and how to actually operate them. Particularly, it will be crucial to establish systems for mobilizing the personnel involved in QA/QC.
In this paper, we report on the following matters from a medical physicist's standpoint with the aim of ascertaining the relationship between MMG image quality and exposure dose : 1) development of a new technique for measuring MMG X-ray exposure dose ; 2) feasibility of a MMG exposure dose standard traceable system that makes use of this new measurement technique ; 3) creation of a database system for QA/QC phantom images in order to control MMG image quality ; 4) the current state of QA/QC in countries that are well advanced in MMG ; and 5) problems related to the operations of a QA/QC system.

View full abstract

It is important to detect early breast cancer by mass screening. The initial mass screening procedure consists of inspection, palpation or cytologic examination of nipple discharge. Three hundred eighty-five patients cases with nipple discharge were underwent nipple discharge cytology at our hospital from 1989 to 1995. Histological examination was done in 84 cases. Breast cancer were identified in 8 of 58 cases in Class I and II (13.8%), 4 of 13 cases (30.8%) in Class III, and 11 of 11 cases (100%) in Class V. The sensitivity, specificity and accuracy rates of nipple discharge cytology were 66.7%, 83.1% and 78.3%, and those for occult blood in nipple discharge were 100%, 44.8% and 68.3%, respectively. Five of 24 cases of breast cancer were TO cases. According to nipple discharge, 19 cases were detected by palpation, mammography and echography. For the detection of TO cancer, nipple discharge cytology is important.

View full abstract

We used fine-needle aspiration cytology (FNAC) to diagnose breast masses and discharge cytology for diagnosing non-palpable breast lesions with nipple discharge. Most palpable breast cancers were confirmed by a triple assessment involving physical examination, breast imaging and FNAC. Our results for sensitivity and specificity of FNAC in the diagnosis of breast mass were 97.6% and 98.2%, respectively, during the past 3 years, and FNAC was more reliable than other conventional diagnostic methods. In contrast, the sensitivity and specificity of discharge cytology for non-palpable breast lesion with nipple discharge were 62.5% and 64.7%, respectively. Since the diagnostic value of discharge cytology is low, duct-lobular segmentectomy (DLS) has been carried out for patients with nipple discharge to obtain a final diagnosis, and recently discharge cytology has been used mainly to select patients who require DLS. There is a great difference between the accuracy of FNAC and that of discharge cytology, and it is important to recognize this fact when diagnosing breast lesions.

View full abstract

Since 1977 mass screening for breast cancer has been conducted in Miyagi prefecture, Japan ; inspection, palpation and cytologic examination of any nipple discharge are part of the initial screening procedures. Nipple discharge cytology was performed on 61, 421 subjects (10.8% of the total). The cytologic diagnosis was positive in 54 subjects, cancer could not be ruled out in 85, epithelial hyperplasia in 1, 386, atypical findings in 7, 409 and no atypical finding in 52, 493. Discharge cytology in 626 cases (190 cancers and 436 benign) histologically identified had 20.5% sensitivity, 91.1% specificity, 69.7% accuracy, 50.0% positive predictive value and 72.5% negative predictive value if “cancer not ruled out” was classified as an error in both cancer and benign cases.
Discharge cytology was useful for detection of non-invasive and/or nonpalpable breast cancers and epithelial hyperplasia.

View full abstract

From 1985 to 1995, an exfoliative cytologic examination of abnormal nipple discharge was carried out on 2, 505 patients at Tokyo Metropolitan Cancer Detection Center. Out of 80 histologically confirmed mammary carcinomas, 38 (47.4%) were correctly diagnosed as cancerous, 9 cases being suspicious, and 33 cases negative. The cytologic diagnosis of discharge from 167 benign cases was positive in 5 cases, suspicious in 5 and negative in 157.
During the study period, 89 cases of nonpalpable breast cancer were detected. Of these cases, 35 had nipple discharge and 25 were diagnosed as malignant by nipple discharge cytology. No abnormal findings were observed by mammography and ultrasonography in 15 of these 25 cases.
The majority of cases of nonpalpable breast cancers detected by nipple discharge cytology were at the early stage. Nine of these 25 cases were noninvasive ductal carcinoma and 11 were papillotubular carcinoma with minimal invasion. Extension of cancer cells to the fat tissue was found in only 3 cases, and all cases were negative for lymph node metastasis.
Cytologica misdiagnosis was mainly attributable to insuffient materials and low atypia of cancer cells. In order to obtain satisfactory results of cytology, improved techniques of obtaining nipple discharge smears are required, and the cytologic diagnosis of cancer must be based on a combination of cellular morphology and the architecture of cell clusters.

View full abstract

Methods and results of cytologic diagnosis of nonpalpable breast cancers presenting nipple discharge were studied. During the period between January 1980 and December 1995, 61 consecutive cases of nonpalpable breast cancer with nipple discharge were reviewed at Sapporo Medical University Hospital.
For smear cytology (S method), the rates of correct diagnosis (Class III b, IV and V) and definitive diagnosis (Class IV and V) were 11.5% and 8.2% respectively, whereas those for ductal washing cytology (W method) were 32.8% and 16.4%. The rates of detection of atypical cells (Class III a, III b, IV, V) for both methods were 54.2%, which were not satisfactory. A fiberoptic ductoscopy system was developed in August 1989, and the diagnostic rates for the S and W methods have decreased since the introduction of endoscopy for diagnosis. However, the rates of correct and definitive diagnosis were 45.5% (10/22) and 36.4% (8/22) respectively for curette cytology under endoscopic observation (C method), which allows successful detection of minimal intraductal cancers in peripheral ducts.
The number of minimal breast cancers detected has recently increased due to the development of diagnostic imaging, and the curette method has become an important new diagnostic procedure for obtaining cell samples from intraductal lesions of the breast. The development of diagnostic imaging and cytologic diagnosis is prerequisite for early detection and diagnosis of nonpalpable breast cancers presenting nipple discharge.

View full abstract

The quality assurance of aspiration cytology for breast cancer in Japan currently lays behind that in Europe. The reason might be that in the United States of America during the last 30 years, aspiration cytology for breast cancer has been avoided by surgeons and cytologists because of medical lawsuit problems. Japanese doctors who were training in America during this period were effected by the social situation, and not by scientific opinion. We emphasize that accurate cytomorphological observation should be examined in more detail. American and Japanese cytology techniques for adenocarcinoma are rather simple and crude, for example, cytological diagnosis is done by the Papanicolaou criteria. However, Papanicolaou criteria for malignancy are set for degenerated squamous cell carcinoma, and not for aspirated fresh adenocarcinoma cells. Every cytological finding has to be investigated from the viewpoint of the degree to which it contributes to diagnosis. There is no absolute item important for smear evaluation. So, the important problem is how to find a combination of cytological criteria that leads to a better conclusion. Further research should be concentrated on this point. In addition, it should be discussed who has to do aspiration handling and make smear preparations. In some countries, surgeons take charge, whereas in other countries, cytologists aspirate and prepare smear specimens. The cytological results differ between the groups.

View full abstract

Since January 1981, mass screening for breast cancer has been carried out in Nagano Prefecture. The methods used for primary examinations have been physical examination combined with ultrasonic examination. Furthermore, since April 1994, occult blood examinations have been introduced for examinees with nipple discharge. One thousand two hundred eighty-nine (1.4%) subjects among approximately 91, 000 participants from April 1994 to March 1996, had nipple discharge. Of these, 1, 135 (88.1%) underwent occult blood examinations, of whom 404 were recommended to receive further examination, and 374 (92.6%) acutually did so. Four (1.1%) of these were diagnosed as having breast cancer by further examination, 5 (1.3%) as intraductal papilloma, 147 (39.3%) as mastopathy, 161 (43.0%) as no abnormality and 57 as other conditions.
The proportions of breast cancers related to the chraracteristics of the nipple discharge were 9.5% (2 of 21) in the bloody discharge group, 0.7% (1 of 137) in the serous discharge group, and 0% (0 of 106) in the milky discharge group. The proportions of breast cancers related to the occult blood reaction were 1.8% (4 of 228) in the blood positive group, and 0% (0 of 73) in both the suspicious positive group and negative group. One of 4 detected cancers was not revealed as a palpable mass.
It is suggested that the introduction of occult blood examination to breast cancer mass screening might be useful, as it was possible to reduce the number of patients requiring further examination in the group with nipple discharge, and the blood reaction of nipple discharge obtained from detected breast cancers was positive in all cases.

View full abstract

The benefit of mass screening for breast cancer becomes small when the sensitivity of screening mammography (MMG) decreases, and thus the benefit in terms of lives saved could be smaller than that of risk due to radiation exposure. In this study we carried out risk-benefit analysis of breast cancer mass screening for screening MMG sensitivites of 88% and 60%. The results showed that the age at which risk was equal to benefit became older with lower sensitivity, although the difference of age was not large. The age for equal risk and benefit was a little lower than the 25-29 year-old group for a sensitivity of 88%, whereas the age was slightly higher than 25-29 yr for a sensitivity of 60%. In conclusion, the age of screenees at which risk and benefit intersect does not change greatly when the senisitivity varies between 60% and 88%, provided that other conditions of mass screening do not change. Thus, the starting age of 35 for breast cancer mass screening in Japan is justified from the viewpoint of risk-benefit analysis.

View full abstract

Recently several kinds of gene related to breast cancer have been identified. These are BRCA1, BRCA2, and the ataxia-telangiectasia mutation gene (ATM). These discoveries raise some issues related to screening MMG, i. e. that persons heterogeneous for ATM may have a higher risk of breast cancer incidence, and also higher radiosensitivity than the general population. Investigations were carried out to clarify the following three issues: 1) Can radiation injury occur to an ATM-hetero person through screening MMG ? 2) Provided that an ATM-hetero person is screened by MMG, what will be the risk-benefit ratio of mass screening for this high-risk group ? 3) What will be the risk-benefit ratio when mass screening is performed on persons other than this high-risk group ? From the present study, the following conclusions were obtained: 1) Screening MMG does not cause radiation injury to an ATM-hetero person. However, quality control which guarantees the upper limit of the radiation dose is necessary because there are large variations in the radiation dose for screening MMG. 2) Screening of breast cancer for the ATM-hetero group by MMG is more effective than in the general population due to the higher incidence of breast cancer. 3) Screening MMG for the remaining population other than the ATM-hetero group has almost the same risk-benefit ratio as that for the general population.

View full abstract

In quality control (QC) procedures for mammography, the image quality of phantoms is assessed and scored. In this paper, a QC database including phantom images is proposed in order to record and evaluate QC data from many institutions. A trial was performed on a Power Macintosh using data obtained in 1991 and 1995 from selected institutions. Phantom images were analyzed basically by NIH Image, and contrast-to-noise ratio was measured for test materials contained in the phantom. Computerized analysis of images facilitated quantitative assessment and comparison of image quality.
QC information services will be offered to registered institutions in the form of QC reports in the near future. The reports will include computerized analysis of phantom images with chronological comparison and comparison with reference images of superior quality. This kind of service for a number of institutions will be conducted effectively through a centralized QC organization.

View full abstract