Screening for women at midlife: we need to remain vigilant

Welcome to this special edition of Climacteric on screening. May I start by thanking the people who have contributed manuscripts to this edition, providing their highly valued perspectives to the journal on this extremely important issue for women at mid-life. Not surprisingly, there is a strong emphasis on screening for cancer, although clearly the concept of screening has a much broader application, as we see in the paper from Tobie de Villiers on screening for osteoporosis¹.

Nevertheless, the focus on screening for cancer fits with the increasing concern women have about cancer as they age. There is also an emphasis on the ‘gender-specific’ cancers of breast (although men develop breast cancer, their risk is much lower than it is for women), cervix and ovary, although we need to remember the important contribution of lung and bowel cancers to cancer deaths in women.

Climacteric is an international journal and the perspective each of us may have on cancer from our own country and own community may be very different from what is seen in other settings.

Cancer is the second leading cause of death world-wide, having caused 8.8 million or one in six deaths in 2015². On a global scale, the most common causes of cancer death are lung, liver, colorectal, stomach and breast. Currently, about 70% of deaths from cancer occur in low- and middle-income countries.

The age-standardized incidence of cancer is increasing in a number of countries³. For example, the age-standardized incidence of breast cancer in Australia has increased by about 50% in the last 20 years. Although screening is contributing to this increase in some developed countries, recent changes in the prevalence of risk factors, such as obesity, are also playing a part. Obesity is now considered the second most important avoidable cause of cancer, after tobacco⁴.

About one-third of deaths from cancer are attributable to five behavioral risk factors: tobacco use, high body mass index, low fruit and vegetable intake, lack of physical activity and alcohol consumption². The proportion of cancer which is potentially preventable is even greater if we include vaccination for infectious causes of cancer including human papilloma virus (HPV) and hepatitis B². In relation to breast cancer specifically, the proportion of disease which is potentially preventable may be as high as two-thirds if a risk minimization approach is taken from childhood⁵.

Despite trends in incidence, in high-income countries age-specific mortality from many cancers is declining³. There have been major advances in the treatment of cancer which now include not only surgery, radiotherapy and chemotherapy but also biological therapy and precision medicine where treatment is tailored to the characteristics of the individual tumor⁶. Despite these improvements, gold-standard therapy may not be available in some countries or access may be possible, but not equitable, in others⁷.

In 1968, Wilson and Jungner published a set of criteria which should be met before the introduction of a screening program⁸. One of those criteria was that ‘the natural history of the condition, including development from latent to declared disease, should be adequately understood.’ The natural history of some cancers is much better understood than others. For example, we know that the cause of the vast majority of cervical cancer is infection with HPV⁹. We also know that most women who are sexually active become infected with HPV and that most clear the infection within a couple of years. Some women who do not clear the infection go on to develop carcinoma in situ and it takes on average about 10 years for a sub-set of carcinoma in situ cases to evolve into invasive cancer¹⁰. This understanding underpins the design of the new screening programs for HPV being introduced around the world, replacing the Pap smear¹¹.

The natural history of other cancers such as breast cancer is more variable. Some breast cancers develop quickly such that, even in a woman who is undergoing regular screening, such a cancer would become symptomatic between screens. Approximately one-third of breast cancers occurring in screened women present clinically between screens^12,13. Conversely, there are indolent breast cancers which develop slowly such that they will not present clinically in the woman’s lifetime and yet they are detectable by screening. The identification of such a tumor is described as ‘over-diagnosis’ because the woman would never have known she had breast cancer if she had not been screened. Furthermore, a woman with this type of indolent tumor who undergoes regular screening has more opportunities to be identified by mammography than a woman whose breast cancer is following a more aggressive course so that, compared with breast cancer which presents clinically, the spectrum of breast cancer identified by screening will have a better prognosis, irrespective of treatment¹⁴.

A woman in the over-diagnosed category is treated for a condition which she would never have known about if she had not been screened. She will live the rest of her life believing that she has a potentially fatal condition and, for her, the treatment is all harm. Her treatment may contribute to her premature death from other causes such as coronary artery disease which is accelerated by external beam radiotherapy¹⁵. There is evidence of multiple biological processes of aging which are accelerated in people who have been treated for cancer that are attributable to their treatment regimens¹⁶. Premature death due to other causes for women who are over-diagnosed and treated for breast cancer may explain why, although screening may contribute modestly to the recently observed reduction in breast cancer-specific mortality, there is no evidence that the introduction of mammographic screening is associated with a reduction in all-cause mortality^17,18.

Many mammographic screening programs now recognize the reality of over-diagnosis and have information about it on their websites^19,20, although the information is inconsistent²¹. There are now also decision aids which include the issue of over-diagnosis so that a clinician can help a woman work through the potential benefits and harms of screening mammography in a process of informed decision-making²². However, some countries are now rethinking their mammographic screening programs. Switzerland²³ and France²⁴ have their programs under review.

Another relevant shift in thinking in relation to breast cancer is the COMET trial. In women with low-risk ductal carcinoma in situ (DCIS), the COMET trial is comparing the risk of subsequent ipsilateral invasive cancer in women with low-risk DCIS managed by active surveillance compared with standard therapy involving surgery and radiotherapy²⁵. This trial is relevant because the diagnosis of DCIS was rare prior to the introduction of mammographic screening²⁶ and support for this trial is coming from patient advocates wanting to reduce overtreatment of DCIS²⁷. Could we rename low-risk DCIS without including the word carcinoma? In a parallel setting, renaming and less aggressive treatment have been suggested for encapsulated follicular variant of papillary thyroid carcinoma²⁸.

All of the contributors to this screening issue have discussed the balancing of benefits and harms of screening. However, both the general public and health practitioners tend to over-estimate the benefits of screening and under-estimate the harms²⁹. Applying any screening test to a large proportion of the population has enormous capacity to do harm. We have a responsibility to remain vigilant in the area of screening and continue to heed the advice of Wilson and Jungner⁸ written half a century ago.

Conflict of interest

No potential conflict of interest was reported by the author.

Additional information

Funding

Nil.