Examining Domestic and International Disparities in Breast Cancer Screening, Treatment, and Outcomes

Screening and Early Detection

First, it is important to clarify and distinguish between the terms “screening” and “early detection.” Screening refers to testing for and identifying cancer in healthy asymptomatic patients. In contrast, early detection refers to rapidly diagnosing a patient who may already be showing signs and symptoms of cancer. At present, organized breast cancer screening programs are largely limited to HICs, whereas screening in LMICs is rarely available. Private clinics may offer fee-for-service screening examinations for those of greatest means, but even strategies for early detection are not universally available in LMICs. This dearth in screening and early-detection opportunities no doubt contributes to the high rates of stage III–IV disease in sub-Saharan Africa, where delays are estimated to be as high as 9 months from symptom presentation to diagnosis.

Based on evidence suggesting a 20% reduction in breast cancer mortality over time owing to mammography, the World Health Organization (WHO) recommends organized population-based screening mammography. Given the potential harms of overdiagnosis, the role and value of screening mammography has been increasingly debated. Nevertheless, most professional organizations recommend that average-risk women over age 50 benefit from screening mammography and advocate for this cancer prevention measure on a population scale. There are varying recommendations for screening of women aged 40 to 49, with some organizations advocating for population screening of this age group and others recommending individual patient choice as an attempt to balance the potential harms of overutilized mammography with the benefits of early detection. These benefits include decreased mortality, which is most evident in the 60- to 69-year-old age group. The harms include temporary procedural pain or discomfort, anxiety and stress, false positives, overdiagnosis, and overtreatment, particularly among women aged 40 to 49, who have fewer screen-detected invasive and noninvasive cancers and more false positives than any other age group.

Cost-effectiveness analyses of various screening strategies using microsimulation models have yielded differing results, depending on the setting and population being examined. In the United States, for example, a screening strategy that starts annual mammography at the age of 45 and switches to biennial screening between the ages of 55 and 75 years was found to be the most cost-effective. Yet, for AA women, in whom premenopausal breast cancer is more commonly diagnosed than among White women, beginning screening at a younger age has been associated with increased survival and decreased breast cancer–specific costs. It is evident that even at the policy level, there is lack of consensus and clarity on an optimal, population-wide approach to screening, and the risks and concerns of women from racial and ethnic minority groups are not always prioritized as part of policy decisions that could significantly affect them.

These conflicting data have contributed to decreased mammography uptake and inconsistent adherence with professional guidelines among some groups of women. Specifically, women of color in the United States, the United Kingdom (UK), and Canada have lower screening participation rates than White women, disparities that are attributed to gaps in knowledge and awareness of screening services, as well as concerns about cost. One study found that more than 20% of women aged 50 to 74 in the United States were paying out-of-pocket costs for mammography. Access to primary care facilities with screening mammography, geography, and insurance status all correlate with screening utilization, particularly among minority groups. Moreover, even when barriers to access are overcome, facilities in underserved US communities tend to have lower-quality screening services with less digital mammography and fewer dedicated breast radiologists. Even in Europe, where most countries have both universal health coverage and organized population screening, there is decreased uptake among women with low socioeconomic status and/or who belong to ethnic and racial minority, rural, and immigrant populations.

Unfortunately, organized, population-based screening mammography is not possible in most low-resource settings. In fact, WHO has stated that screening mammography is not cost-effective in LMICs and that clinical breast examination (CBE) as a means to early detection should be further explored. The International Union Against Cancer’s (UICC) cancer-related beliefs and behaviors survey provides internationally comparable data on cancer risk behaviors, cancer diagnosis and screening, and cancer-related beliefs. It showed that only 2% of female respondents in Southeast Asia and 5% in Africa had had a prior mammogram, as opposed to 49% of respondents in North America and 49% in Northern and Western Europe. Similarly, a prior World Health Survey showed only 2.2% of women aged 40 to 69 from LMICs had ever received any breast cancer screening.

Based on published research and expert opinion, several organizations have formulated guidelines for breast cancer screening and early detection according to resource availability. The Lancet Commission on Health, Equity, and Women’s Cancers suggests that CBE is cost-effective and feasible. The Breast Global Health Initiative, which developed breast cancer screening and treatment guidelines based upon resource availability, also endorses CBE in the lowest-resource settings. The National Comprehensive Cancer Network (NCCN) has also adopted guidelines based on resource stratification and now recommends CBE in basic resource settings. More importantly, 25-year follow-up of a randomized clinical trial in Canada comparing CBE to mammography found no survival benefit to mammography. Additionally, 20-year follow-up of a randomized clinical trial of CBE versus no screening in India showed significant breast cancer downstaging and a 30% mortality reduction in women over age 50 in the CBE screening group. These findings further support CBE as a viable early-detection strategy in LMICs.

Regardless of which screening or early-detection methods are employed, lack of knowledge and awareness about breast cancer remains pervasive among patients, community members, and providers in low-resource settings. This deficit contributes to late presentation, late diagnosis, and overall mortality. Likewise, even in LMICs where screening methods are available in the public health system, awareness and utilization are low. Many studies have suggested that public awareness campaigns are critical to cancer control and prevention in low-resource settings. Furthermore, there is a need for provider education to help minimize diagnostic delay and instead facilitate early detection.

Diagnosis

In addition to insufficient early-detection programs in LMICs and poor screening access and utilization in HICs, other factors contributing to delayed breast cancer presentation include poverty, disease-related taboos, religious beliefs, misconceptions about the disease, and fear of mastectomy. Women’s autonomy in health care decision-making may also be limited in some cultures. Cancer-related stigma—often stemming from myths about disease and its sociocultural implications—drives delay in many settings, contributing to advanced stage at diagnosis. Interestingly, although a randomized controlled trial of CBE screening in the Philippines showed greater than 90% uptake of CBE, only 35% of participants completed diagnostic follow-up and over 40% refused further workup. In addition to such patient-level delays, a systematic review in Africa elucidated several provider- and health system–level contributors to delayed diagnosis and treatment. For example, in some settings, as many as four visits to a health care provider will take place prior to a cancer diagnosis being made, ultimately leading to delays as high as 6 months from symptom to diagnosis. Notably, delays were longer in countries in sub-Saharan Africa compared with those in North Africa. Similarly, studies in the United States show that racial/ethnic minority women have significantly longer time to diagnosis compared with White women.

Once a patient accesses a provider and the provider is concerned a patient may have breast cancer, imaging and tissue sampling are needed to confirm and establish the diagnosis. Yet globally, only approximately 60% of countries have publicly available breast cancer diagnostic services. There is significant geographic variation in this availability based on the income status of individual countries (75% in HICs vs. 35% in LICs), as well as regions (100% in Southeast Asia vs. 30% in Africa).

Arguably, the most important component of breast cancer diagnosis is high-quality pathologic evaluation including tumor receptor status, as biomarker data play a significant role in determining treatment protocols. Yet, few countries in LMICs can provide pathology services to the general public. Thus clinicians are faced with incomplete information to make appropriate treatment decisions. In fact, there are several African countries without a single pathologist, and no country in sub-Saharan Africa has more than five pathologists per million people. In contrast, in the United States and the UK, the ratio of pathologists to the general population is about 1:50.

In parallel, mode of tissue sampling for diagnosis shows significant variation. In HICs, the preferred method is core needle biopsy (CNB), often image guided. However, in LMICs, CNB is often unavailable or too expensive, thus fine-needle aspiration (FNA) is used. However, this method is unable to provide biomarker data or distinguish between in situ and invasive cancer. Excisional biopsy is also very common in LMICs, as less invasive techniques such as CNB or FNA are often unavailable. Notably and in a similar vein, low-income women in the United States are more likely to be diagnosed via excisional biopsies (as opposed to CNB or FNA) than higher income women.

Even after a diagnosis of breast cancer has been made, disparities exist with regards to the postdiagnosis tests and studies performed to guide treatment. For example, genomic assays to help predict chemotherapy benefit are now standard of care for certain types of breast cancer and are even included in the breast cancer staging algorithm for the eighth edition of the American Joint Commission on Cancer (AJCC) staging manual. However, the high cost of these tests is prohibitive for LMICs, thus excluding entire cohorts of breast cancer patients from a diagnostic tool that could significantly alter their treatment plans.

Around the world, there are some notable trends in the distribution of breast cancer subtypes. For example, triple-negative breast cancer (TNBC) is more prevalent in AA women than White women in the United States, Black South African women than White South African women, and West African women than East African women. International comparisons of germline mutations among women of African ancestry (including AA women) from across the globe have demonstrated that the prevalence of BRCA mutations in this cohort ranges from 7% to 56%. In a similar vein, a systematic review and meta-analysis of breast cancer receptor subtype found that globally, women of African ancestry had less estrogen receptor (ER) expression than those of European descent. Furthermore, HER2+ breast cancer is more prevalent among women of Asian ancestry. With the advent of endocrine therapy, as well as targeted anti-HER2 therapy, the predominance of TNBC among women of African ancestry means many will not be able to benefit from our most effective systemic treatments, further limiting opportunities for treatment.