12 Screening women
Screening involves the systematic, population-wide recruitment and application of a test to symptom-free individuals considered to be at sufficient risk of a specified disorder to benefit from further investigation or direct preventive action.1 In other words, screening can be defined as the application of diagnostic tests or procedures to asymptomatic people for the purpose of dividing them into two groups: those who have a condition that would benefit from early intervention and those who do not.2
The most crucial aspect of a screening program, however, is that early detection must enable early intervention in order to reduce morbidity and mortality. If improved outcomes cannot be demonstrated, the rationale for screening is lost. Early diagnosis by itself does not justify a screening program. The only justification for a screening program is early diagnosis that leads to a measurable improvement in outcome.
Breast cancer is the most common malignancy in women, comprising 18% of all female cancers worldwide.3 Approximately one million new cases are diagnosed in the world each year, with the average lifetime risk of breast cancer in the USA at birth being 12%, or one in eight.4
The risk of developing breast cancer increases with age, beginning in the fourth decade of life. The probability of developing invasive breast cancer over the following 10 years is 0.4% for women aged 30–39, 1.5% for women aged 40–49, 2.8% for women aged 50–59, and 3.6% for women aged 60–69.5 This high prevalence, together with studies showing that most women with symptomatic breast cancer could not be cured by local surgery,6,7 prompted the search for a mechanism to screen women for early disease.
While breast cancer mortality is clearly higher in the developed world, it is still the most common cancer in women in developing countries. The epidemiology of breast cancer is complex. Major risk factors include age, geographical variation, age at menarche and menopause, age at first pregnancy, family history, previous benign breast disease and radiation, fat intake and hormone use.4
As the evidence from research became apparent in the 1970s and 1980s, many countries established breast cancer screening guidelines and programs. The UK government initiated screening for breast cancer in 1987 by a single, medial–lateral oblique view of each breast every 3 years for all women aged 50–64 years. The breast screening program in Australia—BreastScreen—began in 1991 and provides two-view mammographic screening at 2-year intervals, mainly for women aged 50–69 years. The US PreventiveServices Task Force revised its recommendations in November 20098 and now recommends that:
Breast cancer screening remains one of the most controversial areas in medicine, despite significant research having been carried out in the last 40 years. This is particularly so because of the large amounts of government money being spent on wide-scale screening programs all over the world and because breast cancer is the most common female malignancy.
CASE STUDY: ‘Like getting her breast slammed in a car door’
Judith was 53 years old. She attended her GP for a repeat prescription of her antihypertensive that she had been taking for 4 years. The GP, checking the history, noted that while Judith had had a routine smear the previous year there was no record of any mammogram results. She asked Judith if she had commenced breast cancer screening. Judith shook her head, answering that her friend who had had a mammogram told her it was like getting her breast slammed in a car door. The GP explained that a mammogram was slightly uncomfortable, as the X-ray required the breast to be compressed between two plates. She also told Judith that current government policy was to recommend mammographic screening to all women between the ages of 50 and 70 and went on to explain the pros and cons of the screening program. Judith said she would think about it and left.
The central controversy remains whether or not mammography is effective at preventing mortality from breast cancer. For those who believe the answer to this question to be yes, the controversies are over how a mammogram should be carried out, how screening programs should be established and what age groups should be targeted.
For those who believe that mammography is ineffective, questions remain about whether women should be actively encouraged and taught to engage in breast self-examination and whether or not clinical breast examination is an effective screening tool.
Until recently, and after heated controversy, there appeared to be general acceptance that the benefit of screening for breast cancer with mammography had been well documented.9 Large randomised trials, covering a total of half a million women, had been carried out in New York, USA10; Edinburgh, Scotland11; Canada12; and Malmö13, Kopparberg14, Östergötland, Stockholm15 and Göteborg16 in Sweden.
These trials focused primarily on comparing mortality from breast cancer in those women invited to be screened with women who had had no intervention. The most quoted study supporting the efficacy of mammography is a meta-analysis published in The Lancet in 199317 of the Swedish trials. This review found the largest reduction of breast cancer mortality (29%) in women aged 50–69 at randomisation. In women aged 40–49, a non-significant 13% reduction was demonstrated. Screening had only a marginal impact in women aged 70–74 years. Another review published in 1995 in the Journal of the American Medical Association was supportive of these figures, quoting a mortality benefit approaching 30% in women over 50 years of age 7–9 years from the start of the trials.18
In 2000, another meta-analysis of the randomised controlled trials in mammography was published in The Lancet, this time by Gøtzsche & Olsen.19 With the perilous title ‘Is screening for breast cancer with mammography justifiable?’, the authors reviewed the methodology of each of the eight trials in meticulous detail. They concluded that there were significant problems in the randomisation procedures of several of the trials and that only two of the eight trials were therefore valid. When the results of these two trials were then pooled, the outcome showed no significant difference in either total mortality or mortality secondary to breast cancer in those screened with mammography compared with controls. This work was subsequently included in the Cochrane database.20 Nystrom et al updated their review of the Swedish RCTs and answered some of the criticisms levelled at these trials in the Cochrane review. Their results confirmed a 21% (RR = 0.79, 95% CI 0.70–0.89) reduction in breast cancer mortality overall, with the reduction being greatest in the 60–69 year age group (33%).21 The most recent review22 has determined that mammography brings about a reduction in breast cancer mortality of 20%; however, the effect is lower in the highest-quality trials and a more reasonable estimate is a 15% relative risk reduction. Based on the risk level of women in these trials, the absolute risk reduction was 0.05%.
Those who argue against mammography point out the fact that as a screening test it results in a significant amount of morbidity. There is an estimated 30% increase in overdiagnosis and overtreatment, or an absolute risk increase of 0.5%.22 Overdiagnosis refers to the detection of abnormalities that will never cause symptoms or death during a patient’s lifetime. Overdiagnosis of cancer occurs when the cancer grows so slowly that the patient dies of other causes before it produces symptoms or when the cancer remains dormant (or regresses). Because doctors don’t know which patients are overdiagnosed, we tend to treat them all. Overdiagnosis therefore results in unnecessary treatment.23 For every 2000 women invited for screening throughout 10 years, only one will have her life prolonged. In addition, 10 healthy women who would not have been diagnosed if there had not been screening will be diagnosed as breast cancer patients and will be treated unnecessarily.24
Table 12.1 shows the morbidity generated by the UK National Health Service (NHS) screening program in 1 year (1997–98).25 These data demonstrate that, while 5% of all women screened are recalled for further assessment, the number of cancers detected are 5.9/1000.
|Number of women invited||1,668,476|
|Acceptance rate (% of invited)||75.1%|
|Number of women screened (invited)||1,252,324|
|Number of women screened (self/GP referrals)||97,780|
|Total number of women screened||1,350,104|
|Number of women recalled for assessment||71,255|
|Per cent of women recalled for assessment||5.3%|
|Number of benign biopsies||2212|
|Number of cancers detected||7932|
|Number of cancers detected per 1000 women screened||5.9|
|Number of in situ cancers detected||1718|
|Number of invasive cancers less than 15 mm||3381|
(From NHS Breast Screening Programme Review25)
The high level of morbidity generated by mammography has been very well demonstrated by Elmore et al.26 They performed a 10-year retrospective cohort study of breast cancer screening and diagnostic evaluations among 2400 women who were 40–69 years old. These women received a median of four mammograms and five clinical breast examinations per woman over the 10-year period. Of the women who were screened, 23.8% had at least one false-positive mammogram, 13.4% had at least one false-positive breast examination, and 31.7% had at least one false-positive result for either test.
The estimated cumulative risk of a false-positive result was 49.1% (95% CI, 40.3–64.1%) after 10 mammograms and 22.3% (95% CI, 19.2%–27.5%) after 10 clinical breast examinations. The false-positive tests led to 870 outpatient appointments, 539 diagnostic mammograms, 186 ultrasound examinations, 188 biopsies and 1 hospitalisation. They estimated that among women who do not have breast cancer, 18.6% (95% CI, 9.8–41.2%) will undergo a biopsy after 10 mammograms, and 6.2% (95% CI, 3.7–11.2%) after 10 clinical breast examinations. For every $100 spent for screening, an additional $33 was spent to evaluate the false-positive results.
Studies have also documented the psychological distress caused by abnormal mammograms and being recalled for further tests. Women with high-suspicion mammograms had substantial mammography-related anxiety (47%) and worries about breast cancer (41%). Such worries affected the moods (26%) and daily functioning (17%) of these women, despite diagnostic evaluation that excluded malignancy.27 In general, women have not been informed about the risk of false-positive results and the accompanying anxiety they generate,28 nor have they been warned about the risk of receiving the diagnosis of carcinoma in situ.29
Women generally exaggerate the benefits and are unaware of the harms of screening and invitations to women to participate in breast cancer screening reinforce this as they are information-poor and biased in favour of participation.31 Box 12.2 lists the points that such invitations and related pamphlets and websites should emphasise to give a more accurate representation of the risks and benefits of mammography.32
(From Jorgensen and Gotzsche32)
The other issues to consider, apart from the morbidity generated by the screening test, is the cost and whether or not the funds would be better directed elsewhere, such as to treatment or at least to research into better treatment. Some have suggested that this should be the case.33
For those countries that have yet to develop and implement a screening program, it must be remembered that mammography is only the start of the process. There need to be in place proper facilities and treatment protocols to implement best practice in treatment if a country is going to go to the bother of screening in the first place. This requires trained personnel, equipment and ongoing quality assurance and evaluation programs to ensure that the program is delivering a quality service.
One that deserves attention is clinical breast examination—that is, examination carried out by specially trained doctors or nurses and carried out in the medical setting. When comparing mammography with clinical breast examination, it is worth considering the outcomes of mammography:
Clinical breast examination should be able to detect lesions >1 cm in size. It is likely that some cancers <1 cm in size will be detectable by clinical examination. This means that if mammography is to be more effective than clinical examination, it has to do so through those 22% of cancers <1 cm in size and the 18% in situ cancers. The question is whether these lesions are clinically significant in terms of causing morbidity or mortality.
The only study to address this issue has been the Canadian National Breast Screening Study.35 It enrolled 40,000 women from 1980–1985 into either clinical breast examination and mammography or clinical breast examination alone. After 13 years of follow-up, there was no sign that mortality was lower in the mammography group.35 The combined group had 107 and the examination-only group 105 deaths attributable to breast cancer, giving a cumulative rate ratio of 1.02 (95% CI, 0.78–1.33).
Although longer follow-up may reveal a benefit, currently there is no evidence to suggest that the detection of non-palpable cancers by mammography contributes to reducing mortality from breast cancer.36 However, the study did highlight the morbidity brought about by mammography, with the rate of biopsy of benign lumps being three times higher in the combined examination/mammography group compared with that in the group undergoing clinical breast examination alone.35
In order to determine whether clinical breast examination can be implemented as a screening tool, however, more research needs to be carried out. This research must be a randomised, controlled trial comparing clinical breast examination on its own with no screening at all. Until this research is carried out, the US Preventive Services Task Force concludes that the evidence is insufficient to recommend for or against routine clinical breast examination (CBE) alone to screen for breast cancer.37
If the implementation of clinical breast examination is also controversial, then what about breast self-examination?
Unfortunately, while breast self-examination (BSE) can detect symptomatic breast cancer at an earlier stage, it does not appear to have any influence on mortality. An American Cancer Society study compared 177,602 women who practised BSE during the preceding 13 years with 272,554 women who did not and found similar breast cancer death rates in the two groups.38 The UK Trial of Early Detection of Breast Cancer (TEDBC)39 involved 300,000 women in eight health districts—two with mammographic screening centres, two where BSE was taught by trained nurses and four where neither form of intervention was available. At 16 years, the relative risk of death from breast cancer in women attending the two screening clinics was reduced by 27% (RR, 0.73; 95% CI, 0.63–0.84), but there was no risk reduction in the two BSE centres (RR, 0.99; 95% CI, 0.87–1.12).
Three randomised trials to evaluate the effect of BSE on breast cancer mortality have been undertaken in St Petersburg and Moscow40 and in Shanghai.41 The final results of the Shanghai study,42 which included more than 250,000 women, found a similar incidence and an identical number of breast cancer deaths among BSE subjects and controls. BSE has however greatly increased biopsy rates, with the number of benign lesions detected in the BSE group being twice those of the controls. These findings indicate that women should be aware of their breasts as part of general body awareness and seek medical help when their breasts look or feel abnormal, but the promotion of regular BSE is not justified.22
Since 1990 there has been a decrease in breast cancer mortality in the order of 30%.43–45 The problem is that this decrease may either be due to screening and early diagnosis or it may be secondary to improved therapy, or both. When the original mammography trials were being undertaken, systemic therapy was not as widespread as it is today, when most women will receive some kind of adjuvant therapy.46
Screening programs now exist in 22 countries around the world. Millions of dollars are being spent on these programs. While the literature remains confused and conflicting, it is perhaps best to continue to encourage women to undergo mammography screening but to be aware that the benefits are more modest than originally believed. Patients need to understand the pros and cons and particularly that there maybe significant morbidity as a result of the screening process.
Compared to breast cancer, cervical cancer is completely different: its aetiology, pathophysiology and development is much better understood and entirely different screening tests (compared with those for breast cancer) exist.
Since the advent and widespread use of the Papanicolaou (Pap) smear—a test that detects asymptomatic preinvasive lesions at the earliest stages—the incidence of cervical cancer has dramatically decreased. With the introduction of the National Cervical Screening Program in Australia, new cases of cervical cancer among women of all ages almost halved, from 13.2 new cancers per 100,000 women in 1991 to 6.9 in 2005. Mortality also halved, from 4 deaths per 100,000 women in 1991 to 2 in 2005.13 However, in many parts of the developing world, cervical cancer continues to cause significant morbidity and mortality.
The greatest burden of disease occurs in developing countries, where unfortunately there are no organised cervical screening programs. In terms of absolute number, the Asian region has an incidence of 265,884 cases of cervical cancer per year.47 In contrast, Australia had 734 new cases of cervical cancer and 221 deaths in 2005.48
Human papillomavirus (HPV) has now been causally related to cervical cancer,49 with HPV DNA detected in at least 95% of cervical cancers (of which HPV types 16 and 18 are the most commonly isolated50). The association between persistent HPV DNA detection and cervical cancer is more than 10 times the association between smoking and lung cancer.51
CASE STUDY: ‘I thought if you weren’t having sex you didn’t need to have a Pap smear.’
Maude was 72 years old; she was a very active lady who enjoyed her tennis and was involved in community groups. She had been widowed for the last 25 years and was surprised when asked by her new GP when the last time was that she had had a smear. ‘Oh, I haven’t bothered with them since my husband died. I thought if you weren’t having sex, you didn’t need to have them any more’, she replied. The GP explained that guidelines recommended that women should continue to have regular smears until the age of 70 and that, as Maude hadn’t had a smear in more than 25 years, she should have at least one. Two weeks later Maude returned for her smear and was pleasantly surprised at how easy it was. ‘They used to hurt when I had them all that time ago,’ she mused. Another 2 weeks passed and the GP found that she had to call Maude in to see her for the results. The smear had shown suspected cancer of the cervix. Maude had not had any symptoms and seemed genuinely surprised at the result. The GP explained that she would need to see a gynaecologist for confirmation of the smear result and organised an urgent appointment. One month later Maude came back to see her GP with a bunch of flowers. She had had surgery to remove the tumour and was requiring further treatment, but wanted to thank her GP for convincing her to have the smear and for ‘saving her life’. Maude’s case illustrates the importance of undertaking smears in ‘underscreened’ women—those who are past their reproductive years and women of non-English-speaking backgrounds.
BOX 12.3 HPV infection: natural history
Until recently, the most effective mechanism for cervical cancer prevention was organised cervical cancer screening with a Pap smear. Pap screening detects cellular changes in the cervix caused by persistent HPV infection. If HPV infection is left untreated, it has been estimated that about 30% of women with high-grade lesions would develop cancer over a 30-year period.58 Detection of high-grade lesions through Pap screening allows treatment prior to the development of cancer and as such is a secondary prevention strategy.
HPV types 16 and 18, worldwide, are responsible for about 70% of cervical cancer cases, 50% of high-grade precancerous lesions and 25% of low-grade lesions.59 HPV 6 and 11 infections are associated with most genital warts and around 8–10% of low-grade cervical lesions.60
Administering the vaccines to the pre-adolescent population maximises the chances of most of the population achieving immunity before HPV exposure (i.e. before sexual activity commences). As a more robust immune response to vaccination is achieved at this age, protection is likely to endure through the years of maximal exposure. However, extended follow-up of populations in clinical trials will help to determine whether a booster is required.61
Favourable estimates of the cost-effectiveness of a catch-up immunisation program for women up to the age of 26 years in economic models adapted to Australian data led to federal government funding of a universal immunisation program for girls aged 12 and 13 in Australia, with a 2-year catch-up program for older adolescent and young adult women.57 This catch-up program, run through schools and general practitioners, ended in 2009. Australia was the first country in the world to roll out a national HPV vaccination program and has established a register of women who received the vaccine. This register will facilitate cross-referencing of vaccination data with information from cervical cytology (Pap smear) or cervical cancer registries for evaluation purposes in the future.
Both Cervarix and Gardasil are registered in Australia for women up to the age of 45. Many question the utility of vaccinating women who are already been sexually active and therefore have already been exposed to HPV. However, the following questions remain to be conclusively addressed61:
And until these issues are resolved, vaccination in women up until the age of 45 can be considered to be like taking out ‘insurance’. Advice for ‘older’ women related to HPV vaccination is given in Box 12.4.
(From Skinner et al61)
Even with increasing age and numbers of sexual partners, most women do not have evidence of past exposure to HPV types 16 or 18. Older women have robust immune responses to the bivalent HPV vaccine, and so should derive benefit from the vaccine if exposed to HPV 16 or 18 in the future.
Is it too late to vaccinate a woman if she has a history of HPV disease shown by clinical evidence such as an abnormal Pap test or genital warts?
Evidence to date indicates that vaccination will have no effect on current or prevalent disease due to any HPV type. However, vaccination would ensure protection from future infection with oncogenic HPV types covered by the vaccine.
Future risk of exposure is difficult to determine accurately on the basis of past and current sexual history. This is because of patterns and changes in sexual behaviour through life, the potential for transmission through successive monogamous relationships, and inaccuracies in predictions about concurrency of sexual partners.
As a woman ages, natural immunity to HPV wanes, but also the incidence of new HPV infection decreases and the time to develop cancerous lesions from HPV must be balanced against the likelihood of other age-related diseases. Both the bivalent and quadrivalent vaccines are licensed for use in women up to 45 years.
No. While women who receive the vaccination who are already sexually active are likely to have been exposed to one or more strains of HPV, it is unlikely they have been exposed to both HPV 16 and 18 and therefore the vaccine will still provide them with some protection. Currently there are no validated, approved and readily available HPV type-specific polymerase chain reaction or serological assays.61 Were they available and used, the process would add considerable expense to an already expensive intervention.61
The HPV vaccine is a prophylactic, not therapeutic, vaccine. It will therefore have no effect on current HPV infection or current or previous dysplasia. Its aim is to prevent the acquisition of new HPV (primarily types 16 and 18, although there is some evidence of cross-protection with other strains).62
One of the most basic questions to ask when implementing a screening test is who should have it done? To answer this question one needs to have an understanding of the risk factors and pathophysiology of cervical cancer. Traditionally we have been told that all women who are having or have ever had sexual intercourse are at risk of cervical cancer, the reason being that they may have acquired HPV as a result of their sexual activity. This is slightly simplistic, however, especially when considering whether virgins, lesbians and women in monogamous relationships should be screened.
The recent introduction of polymerase chain reaction (PCR) testing has revealed that HPV infections are much more common among young, asymptomatic women than previously suspected. The site-specificity of genital HPV led to the assumption that HPVs were primarily transmitted by sexual contact. However, since HPVs have been detected in virgins, infants and children, and after juvenile laryngeal papillomatosis was shown to be caused by these viruses, there has been acknowledgment of the fact that HPV may be transmitted by other non-sexual routes as well.63 Despite this, women who have never engaged in sexual intercourse with a man have a minimal risk of developing squamous cell cancer of the cervix.
The question of when to start screening is slightly more complex, requiring an understanding of the natural history of HPV infection. In Australia the recommendation is to commence screening at 18–20 years of age or 2 years after first intercourse, whichever is later. In America screening guidelines advocate annual smears from the age of 18 until consecutive smears are normal, after which time screening can revert to 2-yearly.