News and views

Hormone therapy and the prevention of chronic diseases

Despite controversy and criticism when its first draft was posted for comment, the final text of the US Preventive Services Task Force's statement on hormone therapy (HRT) for the prevention of chronic conditions has now been published and is still largely discouraging. ¹ As expected, the report concluded that the benefits with combined therapy were only in the risk of type 2 diabetes, fractures and colorectal cancer – with a further benefit from unopposed therapy on the risk of breast cancer. Risks from combined therapy included invasive breast cancer, heart disease, gall bladder disease, stroke, thromboembolism and urinary incontinence. Overall, the Task Force could not recommend HRT (unopposed or combined) for the prevention of any conditions, including osteoporosis, colorectal cancer and cardiovascular disease.

Yet just weeks after the report's final publication a report from a US randomised trial showed that one year's use of transdermal HRT was more effective than placebo in preventing the onset of depressive symptoms in perimenopausal women without depression.² If the findings can be confirmed, say the authors, ‘clinicians may consider prescribing hormone therapy to mitigate the increased risk of depressive symptoms in women in the menopause transition and early postmenopause'.

Results of the study showed that, of the 172 early postmenopausal women included, none of whom had symptoms at baseline, 32% of those given placebo developed clinically significant depressive symptoms, while 17% of those taking transdermal estradiol and intermittent micronised progesterone did so. These benefits appeared to be greater in those with more stressful events in their lives than otherwise. However, the prognostic benefit did come at the cost of worsened menstrual bleeding in the HRT group. It was also noted in an accompanying editorial that the dose of estradiol in the patch (0.1 mg/day continuously) was ‘substantially higher than is currently recommended for treating bothersome hot flashes'.³

This editorial, accompanying the report in JAMA, advised caution in assessing applicability of the results (because of relatively small numbers), as did NHS Choices, but conceded that variability in circulating estradiol concentrations (as well as hot flushes and sleep disturbances) are strongly linked to depressive symptoms.

As background to the study the authors note that the menopause is associated with a 2- to 4-fold increased risk for clinically significant depressive symptoms. They add that a few studies have suggested that HRT can manage existing depression at this time, but results are inconsistent. NICE, in its latest guidance, cites evidence that HRT can treat low mood symptoms associated with the menopause, while a Women's Health Initiative study on combined HRT and quality of life (which included an eight-item scale for depressive symptoms) found no benefit.⁴

The report came just days before a health modelling study forecast that the number of old people with four or more chronic diseases will double by 2035 (from 9.8% prevalence today to 17% prevalence).⁵ And two-thirds of these are calculated to have ‘mental ill-health,' including depression and dementia. The life expectancy gains forecast for women over this period (2.9 years) will be mostly spent with chronic diseases ‘resulting from increased prevalence of rather than longer survival with multi-morbidity'.

In a press release from the researchers, the study's leader, Professor Carol Jagger from Newcastle University's Institute for Ageing also described as ‘worrying' a trend for future young-old adults, aged 65 to 74 years, to be more likely to have two or three chronic diseases than in the past. ‘This is due to their higher prevalence of obesity and physical inactivity,' explained Jagger, who added that such patients need a different approach to their management. ‘A single-disease-focused model of health care is unsuitable for patients with multi-morbidity,' she said. ‘There needs to be a focus on prevention of disease, and a bespoke healthcare service provision for patients with multi-morbidity.'

Thus far, as Louise Newson suggests elsewhere in this issue of the journal, one of the most likely candidates for prevention is cardiovascular disease, where control of risk factors is now unequivocally established. HRT, however, has had a rocky ride in this endeavour since the Women's Health Initiative of 2002 and the consequent earnest advice of organisations such as the American Heart Association to avoid HRT for CVD prevention. But, as Roger Lobo and colleagues argued a couple of years ago, that advice is beginning to look outdated, and that the evidence has now moved on.⁶ Sifting through the most recent evidence – including the WHI's revised time-dependent re-analysis of 2007 – Lobo et al.⁶ also saw a place for HRT as part of prevention strategy for women after the menopause. ‘HRT,' they concluded, ‘clearly reduces bone loss, bone fractures, new onset diabetes mellitus, CHD and all-cause mortality … particularly when initiated in women <60 years of age and/or <10 years-since-menopause.'

US Preventive Services Task Force. Hormone therapy for the primary prevention of chronic conditions in postmenopausal women. JAMA 2017; 318: 2224–2233.

Early menopause: Risk of CVD and association with body weight

When the UK Biobank began collecting data in 2006 its aims were to provide a research database for those investigating the environmental and genetic contributions to disease. The data comprised personal, medical and lifestyle details, with the addition of blood and urine samples, of a prospective cohort of 500,000 people aged 40–69 years. Since 2012 the data has been available to research groups hoping that the real-life evidence will reveal patterns of disease not so far clearly evident.

One such group from Oxford has just reported an association between several reproductive factors in women – including an early menopause – and an increased risk of cardiovascular disease in later life. ¹ During a monitoring period of around seven years, 9054 cases of CVD were recorded, a third (34%) of which were in women (who comprised around half the volunteers). The average age of the women at the start of the study was 56, and around half (51%) came from the most affluent third of the UK population. Six out of 10 had never smoked.

Results showed that those who went through the menopause early (before the age of 47) had a 33% heightened risk of CVD, rising to 42% for their risk of stroke, after controlling for other potentially confounding factors. The study also found that early menarche, complications of pregnancy and hysterectomy were associated with a raised risk of CVD.

As an explanation for their early menopause findings, the authors note that a higher risk of CVD has commonly been attributed to the adverse effects of the loss of ovarian function. However, they add that the reverse might be true – that it is not menopause per se that adversely affects this risk but rather that risk factors for CVD determine age at menopause, possibly through direct effects on the endocrine system or by inducing ischaemic damage in the ovaries.

Meanwhile, an unconnected long-running US study of nearly 80,000 women has found that those underweight or who were underweight as teenagers are at greater risk of early menopause than lean or normal-weight women.² The results were from the Nurses' Health Study, now said to be one of very few studies to follow women prospectively from as young as 25 to menopause, and the largest ever to do so, with much information gathered from teenage years.

The study analysed body mass index, weight distribution, weight change and timing of menopause in 78,759 premenopausal women aged between 25 and 42 who joined the Nurses' Health Study II in 1989. The women provided information on their medical history and health-related behaviours, such as smoking and exercise, at baseline, and updated the information by questionnaire every two years thereafter. The researchers followed them until 2011, by which time 2804 women had reported early natural menopause.

Results showed that women who were underweight at any age (BMI of less than 18.5 kg/m²) had a significant 30% higher risk of early menopause than lean or normal weight women (BMI 18.5–22.4 kg/m²). Overweight women with BMIs between 25 and 29.9 kg/m² had a 21–30% lower risk of early menopause than normal weight women. And women who were underweight at 18 with a BMI of less than 17.5 kg/m² had a 50% higher risk of early menopause than lean or normal weight women.

The authors defined early menopause as the cessation of ovarian function before age 45, and they too recognise an association with a higher risk of CVD, cognitive decline, osteoporosis and premature mortality. Though genetic factors account for some of the increased risk of early menopause, modifiable lifestyle, reproductive and environmental risk factors may also play a role, in which ‘substantial weight loss' may be a factor. Women who lost 20 pounds or more from age 18 to 35 had a higher risk of early menopause than those who gained 5–15 pounds. These patterns were also seen in a recent meta-analysis of nine studies.³

The study did adjust for physical activity, which at high exercise levels has been commonly associated with menstrual cycle disturbance and amenorrhea. Women in this study, however, were said to have generally low to moderate levels of physical activity, ‘with very few exercising at levels that would disrupt menstrual cycle function'.

Peters SAE and Woodward M. Women's reproductive factors and incident cardiovascular disease in the UK Biobank. Heart. Epub ahead of print 15 January 2018. 2018. DOI: 10.1136/heartjnl-2017-312289.

‘Liquid biopsies' and the non-invasive detection of common cancers

There was much interest in a January report in the journal Science suggesting that a blood test might detect and localise tumours at an early stage of their progression. ¹ The test described, which works by analysing cancer-related DNA and proteins in blood, achieved a 70% positive result in eight common cancer types in 1005 patients whose tumours had not yet spread – apparently, according to Science, among the best performance yet for a universal cancer blood test. In a commentary, the journal added that a test of this type might ‘one day lead to a tool for routinely screening people and catching tumors before they cause symptoms, when chances are best for a cure'.

There have in the past few years been several studies of this type – ‘liquid biopsies' – designed to detect a single kind of cancer by spotting tumour-associated mutations in DNA sequences found floating freely in the blood. This latest method, known as CancerSEEK, adds certain proteins to the DNA analysis for the detection eight different cancers.

The study itself, performed mainly by a huge team of investigators from Johns Hopkins University in Maryland, applied the test to 1005 people already diagnosed with one of eight non-metastatic cancers – ovarian, liver, stomach, pancreatic, oesophageal, colorectal, lung or breast – but excluded those with later-stage disease to focus on early development. Significantly, say the authors, there are no screening tests available in most of these cancers for average-risk individuals.

Study results showed that the effectiveness of this particular test was greatest in ovarian cancer, with a detection rate of 98%. However, only 33% of breast cancer cases were identified, and overall detection rate was 63%. One of the concerns of applying these tests, especially in large ‘healthy' populations, is the stress (and unnecessary treatment) prompted by a false-positive result. However, the case for early detection, as the authors of this study emphasise, is that tumours identified before metastasis can usually be cured by surgery alone. ‘Once distant metastasis has occurred,' they add, ‘surgical excision is rarely curative.' The test also narrowed the origin of the cancer to two possible sites in about 80% of patients.

The investigators initially explored several hundred genes and 40 protein markers, whittling the number down to segments of 16 genes and eight proteins. They point out that this molecular test is solely aimed at cancer screening and, therefore, is different from other molecular tests, which rely on analysing large numbers of cancer-driving genes to identify therapeutically actionable targets.

‘The use of a combination of selected biomarkers for early detection has the potential to change the way we screen for cancer, and it is based on the same rationale for using combinations of drugs to treat cancers,' said lead investigator Nickolas Papadupoulos.

However, much expert comment on the report was cautious, while acknowledging that liquid biopsy represents the most promising approach for non-invasive cancer screening. In its commentary, the journal Nature said ‘it will take time to figure out whether widespread screening of healthy people with a universal blood test can reduce cancer deaths without doing harm'. The commentary quoted one UK expert who said: ‘If people expect to suddenly catch all cancers, they'll be disappointed. This is exciting progress, but evaluating it in the real world will be a long process.'