Optimising Risk-based Breast Cancer Screening in Hong Kong

Hong Kong J Radiol 2023 Dec;26(4):260-5 | Epub 12 Dec 2023

 

 

Department of Radiology, Queen Mary Hospital, Hong Kong SAR, China

 

  Full Article in PDF

Female breast cancer had the highest incidence among all cancers diagnosed in women in 2020, with an estimated 2.3 million new cases (11.7%). It accounted for one in four cancer cases and for one in six cancer deaths.[1] The aim of breast cancer screening programme is to reduce breast cancer mortality through early detection and treatment. The World Health Organization recommends population-based mammography screening biennially for women aged 50 to 69 years at average risk for breast cancer in countries with adequate resources.[2]

 

Although breast cancer is less common in Asian countries compared to the United States and Europe, its incidence has been increasing substantially over the past decades.[1] [3] [4] In Hong Kong, breast cancer is the most common cancer and the third leading cause of cancer mortality in women, with age-standardised incidence and mortality rates of 65.5 and 10.2 per 100,000 population, respectively, in 2019.[5] Most Western countries have well-established population-based mammogram screening programmes based on age.[6] [7] Our Asian counterparts (i.e., South Korea, Singapore, Taiwan, and Japan) have also introduced screening programmes in the past two decades.[8] In 2021, the Hong Kong Government launched a risk-based pilot screening programme based on the revised recommendations of the Cancer Expert Working Group on Cancer Prevention and Screening (CEWG) under the Centre for Health Protection.[9]

 

This article provides an overview of breast cancer screening, briefly discusses the background and updated recommendations, and focuses on the supplementary screening tools and future directions in risk-based screening in Hong Kong.

 

The goal of breast cancer screening is to detect breast cancer at its pre-clinical stage, so that it can be treated early to reduce disease- and treatment-related morbidity and mortality. Evidence has shown that detecting breast cancers at this early stage is associated with better outcomes.[10] The risk of metastasis and death increases with tumour size and number of involved axillary lymph nodes at detection.[11] [12]

 

The 5-year survival rate drops from 86% to 99% in localised and regional disease to 29% in distant disease.[13] Treatment of early cancers requires less extensive breast tissue resection and axillary lymph node dissection, and hence fewer complications and side-effects. It also reduces overall treatment costs and financial burden on the healthcare system. Potential harms, which lead to controversies of effectiveness of screening programmes, include false-positives, overdiagnosis, overtreatment, psychological stress of participants; and lead-time bias which leads to seemingly increased survival.

 

Careful consideration of the cost-benefit balance, and adherence to the World Health Organization principles outlined by Wilson and Jungner,[14] are important when implementing breast cancer screening programme With numerous randomised controlled trials demonstrating a 20% to 30% decrease in mortality from breast cancer, numerous countries have implemented population-based screening programmes.[15] Most well-established population-based breast cancer screening programmes offer biennial mammography to women aged 40 to 50 years to 69 to 74 years.[16] Subsequent studies with data generated from screening programmes have provided further evidence that screening mammography is beneficial.[17]

 

In 2010, the CEWG had adopted a simple and rather restrictive set of risk stratification criteria based on the presence of BRCA1 or BRCA2 genes, family history, and selected personal risk factors that led to only a small group of high-risk eligible women being screened.[18] For the last few decades, elective screening has been practised in Hong Kong in the private sector. Women attending such screening services are self-referred. The largest self-financed and self-referred mammography screening programme was organised by the Tung Wah Group of Hospitals, which has been offering mammography screening since 1993 with the number of examinations continuing to increase throughout the years.[19]

 

Recently, there has been a transition from elective screening to a broader risk-based approach. Based on the revised recommendations of the CEWG, the Hong Kong Government has rolled out a risk-based breast cancer screening pilot programme to provide screening services for eligible women over a period of 2 years. A local breast cancer risk stratification model for the Hong Kong Chinese population, developed by The University of Hong Kong based on the identified local risk factors for breast cancer, is employed.[9]

 

The new CEWG recommendations result in two additional groups of women being recommended for mammographic screening every 2 years. These groups are made up of women at moderate risk with a relevant defined family history, and women aged 44 to 69 with certain combinations of personal risk factors (including a history of breast cancer among first-degree relatives, a prior diagnosis of benign breast disease, nulliparity and late age of first live birth, early age at menarche, high body mass index, and physical inactivity) putting them at increased risk of breast cancer.[9]

 

Population-based screening programmes based on different age groups have long been implemented in Western countries. They have proven to reduce breast cancer–related mortality effectively.[20] On the other hand, there was increasing evidence in favour of advocating for risk-based breast cancer screening due to potential higher cost-effectiveness in concentrating the resources on screening women with increased risk.[21] [22] Some Western countries are also transitioning to risk-based screening approaches that do not rely on age alone.[23] [24]

 

Risk-based programmes are aimed at women who are more likely to benefit, thus reducing the risk of causing harm to women at lower risk, and allow resources to be used more efficiently. In Hong Kong, although the age-standardised incidence rate of breast cancer is rising, it still remains relatively low when compared to Western countries. It is known that performing screening mammography in populations with relatively low disease prevalence would lead to higher false-positive rates and hence unnecessary biopsies, creating potential complications and psychological distress.[25] Therefore, under these circumstances, a personalised risk-based approach may be more cost-effective than universal age-based screening for Hong Kong Chinese women.[26]

 

The effectiveness of a risk-based screening programme depends much on the accuracy of individual risk estimation. In terms of risk stratification, the Gail model is one of the earliest breast cancer risk assessment tools that has been developed, validated and calibrated to be deployed in different populations.[27] [28] More recently, the Breast Cancer Surveillance Consortium model and the Tyrer-Cuzick model have included breast density in risk assessment, which demonstrated modest performance improvement.[29] [30] Histories of hyperplasia and lobular carcinoma in situ are also included.[29] [30] With more local radiological and biomarker data available, the Hong Kong breast cancer risk stratification model can be improved to cover additional risk factors and identify women with increased risks effectively in the future.

 

Increased mammographic breast density is an independent moderate risk factor for breast cancer. Women with extreme density are 4 to 6 times more likely to develop breast cancer than those with fatty breasts. Furthermore, extreme breast density is a solitary risk factor that puts women into higher lifetime and 10-year risk categories for breast cancer.[31] It is regarded as the same risk category (relative risk of 2.1 to 4) with ductal carcinoma in situ, high endogenous postmenopausal hormonal levels, high-dose radiation to the chest, and two or more first-degree relatives with breast cancer.[32]

 

Mammography is known to have lower sensitivity in women with dense breasts, including the Chinese population, younger and premenopausal women, and those with genetic predispositions to breast cancer, due to the increased mammographic density masking the radiological features of early breast cancer.[33] Therefore, it leads to more interval cancers and higher cancer stages at diagnosis. In addition, superimposed glandular tissue can also mimic the presence of a lesion, resulting in reduced specificity, increased recall rates, and unnecessary investigations.

 

In view of the associated increased cancer risk and mammographic masking effect in the relatively denser breast tissue in Chinese women, some recommend the use of supplementary screening tools to increase screening sensitivity and specificity.[34] Digital breast tomosynthesis (DBT) generates pseudo–three-dimensional images, which can resolve superimposition of breast tissue, thus increasing lesion visibility and reducing unnecessary recalls due to summation artifacts. It has been shown to have higher sensitivity and specificity compared to traditional digital mammography.[35] Ultrasound has been shown to reduce interval cancer rates for women with dense breasts when added to mammography. However, it is only suggested to be considered as a supplementary tool on a case-by-case basis due to its high false-positive rate.[36]

 

Breast contrast-enhanced magnetic resonance imaging (MRI) provides physiological parameters related to tumour angiogenesis in addition to anatomical assessment. It has been widely accepted for screening women who are at high risk for breast cancer, such as confirmed carriers of BRCA1 or BRCA2 deleterious mutations in genetic testing and those who had radiation therapy to the chest between 10 to 30 years of age for Hodgkin lymphoma.[9] Kuhl et al[37] found that MRI could identify an additional 15.5 cancers per 1,000 cases in women at average risk of breast cancer. The DENSE (Dense Tissue and Early Breast Neoplasm Screening) trial has shown that using supplemental MRI screening in women with extremely dense breasts resulted in significantly fewer interval cancers than mammography alone.[34] However, due to long imaging time and limited availability, MRI is restricted to screening a limited number of high-risk women. Abbreviated MRI (AB-MRI) is a shortened version of the standard MRI protocol. By retaining a dynamic contrast-enhanced MRI sequence and one to two other sequences (depending on individual institutions), the examination time is shortened to 10 minutes. According to the American College of Radiology Imaging Network EA1141 trial,[38] AB-MRI was superior to DBT in detection of both invasive breast cancer and ductal carcinoma in situ, with 2.4 times higher detection rate in women with dense breasts. The positive predictive values of AB-MRI and DBT were shown to be similar.[38] Current evidence demonstrates that this technique has the potential to supplement mammography screening in women with dense breast tissue and increased risk of breast cancer. The shortened AB-MRI protocol and examination time increase the availability of MRI, allowing more women to be screened. It may potentially be included in the future screening programme.

 

Contrast-enhanced digital mammography (CEDM) is one of the latest advances in breast imaging. It uses a dual-energy technique performed after intravenous administration of iodinated contrast to identified enhancing lesions. The underlying principle is based on tumour physiology—tumoral angiogenesis increases vascular permeability, resulting in enhancement. Studies have shown that CEDM can improve diagnostic accuracy in evaluation of screening recalls. Initial findings evaluating the application of CEDM in high-risk screening have shown comparable specificity and positive predictive values with MRI. This suggests that CEDM may be useful as an alternative when MRI cannot be performed because of patient contraindication or inaccessibility.[39]

 

To improve the effectiveness of a risk-based screening programme, accurate risk stratification is essential. Incorporating mammographic breast density in the Hong Kong risk prediction model when more radiological data are available in the future can further enhance the discriminative power of the model to identify women who would benefit from screening.

 

Diagnostic accuracy depends on interpreter training, skills, and experience. Regular feedback from outcome of screening through assessment, follow-up results and radiological-pathological correlation can enhance the performance of radiologists. Measures such as regular audits and review of interval cancers should be implemented. Double reading is practised in some screening programmes to increase screening performance. Computer-aided detection algorithms can identify areas of abnormal density, morphology, and calcifications and mark them on an overlay image. It is most frequently used as a prompt to radiologists for special consideration during interpretation. Ongoing studies have evaluated it as a surrogate for a second reader.[40] A meta-analysis of five retrospective studies demonstrated better performance in machine learning mammographic breast cancer detection (area under the curve = 0.89) than radiologists (area under the curve = 0.85).[41] Other applications of artificial intelligence in breast cancer screening include lesion characterisation, determination of lesions’ malignant probability, and triage of the worklist to streamline workflow.[42] [43] A previous study reported that machine learning reduced the number of mammography reads by radiologists by 17% to 91% with 0% to 7% cancers missed.[41] Local regulations, guidelines, and recommendations should be adopted to ensure the quality assurance of mammography screening.

 

The level of participation and compliance in screening are influenced by personal, socio-economic, and cultural factors. Informed decision-making is important since screening has both positive and negative impact for individuals. Women should be fully informed about the benefits, limitations, and harms under both ethical and legal considerations. Measures can be taken to address the psychological consequences of mammography screening (such as hotlines, follow-up clinics, etc.) to alleviate patients’ psychological distress and anxiety. Finally, participation can be influenced by how the screening invitation is made, how access to screening is organised, and how effectively breast cancer awareness is promoted, which are all amendable.

 

The Hong Kong risk-based breast screening programme is in its early phase. Its cost-effectiveness, which depends on multiple modifiable factors, requires continuous evaluation and improvement. The effectiveness of a risk-based screening programme is determined by the discriminative power of the risk stratification model. Inclusion of individual factors such as breast density, histology results of breast biopsy, and biomarkers may modify it to a more comprehensive model. Personalised screening tailored to individuals’ risks and preferences maybe the future direction in Hong Kong and worldwide.